Anecdotes - PJREVAT Avatar 2 - Vidéo Dailymotion

Durendal

Un retour détaillé sur la réalisation du dernier PJREVAT de la chaine, sur Avatar 2. Anecdotes, infos, astuces, et tous les petits secrets !  0:00 La chaine passe en 2.0 2:29 Vos réactions 5:24 Le projet 15:32 Le décor 3D : l'origine 27:34 Le décor 3D : Poser 12 40:07 Le décor 3D : le développement 58:32 Le décor 3D : s'ajouter des emmerdes 1:12:12 Les sketchs du début 1:15:42 L'habillage 1:16:27 Le lipsync 1:20:20 Volonté non-réalisée & animation 4 côtés 1:25:42 Jeu d'acteur 1:80:00 Les split-screens 1:30:11 WHYYYYYYYYYYYYYYYYYYYYY ? 1:38:26 "Ils sont complètement débiles !" 1:39:48 Les incrustations d'images/vidéos 1:40:21 Intonations 1:42:15 Une, ou deux parties ? 1:45:25 "Désolé pour ma femme..." 1:46:34 KIRI KIRI KIRIIIII ! 1:49:32 Sous l'océan 1:50:24 Articulation 1:51:22 PHILIPPE !!!!!! 1:54:04 Réécritures 1:54:44 Sketch annulé 1:55:44 Les jeux de mots 1:56:30 L'outro 1:57:37 Conclusion  Tipeee : https://www.tipeee.com/le-cinema-de-durendal Facebook : https://www.facebook.com/LeCinemaDeDurendal Twitter : https://twitter.com/CinemaDurendal

Transcript

00:00:00 *sigh*

00:00:01 Hello everyone! How are you?

00:00:03 It's been a while since I've stopped in front of a camera,

00:00:05 since I've been filming less and less lately.

00:00:07 Hello everyone!

00:00:09 Welcome to this video, which will be about

00:00:11 the "Why I'm Right and Wrong" on Avatar 2,

00:00:15 the voice of water.

00:00:17 A different version of what I did before,

00:00:19 I'll include the failed scenes in this one,

00:00:21 and a general making-of rather than a shooting anecdote,

00:00:23 to explain things in more detail.

00:00:25 So, this video, Avatar 2, was released,

00:00:27 and I wanted to do a debrief,

00:00:29 because...

00:00:31 well, the channel is back to its new rhythm.

00:00:33 Now, there can be a few months,

00:00:35 a few weeks, without any video,

00:00:37 to be able to prepare the big videos,

00:00:39 which require a lot more time and investment.

00:00:41 I did it on purpose, to be relatively...

00:00:43 ...pure, during the editing of this video,

00:00:45 to see how much time I needed,

00:00:47 while being really calm,

00:00:49 and not really putting myself in the...

00:00:51 not completely emptying my energy,

00:00:53 with other things on the side,

00:00:55 and not dissipating myself

00:00:57 on other things.

00:00:59 I didn't work on any other video,

00:01:01 I didn't do anything else,

00:01:03 and so, I thought it was interesting

00:01:05 to see how much time I needed,

00:01:07 just to make a perfect and complete video.

00:01:09 I admit I didn't talk about 2 months,

00:01:11 but it's because a lot of things changed in this video.

00:01:13 I'll come back to it as I go along.

00:01:15 So, I spent about 2 months

00:01:17 to release a "Why I took so long",

00:01:19 and only doing that.

00:01:21 From now on, there will be other videos

00:01:23 to release on the channel.

00:01:25 They will be 5 or 6 months late,

00:01:27 one of the first ones I'll release

00:01:29 will be with "Knock at the Cabin",

00:01:31 I know a lot of people won't like it,

00:01:33 because they'll ask me for advice

00:01:35 on other things to release next,

00:01:37 like Super Mario, etc.

00:01:39 It'll come, but I'll take my time,

00:01:41 I'm calm, I'm laid back.

00:01:43 So, from now on, I'll start

00:01:45 releasing other videos,

00:01:47 and maybe, in the meantime,

00:01:49 it'll slow me down on the other releases,

00:01:51 I don't really feel like it,

00:01:53 so I'll spend 2 or 3 weeks,

00:01:55 maybe even a month,

00:01:57 to make sure it'll free me from the planning,

00:01:59 and not to be a thing that takes care of me,

00:02:01 and prevents me from working on other things.

00:02:03 So, with 2 months without doing anything,

00:02:05 I'd say that 3 or 4 months,

00:02:07 3 months, let's say,

00:02:09 to make a "Why I took so long",

00:02:11 if I make other videos at the same time,

00:02:13 it seems like a not-so-ugly planning

00:02:15 to have.

00:02:17 In a month, I can,

00:02:19 I really can't. Videos are too long,

00:02:21 I can't make a video like that in a month,

00:02:23 unless there's no work on the staging,

00:02:25 but that's where we'll get into

00:02:27 the details.

00:02:29 I wanted to say that,

00:02:31 as it's the first test of this new method,

00:02:33 I am...

00:02:35 You can't even imagine how happy I am

00:02:37 to know how much you liked this video.

00:02:39 It's... Of course,

00:02:41 when each video comes out,

00:02:43 it's the satisfaction of a YouTuber,

00:02:45 it's seeing his work rewarded,

00:02:47 it's seeing the editing effects you've put in,

00:02:49 the jokes you've thought about,

00:02:51 making flies,

00:02:53 hitting right and provoking

00:02:55 exactly what you wanted to do.

00:02:57 The reactions you've had, they were crazy.

00:02:59 I think there wasn't

00:03:01 a single negative reaction,

00:03:03 except for people who didn't like it at first.

00:03:05 Not even a comment like "it was a bit too fast at that moment",

00:03:07 etc.

00:03:09 Maybe a comment about the articulation,

00:03:11 because there's a moment where I was surprised

00:03:13 that I didn't get more feedback than that

00:03:15 about the articulation in some places,

00:03:17 but there was really no problem.

00:03:19 So I really have to thank you,

00:03:21 because you've just completely validated

00:03:23 my new rhythm.

00:03:25 It was really the risk.

00:03:27 I can ask you the question in December,

00:03:29 "Do you agree if I slow down the rhythm?"

00:03:31 And then I slow down the rhythm and you're not there anymore.

00:03:33 It's already been done on the Internet,

00:03:35 there are tons of channels that are dead like that

00:03:37 with a change of rhythm or something like that.

00:03:39 And so it was really risky.

00:03:41 And now to see that it pays off to this point,

00:03:43 the "Why I'm Doing This" has 350,000 views

00:03:45 in a year and a half.

00:03:47 It's been years since I've had

00:03:49 that many views so quickly,

00:03:51 so easily,

00:03:53 I'm still surprised.

00:03:55 I even had competitors who launched videos

00:03:57 on Avatar the same day.

00:03:59 We both released a video on Avatar the same day.

00:04:01 Obviously, we competed with each other.

00:04:03 But even with that, the video

00:04:05 worked incredibly well, your reactions

00:04:07 are absolutely crazy, and you're all saying

00:04:09 in the comments, "But man, if it's to have videos like that,

00:04:11 you can put six months to release them,

00:04:13 six months without anything, if it's for videos like that, it's perfect."

00:04:15 And so, well...

00:04:17 It's a dream, to have reactions

00:04:19 so positive and so cool, it's a dream,

00:04:21 truly a dream.

00:04:23 So it was an ideal release.

00:04:25 But really, it was an ideal release.

00:04:27 I had no problem

00:04:29 with this video.

00:04:31 I wasn't detected by Content ID

00:04:33 for copyright and stuff like that,

00:04:35 neither for the movie nor for the sound.

00:04:37 I had no problem with export,

00:04:39 with a single shot, I export and the video,

00:04:41 I don't know, a bug somewhere, a sound that doesn't work well,

00:04:43 or that doesn't work a few seconds later, or that shifts, etc.

00:04:45 I saw that in videos about Oscars, for example,

00:04:47 my video about Oscars, where I present the movies.

00:04:49 At the end, I do a little presentation

00:04:51 of the music. There's a shift

00:04:53 of one or two seconds that was added, but

00:04:55 to the rendering, to the export of the video.

00:04:57 I didn't do it myself like that in the editing.

00:04:59 It's just to the export, and just...

00:05:01 I had to go to the end to check if it was good,

00:05:03 and I just had to skip those five seconds where there's a little shift.

00:05:05 That's what's very complicated, it's that really,

00:05:07 when you export videos like that, if you don't watch them,

00:05:09 every second of the thing, you can miss a problem.

00:05:11 And I didn't even have that kind of problem

00:05:13 with this video.

00:05:15 I didn't have any stupid export problem,

00:05:17 I didn't have anything like that.

00:05:19 It was a perfect release.

00:05:21 Perfect.

00:05:23 It was very, very cool.

00:05:25 So, to come back a little bit on the idea,

00:05:27 it's the first Why Jordan 4

00:05:29 since a year and a half, and so I wanted

00:05:31 to make sure I didn't miss it. I wanted to make sure

00:05:33 to do a Why Jordan 4 that I would consider

00:05:35 successful, and that you would consider successful.

00:05:37 And so, I wanted to put

00:05:39 all the chances on my side, especially

00:05:41 I didn't want to take back Why Jordan 4

00:05:43 that are already in progress and not finished yet.

00:05:45 You know, there's the Why Jordan 4

00:05:47 on Final Fantasy that's been in progress for...

00:05:49 two years, I think, since COVID,

00:05:51 it seems to me, so since 2020, three years now,

00:05:53 that

00:05:55 is now stopped at Final Fantasy 7,

00:05:57 I have to continue with the first movie,

00:05:59 Final Fantasy 10,

00:06:01 maybe even 11 and 12,

00:06:03 then I have another game coming with Elden Children,

00:06:05 I'll talk about 13 and 14

00:06:07 at that time, and then I have a last video

00:06:09 on 15, so there are still the games on

00:06:11 both movies and on Final Fantasy 15 to do.

00:06:13 I could have gone back on that,

00:06:15 but I told myself

00:06:17 I know that there are a lot of people

00:06:19 who are very afraid since these videos

00:06:21 came out that cinema is no longer interesting to the channel.

00:06:23 That's also why

00:06:25 I did a lot of vlogs next to it, because I thought

00:06:27 if I do a Why Jordan 4

00:06:29 on a video game, already that takes time to release,

00:06:31 people will be fed up with me not talking about cinema,

00:06:33 so I really have to talk a lot about cinema,

00:06:35 and so, since it's been a year and a half

00:06:37 since the one on Final Fantasy 7 came out,

00:06:39 I thought if after a year and a half I come back

00:06:41 with a Why I'm Right, You're Wrong,

00:06:43 and it's not on a movie,

00:06:45 I'm going to lose all my audience, you're all going to break down.

00:06:47 And it's not like

00:06:49 I really wanted to talk about Final Fantasy

00:06:51 now, it's something

00:06:53 I love, I can talk about it at any time of my life,

00:06:55 so it's not like it was urgent.

00:06:57 So that's why I thought Final Fantasy

00:06:59 and its sequels,

00:07:01 maybe not now,

00:07:03 because there might be people who won't appreciate it.

00:07:05 I also decided

00:07:07 not to continue

00:07:09 another Why I'm Right, You're Wrong

00:07:11 I started last year. If it's been a year and a half

00:07:13 since the last Why I'm Right, You're Wrong,

00:07:15 it doesn't mean it's been a year and a half since I wrote nothing,

00:07:17 I didn't do anything. There's a Why I'm Right, You're Wrong

00:07:19 that's been around since the end

00:07:21 of last summer, I started writing it in May

00:07:23 and I had to finish it in August.

00:07:25 People on Tipeee know what the subject is,

00:07:27 because they were aware at the time,

00:07:29 they saw the whole process, they saw me struggling for months,

00:07:31 they saw me giving up in the end, etc.

00:07:33 And I thought, if I go back to that thing,

00:07:35 there's so much trauma around

00:07:37 Why I'm Right, You're Wrong, I might get bored

00:07:39 myself. So I really thought

00:07:41 I'm going to start on something

00:07:43 that's not

00:07:45 tinted with bad waves

00:07:47 and bad experiences. That's why

00:07:49 I decided to start on Avatar,

00:07:51 knowing that I started making a video on James Cameron

00:07:53 a long time ago, but more specifically on Terminator,

00:07:55 because I hadn't started making the movie

00:07:57 yet. I had seen it,

00:07:59 but I hadn't started writing the text yet, and I wanted to start

00:08:01 on Terminator at the time, and I had stopped

00:08:03 in the middle, but it wasn't James Cameron himself,

00:08:05 and it wasn't Avatar.

00:08:07 And the video was centered on Avatar.

00:08:09 So I thought, this is a subject that's well-defined,

00:08:11 that's calm, I don't have trauma around

00:08:13 the subject, etc. And it's a subject

00:08:15 that was close to my heart, because when

00:08:17 Avatar 2 came out, I wanted to release my video on Avatar 1

00:08:19 to remind people that

00:08:21 it wasn't just nice special effects

00:08:23 and an empty story, etc.

00:08:25 The fact that it looks like "Dance with the Wolves", "Pocahontas",

00:08:27 etc. wasn't a problem,

00:08:29 it offered enough things to

00:08:31 be interesting enough.

00:08:33 And that's precisely why I had

00:08:35 a very different relationship with 2,

00:08:37 because

00:08:39 knowing what 1 did, and seeing what 2 did,

00:08:41 I thought, "No, this is just a repetition,

00:08:43 there's nothing else,

00:08:45 except nice graphics."

00:08:47 It's the first movie, re-made, with a few

00:08:49 different things, and that's it.

00:08:51 That's why I got a little angry. By the way, I have comments

00:08:53 from people who sometimes tell me, "Yeah, no,

00:08:55 I still like 2, because it's nice,

00:08:57 etc. But the first one was just a re-make

00:08:59 of a bunch of other movies, and if you don't like 1

00:09:01 because it's a re-make of other movies, how can you like 2

00:09:03 which is a re-make of 1, which is a re-make of other movies?

00:09:05 At some point, you're not logical, dude!

00:09:07 At least, hate both,

00:09:09 but don't say you like 2 and not 1,

00:09:11 it doesn't make any sense!

00:09:13 So, there you go.

00:09:15 The other reason I decided to talk about Avatar 2,

00:09:17 more than anything else, was

00:09:19 the opportunity, in terms of

00:09:21 analysis,

00:09:23 of the video's construction,

00:09:25 to make a text

00:09:27 which was both fresh,

00:09:29 to try to renew my style a little bit,

00:09:31 to find more jokes,

00:09:33 things like that. I'll come back to that a little bit

00:09:35 in the analysis of the video, but there are some

00:09:37 points where I said to myself, "I can't do the same thing

00:09:39 I've been doing for years, I need to find a variation,

00:09:41 I need to find something different." And at the same time,

00:09:43 I wanted to make a classic of the channel.

00:09:45 Because the videos which have

00:09:47 the most market on my channel,

00:09:49 often the ones I have the most pleasure to make,

00:09:51 are videos which are in

00:09:53 the format

00:09:55 I made for Avatar 2, that is to say,

00:09:57 the format I made for Prometheus, for Alien Covenant,

00:09:59 for Star Wars 8, or for

00:10:01 "Au temps n'importe le vent", for "La machine à explorer le temps", etc.

00:10:03 Most of the "Why I'm a Retard" videos which really

00:10:05 marked people, apart from the big retrospectives

00:10:07 on really important subjects,

00:10:09 are videos which come back on a movie from A to Z,

00:10:11 where I take the movie at the beginning, I go until the end,

00:10:13 and like that, I can come back on each problem,

00:10:15 each time it comes back in the movie, etc.

00:10:17 Which doesn't necessarily

00:10:19 take more time than making a thematic plan,

00:10:21 because I have a particular writing style,

00:10:23 it could take as much time to make it in thematic

00:10:25 as in chronological, but I think it

00:10:27 marks more how the flaws of the movie

00:10:29 are present when, instead of saying "once",

00:10:31 "there are 36 times this problem", and enumerate the 36 times,

00:10:33 to say "there is this problem",

00:10:35 you present it once, the first time, and then,

00:10:37 each time it comes back, you say "ah yes, there is this problem which came back",

00:10:39 "ah, and there is this problem which came back", "ah, and there is this problem which came back",

00:10:41 I think it works, it shows that it's not

00:10:43 just a problem here and there, it comes back constantly

00:10:45 in the movie, it's not just a little problem,

00:10:47 or a little delirium on my part, or something like that.

00:10:49 And above all, it's a way of doing things

00:10:51 which allows me to make a lot of jokes, because I can

00:10:53 base myself on the text itself, to

00:10:55 make jokes afterwards, etc, like

00:10:57 "excuse me for my wife, I don't know how she got out of the kitchen",

00:10:59 etc, it's the kind of thing you can

00:11:01 much more easily come up with when you're

00:11:03 in the middle of the story, summarizing it,

00:11:05 than when you're doing something which summarizes by theme

00:11:07 the different problems that could be there.

00:11:09 Because at that moment, if I do it by theme,

00:11:11 I'm going to link all the problems

00:11:13 of, for example, the way of treating Nettiri

00:11:15 in the story compared to the first movie,

00:11:17 and so,

00:11:19 as I link them, I can't really take a break

00:11:21 to make a joke on each scene,

00:11:23 that's why it's much more fun to do it in that order.

00:11:25 I knew that it was...

00:11:27 if I wanted to please my audience,

00:11:29 and if I wanted to please myself in terms of writing,

00:11:31 I had to do this way, this presentation,

00:11:33 this development plan,

00:11:35 knowing that the "Why I'm Right"

00:11:37 which will come after,

00:11:39 the one I had abandoned last summer,

00:11:41 will be something much more analytical.

00:11:43 It will be less focused on the sketches,

00:11:45 on the fact of making jokes, etc.

00:11:47 It will be much more about

00:11:49 the staging, what it allows to understand on the movie,

00:11:51 why the comments that were made weren't necessarily right, etc.

00:11:53 That's the next topic.

00:11:55 I haven't completely changed the style of the videos,

00:11:57 but usually, I like to alternate between the two.

00:11:59 I like to do things that are more analytical,

00:12:01 things that are more for fun,

00:12:03 that have analysis too, but that are more for fun.

00:12:05 So, I thought it was a good way

00:12:07 to resume the work.

00:12:09 And I also wanted that in this video,

00:12:11 to put all the things on my side,

00:12:13 I wanted that the work

00:12:15 would really be visible on the screen.

00:12:17 There are a lot of things

00:12:19 that I did in a very different way

00:12:21 in this video,

00:12:23 because, unlike another video

00:12:25 where I could have the desire to do it

00:12:27 before with my old rhythm,

00:12:29 and I would have said to myself

00:12:31 "If I start doing this, it's going to take a lot of time,

00:12:33 I'm going to take 4 hours to finish this thing instead of 1 hour,

00:12:35 and so it's as much time as I don't have to do

00:12:37 on the rest of the editing,

00:12:39 and tomorrow I have a vlog to shoot,

00:12:41 and a thing to edit, etc.

00:12:43 Here, I wanted to allow myself,

00:12:45 by changing the rhythm,

00:12:47 to show what it can bring.

00:12:49 To show, in terms of aesthetics,

00:12:51 what it can bring differently.

00:12:53 I always have the same presentation

00:12:55 as in my other videos,

00:12:57 for example, the fact of putting the posters,

00:12:59 all the posters, with the little editing

00:13:01 with the photos of the directors,

00:13:03 when I do the history of how the project

00:13:05 is done, I do it in a way that is more aesthetic.

00:13:07 A stupid thing, for example,

00:13:09 is that every time there is this passage,

00:13:11 there is a blurry background behind it.

00:13:13 And this blur, in principle,

00:13:15 would be what we call a Gaussian blur,

00:13:17 which is just a blur.

00:13:19 And so, for this video, I said to myself

00:13:21 "Let's jump the step, let's find a plugin

00:13:23 that will do it for me",

00:13:25 because that's the problem with Premiere,

00:13:27 there is no integrated plugin in Premiere

00:13:29 that does this kind of blur,

00:13:31 the blur of photo lenses that I put in this thing.

00:13:33 So, to find a plugin that would allow you to do it,

00:13:35 and to buy the plugin, because it costs quite a lot,

00:13:37 and to find one that would do it quickly enough

00:13:39 so that it would work in terms of rendering,

00:13:41 it's a stupid decision to put blur on a background image,

00:13:43 but it takes a lot of study.

00:13:45 And I will tell you, even there,

00:13:47 the problem is that my graphics card is not made

00:13:49 to work in 4K.

00:13:51 And to get the effect I wanted to have on these images,

00:13:53 I'll tell you about it when I'm in front of the video,

00:13:55 I had to work in 4K.

00:13:57 And so it was completely slowing down my pipeline.

00:13:59 Every time you do an exploration with a 4K source,

00:14:01 it's really complicated.

00:14:03 That's why the videos, the camera faces,

00:14:05 when they are shot in 4K, they are really long to render.

00:14:07 Even if I only do 1080p.

00:14:09 And even having found a plugin that was very fast at the base,

00:14:11 as it's 4K, it took a lot of time,

00:14:13 so I had to pre-render all these backgrounds

00:14:15 that you have in all these things, in other video formats,

00:14:17 which allow me to have them as they are,

00:14:19 with the blur well put, etc.

00:14:21 and in 4K,

00:14:23 and then to be able to simply put them in the image,

00:14:25 and with a rendering that happens extremely quickly, etc.

00:14:27 Even a thing like that, it takes time

00:14:29 to make a decision as stupid as that.

00:14:31 Do I put Gaussian blur or a blur of the camera

00:14:33 on the background of my image?

00:14:35 Stupid things like that,

00:14:37 I thought, this time I'm going to take the time

00:14:39 and I'm going to put it in, and I'm going to go all out,

00:14:41 and I want it to be awesome.

00:14:43 Most of the memes I made, it's almost all the time 3D text,

00:14:45 except for the "why", because there was another little modification

00:14:47 on the "why" that was going to come back on it.

00:14:49 Make sure that the videos are

00:14:51 definitely more beautiful aesthetically,

00:14:53 and that the time that was put in addition is useful.

00:14:55 I found that

00:14:57 it was an essential thing in this feedback video,

00:14:59 because it must show

00:15:01 both celebrating the feedback

00:15:03 for a year and a half, and at the same time showing that

00:15:05 this new style will allow you to do something new,

00:15:07 this new use of time.

00:15:09 So here I am, I look at my notes quickly,

00:15:11 but I think that's all I noted for the introduction.

00:15:13 So from now on, we'll be able to start

00:15:15 to redo the video quietly.

00:15:17 So I'm going to do it again quietly, and then I'll stop

00:15:19 every time I have something to say.

00:15:21 You'll see, we'll stop very quickly, for example, from the beginning,

00:15:23 so that I can tell you about the background, etc.

00:15:25 And then I'll tell you anecdotes at some point,

00:15:27 to tell you how or when I found some jokes, etc.

00:15:29 I hope you'll like it.

00:15:31 Here we go!

00:15:33 [Music]

00:15:35 [Music]

00:15:37 [Music]

00:15:39 [Music]

00:15:41 [Music]

00:15:43 Here we go!

00:15:45 This is the first thing we need to start with,

00:15:47 the famous virtual background.

00:15:49 But the thing is, at the beginning, as I wanted this video

00:15:51 to mark a renewal on the channel,

00:15:53 the first idea I had

00:15:55 was not to push all the cursors to the max,

00:15:57 it was "I need to get it out

00:15:59 as quickly as possible".

00:16:01 That was my goal when I started the video.

00:16:03 And that's why I came up with a tweet at the beginning,

00:16:05 saying "I've already written my text in full,

00:16:07 there's no way this video will take me

00:16:09 more than a month to edit, it's going to be perfect, etc."

00:16:11 Because at the beginning, the plan was this.

00:16:13 The plan was... Well, the plan.

00:16:15 The plan for the video was to have this plan.

00:16:17 Well, with the higher camera,

00:16:19 traditional "Why am I recording?" plan,

00:16:21 in front of the shelf, where I'm talking, etc.

00:16:23 What I do is I do a first draft of the text,

00:16:25 and then I do all the corrections in the oral.

00:16:27 I never do written text corrections,

00:16:29 because I hate doing them in writing.

00:16:31 I can't have a vision of my whole idea,

00:16:33 of my thing, and so I prefer

00:16:35 to listen to my whole voice-over

00:16:37 and see what I want to correct next.

00:16:39 And while I was doing this work,

00:16:41 I thought to myself, "Oh, come on,

00:16:43 I'm going to talk about Avatar 2,

00:16:45 there's a whole thing about

00:16:47 technical accomplishment

00:16:49 that hides a big void."

00:16:51 And I thought,

00:16:53 "It would be better

00:16:55 if I tried to do a minimum

00:16:57 of technical accomplishment

00:16:59 in my video, because it's a little bit

00:17:01 in the theme, it would be a little bit

00:17:03 meta, finally, with the stuff."

00:17:05 And so the first idea I had was

00:17:07 to shoot in Navi.

00:17:09 With the idea of making a blue Navi painting,

00:17:11 so white, and all the mess,

00:17:13 with the little dots and all the stuff

00:17:15 that follows, and to shoot

00:17:17 all of this in front of the camera

00:17:19 with a wig, etc.

00:17:21 So I bought the wig at that time.

00:17:23 And then, thinking about it, I said to myself,

00:17:25 "The problem is that the make-up,

00:17:27 I'll have to do it myself,

00:17:29 I'm not really good at it,

00:17:31 I don't want to call a make-up artist

00:17:33 to come and do this thing,

00:17:35 because it would imply that I absolutely

00:17:37 have to shoot a very specific day,

00:17:39 the day this person comes to do my make-up,

00:17:41 and I know by definition

00:17:43 that it doesn't always happen like that,

00:17:45 but there are so many days

00:17:47 where I wake up thinking, "Today I'm shooting,

00:17:49 and I wake up with a huge migraine,

00:17:51 which lasts for two days,

00:17:53 and so you think, "Well, no,

00:17:55 not today." And if you have someone else

00:17:57 involved in the work with you,

00:17:59 at that moment, you don't feel it anymore.

00:18:01 I say that because I'm a chronic migraineur,

00:18:03 it happens to me often to have migraines,

00:18:05 so it's something that can often happen to me,

00:18:07 and by the way, I had this problem this year,

00:18:09 with this video, it's been so long since I did it,

00:18:11 that I have to say it now,

00:18:13 I had a huge metal bar on my forehead,

00:18:15 so it was impossible to shoot that day,

00:18:17 and so it was a little more complicated.

00:18:19 And so I thought,

00:18:21 it's going to be a bit laborious

00:18:23 with the paint, etc.

00:18:25 So, evolution of the idea,

00:18:27 I'm going to shoot on a green background,

00:18:29 and I'm going to

00:18:31 modify the colors

00:18:33 of what I would have shot on a green background,

00:18:35 once the green background is blurred.

00:18:37 Once you've blurred your green background,

00:18:39 you can modify what's left once the green background is removed,

00:18:41 and there's something that shifts all the colors,

00:18:43 but logically, one by one, compared to the others.

00:18:45 And I know,

00:18:47 naturally, that if you take human skin

00:18:49 and you use this chromatic wheel,

00:18:51 at some point it's going to turn blue.

00:18:53 So I said to myself, rather than putting on makeup,

00:18:55 which could be annoying when it comes to shooting,

00:18:57 and could cause me problems later,

00:18:59 I'm going to do a blue background,

00:19:01 and then, well,

00:19:03 a green background, rather, and change the color

00:19:05 so that I'm blue, then.

00:19:07 It was at that moment, by the way, that I decided to wear

00:19:09 this thing, which is something I bought,

00:19:11 I bought some clothes, I try to buy clothes from time to time,

00:19:13 but I always find them ugly, I don't wear them, or I regret it,

00:19:15 I have a hard time finding clothes that I like.

00:19:17 And I found this thing,

00:19:19 and I like these ways of closing the collar,

00:19:21 I find it nice,

00:19:23 but this one I found a little too hippie,

00:19:25 close to nature,

00:19:27 you feel like the thing is in a chamber, or I don't know what.

00:19:29 I don't even know what material it is,

00:19:31 it's not particular.

00:19:33 And so I said to myself, I'm going to be able to use this thing,

00:19:35 because, by definition,

00:19:37 the more my skin turns blue,

00:19:39 the more the blue

00:19:41 gets closer to the color of my skin.

00:19:43 And so I knew that if I wore this thing,

00:19:45 it would be brown,

00:19:47 which is the thing of military uniforms

00:19:49 that they have when they go to the jungles of Pandora, for example,

00:19:51 things like that.

00:19:53 All of this was very thought-out, I knew exactly

00:19:55 that I wanted this color for the thing,

00:19:57 because if I didn't have this with this color,

00:19:59 I would have bought another one, or I would have found another thing,

00:20:01 I have other things that are a little blue,

00:20:03 all my clothes are stored behind here,

00:20:05 so I have other things like that, blue t-shirts and stuff,

00:20:07 I would have put another one, but here, with the little collar,

00:20:09 it was perfect.

00:20:11 But even the color, it was very chosen,

00:20:13 the way to do this thing.

00:20:15 But I said to myself at that moment,

00:20:17 if I don't give a fuck about shooting on a green background,

00:20:19 wouldn't it be stupid

00:20:21 to do all this

00:20:23 just to put myself in front of my shelf?

00:20:25 It would be half the effort

00:20:27 only at the visual level,

00:20:29 you made all the effort to make an incrustation,

00:20:31 and you just put yourself in front of your shelf,

00:20:33 and it would have stayed the same color,

00:20:35 it would have sold the effect of skin color change

00:20:37 really well, I think,

00:20:39 but in terms of rendering,

00:20:41 there was nothing special, it was just me in blue,

00:20:43 and you say, well, at that moment, it's just to be in blue,

00:20:45 obviously, indeed, as much as you can

00:20:47 start with blue before shooting something, and it's good.

00:20:49 Although I really didn't want to use

00:20:51 paint, because it always makes

00:20:53 traces and it's a little disgusting, etc.

00:20:55 I didn't want it to be disgusting, I wanted it to feel like a real blue skin.

00:20:57 So at that moment, I said to myself,

00:20:59 if it's just to be

00:21:01 incrustated like that in blue,

00:21:03 would there be another thing that would

00:21:05 maybe make me feel

00:21:07 more disgusted? It wouldn't be

00:21:09 to order

00:21:11 from an infographicist

00:21:13 an avatar

00:21:15 of Navi, with

00:21:17 the goal of using

00:21:19 Unreal Engine, because they

00:21:21 just released a whole bunch of demos

00:21:23 of technology they use, and frankly,

00:21:25 if it can become accessible to the common

00:21:27 mortals, it would be extraordinary.

00:21:29 They showed a video where they do

00:21:31 face and body motion capture,

00:21:33 both at the same time, without any

00:21:35 motion capture ball,

00:21:37 without any reception trick, just

00:21:39 a visual analysis by an

00:21:41 artificial intelligence, which then reproduces

00:21:43 all your movements on

00:21:45 your 3D character, your avatar.

00:21:47 And so I said to myself,

00:21:49 perfect, I'm going to use that.

00:21:51 And so I was contacted by an infographicist, who has already contacted me

00:21:53 for some time, he did the

00:21:55 passages where I am

00:21:57 in a SD polygon in Final Fantasy 7,

00:21:59 in my project on Final Fantasy 7, where I'm

00:22:01 riding on Chocobos.

00:22:03 And so he knows himself extremely well,

00:22:05 because he worked on a whole bunch of video games,

00:22:07 he also works in infography on a whole bunch

00:22:09 of movies. I'll tell you

00:22:11 what's written on his

00:22:13 Twitter handle, but he worked

00:22:15 on Resident Evil 7,

00:22:17 Horizon Zero Dawn,

00:22:19 and... what's that? LDNR?

00:22:21 I don't know the game.

00:22:23 But he worked

00:22:25 on very big projects, big things,

00:22:27 he's a real infographicist,

00:22:29 who really knows what he's talking about. And especially, his speciality,

00:22:31 and I didn't even know

00:22:33 when I was going to talk to him, but his speciality is

00:22:35 face motion capture.

00:22:37 So it was perfect!

00:22:39 Yes, but there are

00:22:41 some little problems.

00:22:43 When you're a professional and you want to pay people

00:22:45 correctly, etc., you ask them

00:22:47 how much it would cost. And so I went to see him

00:22:49 and I told him, "How much would it cost me

00:22:51 to ask you to do a model of me

00:22:53 with an avatar model, etc.

00:22:55 And he told me, "My price

00:22:57 for a project that's not

00:22:59 professional, you're still an amateur,

00:23:01 you're on YouTube, you don't have a revenue, etc.

00:23:03 A very good quality

00:23:05 model for a project like that

00:23:07 is 2000€.

00:23:09 I asked him if he agreed,

00:23:11 he agreed to let me broadcast the game,

00:23:13 but maybe he wouldn't have spent 2000€ if you were a big company

00:23:15 or something like that. It's just what he gave me.

00:23:17 And I told him,

00:23:19 "My dear Maxime,

00:23:21 2000€,

00:23:23 I'm not even sure I can do that by releasing the video.

00:23:25 And at the moment

00:23:27 I'm making this video, so it has

00:23:29 350,000 views, I've barely exceeded

00:23:31 1,000€, I'm at 1,100, I think. And it's going to

00:23:33 get more and more laborious because every day

00:23:35 that passes brings less and less money from now on.

00:23:37 I'm going to have a hard time getting to 2000.

00:23:39 So when he tells me that, I tell him,

00:23:41 "Yeah, no, it's going to be complicated." And he even had

00:23:43 ideas for me for what was motion capture,

00:23:45 faces, bodies, etc. Things that were

00:23:47 a little complex to do, but I thought we could still

00:23:49 do it, especially if he helped me. The problem is that

00:23:51 if he helps me and I have to pay him, I won't be able to do it.

00:23:53 I don't have the budget, I don't have the means.

00:23:55 And I'm not going to be my YouTuber asshole who says,

00:23:57 "I'm going to pay you in public images

00:23:59 and stuff like that.

00:24:01 In exposure, that's what I was looking for.

00:24:03 So I told him, "No, dude,

00:24:05 it's not worth it. If it's 2000€,

00:24:07 it's way beyond my budget.

00:24:09 I would never pay for all of that, even if the video

00:24:11 is a huge hit. I already knew that in advance.

00:24:13 So at that moment, I started thinking

00:24:15 "Let's get into a jungle."

00:24:17 But at that point, the idea was

00:24:19 "Let's go to Google Images,

00:24:21 find a nice jungle image,

00:24:23 and then let's go around the ferns,

00:24:25 the plants, and stuff like that.

00:24:27 Then you look, and you realize that

00:24:29 you already found exactly what you wanted.

00:24:31 I wanted to be able to get into the set.

00:24:33 I didn't just want to be in front of

00:24:35 a virtual set.

00:24:37 I wanted to be inside.

00:24:39 So I wanted a background and a foreground

00:24:41 behind which I could be.

00:24:43 And already finding a photo like that,

00:24:45 exactly as you saw it in Google Images,

00:24:47 it's not that easy.

00:24:49 You can find a small clearing in front of some trees,

00:24:51 some shrubs, and stuff like that.

00:24:53 But with the foreground in front,

00:24:55 it's a bit more complicated.

00:24:57 And I thought, "Dumbass, if you find a photo like that,

00:24:59 the foreground will be blurry,

00:25:03 and if it's blurry, it's going to be

00:25:05 a pain to blur."

00:25:07 It's one of the most annoying things

00:25:09 in the world, to blur something

00:25:11 that's blurry compared to a background behind it.

00:25:13 So I thought,

00:25:15 "It's going to be a disaster, it's going to be ugly,

00:25:17 and I'm never going to be able to do what I want."

00:25:19 So,

00:25:21 I kept thinking,

00:25:23 and I ended up thinking,

00:25:25 "A few years ago,

00:25:27 there was a 3D program that I knew how to use.

00:25:29 It's not the best,

00:25:31 but it might be able to do what I want."

00:25:33 And this program came back to me

00:25:35 because I started working on it

00:25:37 for the "Why I'm Right, They Were Wrong"

00:25:39 that I gave up last year.

00:25:41 There are plans

00:25:43 for which I need special effects,

00:25:45 for the "Why I'm Right, They Were Wrong"

00:25:47 that will come later this year.

00:25:49 I absolutely needed a 3D program to do them,

00:25:51 and at that moment, I thought,

00:25:53 "I'm going to get a 3D program that I used

00:25:55 when I was a teenager."

00:25:57 And I realized that it had had a lot of evolution

00:25:59 in the meantime,

00:26:01 including the 3D rendering engine

00:26:03 of a pretty well-known 3D program called Blender,

00:26:05 which is free and open-source,

00:26:07 and all that crap,

00:26:09 and it's a gas plant to use.

00:26:11 I've had a technique since I was a teenager.

00:26:13 When I want to do something,

00:26:15 I take a program that helps me do it,

00:26:17 and I learn how to use the program.

00:26:19 How do you do it?

00:26:21 You open the program,

00:26:23 and you press all the buttons,

00:26:25 and you see what it does.

00:26:27 And after a while,

00:26:29 you try to find,

00:26:31 "I'm going to associate this one,

00:26:33 this one, this one,

00:26:35 this one, this one,

00:26:37 this one, this one, this one."

00:26:39 Blender works almost entirely with shortcuts.

00:26:41 Before doing an action in Blender,

00:26:43 you have to know

00:26:45 exactly what action you want to do,

00:26:47 and what shortcuts will allow you to do it.

00:26:49 Because, like,

00:26:51 clicking on an object with your mouse

00:26:53 won't select the object.

00:26:55 It won't select the Y-curve

00:26:57 on which to make a rotation

00:26:59 on the object,

00:27:01 or it will do this,

00:27:03 or it will do that.

00:27:05 It's not that simple.

00:27:07 You have to press a shortcut,

00:27:09 and then you click,

00:27:11 and you move your mouse,

00:27:13 and it will enlarge the object

00:27:15 depending on the movement of your mouse.

00:27:17 But you have to know what the shortcut is.

00:27:19 If you want to move the object,

00:27:21 it's another key, etc.

00:27:23 You have to know what this other key is.

00:27:25 And so, in fact, Blender is a nightmare.

00:27:27 I don't even dare to put myself on it,

00:27:29 because it's really the kind of thing

00:27:31 you have to spend three months

00:27:33 to learn. When you're an adult,

00:27:35 it's not the kind of thing that's easy to do.

00:27:37 And, on the contrary,

00:27:39 the program I use, called Poser,

00:27:41 which has been around for a long time,

00:27:43 we're on version 12 now,

00:27:45 is ultra user-friendly.

00:27:47 But really,

00:27:49 it's "what you see is what you get".

00:27:51 It's an old code term

00:27:53 we used before.

00:27:55 You press a button,

00:27:57 the action that will be done

00:27:59 and the action that is marked on the button.

00:28:01 You have a ball on the side,

00:28:03 with a ball that allows you to rotate

00:28:05 around an axis for the cameras

00:28:07 that are focused around a point.

00:28:09 And you have a whole bunch of other options

00:28:11 to move the camera on the sides,

00:28:13 move the camera forward, move the camera

00:28:15 on top and on the bottom.

00:28:17 And you don't have a whole bunch of shortcuts

00:28:19 to use, etc.

00:28:21 Same for textures, it's a lot easier.

00:28:23 In part because there are a lot of options

00:28:25 that are not available, but it's a lot easier

00:28:27 to manage, to add textures, to add

00:28:29 displacement maps that make a real

00:28:31 relief from something that is supposed to be flat

00:28:33 and from a texture, with height information.

00:28:35 And so, having already started

00:28:37 using this program,

00:28:39 being a little familiar with it last year,

00:28:41 I said to myself,

00:28:43 "You're going to take your composer

00:28:45 and you're going to create a 3D jungle

00:28:47 decor".

00:28:49 And I didn't really realize

00:28:53 the difficulty of the exercise

00:28:55 I had given myself.

00:28:57 At that moment, he told me,

00:28:59 "If you can manage it on your own,

00:29:01 there's something I can do for you".

00:29:03 He helped me, we

00:29:05 stole

00:29:07 the Epic website.

00:29:09 Because Epic

00:29:11 made a website with a whole bunch of 3D models

00:29:13 that are scanned

00:29:15 by photogrammetry. So we take a whole bunch of

00:29:17 real objects and then we convert them

00:29:19 thanks to artificial intelligence into 3D objects

00:29:21 with a few polygons, etc.

00:29:23 So it's very easy to get

00:29:25 a very photorealistic result

00:29:27 if we take the most detailed version.

00:29:29 And at the same time, a very basic version

00:29:31 that goes through any video game engine with very little RAM memory.

00:29:33 And I really

00:29:35 went to the pages

00:29:37 and I said to him, "I want this, I want that, I want that,

00:29:39 I want that, I want that, I want that, I want that".

00:29:41 And he sent me a pack with

00:29:43 like, I must have stolen about thirty

00:29:45 or forty... I say stolen, but it's

00:29:47 public, it's free, it's for...

00:29:49 I didn't do anything illegal, you know.

00:29:51 You just need to have an Epic account, and he has an Epic account,

00:29:53 and he sent me a file with

00:29:55 forty models of

00:29:57 vegetation, plants,

00:29:59 trees, and stuff like that. And he himself

00:30:01 already had a few assets, since he obviously works

00:30:03 in video games and stuff like that, he already had a few

00:30:05 jungle assets, nature assets,

00:30:07 some ferns, some trees,

00:30:09 and especially a tree with, you know, the vines

00:30:11 that hang in the middle, which is the big tree we put in the middle

00:30:13 of the thing.

00:30:15 But there was a little quality gap

00:30:17 between the assets he had

00:30:19 that were made for video games, so in absolute low poly,

00:30:21 and the rest of the assets that were

00:30:23 rather in good quality. And the thing is,

00:30:25 since you're doing a virtual plan

00:30:27 to evoke Avatar, the goal is

00:30:29 more to use the high quality assets

00:30:31 than the rest, you know. The thing is,

00:30:33 I knew personally that the assets in very good quality

00:30:35 take up a lot of space, and that it was going to be

00:30:37 a hassle to render

00:30:39 the scene afterwards. So

00:30:41 I didn't go for the top mega

00:30:43 quality, but I wanted a higher quality.

00:30:45 So there are some things I did, especially

00:30:47 his tree with the vines. In fact,

00:30:49 all the branches are

00:30:51 kind of semi-domes,

00:30:53 of...

00:30:55 balls cut in the middle, with the

00:30:57 plants that are in a flat texture on top.

00:30:59 And so, depending on the angle,

00:31:01 either it can go very well, or you realize

00:31:03 that it's a whole bunch of turtle shells

00:31:05 stacked on top of each other with

00:31:07 a weird transparency.

00:31:09 So I went looking for other assets for some things,

00:31:11 especially the big central tree, and in fact

00:31:13 I didn't find it quite well-placed,

00:31:15 so I mixed the two. In fact, in this

00:31:17 big central tree, there's both the asset

00:31:19 very low-poly, and there's a tree

00:31:21 much higher quality.

00:31:23 And the two are mixed together, because

00:31:25 the leaves of this tree are much more realistic,

00:31:27 much more credible, etc. And so it brings

00:31:29 all the credibility to this tree.

00:31:31 And then the rest, I started

00:31:33 to dress it up quietly, and in fact I found

00:31:35 a whole system, there are

00:31:37 mathematicians who found a geometric shape

00:31:39 that can

00:31:41 be repeated

00:31:43 infinitely, without ever giving the impression of repetition, because

00:31:45 the shape itself is so complex that when you repeat it, it looks very weird.

00:31:47 But I think you also have to turn it in other

00:31:49 directions, make it rotate on itself,

00:31:51 etc. In the end, it didn't work that well,

00:31:53 but in fact what I did was create

00:31:55 a vegetation plate with

00:31:57 a bush. So at first I started with the bush thing,

00:31:59 and I created

00:32:01 a whole forest

00:32:03 surface, which in the end

00:32:05 was a little flat. I'll show you,

00:32:07 because I still have some pictures, it looked like this at first.

00:32:09 You see that it's a little bit particular,

00:32:11 you get the impression that it's trees that grow very flat, or that it's been cut

00:32:13 by the Paris Commune or I don't know what,

00:32:15 with things to cut trees and all that.

00:32:17 It's a little bit peculiar. And it was very long

00:32:19 to render too, the file was

00:32:21 extremely big, it was more than 2 GB

00:32:23 I think, in total, the file.

00:32:25 And

00:32:27 in the end, the rendering time

00:32:29 was so long that if I wanted to do

00:32:31 my 10 seconds of intro shots, it would have taken me

00:32:33 almost two weeks to render.

00:32:35 So at that moment,

00:32:37 I thought, well,

00:32:39 it's a failure for the jungle,

00:32:41 so we're going to go for something else.

00:32:43 And I was going to give up.

00:32:45 And still, it pissed me off.

00:32:47 It pissed me off to tell myself that.

00:32:49 I started to discuss with Maxime,

00:32:51 the graphic designer,

00:32:53 "Do you think I'm going to switch to Unreal Engine?"

00:32:55 Because it's a bit laborious.

00:32:57 At that moment, he blew a cable because he hates Unreal

00:32:59 and he prefers Unity because he says that Unreal is shit,

00:33:01 it's super heavy,

00:33:03 it's always buggy,

00:33:05 you load the slightest FBX,

00:33:07 so it's a 3D file,

00:33:09 the 3D model of whatever,

00:33:11 and apparently the plant file, etc.

00:33:13 But at the same time, the rendering quality is crazy

00:33:15 with Unreal Engine 5.

00:33:17 But the thing is that Unreal Engine 5,

00:33:19 the rendering quality is also crazy

00:33:21 because it does ray tracing in terms of light.

00:33:23 That's what gives it,

00:33:25 well, the quality of the models is crazy,

00:33:27 but it's especially the light that makes it so realistic.

00:33:29 And you can now have jungle rendering

00:33:31 in a video game that can look as realistic

00:33:33 as in an avatar,

00:33:35 simply with low poly and stuff like that.

00:33:37 And the thing is that my PC is not made to do ray tracing.

00:33:39 I can't use Unreal Engine 5 on my PC

00:33:41 because I'm still with a GTX

00:33:43 in terms of my GeForce,

00:33:45 my graphics card,

00:33:47 which doesn't handle ray tracing.

00:33:49 So I think it could still make the thing run,

00:33:51 but I wouldn't have any speed gain

00:33:53 in rendering that brings the ray tracing

00:33:55 and Unreal Engine 5 with the graphics card

00:33:57 that I have here.

00:33:59 So I thought, "No, I can't do it on Unreal Engine,

00:34:01 it's useless."

00:34:03 And I was still talking to Maxime,

00:34:05 who said, "The avatar is still not feasible."

00:34:07 And then after a while, I said to myself,

00:34:09 "Find yourself

00:34:11 a tree

00:34:13 that has a very good quality rendering

00:34:15 despite

00:34:17 very, very few polygons."

00:34:19 Because usually, the more you reduce the polygons, the more you can see it.

00:34:21 But there are some that can be organized

00:34:23 in a way that you don't even realize

00:34:25 that there is a geometric shape

00:34:27 behind the texture you have behind,

00:34:29 and that it's not really leaves growing at the end of branches,

00:34:31 but that there's a whole bunch of grooves behind.

00:34:33 So I ended up finding

00:34:35 a tree that was

00:34:37 ultra credible and ultra detailed,

00:34:39 that had leaves that were well grown,

00:34:41 that had a good thing, etc.

00:34:43 And that at the same time was super light.

00:34:45 I deleted all my

00:34:47 2 giga-octave

00:34:49 bush plates that I had made in the thing before,

00:34:51 I deleted everything.

00:34:53 I did, in the end, what I should have done from the beginning.

00:34:55 I placed trees by hand

00:34:57 everywhere on this map.

00:34:59 Once or twice, I think I changed the tree model,

00:35:01 it's always the same, and I especially

00:35:03 modified a little bit

00:35:05 the width or the height to give

00:35:07 a lot more relief

00:35:09 to this map, and that we don't have the impression that it's a flat thing

00:35:11 like the first version I had.

00:35:13 And so I ended up

00:35:15 with a file that was

00:35:17 about 800 mega-octave, so we reduced

00:35:19 the 2 giga that was at the base anyway,

00:35:21 and that had a much

00:35:23 better rendering and that rendered much more

00:35:25 quickly. So a gain

00:35:27 everywhere, it's lighter, it's rendering faster,

00:35:29 it works well, it's perfect.

00:35:31 So I finally have a rendering that's going on,

00:35:33 I think it would take me 4-5 days to

00:35:35 make the total rendering

00:35:37 of the video, but

00:35:39 4-5 days with the quality that I'm used

00:35:41 to using with my PC.

00:35:43 My PC, which I call

00:35:45 a GTX 1070.

00:35:47 I'm 4 generations behind the GeForce that are

00:35:49 currently released. So when I do these renders,

00:35:51 it's not the maximum quality renders.

00:35:53 There's still a lot of stuff to do

00:35:55 and a lot of things to refine. And if I put myself

00:35:57 in the maximum quality, I realized

00:35:59 "Oh, it's too late". It wasn't really the maximum quality.

00:36:01 There's still a quality above what I did there.

00:36:03 But to really get the maximum quality and not see the grain

00:36:05 on the image, so that the

00:36:07 encrustation goes better too, I had to

00:36:09 make a much better quality rendering.

00:36:11 And this better quality rendering,

00:36:13 of course, took twice as long and I

00:36:15 ended up with the same

00:36:17 problem as before I redo

00:36:19 my whole map and change all the trees

00:36:21 that were present on the thing, etc.

00:36:23 So it could have been a bit laborious, but

00:36:25 at that moment, I went to talk to one of my

00:36:27 friends, Gorkab, who I talk to regularly.

00:36:29 He has a channel that is on the analysis of

00:36:31 the history of special effects

00:36:33 from the 80s, 70s, 80s,

00:36:35 until 90s, when they really started

00:36:37 to democratize. He told me

00:36:39 that he had just got a brand new PC with

00:36:41 a GeForce 4090.

00:36:43 And so I said to him "Wait, you have the best graphics card in the world

00:36:45 at the moment, in your PC?

00:36:47 At any rate, could I

00:36:49 come to your house and see how long it would take

00:36:51 to render my video at your house?

00:36:53 Because a single high definition image

00:36:55 1080p would take me

00:36:57 like...

00:36:59 15 minutes

00:37:01 to 30 minutes to get home

00:37:03 to do the final rendering.

00:37:05 I went to his house, I started the rendering, in 1 minute 10, it was over.

00:37:07 The change of a sudden!

00:37:11 And so I could do my rendering at his house.

00:37:13 Knowing that it improved the rendering a lot,

00:37:15 because I realized something. I tried to do,

00:37:17 since the beginning of this video,

00:37:19 in these 10 seconds of camera

00:37:21 flying above the jungle before going down,

00:37:23 an atmosphere effect. Because that's what sells

00:37:25 the realism of this shot,

00:37:27 the atmosphere effect, the light that goes through the leaves,

00:37:29 that creates the divine rays

00:37:31 and all the mess.

00:37:33 And so I really wanted that, because

00:37:35 it was what gave a good rendering.

00:37:37 And in fact, on my PC, I kept trying to work on the thing,

00:37:39 and I didn't understand how

00:37:41 the system worked, because it was never

00:37:43 as dense as it should be.

00:37:45 It took the light,

00:37:47 but it wasn't as opaque and dense,

00:37:49 there was something that didn't work at all

00:37:51 with this atmosphere effect.

00:37:53 And I had struggled for several

00:37:55 hours and several days to find

00:37:57 a rendering that was dense enough, and it still wasn't what I wanted.

00:37:59 And I arrive at GORECAB,

00:38:01 I put the maximum quality that the program

00:38:03 offered me, which is not the real maximum quality that we can have,

00:38:05 because I find it very stupid from the program,

00:38:07 you should have a rendering like THE maximum quality,

00:38:09 the finished quality. But no,

00:38:11 I just noticed an ultra mode somewhere.

00:38:13 And there, I suddenly see all my

00:38:15 atmosphere effect, which takes twice as much

00:38:17 intensity, which is twice as full, which does twice as much stuff.

00:38:19 Because in fact,

00:38:21 the quality of this atmosphere effect,

00:38:23 it depends on the quality of your rendering.

00:38:25 The lower the quality of your rendering,

00:38:27 the less dense the atmosphere is.

00:38:29 Which seems to me to be the most absurd

00:38:33 thing in the world. I don't know how this program

00:38:35 is coded, and if it makes sense in terms of coding

00:38:37 for the rendering, but you do all

00:38:39 the settings in your textures, in your

00:38:41 volumetrics, in your stuff like that, in the parameters

00:38:43 of your program, it's not for it to change

00:38:45 depending on the rendering mode you're asking for.

00:38:47 Basically, whatever

00:38:49 the rendering mode you're asking for, you should have,

00:38:51 in the texture section

00:38:53 of your 3D program,

00:38:55 or in the volumetric section,

00:38:57 you shouldn't have a different rendering

00:38:59 depending on whether you put the maximum rendering or not.

00:39:01 Indeed, more or less granular

00:39:03 or not, yes, but more or less dense,

00:39:05 which captures more or less well the light,

00:39:07 etc. It was completely stupid,

00:39:09 it didn't work, I don't understand.

00:39:11 So I realized this thing too,

00:39:13 and so, the video has a good quality

00:39:15 in the end thanks to that too, because

00:39:17 the atmosphere rendering worked infinitely

00:39:19 better with the GORECAB PC,

00:39:21 and so it allowed you to make much better shadows,

00:39:23 nice gradients, etc., and make a rendering that

00:39:25 looked realistic and not too

00:39:27 uniform, because if you don't put this rendering there,

00:39:29 all the plants have

00:39:31 exactly the same hue everywhere, in every place,

00:39:33 nothing really changes

00:39:35 in the thing, it's always the same contrast

00:39:37 everywhere, etc., and it's much less

00:39:39 credible and realistic. It's not something

00:39:41 that can work, but usually you need

00:39:43 a minimum of atmosphere

00:39:45 somewhere in a scene. And so,

00:39:47 thanks to all this work,

00:39:49 I ended up with a

00:39:51 jungle entirely mod...

00:39:53 well, not modeled, because I didn't

00:39:55 model it, but entirely created in 3D,

00:39:57 in which I can walk my camera

00:39:59 and in which I can choose

00:40:01 exactly what I want as a shot,

00:40:03 where I want to have vegetation in front of the image,

00:40:05 etc., etc., and so

00:40:07 I can take advantage of that to make

00:40:09 the video as I want. I tried

00:40:11 to calculate the shot

00:40:13 so that I could

00:40:15 shoot with my usual shot value,

00:40:17 because I didn't want to

00:40:19 bother with the green background this time.

00:40:21 Making a green background at home is always a mess,

00:40:23 because my apartment is tiny, I'll show you in a few seconds.

00:40:25 And I wanted to try

00:40:27 to shoot with my traditional shot,

00:40:29 why I had the reason to do it before, front camera,

00:40:31 with the green background behind, which meant shooting from there.

00:40:33 The problem is that I wanted,

00:40:35 as we were in Avatar, I wanted

00:40:37 at least an idea that the environment was a bit bigger,

00:40:39 and for that I wanted the shot to be a bit wider.

00:40:41 And so I had planned

00:40:43 that the two blue plants

00:40:45 you see, which weren't blue at first, I was the one who

00:40:47 modified the texture, they were green at first,

00:40:49 they were a little bit more on the side,

00:40:51 because the goal was to completely hide the middle of the image

00:40:53 so that we don't see my legs,

00:40:55 we only see my stomach protruding,

00:40:57 and that I can shoot all my shots

00:40:59 in front camera from there,

00:41:01 but that in the image it looks

00:41:03 more or less coherent because the bottom of my body

00:41:05 is hidden behind the plant.

00:41:07 I still had to move away a bit,

00:41:09 and I shot in a shot...

00:41:11 I always reverse if it's

00:41:13 American or Italian when it's just above the knee.

00:41:15 I think it's American because it's Western.

00:41:17 But despite everything, I still used my

00:41:19 original technique thanks to that, because

00:41:21 there are a lot of details

00:41:23 of my green background that are disgusting

00:41:25 and that are hidden by the plant.

00:41:27 Behind the plant, behind the

00:41:29 roses that are on the sides and all the mess,

00:41:31 there is the edge of my

00:41:33 kitchen counter,

00:41:35 because I'm in a very small apartment,

00:41:37 I'll show you, and the edge

00:41:39 of this thing appears on the edge of the image,

00:41:41 and I have to remove it, of course.

00:41:43 There are also some folds of the sheet

00:41:45 that are difficult to remove, given how it was put,

00:41:47 and they are also in these places,

00:41:49 so it was perfect. But in the end,

00:41:51 I had to mess up my apartment to be able to shoot in front.

00:41:53 So, I'm going to show you a picture,

00:41:55 but I'm going to be a little honest.

00:41:57 I'm almost ashamed to show you this picture,

00:41:59 because I think you can't imagine

00:42:01 what apartment I live in,

00:42:03 and in what surface I live.

00:42:05 This is what my apartment looks like

00:42:07 with all the messed up stuff for the green background.

00:42:09 So, there is no

00:42:11 basic place

00:42:13 to put my sofa

00:42:15 and a green background in the same place.

00:42:17 So if I make a green background, I have to move the sofa.

00:42:19 If I don't need

00:42:21 to film myself from too far away,

00:42:23 I can just

00:42:25 install the green background

00:42:27 on the sofa,

00:42:29 and I can just put it on top of each other.

00:42:31 So I have a green background behind me,

00:42:33 and it's fine. But if I want to see myself from far away,

00:42:35 I have to move the camera to a place

00:42:37 where I can't move it.

00:42:39 It's right behind the TV.

00:42:41 So, it was a real hassle

00:42:43 to shoot, because this apartment is very small,

00:42:45 and if I do something like that,

00:42:47 it's a real hassle.

00:42:49 So, for example,

00:42:51 the basic plan was...

00:42:53 I already did this for the shoot

00:42:55 I did for Stanley Kubrick,

00:42:57 because I also had a lot of green background

00:42:59 to do at that time, and the green background was very laborious

00:43:01 to do, and to put in place,

00:43:03 and there was the light that I had already done in advance, the pre-light,

00:43:05 to be able to just turn on the light,

00:43:07 and put myself in front of the green background and shoot the next day,

00:43:09 because it takes a long time to put a green background in place,

00:43:11 to be sure that the light is good, that it doesn't make

00:43:13 unpleasant shadows, etc. It's really complicated.

00:43:15 And so, when everything is a mess like that,

00:43:17 I have to sleep on the floor.

00:43:19 Even on the floor. Because usually,

00:43:21 the sofa is occupied by something.

00:43:23 Typically, there was a foot of the green background

00:43:25 that I put on it. If ever

00:43:27 the foot of the green background broke,

00:43:29 it was useless, it would take a long time

00:43:31 to put everything in place the next day, etc.

00:43:33 It's not a crazy time, it wouldn't take me the whole day,

00:43:35 but it's laborious.

00:43:37 You can't just take apart your green background and put it back on.

00:43:39 When you're going to put it back on,

00:43:41 there will be changes that will make the light better,

00:43:43 you'll also have to change the light, etc.

00:43:45 It's a whole lot of shit.

00:43:47 So, basically, I was going to sleep on the floor.

00:43:49 The time before, I had better prepared,

00:43:51 and I had a whole bunch of quets that were already ready.

00:43:53 So I could stack a number of them on the floor,

00:43:55 and I had made a real little mattress,

00:43:57 a little duvet somewhere.

00:43:59 So I didn't sleep too badly at that time, for Kubrick.

00:44:01 This time, I was a little less well prepared,

00:44:03 I didn't have as many sheets that were clean,

00:44:05 and all the mess.

00:44:07 So I was really on the floor,

00:44:09 and after an hour and a half,

00:44:11 I had back pain, neck pain, etc.

00:44:13 And I ended up sleeping on the sofa,

00:44:15 but like, on the three quarters of the sofa.

00:44:19 There was a quarter that was taken by the foot of the green background,

00:44:22 and I was in a ball, in three quarters,

00:44:24 lying on the rest.

00:44:26 And so, it's not my best night.

00:44:28 It's clear and clear, it's not my best night.

00:44:30 But I still managed to organize everything,

00:44:32 and put everything in place,

00:44:34 to be able to turn this green background as I wanted to do it,

00:44:36 and be able to embed myself in it as I wanted,

00:44:38 and have the virtual setting that I wanted.

00:44:40 Two little things I'm going to talk about,

00:44:42 and then we'll move on to the rest of the video.

00:44:44 I think it's been an hour since we did this thing.

00:44:46 I started by making the final plan.

00:44:49 So the plan that's behind the plants, etc.

00:44:51 And there are a lot of things in this background,

00:44:55 but that weren't used in the end.

00:44:57 The background has changed a bit, the tree has become much bigger.

00:44:59 At first, it was much smaller than that,

00:45:01 and I thought we couldn't see the scale ratio

00:45:03 between the rest of the trees and this huge tree in the middle.

00:45:05 And so, I enlarged it a lot more.

00:45:07 And in fact, behind the tree,

00:45:09 there's a whole bunch,

00:45:11 there's a little wall of rocks,

00:45:13 there are a few plants,

00:45:15 one of the blue plants, for example,

00:45:17 is here, there's another one behind.

00:45:19 There are some ferns,

00:45:21 there are some things like that,

00:45:23 there are some simply grass tufts.

00:45:25 There was a moment when there was a plate

00:45:27 to make a reflective water surface,

00:45:29 which I ended up removing

00:45:31 because it's typically in terms of ray tracing,

00:45:33 it takes a lot of calculations

00:45:35 with my graphics card.

00:45:37 So I couldn't have seen it all the way through.

00:45:39 But at the base, there should have been a small water surface,

00:45:41 a small pond somewhere behind the thing.

00:45:43 And in fact, I didn't delete things

00:45:45 as I hid them,

00:45:47 because it adds to the "touffu" aspect of the environment.

00:45:49 I made sure that the environment

00:45:51 is extremely rich, extremely "touffu",

00:45:53 that there's never a place where there's just a hole

00:45:55 and you can see the sky behind it.

00:45:57 Because that's how nature works,

00:45:59 there are no holes like that in nature,

00:46:01 except for small lighthouses,

00:46:03 but in principle, you see nature everywhere around you.

00:46:05 You never see...

00:46:07 When you're in a forest, you don't see the city behind the forest,

00:46:09 you only see holes everywhere and leaves.

00:46:11 And so, I was very happy in the end

00:46:13 to make the environment very "touffu",

00:46:15 because I didn't have to worry about plants

00:46:17 getting on top of each other.

00:46:19 The little thing with the blue water lily

00:46:21 that has its plant

00:46:23 that's just between two leaves

00:46:25 of the plant behind it,

00:46:27 one above, one below, etc.

00:46:29 I don't even know how I managed to get out of it

00:46:31 to make this 3D game of "fence"

00:46:33 because I think that if I had tried to do it,

00:46:35 I wouldn't have succeeded.

00:46:37 I moved the plant and by chance,

00:46:39 it ended up like this in the rendering.

00:46:41 I think I just had to lower it a little bit

00:46:43 so that it would fit in the leaf.

00:46:45 But it was a total luck.

00:46:47 I would have wanted to do it on purpose,

00:46:49 I couldn't have. But really,

00:46:51 that's what I wanted to work on,

00:46:53 that it's really "touffu" and that it works well.

00:46:55 And I have the impression that it works well,

00:46:57 that it's rather organic.

00:46:59 And the thing that was very laborious to do too

00:47:01 was the light.

00:47:03 Because you needed a light

00:47:05 that was a little realistic,

00:47:07 that wasn't too complicated for me to reproduce either,

00:47:09 because there are styles of light

00:47:11 that are more complicated than others to do.

00:47:13 So in fact, from the beginning,

00:47:15 I built the lighthouse

00:47:17 in this little place

00:47:19 so that once the tree was above,

00:47:21 with the light coming in,

00:47:23 we have a light area right on the plants

00:47:25 that are in the foreground

00:47:27 and the rest is a little more in the shade,

00:47:29 the tree in particular.

00:47:31 If only to allow me to detach myself better in this thing,

00:47:33 to also do something that is more realistic in terms of light, etc.

00:47:35 And in fact, the light,

00:47:37 I'm quite surprised, but I really managed to reproduce it.

00:47:39 It's actually,

00:47:41 quite stupidly, I took one of my lights,

00:47:43 one of them is connected and the other is super far away.

00:47:45 What am I going to do?

00:47:47 So that's what I use

00:47:49 to be able to make my lights.

00:47:51 So these are small LED lights.

00:47:53 I can't turn it on because it's not connected.

00:47:55 But in fact,

00:47:57 it's this thing that I have,

00:47:59 and that's the foot on which I can connect it.

00:48:01 And in fact,

00:48:03 I put this foot on the shelf,

00:48:05 but really in balance

00:48:07 between the end of the shelf, the wall

00:48:09 and the metal bar of the green background,

00:48:11 as you can see in the photos.

00:48:13 And suddenly, I admit that I was a little anxious

00:48:15 during the shooting with the drops of sweat

00:48:17 and I was thinking,

00:48:19 "Will it fall on my face during the shooting?"

00:48:21 Because there is a good part of the weight

00:48:23 that is on the whole metal mat that holds the projector,

00:48:25 but the projector itself has weight too.

00:48:27 And then, in addition, it is connected to

00:48:29 a small adapter

00:48:31 to be powered and all the mess.

00:48:33 So it's also weight that adds to the thing

00:48:35 and it happens that the thing will end up tipping over

00:48:37 and fall on my face during the shooting.

00:48:39 So I had this anxiety during the whole thing.

00:48:41 I slept at night without the light.

00:48:43 It's stupid because it was the pre-light

00:48:45 and it's one of the most important things.

00:48:47 But there aren't many places where I can

00:48:49 wedge this projector compared to

00:48:51 everything that is above the shelf.

00:48:53 I could move a few to put the thing

00:48:55 and there was just enough space

00:48:57 to put the metal foot

00:48:59 well folded and well packed on itself.

00:49:01 There wasn't a centimeter more.

00:49:03 So by pulling the light,

00:49:05 I knew that I could only put it

00:49:07 in one place and that it would be

00:49:09 one type of lighting.

00:49:11 But I didn't even think to succeed

00:49:13 to reproduce the sunlight

00:49:15 because it's not a light that is projected

00:49:17 with an extremely wide angle

00:49:19 like a thing.

00:49:21 And it was going to be close to my head

00:49:23 so I thought it might be a bit laborious

00:49:25 but I managed to do the thing.

00:49:27 And in fact, to perfectly reproduce the light,

00:49:29 I included in the jungle rendering

00:49:31 a little character, a little man,

00:49:33 there are little wooden models

00:49:35 included in the 3D program I have

00:49:37 that allowed me to see

00:49:39 how the light should be taken

00:49:41 to be realistic with this thing.

00:49:43 And so I saw that I always had to have

00:49:45 the shadow of my head on the front of my body,

00:49:47 that the sun was always behind me,

00:49:49 that I didn't really have it on the front

00:49:51 except if I raised my head like that,

00:49:53 at that moment it should be lit

00:49:55 but there is really an angle

00:49:57 according to which the light comes

00:49:59 and I managed to reproduce it.

00:50:01 There are people who commented

00:50:03 that I would be badly calibrated

00:50:05 that there would be a calibration flaw

00:50:07 in the video. I'm not badly calibrated,

00:50:09 I did it on purpose to be calibrated

00:50:11 exactly like the darkest black

00:50:13 of this image and the lightest

00:50:15 of this image to be able to

00:50:17 fit it correctly. That's how you

00:50:19 succeed in fitting correctly in a setting.

00:50:21 As there is an atmosphere effect in this setting,

00:50:23 there is necessarily a slightly opaque thing

00:50:25 in front of me that suddenly removes

00:50:27 the darkest black and in fact it is

00:50:29 on the contrary, it was a big job

00:50:31 on my part to try to fit perfectly

00:50:33 what I could see as the darkest black

00:50:35 at that level of the plan

00:50:37 with the plants that are next to it,

00:50:39 not the fern but the

00:50:41 kind of rose or I don't know what

00:50:43 to know what deep black I can reach

00:50:45 to the maximum and what

00:50:47 global lighting I can have on my character

00:50:49 so that it goes correctly in this setting.

00:50:51 And on the contrary,

00:50:53 it is to help the integration that it turned out like that.

00:50:55 There is no calibration defect, I would go like that

00:50:57 in this setting if I was in it.

00:50:59 And where I was very lucky

00:51:01 is with the camera I use.

00:51:03 I have been using a Blackmagic for years,

00:51:05 a production camera 4k that is starting to

00:51:07 be outdated now, I would like to change it a little bit but

00:51:09 it still costs 2000 bucks like the camera

00:51:11 but suddenly it has

00:51:13 a dynamic range

00:51:15 quite vast.

00:51:17 The dynamic range is the gap

00:51:19 that it can manage between

00:51:21 the brightest thing in the image

00:51:23 and the darkest thing in the image.

00:51:25 There is inevitably a moment when something is too bright

00:51:27 and it will completely burn the image.

00:51:29 Something that will be too dark

00:51:31 and it will necessarily be entirely black.

00:51:33 But with this Blackmagic, there is a

00:51:35 great dynamic range and suddenly

00:51:37 there was a big risk, it was that as the light

00:51:39 was resting on the green background,

00:51:41 suddenly the light directly illuminates the green background

00:51:43 and it's the kind of thing that makes you

00:51:45 a hassle for the incrustation of the green backgrounds

00:51:47 because suddenly you have a big difference between

00:51:49 the green of this area of the image

00:51:51 which is extremely

00:51:53 illuminated and therefore close to white

00:51:55 and the rest of the green of the green background

00:51:57 which is more classic, sometimes even darker

00:51:59 etc. So it can create things

00:52:01 that are very complicated to

00:52:03 to distort and to

00:52:05 do the incrustation. And in fact

00:52:07 thanks to the dynamic range of my

00:52:09 Blackmagic, the green background

00:52:11 has never been completely broken white

00:52:13 and like, but I had to do a single

00:52:15 click with my

00:52:17 green background incrustation tool

00:52:19 to have a

00:52:21 a matte, a completely clean

00:52:23 layer.

00:52:25 Almost completely clean.

00:52:27 There were one or two small defects, some little things to remove

00:52:29 here and there, the things that were on the table as I told you

00:52:31 but I had a clean map

00:52:33 in an extremely simple way, although I thought

00:52:35 I was going to struggle with this fucking light spot

00:52:37 behind me.

00:52:39 There is just a small detail,

00:52:41 it's that

00:52:43 despite everything, so it still seems weird to me because

00:52:45 you know, so I have this wig

00:52:47 with braids on the front

00:52:49 and the thing is that the braid, it's a braid

00:52:51 with black hair, by definition

00:52:53 there is a crazy contrast

00:52:55 compared to the green background and especially a green background

00:52:57 well lit. But

00:52:59 weirdly, the closer I got

00:53:01 to this area of the green background

00:53:03 and the more the braids

00:53:05 disappeared on the image.

00:53:07 And in fact the thing is that

00:53:09 at one point I delayed the video for a week

00:53:11 just to manage the incrustation on the green background

00:53:13 because it had one or two defects

00:53:15 that were not huge but that were a bit noticeable

00:53:17 and honestly I had a hard time managing them

00:53:19 and there are even some that I let pass.

00:53:21 I wanted, so when I saw that in this rendering

00:53:23 there were moments where

00:53:25 the braids were disappearing

00:53:27 I corrected my green background incrustation

00:53:29 to be sure that

00:53:31 whatever the moment

00:53:33 the braids were always well cut

00:53:35 on this side. I finished

00:53:37 I had a super clean map

00:53:39 with my braids constantly displayed

00:53:41 and all the mess. Only problem

00:53:43 to manage to change the green background

00:53:45 enough to see those braids, the side effect

00:53:47 was that there was a very light line

00:53:49 around

00:53:51 some

00:53:53 especially hair because I didn't have it on the rest of the body

00:53:55 but at the top of the hair there was a little white line

00:53:57 because hair is not a very very clear thing

00:53:59 and so you often have a little blurry part

00:54:01 that the program doesn't know how to handle

00:54:03 and so it creates a little red line

00:54:05 or white on the sides of the braids.

00:54:07 And so in fact it made even more

00:54:09 the green background

00:54:11 because finally we could see perfectly

00:54:13 the hair's detour in the green background

00:54:15 and it made it even uglier

00:54:17 so I took a week

00:54:19 to correct it by adding the braids

00:54:21 and once I added the braids I realized that it was worse

00:54:23 it was uglier than

00:54:25 my previous version

00:54:27 and so I had to use

00:54:29 the more "rough" version

00:54:31 that

00:54:33 for example when the braids move

00:54:35 too fast they disappear completely

00:54:37 there are moments where the braids are not even visible anymore

00:54:39 because they are blurry

00:54:41 and so it changes the way you do the thing

00:54:43 but I had to use this more "rough" version

00:54:45 to have a cleaner rendering

00:54:47 in the end. What was very practical

00:54:49 is that once I had correctly corrected

00:54:51 the layout of

00:54:53 my camera face

00:54:55 I'll come back to it at the end

00:54:57 my braids were a little confused with the background of the image

00:54:59 especially the darkest part

00:55:01 and that's where it saved me

00:55:03 because as they are almost naturally confused

00:55:05 with the rest of the background

00:55:07 they disappear and you can't see them anymore

00:55:09 especially when there is movement and you focus on my face and my eyes

00:55:11 and all that mess. So I was saved

00:55:13 at the last moment by this thing

00:55:15 but I spent two full days to change something

00:55:17 the Blackmagic

00:55:19 has a debayerization filter

00:55:21 so it's complicated

00:55:23 on a sensor you have

00:55:25 a photosensitive cell

00:55:27 a little hole that receives light

00:55:29 by color, there is one for green

00:55:31 one for red and one for blue

00:55:33 sometimes there are two of certain colors because you have to do something a little square

00:55:35 so it's called a Bayer filter

00:55:37 and when the camera

00:55:39 processes all the information

00:55:41 it receives, it receives for each pixel

00:55:43 four information

00:55:45 red, green, blue and some

00:55:47 double colors. And the Blackmagic when it receives

00:55:49 this information, it processes the information

00:55:51 saying on this place

00:55:53 on this pixel I have

00:55:55 three color information, I have to bring them together

00:55:57 and depending on how

00:55:59 the Bayer filter is organized

00:56:01 on the sensor

00:56:03 there are sometimes results that are not

00:56:05 particularly good and in particular

00:56:07 there is a part of my glasses here

00:56:09 the edge of my glasses

00:56:11 which when it captures a very

00:56:13 very strong light, and the light was

00:56:15 strong because I had a false sun above me

00:56:17 I had put the LED projector to the maximum to be sure

00:56:19 that it was going to burn a part of my face

00:56:21 it creates a very very big reflection

00:56:23 very very white here

00:56:25 and as it's very very white

00:56:27 it's really almost aggressively white

00:56:29 the debayerization filter bugs a little

00:56:31 and it happens to put some green lines

00:56:33 red, blue, etc.

00:56:35 on some places that are a little too much

00:56:37 and in fact when you make a green background

00:56:39 it makes you have a whole line here

00:56:41 it can blink in a very annoying way

00:56:43 and so in fact the version

00:56:45 that you have there, I spent two days

00:56:47 to encrusted behind

00:56:49 each of my green background renders

00:56:51 a small white layer

00:56:53 which moves following my glasses

00:56:55 to be sure that my glasses are never

00:56:57 transparent and that it never makes a visual bug

00:56:59 on the thing. You really have to realize

00:57:01 that as soon as you do something in special effects

00:57:03 like that, it's always complicated

00:57:05 there are always problems that add up

00:57:07 and in particular there is something, I make a video

00:57:09 that is 1h28, I have in total

00:57:11 on this 1h28, I have 20 minutes

00:57:13 of front camera

00:57:15 which

00:57:17 transcribe in the end

00:57:19 in 300, 400 shots

00:57:21 on my face, something like that

00:57:23 and in fact you have to find methods

00:57:25 of operation that are not those of the cinema

00:57:27 that would allow to have perfect shots

00:57:29 for each thing, etc. But in fact to have perfect shots

00:57:31 for each thing, you have to do each shot independently

00:57:33 of each other. In principle when I go from a big shot

00:57:35 to a medium shot, a wide shot, I would have to have a different way

00:57:37 to treat my encrustation

00:57:39 to be sure

00:57:41 to always manage it

00:57:43 in the best way and it's not doable

00:57:45 with my

00:57:47 quantity of things I have to produce

00:57:49 and my way of editing, my way of doing, etc.

00:57:51 So you have to find

00:57:53 things a little laborious

00:57:55 like for example to shoot everything in 4K

00:57:57 and then shoot to make the wide shots

00:57:59 which was not very practical with this shot

00:58:01 because I was shooting from further than usual

00:58:03 so American or Italian shot

00:58:05 I don't remember, just above the knees

00:58:07 so my face was filmed from closer

00:58:09 so when I shoot the very wide shots on my face

00:58:11 the quality was a little less good

00:58:13 I added a filter after

00:58:15 on the very wide shots

00:58:17 the biggest shots of these front cameras

00:58:19 because otherwise they were blurry, it was too visible

00:58:21 There are things I know I can't do

00:58:23 in encrustation, I have to find

00:58:25 cheap solutions that at the same time

00:58:27 are not too bad

00:58:29 and I think I got away with it

00:58:31 with all of this

00:58:33 it's not too bad

00:58:35 and at the same time it was not too complicated

00:58:37 not too hard to do

00:58:39 To finish on this section

00:58:41 virtual background

00:58:43 which will take up the whole video

00:58:45 the thing is that the program

00:58:47 I used to make this 3D background

00:58:49 Poser

00:58:51 it's not at all made

00:58:53 to do what I did with it

00:58:55 I pushed it but in its clashes

00:58:57 the program is not

00:58:59 at all

00:59:01 made to make scenes

00:59:03 so complex, to make renders

00:59:05 so crazy, to make games with light

00:59:07 things like that

00:59:09 I exploded the program

00:59:11 it's not surprising that it took me so long

00:59:13 to make the video renders

00:59:15 even at Gorkab

00:59:17 it took 2-3 days to make the total render

00:59:19 of the sequence

00:59:21 but the program is not made for that

00:59:23 Poser is made to

00:59:25 prepare the pose of the characters

00:59:27 and to work with humans

00:59:29 it's made to

00:59:31 take human models

00:59:33 put some clothes on them

00:59:35 maybe a rudimentary decor

00:59:37 and that's it

00:59:39 if you go on

00:59:41 websites like

00:59:43 Entai, Yaoi

00:59:45 you may have seen

00:59:47 that they created characters in 3D

00:59:49 it's usually with Poser

00:59:51 or with a new evolution of this program

00:59:53 called Daz 3D

00:59:55 I say more recent but Poser just

00:59:57 released a new version 13

00:59:59 it's also a recent program

01:00:01 it's the same thing

01:00:03 and Daz produced a lot of

01:00:05 3D objects for Poser

01:00:07 in time, I don't know why

01:00:09 they started to make their own 3D program

01:00:11 so yeah, I used a program that is not made

01:00:13 to do what I do

01:00:15 it manages the render cycle of Blender

01:00:17 which is a particular style of rendering

01:00:19 it is able to do subsurface scattering

01:00:21 to make light that goes deep

01:00:23 in an object and not just light

01:00:25 a surface

01:00:27 so it can do that, it can do reflections

01:00:29 it can do godrays

01:00:31 and atmosphere effects

01:00:33 but it's not made to do it

01:00:35 the way I did it

01:00:37 I filled the map with vegetation

01:00:39 I put assets everywhere

01:00:41 I put textures in very high resolution

01:00:43 it's not made to do that

01:00:45 and in my work process

01:00:47 it's obvious

01:00:49 because I didn't know how to organize

01:00:51 the thing to make a pro rendering

01:00:53 in principle, a pro

01:00:55 would have rendered

01:00:57 the background

01:00:59 the sky

01:01:01 a first forest plan

01:01:03 which would go until

01:01:05 everything behind the plants

01:01:07 once you get to the final plan

01:01:09 and then the rest of the forest in front

01:01:11 I didn't know how to do that

01:01:13 but it's a rendering method used by infographers

01:01:15 because if you have a bug in one of the elements

01:01:17 you don't have to change all the elements

01:01:19 you can just correct the one with a bug

01:01:21 and I didn't know how to do that with Poser

01:01:23 it's very complicated

01:01:25 you can ask some objects to appear

01:01:27 or not to the camera

01:01:29 you can ask them to not appear

01:01:31 but to consider their shadows

01:01:33 and their interactions with the rest

01:01:35 so that the raytracing is correctly calculated

01:01:37 even if the object itself is not on the screen

01:01:39 I did a good job

01:01:41 because I had a big flaw

01:01:43 in the first version of the video

01:01:45 we did a first export and there was a bug

01:01:47 and it was so big that we had to redo everything

01:01:49 the big floating rock

01:01:51 on the side of the image

01:01:53 it's just a 2D texture

01:01:55 it's maybe a bit visible

01:01:57 I thought it would be much more visible

01:01:59 in the image and the perspective

01:02:01 would hide all these flaws

01:02:03 but it was still a bit too close

01:02:05 it was really close the first time I did it

01:02:07 and in fact, it was so close

01:02:09 that it projected a shadow

01:02:11 on the forest itself

01:02:13 which means that even if I had

01:02:15 separated the background with the sky and the rock

01:02:17 and then the big tree

01:02:19 and the forest underneath

01:02:21 I would still have to redo it

01:02:23 and I would have to redo it twice

01:02:25 because the thing was so close

01:02:27 that it influenced another part of the frame

01:02:29 that wasn't influenced at all

01:02:31 and it was causing a lot of complaints

01:02:33 so it betrayed the perspective effect

01:02:35 that I was trying to create

01:02:37 so I took it a lot further

01:02:39 and I had to redo the whole image

01:02:41 because of that

01:02:43 we also had a bug with the main tree

01:02:45 which is called a "capok"

01:02:47 there are leaves everywhere

01:02:49 and they were much more realistic

01:02:51 with this thing

01:02:53 so I did the entire rendering at GORECAB

01:02:55 they send me the files, I watch the videos

01:02:57 I think "oh it's cool"

01:02:59 and at some point I stop and I think

01:03:01 "there's a bug on the tree leaves"

01:03:03 there were the tree leaves

01:03:05 and there was a little black thing

01:03:07 around the tree leaves

01:03:09 around each leaf

01:03:11 it wasn't a big black square

01:03:13 a big polygon or something like that

01:03:15 but there was a weird black thing

01:03:17 around the leaves

01:03:19 I thought "that's weird, what's that?"

01:03:21 I want to check in my 3D program

01:03:23 and the thing is that Poser

01:03:25 being not made to do things a bit too elaborate

01:03:27 doesn't know how to handle

01:03:29 the PNG with transparency

01:03:31 and the texture of the leaves

01:03:33 the guy who created the tree

01:03:35 he did it on a program that knew how to handle that

01:03:37 so it was a PNG with transparency

01:03:39 so the program was supposed to create transparency

01:03:41 but since my program doesn't know how to handle that

01:03:43 so there was a black line

01:03:45 around each leaf

01:03:47 which should have been removed by transparency

01:03:49 but was added by the program

01:03:51 because it doesn't know how to handle transparency

01:03:53 so I had to also re-render

01:03:55 the whole tree thing

01:03:57 because all the leaves were disgusting

01:03:59 and I had to remove them

01:04:01 holy shit

01:04:03 it took 3 or 4 more days

01:04:05 to render them

01:04:07 that's why the video was so slow

01:04:09 because of that

01:04:11 it took 2 or 3 weeks to render

01:04:13 but the rendering itself was very elaborate

01:04:15 so yeah

01:04:17 it was an adventure

01:04:19 this first shot

01:04:21 knowing that

01:04:23 even the camera movement

01:04:25 is a bit more complex

01:04:27 than what you see here

01:04:29 because in fact

01:04:31 3D cameras often have

01:04:33 very very very

01:04:35 boring ways to handle their movement curves

01:04:37 so in fact

01:04:39 you put the camera somewhere here

01:04:41 you ask it to go down here

01:04:43 it will move the camera

01:04:45 very slowly

01:04:47 and bring it here

01:04:49 if once it's here you say

01:04:51 I want you to slow down here

01:04:53 from this point to that point

01:04:55 it's not worth trying

01:04:57 the 3D program doesn't know how to do that

01:04:59 it doesn't know how to create a curve

01:05:01 at least the one I use doesn't know how to do it

01:05:03 maybe Blender or something like that can do it

01:05:05 but I can't

01:05:07 so the thing that could have ruined

01:05:09 the realistic side of this shot

01:05:11 is the camera having a very virtual movement

01:05:13 so

01:05:15 we had to work on the end of the movement

01:05:17 so that the movement slows down

01:05:19 progressively and not stop abruptly

01:05:21 or suddenly goes from

01:05:23 20 km/h to 10 km/h

01:05:25 and it feels like a speed change

01:05:27 at first I just made a movement

01:05:29 the camera starts here and it arrives here

01:05:31 at first it had to stay in front

01:05:33 and then go down in the foliage

01:05:35 but I realized

01:05:37 that the tree was disappearing

01:05:39 very quickly, the big central tree

01:05:41 and I wanted to focus on it

01:05:43 because it was the thing that gave the whole

01:05:45 grandiose aspect of the shot

01:05:47 so I decided to keep the same movement

01:05:49 but I made a panoramic

01:05:51 from the camera to the tree

01:05:53 and when it gets closer it goes down

01:05:55 until it gets to the position I had in mind

01:05:57 but at the end

01:05:59 my editing program takes over

01:06:01 and takes the last parts

01:06:03 when it leaves the foliage

01:06:05 because if I was in the foliage it would have made ugly artifacts everywhere

01:06:07 so once I leave the foliage

01:06:09 I get the reading curve

01:06:11 of my video

01:06:13 and I slowly bring it to a gradient

01:06:15 but even there

01:06:17 it's a pain

01:06:19 because you don't know why

01:06:21 it's supposed to be one of the most common effects

01:06:23 to do

01:06:25 not the most simple, but the most common

01:06:27 effects that most people want to do

01:06:29 a quiet stop

01:06:31 but for a reason or another

01:06:33 it's a pain to do that in editing programs

01:06:35 at least they can understand

01:06:37 the more you get closer

01:06:39 to this key frame

01:06:41 to this key point of the movement

01:06:43 the faster you have to go

01:06:45 or the more you have to

01:06:47 reduce the space between the different steps

01:06:49 the editing program is able to understand that

01:06:51 but

01:06:53 it's going to be very laborious

01:06:55 because you could tell it

01:06:57 from here you start to slow down

01:06:59 from there I want it to be this thing

01:07:01 but no, it's a pain too

01:07:03 you make a point

01:07:05 and your point creates a curve

01:07:07 and to manage this curve

01:07:09 you have a big bar

01:07:11 and you have to manage

01:07:13 the size and the inclination

01:07:15 and that changes the movement curve

01:07:17 and to manage to do a good slow down

01:07:19 with just the images that slow down more and more

01:07:21 as you read

01:07:23 but it's a pain

01:07:25 I don't understand why it's still a pain

01:07:27 to do slow down effects

01:07:29 of game on speed

01:07:31 why can't you just

01:07:33 put a key frame at this frame

01:07:35 and another at this point

01:07:37 and say I want it to be managed like that

01:07:39 no it's not like that

01:07:41 on Premiere it's

01:07:43 a precise key frame

01:07:45 to start to change the time

01:07:47 and at this point it creates

01:07:49 a marker like that

01:07:51 not really a heart, it's more like that

01:07:53 a marker like that

01:07:55 and you can move both parts of the marker

01:07:57 you asked to create it here

01:07:59 but you can put the beginning more to there

01:08:01 and the end more to here

01:08:03 and you can only do that

01:08:05 you can only stretch this marker

01:08:07 and when you reach a certain distance

01:08:09 the program refuses to continue to stretch

01:08:11 and you have to do a lot of...

01:08:13 it's a pain, it's a pain

01:08:15 I hated it, I think I had more trouble

01:08:17 to do the slow and progressive stop

01:08:19 of the camera at the end of the shot

01:08:21 than the shot itself

01:08:23 because honestly it was boring

01:08:25 to do the 3D virtual shot

01:08:27 but it wasn't a pain

01:08:29 when I wanted to do something I knew how to do it

01:08:31 when I wanted to shoot an object I shoot it

01:08:33 depending on the axis, the size, the thing

01:08:35 it's quite easy and direct

01:08:37 but there, managing to negotiate

01:08:39 with the editing program

01:08:41 to make the progressive stop

01:08:43 of the video's reading

01:08:45 as you want it, it was a pain

01:08:47 and I think

01:08:49 that there are two slowdown effects

01:08:51 I'll check my...

01:08:53 yes, I remember, there are two slowdown effects

01:08:55 there is a first one

01:08:57 that already reached the limits

01:08:59 of slowdown with which the program

01:09:01 could read the original video

01:09:03 and add images in the meantime

01:09:05 an image interpolation

01:09:07 between the key images

01:09:09 that already exist and the one it has to create

01:09:11 and so I had to do

01:09:13 this first slowdown

01:09:15 and above this slowdown I add

01:09:17 a second slowdown of the program

01:09:19 that allowed to specifically add

01:09:21 a slowdown on the end of the video

01:09:23 [Slowdown]

01:09:41 So there you go, I just took

01:09:43 more than 1h20

01:09:45 I think, just to sum up

01:09:47 how I made this shot, so it's going to be a bit complicated

01:09:49 for the rest of the video, we'll continue

01:09:51 on the technical anecdotes

01:09:53 and writing, and we'll

01:09:55 be able to start

01:09:57 to get into the heart of the matter. Oh yes, last detail

01:09:59 that I forgot, there is

01:10:01 a depth effect that has to be created, and to help

01:10:03 the depth effect and also the movement effect

01:10:05 because the thing is that the 3D program

01:10:07 that I use is not made

01:10:09 to do what I'm doing here

01:10:11 and so

01:10:13 I would have liked the vegetation to move

01:10:15 but it's not possible

01:10:17 with the program I have, it's not able

01:10:19 to create this kind of movement

01:10:21 with the wind or

01:10:23 things like that. It can do it

01:10:25 there is a whole thing, a simulation

01:10:27 of fabric, and there is

01:10:29 a force simulation

01:10:31 wind, a wind tunnel

01:10:33 that can be used by the poser, it is able

01:10:35 to do this kind of thing, but it can't

01:10:37 because all these assets are made for

01:10:39 video games, so normally they have a movement

01:10:41 that is included in the thing, they are supposed to swing

01:10:43 a little bit with the wind or something like that, or be able to do it

01:10:45 and the thing is that to do it

01:10:47 I would have to transform

01:10:49 all the plants

01:10:51 that are on this thing, in a simulation

01:10:53 of fabric or clothing, I find

01:10:55 a way to create a wind tunnel and I render the thing

01:10:57 and the thing is that if it moves

01:10:59 it means that you have to

01:11:01 calculate the image and render it

01:11:03 on several seconds to try to have

01:11:05 a loop that is not seen too much

01:11:07 that is repeated a little easily in the thing with the movements

01:11:09 etc. Which means that it would have added

01:11:11 at least one or two weeks of rendering time

01:11:13 to render the moving vegetation

01:11:15 and in the end I thought it was not necessary

01:11:17 Nature is much more

01:11:19 fixed than we think, nature moves

01:11:21 much less than we think in reality

01:11:23 if you go in nature, there will be very few things

01:11:25 that will move unless there is a big wind

01:11:27 but if there is no wind, there is nothing that moves

01:11:29 and I thought to still add movement

01:11:31 there will be me, already at the base

01:11:33 but I will add a lot of particles in the air

01:11:35 small dusts

01:11:37 small things etc. that will move

01:11:39 that will give a feeling of movement and that will be visible

01:11:41 and that will also help the depth of field

01:11:43 because you will notice that the more we

01:11:45 oh yes, it's a small detail too

01:11:47 at each

01:11:49 zoom of the image, of the plane value

01:11:51 I modified the background

01:11:53 the background and the foreground

01:11:55 so that it is more blurry

01:11:57 with a real blur of camera lens

01:11:59 and by the way, you will notice

01:12:01 it's very subtle, I try not to do something ultra blurry

01:12:03 because it's the kind of thing that is ugly and that makes it look like it's not

01:12:05 that it's fake, it has to be quite subtle

01:12:07 but suddenly, the closer we get to my face

01:12:09 the more the depth of field is restricted

01:12:11 and so the more the background and the foreground are blurry

01:12:13 and to help this effect of depth

01:12:15 and movement, there are also these dusts

01:12:17 that move, these small particles that move in the atmosphere

01:12:19 which help to have

01:12:21 something much more dynamic

01:12:23 and realistic.

01:12:25 With that, normally I have

01:12:27 all the anecdotes of this virtual background

01:12:29 In my green background, the way to embed the thing

01:12:37 I had planned to play with the background

01:12:39 and the fact that I couldn't get to the right place

01:12:41 I didn't want to get up because I didn't want to just look classy

01:12:43 and do the thing

01:12:45 like "it's too cool, I live in the jungle"

01:12:47 I wanted to have the side

01:12:49 I just crawled in the jungle, I just tried to pass the bushes

01:12:51 to get where I wanted

01:12:53 What was added was the virtual background sketch

01:12:55 I showed this video

01:12:57 at the beginning

01:12:59 without the camera faces, there was already the rendering of the jungle

01:13:01 but there was no me embedded in it

01:13:03 and I showed this video like that to Gorkab

01:13:05 and to Arkady

01:13:07 and when I showed it to Arkady

01:13:09 he told me "you know, you should put something in your video

01:13:11 to show that you did the background

01:13:13 because I'm sure that most people

01:13:15 won't even notice"

01:13:17 and the thing is

01:13:19 the job of a YouTuber

01:13:21 has changed a lot

01:13:23 having YouTubers who have editors

01:13:25 sometimes an army of editors

01:13:27 who work for them to do the things, the clothes, etc

01:13:29 that's not how YouTube was thought

01:13:31 in principle

01:13:33 we had to do that

01:13:35 because of the rhythm imposed by YouTube

01:13:37 there's a moment when you have to upload so many videos

01:13:39 and do so many things

01:13:41 that you need to delegate work

01:13:43 that's what almost killed me, I don't delegate my work

01:13:45 because I'm a former YouTuber

01:13:47 and it's true that

01:13:49 because of this habit on YouTube

01:13:51 it would have been very easy for people to get on my video

01:13:53 see that and think

01:13:55 "oh that's good, someone did a virtual background

01:13:57 or you went to see an infographic box to do things"

01:13:59 and he said

01:14:01 "you have to find a way for people to realize

01:14:03 that you did this background and you did it alone and you fucked up"

01:14:05 and I said

01:14:07 "oh yeah, that would have pissed me off

01:14:09 to have comments in the video saying

01:14:11 "you found a beautiful background on the internet"

01:14:13 no no, I was pissed off to create everything alone, shit

01:14:15 and so that's when

01:14:17 he told me

01:14:19 he didn't tell me how to do it, but he told me

01:14:21 "you have to find a way to say it"

01:14:23 and so at that moment I had the idea of

01:14:25 "hey, that's not bad" and so the big huge bug

01:14:27 that makes everything mess up

01:14:29 that makes you see that I have the wireframe

01:14:31 to control all these backgrounds and that I'm the one controlling it

01:14:33 I'm the one who has access to all these things

01:14:35 and I have access to

01:14:37 the pre-rendered, to everything that exists before it exists

01:14:39 and so that was the idea of this sketch

01:14:41 to show that

01:14:43 I did it

01:14:45 a little sketch that was found like that

01:14:47 thanks to Arcadi, there are little anecdotes like that

01:14:49 that will come back, especially on one of the best sketches

01:14:51 of why I did it wrong

01:14:53 we'll come back to it in time

01:14:55 when I typed on the thing

01:14:57 I knew there would be this plant around me

01:14:59 and I didn't know where, and I thought

01:15:01 if I try to aim at a specific place

01:15:03 I'll make 17 takes to always hit the empty space

01:15:05 at the wrong place, etc.

01:15:07 so I made sure to have a good shot

01:15:09 where I was hitting correctly

01:15:11 I was sure that my point was where it had to be

01:15:13 in front of the desk and all that

01:15:15 the work plan

01:15:17 and in fact, what you see on the screen

01:15:19 it jumps, you don't even notice it

01:15:21 but when I do that

01:15:23 I was supposed to lean

01:15:25 way forward to hit the plant

01:15:27 because I thought it would be lower

01:15:29 and in fact it was way higher than that

01:15:31 when I lower the point and it should be there

01:15:33 I lift the whole green background layer

01:15:35 and so

01:15:37 my face that was going down

01:15:39 following the point, stays at the same place

01:15:41 from your point of view

01:15:43 and my point gives the impression

01:15:45 of hitting way higher, whereas before it was hitting way lower

01:15:47 and so

01:15:49 that's how I saved this shot

01:15:51 because otherwise the shot wouldn't stick at all

01:15:53 I had 5 takes, none of them stuck

01:15:55 and I had to do this little trick

01:15:57 on this thing

01:15:59 so here we have a first edit like this

01:16:01 with a poster and a background

01:16:03 and you'll notice that the background

01:16:05 I told you I put a blur of the camera lens

01:16:07 but I didn't just put a blur of the camera lens

01:16:09 I put a blur of the camera lens

01:16:11 anamorphic

01:16:13 and that, in principle, no program

01:16:15 would allow me to do

01:16:17 it's just that I found a trick

01:16:19 because I know how anamorphic works

01:16:21 and so, knowing how the plugin works

01:16:23 I say, if I modify this

01:16:25 this way and then I do this

01:16:27 it's going to make anamorphic blur

01:16:29 and it's a stupid thing that makes

01:16:31 this video look better

01:16:33 than the videos I made before

01:16:35 because simply the blur rendering is much more

01:16:37 aesthetic and pretty

01:16:39 than in all the videos I made before

01:16:41 where it was just blur

01:16:43 it took me some time to convince you it was shit

01:16:45 but I got there and I can hear the explosion again

01:16:47 and so that's typically

01:16:49 the kind of anecdote

01:16:51 that I noticed

01:16:53 that there was a big problem at that moment

01:16:55 because

01:16:57 I had written this text

01:16:59 with, at the beginning, I'm living a Prometheus effect

01:17:01 period

01:17:03 it took time to convince people that Prometheus was shit

01:17:05 but I got there and 10 years later

01:17:07 I can hear the explosion again

01:17:09 and in fact I left this text like that

01:17:11 during almost all the work

01:17:13 of the video and when I shot the image

01:17:15 I said to myself, but in fact

01:17:17 Prometheus was a little too specific, Star Wars 7

01:17:19 it was specific to me, I mean

01:17:21 to my channel, Prometheus, Star Wars 7

01:17:23 more people are aware of the problems there are

01:17:25 of repetition and other problems

01:17:27 so I had to evoke Star Wars 7 more

01:17:29 but I still wanted to evoke Prometheus because it was my channel

01:17:31 and this movie was a big disappointment

01:17:33 so there was a lot of allusions to why

01:17:35 I was right and so on

01:17:37 and so during the shooting, I shot

01:17:39 I'm living a Prometheus effect or Star Wars 7

01:17:41 but then after, the rest of the sentence

01:17:45 and 10 years later I can hear the explosion again

01:17:47 it doesn't stick anymore

01:17:49 because my video on Star Wars 7

01:17:51 wasn't released 10 years ago

01:17:53 and so there's something wrong

01:17:55 and I didn't realize it during the shooting

01:17:57 and so

01:17:59 I had to find a way

01:18:01 during editing

01:18:03 to make it disappear 10 years later

01:18:05 and to transform the sentence

01:18:07 into "and I can hear the explosion again"

01:18:09 in general, to avoid

01:18:11 a temporary reference

01:18:13 I got in deep shit all by myself like an idiot

01:18:15 and so 10 years later, I should have known

01:18:17 that I had to remove it during the shooting

01:18:19 but you don't realize it sometimes

01:18:21 and so if you look closely

01:18:23 my lip movements don't match what I'm saying

01:18:25 because the sentence I was saying at the beginning

01:18:27 was "and 10 years later, I can hear the explosion again"

01:18:29 and finally it became

01:18:31 "and I can hear the explosion again"

01:18:33 but I got there

01:18:35 and 10 years later, I can hear the explosion again

01:18:37 but I got there

01:18:39 and I can hear the explosion again

01:18:41 and I managed to get through

01:18:43 but if you really

01:18:45 pay attention to this cut

01:18:47 I'm sure you'll realize that there's a big problem

01:18:49 that my lips make sounds

01:18:51 that are not pronounced

01:18:53 "and I can hear

01:18:55 the explosion again"

01:18:59 How can you correct a sentence

01:19:01 like that during the shooting

01:19:03 when you've been careful to correct your text for 2 months

01:19:05 but at the last moment you realize you've missed something

01:19:07 and it's simply because

01:19:09 I think I only did one take where I said "Prometheus" or "Star Wars 7"

01:19:11 and I didn't adapt this take

01:19:13 but all my other takes

01:19:15 I said "an effect Prometheus"

01:19:17 so it was normal to say "and 10 years later, I can hear the explosion again"

01:19:19 because I only talked about each take

01:19:21 with a Prometheus effect that was done 10 years ago

01:19:23 but it's just the fact of having added

01:19:25 on one take "Star Wars 7" and I wanted to add it

01:19:27 because it makes more sense and it's necessary for the rest

01:19:29 but it suddenly made the sentence look like an ending

01:19:31 and I had to find a visual trick to correct it

01:19:33 and in fact

01:19:35 there were 7 takes and there was only one

01:19:37 where the movements of my face

01:19:39 when I start to say the sentence

01:19:41 could potentially

01:19:43 match the ending of the sentence

01:19:45 when I finished the take

01:19:47 because otherwise I would close my eyes too much during the take

01:19:49 and when I started the take with my eyes open

01:19:51 and I continued with my eyes closed

01:19:53 it was a pain, it was a nightmare

01:19:55 but I think it's ok, I think you all missed it

01:19:57 normally, I didn't get any comment on that

01:19:59 and by the way, I forgot to say

01:20:01 and I can hear the beginning of the sentence

01:20:03 you don't know but

01:20:05 there are a lot of sentences like that

01:20:07 where I re-record some parts of the sentence

01:20:09 and typically, it's a camera face

01:20:11 and I didn't have a camera face where I could hear the beginning of the sentence

01:20:13 so I had to record separately

01:20:15 and hear the beginning of the sentence

01:20:17 instead of "and 10 years later I can hear the beginning of the sentence"

01:20:19 and I put this new "and I can hear the beginning of the sentence"

01:20:21 on the original sentence

01:20:23 and that's why the lip sync doesn't match

01:20:25 it doesn't match what I'm saying

01:20:27 I recorded it separately, this part of the sentence

01:20:29 but just the beginning of the sentence

01:20:31 and I can hear it well and it's with the "well" that I make a transition to the original sentence

01:20:33 and I can hear the beginning of the sentence

01:20:35 there are a lot of tips like that

01:20:37 to save sentences or small parts

01:20:39 in the voice-overs generally but sometimes also in the face-cam

01:20:41 I had planned my set

01:20:43 to have space on the sides

01:20:45 and to play with the fact that

01:20:47 the fern was going to pass in front of a lot of things

01:20:49 I wanted it to be included in the set

01:20:51 I didn't go all the way

01:20:53 in my delirium because if I really had been

01:20:55 all the way, all the way, all the way

01:20:57 really where I wanted to go

01:20:59 I would have embedded the poster

01:21:01 of Terminator Dark Fate there

01:21:03 or other elements like articles that I put there at some point

01:21:05 really in the 3D set

01:21:07 I would have launched a new 3D rendering on my program

01:21:09 to have at least one image

01:21:11 with the thing perfectly included in the set

01:21:13 with the same mist above

01:21:15 and maybe even a shadow projected on the fern in front

01:21:17 because as the sun is a little behind

01:21:19 it would have probably put a shadow

01:21:21 a little on the fern

01:21:23 and basically that's why the fern is here

01:21:25 because I wanted a shadow to be projected on the fern

01:21:27 when I put new objects or things like that

01:21:29 but it was a step above my means

01:21:31 I need a PC that allows me to easily render

01:21:33 this shot and it's not easy to do

01:21:35 So, I didn't go all the way

01:21:39 However

01:21:41 I made a big delirium on this

01:21:43 I wanted it to be a fresh video

01:21:45 and a little different, I wanted to talk about the backgrounds

01:21:47 when I put the posters of the movies, etc

01:21:49 There is something I wanted to do fresh with this one

01:21:51 I wanted to make a little animation, a little funny and a little cartoon

01:21:53 but that would be entirely made by me

01:21:55 that would not be a thing, everything prepared

01:21:57 a plugin, a thing like that

01:21:59 I've never done anything like that before

01:22:01 I mean, you know

01:22:03 when a poster arrives in the frame

01:22:05 I have a plugin that automatically

01:22:07 transitions between the poster that is there

01:22:09 that arrives in the middle of the frame and then goes to the other side

01:22:11 when I make the images, the thumbnails of my vlogs

01:22:13 it's the same plugin

01:22:15 I wanted to do something that animated the thing

01:22:17 but really animated it in a comic way

01:22:19 cartoonish and with the limit

01:22:21 I wanted to force myself the limit

01:22:23 of a 4-corner animation

01:22:25 There are several ways to animate a square like that

01:22:27 there is a 4-corner animation

01:22:29 and an 8-corner animation

01:22:31 4-corner is just the 4 sides of the cube

01:22:33 and 8-corner adds a little to the middle

01:22:35 and it allows you to make curves, to make things a little more deformed

01:22:37 it's not a very pretty curve, but it allows you to make things a little more deformed

01:22:39 I told myself

01:22:41 I want to find a way to make a comic animation

01:22:43 on this poster that runs away when I write on it

01:22:45 just with the 4-corner animation

01:22:47 and that's what I did

01:22:49 it's a 4-corner animation, there are the feet that are a little stable

01:22:51 in each thing

01:22:53 there are the sides of the top that make

01:22:55 the hands of the person screaming

01:22:57 like "AAAAAAA"

01:22:59 and that's it, you have the 4 corners that are moving like that

01:23:01 I had to animate the fact that

01:23:03 it's animated keyframe by keyframe

01:23:05 so I move forward

01:23:07 of a frame in my timeline

01:23:09 I change the thing, I move forward

01:23:11 I change, I change, I move forward

01:23:13 and so, at each move, you had to take into account that

01:23:15 the more he screams, the more he had to be

01:23:17 stretched on the top

01:23:19 before coming back below

01:23:21 and then there are the different bounces

01:23:23 with the bottom that stretches a little more

01:23:25 and the top that stays pretty much normal

01:23:27 that then goes up and does something

01:23:29 and then it stretches, so we arrive on the top

01:23:31 that will fall back on the ground by stretching again

01:23:33 and all that is animated by hand

01:23:35 and with the final little key

01:23:37 which is the motion blur that I added to the thing

01:23:39 with a plugin that I had not heard of for a long time

01:23:41 called RealSmart Motion Blur

01:23:43 and it's

01:23:45 the best motion blur plugin

01:23:47 I've ever seen in my life, it does everything

01:23:49 ultra quickly, it's calculated but in real time

01:23:51 with a GPU or CPU assistance

01:23:53 and

01:23:55 it's a very, very, very, very good quality motion blur

01:23:59 it's very used by a lot of people

01:24:01 who do let's play, video game stuff

01:24:03 and who want to give a little dynamic and fun side

01:24:05 to a video game, they will add a lot of motion blur

01:24:07 but on the whole thing, while the game has no motion blur at the base

01:24:09 just because it makes a little accelerated effect

01:24:11 or a little nice thing at times

01:24:13 and I think it looks great

01:24:15 I also used it for the foreground, oh yes, because I completely forgot to say it

01:24:17 the motion blur is

01:24:19 extremely expensive to render

01:24:21 in a 3D program, so the program

01:24:23 the foreground was rendered without motion blur

01:24:25 all the motion blur

01:24:27 all the movement blur

01:24:29 that you have on the foreground was added in post processing

01:24:31 and even with a particular thing

01:24:33 because at some point the camera goes through the leaves

01:24:35 and I had to distinguish the foreground

01:24:37 which were the different leaves that go through the camera

01:24:39 and the background which is the lighthouse

01:24:41 that we will end up arriving at

01:24:43 because otherwise the program did not understand

01:24:45 and everything was in motion blur

01:24:47 in motion blur in the same way

01:24:49 and you could not distinguish the leaves from the rest of the thing

01:24:51 and it created a disgusting effect

01:24:53 and it creates artifacts and defects in the movement of the leaves

01:24:55 because it moved so fast with so much movement

01:24:57 that it could have noticed the plugin

01:24:59 so I had to help him with a little visual guide

01:25:01 which told him where the leaves were

01:25:03 that were passing in front of the image

01:25:05 compared to the rest of the background

01:25:07 but here are just these two little plugins

01:25:09 the blur plugin

01:25:11 and the motion blur plugin

01:25:13 these two little things

01:25:15 that I wanted to have for years

01:25:17 but I never bought because I had to have time

01:25:19 to study the question a little

01:25:21 to test the plugin to see if it worked correctly

01:25:23 and if I really wanted to buy it

01:25:25 it still costs more than 100 euros each

01:25:27 it's the kind of complications you do not want to put yourself

01:25:29 and there I said to myself for once I have time

01:25:31 it's something I'm going to do now

01:25:33 and it will improve the quality of my videos

01:25:35 on the 20 years to come

01:25:37 it was an investment to make

01:25:39 and I think it looks very good

01:25:41 this little crazy with the poster that escapes

01:25:43 on the other hand no one told me in the comments

01:25:45 it's a little thing in the comments

01:25:47 the comments that make me the happiest

01:25:49 it's the comments with a timecode that redirects me to the joke that made people laugh

01:25:51 and suddenly generally when I make jokes

01:25:53 I wait to see if someone mentioned it

01:25:55 and this one I think no one mentioned it anywhere

01:25:57 there are plenty that people have mentioned I wanted to say as and when

01:25:59 but this one no one talked about it and yet I'm still super happy

01:26:01 I'm really super good

01:26:03 I just prefer my animation

01:26:05 the film starts with a shot in the mist of the Pandorian jungle

01:26:07 oh sorry

01:26:09 excuse me I'm Buenay I put the beginning of the first movie

01:26:11 I wonder how I could have been wrong

01:26:13 so this replica

01:26:15 it was very very hard to re-record

01:26:17 there are a lot of things that are complicated to record

01:26:19 I will tell you about it a little further

01:26:21 but when I record the text at the base

01:26:23 I read everything at once

01:26:25 the face cameras, the voice-over and all the mess

01:26:27 I just read my text in full

01:26:29 and then it's at the editing that I finish

01:26:31 what is a voice-over, what is face camera

01:26:33 sometimes there are things that were planned voice-over that go face camera

01:26:35 etc.

01:26:37 the fact is that suddenly I do a lot of editing

01:26:39 and so I'm in the same rhythm

01:26:41 from start to finish

01:26:43 I do all the text on a single throw

01:26:45 and I know where I come from

01:26:47 and so in fact this replica was very complicated to do

01:26:49 because it is fundamentally made to be in the flow of the story

01:26:51 we have to feel

01:26:53 that it is the progression of the same sentence

01:26:55 there must not be a break between the two

01:26:57 and in fact when I re-recorded it

01:26:59 it was very complicated

01:27:01 to succeed in recovering this rhythm

01:27:03 from the base of the sentence

01:27:05 and succeed in keeping it and expressing it correctly

01:27:07 because you're supposed to play an emotion

01:27:09 at the base when you do this thing

01:27:11 that's what's complicated with this replica

01:27:13 I'm supposed to pretend to laugh

01:27:15 and even if the laugh is false

01:27:17 it was very complicated to laugh correctly

01:27:19 I'm going to put some failed scenes like that

01:27:21 Oh sorry, excuse me, I'm Benet

01:27:23 I launched the first film

01:27:25 I wonder how I could be wrong

01:27:27 it's weird, why can't I do it again?

01:27:29 from the Pandorian jungle

01:27:31 Oh sorry, excuse me, I'm Benet

01:27:33 I put the beginning of the first film

01:27:35 Oh sorry, excuse me, I'm Benet

01:27:37 I put the beginning of the first film

01:27:39 I wonder how I could be wrong

01:27:41 it didn't give the impression

01:27:43 to be in the sequence of the sentence

01:27:45 it didn't give the impression

01:27:47 to be the joker side

01:27:49 that I wanted for this replica

01:27:51 and it was very complicated to succeed in reproducing it

01:27:53 there are things that I didn't succeed in reproducing at all

01:27:55 and that are from my original voice-over

01:27:57 I'll show you when we get there

01:27:59 and there typically I absolutely wanted to get the sentence out

01:28:01 on this rhythm, I wanted the "I wonder how I could be wrong"

01:28:03 I wanted the "I wonder how I could be wrong"

01:28:05 and every time I read the sentence in my text

01:28:07 or on my timeline when I heard it

01:28:09 I saw myself doing that

01:28:11 and it was very complicated

01:28:13 to find the sentence that allowed me to finish

01:28:15 "I wonder how I could be wrong"

01:28:17 how do you get to this thing first?

01:28:19 anyway, it was complicated to play

01:28:21 so that's something I find very important

01:28:23 in terms of staging

01:28:25 it was a decision I knew I was going to make

01:28:27 from the text, I had marked it in my text

01:28:29 to make split-screen everywhere

01:28:31 it showed that there are things that are exactly the same

01:28:33 especially from the first shot of the movie

01:28:35 there is this main tree

01:28:37 it's exactly the same tree model

01:28:39 from one movie to the other

01:28:41 and we can see that the camera

01:28:43 passes above exactly at the same time

01:28:45 in both shots, etc.

01:28:47 but if you don't put it side by side

01:28:49 you can't realize it

01:28:51 the tree is so lost in the mist that it's complicated

01:28:53 there we really realized it was the same tree

01:28:55 and that there was a big comparison between the things

01:28:57 and really, in general

01:28:59 split-screen is really important to me

01:29:01 because you realize that

01:29:03 it's not just the same scenes

01:29:05 it's the same gestures

01:29:07 when we train the oak to breathe

01:29:09 and you have Tyria

01:29:11 who teaches him

01:29:13 to press on the chest to show him where to breathe

01:29:15 it's exactly the same

01:29:17 hand positions and the same

01:29:19 movement style as what Nettiri does

01:29:21 with Jake in the first movie

01:29:23 where she also touches his chest to tell him

01:29:25 where to breathe, how to hold on

01:29:27 how to pull the bow, etc

01:29:29 it's not just the same scene spirit

01:29:31 it's exactly the same movements

01:29:33 it goes that far

01:29:35 so for me it was really important to do split-screen

01:29:37 from the beginning to the end

01:29:39 and I'm happy because nobody told me it was unreadable

01:29:41 I think there is one scene where there are 4

01:29:43 that should be a little unreadable when I compare

01:29:45 Terminator 1 and 2 and Avatar 1 and 2

01:29:47 and I see that on the timeline there are a lot of people who read this scene

01:29:49 so this one should be a little more complicated to follow

01:29:51 I totally agree

01:29:53 but it's still the best way to show it

01:29:55 I just talked about it on Terminator

01:29:57 by just showing the image in full screen and transitioning between scenes

01:29:59 the editing would have been incomprehensible

01:30:01 here, typically

01:30:03 the parallel between Jake and Tua Rich's waking scenes

01:30:05 it's ultra-parallel when you put them face to face

01:30:07 you have exactly the same shot

01:30:09 that starts on the light of the operating table

01:30:11 seen in a subjective shot

01:30:13 it ends with a moment

01:30:15 where the avatar

01:30:17 sticks to a window

01:30:19 so you put the two shots side by side

01:30:21 it's exactly the same

01:30:23 as the mirror in the room

01:30:25 where you put your hand and you adore yourself

01:30:27 because you see your reflection in the mirror

01:30:29 and it means something about you that comes out of the background

01:30:31 that's why it was important to put them in full screen

01:30:33 Why did you write your reboot so badly

01:30:35 when you had everything

01:30:37 under your hand to do it well, Jimmy?

01:30:39 Why?

01:30:41 Why?

01:30:43 Why?

01:30:45 So, the why...

01:30:47 I have to explain you the why

01:30:49 because a lot of people don't know

01:30:51 where it comes from

01:30:53 First, I'm really happy because this sketch went really well

01:30:55 and I was really afraid it wouldn't

01:30:57 because it's a running gag, it's a sketch that's repeated

01:30:59 six times, I think, in the video

01:31:01 and I thought that after a while

01:31:03 some people would get angry

01:31:05 and indeed, there are always people

01:31:07 who get angry when it's too much

01:31:09 when it comes back too often

01:31:11 I managed to make my reference to the why work

01:31:13 So, first, where does it come from?

01:31:15 It comes from a musical comedy

01:31:17 a rock opera, I think

01:31:19 called "Jesus Christ Superstar"

01:31:21 which is a reinterpretation

01:31:23 of Jesus Christ's life in a musical comedy

01:31:25 and there's a whole song

01:31:27 called "Get the Man"

01:31:29 in which Jesus Christ

01:31:31 speaks to God and asks

01:31:33 God to justify him

01:31:35 why he should die for the sins of humanity

01:31:37 I have to know, have to know my Lord

01:31:39 I have to know, have to know my Lord

01:31:41 Why should I die?

01:31:43 And this song was written

01:31:51 since we're talking about God

01:31:53 with the idea of adding

01:31:55 a little thing in particular

01:31:57 there's a note, one of the most difficult

01:31:59 to reach for a male singer

01:32:01 from what I understood, basically

01:32:03 you can reach it or you can't reach it

01:32:05 your vocal cords are made to reach

01:32:07 as high as you can, or not

01:32:09 you can't train yourself to reach it, etc

01:32:11 so, it's already reserved for some people

01:32:13 I think it's a G

01:32:15 in the American notation, which would be equivalent to a G

01:32:17 in France, in European notation

01:32:19 but I'm not 100% sure

01:32:23 about the notes and stuff like that

01:32:25 I know it's a note

01:32:27 that's very difficult to reach

01:32:29 and so, that's the challenge of this song

01:32:31 Jesus asks God

01:32:33 to explain to him why he should die

01:32:35 and so, he arrives and says

01:32:37 I have to know, I have to know my Lord

01:32:39 Why should I die?

01:32:41 and so, they have to hold

01:32:43 a very high note, almost impossible to play

01:32:45 to do the

01:32:47 Why should I die?

01:32:49 etc

01:32:51 Why should I die?

01:32:53 and so, I just got the "why"

01:32:55 because this way of shouting "why" like that

01:32:57 is a bit what I felt in front of the movie

01:32:59 but why did you do such a bad thing?

01:33:01 You have everything under your hand to do a good job

01:33:03 You're James Cameron

01:33:05 Why did you do such a bad thing?

01:33:07 and so, the "why" aspect, I really wanted to bring it out

01:33:09 So, how did I discover this song?

01:33:11 It's thanks to Lindsay Ellis

01:33:13 who is a very good YouTuber

01:33:15 who stopped YouTube now because she ate too much

01:33:17 scandals

01:33:19 I think she got cancelled once too much

01:33:21 she did some things at some point

01:33:23 that are not very appreciated by

01:33:25 some communities, but still

01:33:27 she is very tolerant and open

01:33:29 and so, she left YouTube

01:33:31 but before leaving YouTube, she did a lot of good videos

01:33:33 she did a very good video on The Hobbit by the way

01:33:35 she is a big fan of musicals and she did a lot of videos on musicals

01:33:37 and she has on her channel

01:33:39 an editing called "Why"

01:33:41 and it's an editing of

01:33:43 all the versions she could find at the time

01:33:45 of this song "Get the Man"

01:33:47 related to the one of Lin-Manuel Miranda

01:33:49 who was starting to get known at that time

01:33:51 with the thing on Hamilton

01:33:53 he had done "Get the Man"

01:33:55 once in a presentation for something

01:33:57 I don't know what it was

01:33:59 it was a school thing or something

01:34:01 it looks like a little noise, it's filmed on a VHS

01:34:03 or something, it's disgusting

01:34:05 but indeed, Lin-Manuel Miranda

01:34:07 is not made to reach this note

01:34:09 and so, when he does the "Why"

01:34:15 it's a bit ridiculous

01:34:17 and so, she has a whole editing of the "Why should I die"

01:34:19 with a lot of singers who do it

01:34:21 and him who is bad

01:34:23 and so, I thought

01:34:25 I'm going to reuse this sketch

01:34:27 and I'm going to use it to make

01:34:29 the redundancy more funny

01:34:31 in this list, there are a lot of people

01:34:33 who can reach the note perfectly

01:34:35 and there are a lot of them who are not exactly

01:34:37 at the level

01:34:39 either because they don't have the level

01:34:41 or because they want to laugh

01:34:43 there is one of the singers I put

01:34:45 he does this in a bar

01:34:47 he has a big smile when he does that

01:34:49 and you see that he's having fun

01:34:51 doing a bit of anything with his voice

01:34:53 he doesn't try to reach the note

01:35:05 he tries a little bit but he doesn't try to reach it

01:35:07 and so, the thing is

01:35:09 I said to myself, I'm going to put the "Why" several times

01:35:11 each time he really did a big bullshit in the script

01:35:13 but I'm going to try to make it interesting

01:35:15 by modifying it a little bit

01:35:17 so the first one I put, for example

01:35:19 is for me one of the best versions of this note

01:35:21 it's one of the singers who can reach it the best

01:35:23 for me, the best one is

01:35:25 Glenn Carter, I had written it down

01:35:27 I had made a little bit of notes

01:35:29 it's the version of Glenn Carter, I think it's the one who can reach the note the best

01:35:31 and I put it in second or third I think

01:35:43 and then, the ones I put at the end

01:35:45 are the ones who really have a hard time

01:35:47 there's one who puts a lot of auto-tuning

01:35:49 and then I put the guy who has a bit of fun

01:36:01 in his concert in a bar

01:36:03 and the last one I put is a guy who does that in a kind of cathedral

01:36:05 I don't know if it's supposed to be a professional singer

01:36:07 or if it's supposed to be

01:36:09 like, I don't know, a casting call

01:36:11 for a The Voice or something like that

01:36:13 because I know there are some like that in the list

01:36:15 but I think he has the most ridiculous way

01:36:17 of doing the note

01:36:19 you really feel like

01:36:21 you stepped on his foot before he did the note

01:36:23 it's not a "why"

01:36:25 that's addressing God, it's a "why"

01:36:27 it's a weird version

01:36:29 I told myself that

01:36:39 to make the sketch evolve little by little

01:36:41 every time I made it come back, to make it more and more ridiculous

01:36:43 it was going to help

01:36:45 this repetition effect for this running gag

01:36:47 to be better with people

01:36:49 and I feel like it went pretty well for most of you

01:36:51 because you really enjoyed this sketch

01:36:53 and so I'm really happy

01:36:55 because it went well

01:36:57 and so there are a lot of people who are curious and who say "where does this come from?"

01:36:59 but mainly it made us laugh and that was the main goal of this thing

01:37:01 the animations also evolved a bit

01:37:03 at the beginning, each video

01:37:05 was supposed to be just the "why" that goes through

01:37:07 and as I watched the video again

01:37:09 I told myself "it's too similar every time"

01:37:11 and it makes the thing a bit more uniform

01:37:13 than what I wanted

01:37:15 and so I told myself

01:37:17 I would say 2 weeks before the video came out

01:37:19 2 weeks before the video came out

01:37:21 I told myself "ok let's go"

01:37:23 I take them all one by one

01:37:25 and I animate on each thing, and it's animated by hand

01:37:27 so each "why" varies in height

01:37:29 pretty quickly

01:37:31 depending on the singer's tremolo

01:37:33 and I added on top of that

01:37:35 a size change

01:37:37 of the text

01:37:39 depending on the height

01:37:41 or the intensity

01:37:43 that he could reach

01:37:45 and in fact, these text changes

01:37:47 make the things a lot more different

01:37:49 and it also makes it a bit funnier

01:37:51 because you transcribe a bit more

01:37:53 the idea of screaming

01:37:55 you feel like it's a volume thing

01:37:57 all these "i"s that are passing by

01:37:59 it added to the sketch

01:38:01 and I thought it was a lot better like that

01:38:03 so yeah, that was a sketch

01:38:05 it came out of nowhere

01:38:07 it's a reference I've known for 5-6 years

01:38:09 I don't know when the initialist came out

01:38:11 but I think I saw it when it came out

01:38:13 and I never thought it would be in a video

01:38:15 but now, as I had time to develop sketches

01:38:17 as I had time to listen to my text

01:38:19 and to tell myself it was missing jokes

01:38:21 here and there

01:38:23 I had all the time to find this sketch

01:38:25 and to reuse it

01:38:27 I don't know if... I was wondering

01:38:29 I wrote it but only

01:38:31 for the moment where Spiders save Quaritch

01:38:33 it was the only moment where there should be a "y"

01:38:35 basically

01:38:37 and in fact, after placing this one

01:38:39 I thought it was so funny, I had to place it somewhere else

01:38:41 and that's when the idea came

01:38:43 to repeat the sketch

01:38:45 and to take versions of it

01:38:47 from the worst to the worst as the video progresses

01:38:49 THEY ARE

01:38:51 COMPLETELY

01:38:53 STUPID

01:38:55 I didn't plan it like that

01:38:57 they are completely stupid

01:38:59 I wanted to do something traditional

01:39:01 I wanted to change it

01:39:03 with this new version

01:39:05 why did I put it in this new version

01:39:07 at first I planned to do my classic

01:39:09 start with a wide shot

01:39:11 then get closer to the chest and shoulder shot

01:39:13 and then make a very big shot on my face

01:39:15 at the maximum

01:39:17 for the "stupid"

01:39:19 and I thought it would be much funnier

01:39:21 as I have a virtual background

01:39:23 to take advantage of the virtual background

01:39:25 and I took the shot from the beginning

01:39:27 I took it like 4 or 5 times

01:39:29 and I decided instead of finishing on a big shot

01:39:31 to finish on a landscape shot

01:39:33 a super wide shot

01:39:35 because I scream so loud

01:39:37 that the "stupid" can hear himself very far

01:39:39 and that's what's funny

01:39:41 I even added a bird's nest

01:39:43 and I shot it to be in a big shot

01:39:45 and at the end I thought

01:39:47 it would be much funnier to play with a camera's back

01:39:49 rather than getting closer

01:39:51 because it's something I can't do in my other videos

01:39:53 and here I can

01:39:55 shoot a shot with a virtual background

01:39:57 so let's use the virtual background

01:39:59 that's a little trick

01:40:01 but it changes

01:40:03 the traditions, the visual codes of my channel

01:40:05 and the video looks newer compared to the rest

01:40:07 for people who have been following me for 10 years

01:40:09 just by changing a little code like that

01:40:11 "What Rich?"

01:40:13 "What Rich?"

01:40:15 that's the kind of thing

01:40:17 I would like to do more often

01:40:19 and I like to include the video in the background

01:40:21 because I like to be able to get excited

01:40:23 in front of the video

01:40:25 to say "look at that, it's disgusting, it's stupid"

01:40:27 and to have the thing next to you in the picture

01:40:29 and to be able to interact with it

01:40:31 so you can point in the right direction

01:40:33 I think it adds a funny and nice side

01:40:35 and it allows you to have the picture

01:40:37 and the comment in the same shot

01:40:39 it's pretty funny

01:40:41 but I like how it looks, it makes me laugh

01:40:43 it's not much but it makes me laugh

01:40:45 and so the right solution at this point

01:40:47 would be to set up an operation to get Spider

01:40:49 not to run away between the legs

01:40:51 that's where it would be interesting to know

01:40:53 did you notice something weird about this ending?

01:40:55 not to run away between the legs

01:40:57 compared to the previous sentence

01:40:59 and so the right solution at this point

01:41:01 would be to set up an operation to get Spider

01:41:03 not to run away between the legs

01:41:05 did you notice a difference?

01:41:07 because the thing is

01:41:09 once again, when you record this kind of voice-over

01:41:11 and you do a passage like that

01:41:13 a enumeration

01:41:15 going from "if" to "that"

01:41:17 you have to make it clear in your voice

01:41:19 that the sentence has never been finished

01:41:21 and so I had recorded

01:41:23 this line in a way

01:41:25 "not to run away between the legs"

01:41:27 like that, finishing the sentence

01:41:29 and so the sentence is never finished

01:41:31 and I re-recorded this line

01:41:33 at least 12 times

01:41:35 with different intonations

01:41:37 not necessarily 12 different takes

01:41:39 but when I record it, I do it 12 or 13 times

01:41:41 because it's super complicated to reach

01:41:43 the right tone at the end of the sentence

01:41:45 "not to run away between the legs"

01:41:47 yes, I just missed it

01:41:49 "not to run away between the legs"

01:41:51 like that, he goes up like that

01:41:53 to put a little suspension

01:41:55 and if you hear

01:41:57 I'm not quite playing my line

01:41:59 and so the right solution

01:42:01 at this point would be to set up an operation

01:42:03 to get Spider, not to run away

01:42:05 between the legs, we feel

01:42:07 I'm a bit too focused on how

01:42:09 I'm finishing my sentence

01:42:11 and maybe you felt a slight shift

01:42:13 between this line and the one before

01:42:15 which is a bit of a shame

01:42:17 but I never really managed

01:42:19 to do a take again

01:42:21 that went better than that

01:42:23 and so I was wondering if you were able

01:42:25 to hear these little passages

01:42:27 that I do again to reach

01:42:29 the right tone, etc

01:42:31 if you feel the difference compared to the one before

01:42:33 this one, I think I play my line less

01:42:35 in the rest of the video

01:42:37 I'm a bit afraid it goes a bit less well

01:42:39 We have 2h13 left

01:42:41 of "Avatar and the Wet Shaman"

01:42:43 and they don't have a long

01:42:45 summary because it's the first movie

01:42:47 but nothing happens

01:42:49 The release in several chapters

01:42:51 in several parts

01:42:53 doesn't help in the algorithm

01:42:55 to get the video correctly recommended

01:42:57 so

01:42:59 I decided this year to test

01:43:01 with this Procrastinator to do it in one part

01:43:03 which was useful because

01:43:05 I feel I made the same income as if I had released

01:43:07 several separate parts and it works very well

01:43:09 in the recommendations, in a few weeks it got 350k views

01:43:11 and it's not often that my videos do that

01:43:13 so I think the fact that it was released

01:43:15 in one part helped the video more than anything else

01:43:17 but there's also

01:43:19 something I wanted to do

01:43:21 I wanted each part to have a different background

01:43:23 I wanted it to be the forest

01:43:25 for the first part, since it was "Avatar 1"

01:43:27 and the second part, when we arrive

01:43:29 at the Met Cahina, I wanted to make

01:43:31 a background where I was under the ocean

01:43:33 I wanted to make an ocean floor with sand

01:43:35 I had been loaded on the Epic website

01:43:37 with a lot of algae assets

01:43:39 and stuff like that

01:43:41 and I wanted to make a new background

01:43:43 knowing that I could have done this one without

01:43:45 the first shot above me

01:43:47 because it's just a sea floor

01:43:49 it's much easier to have several shots

01:43:51 of depth

01:43:53 in the sea floor

01:43:55 things are usually quite low in the water

01:43:57 the algae and stuff are very projected by the water

01:43:59 so it's complicated to have the same kind of effect

01:44:01 as what we have here

01:44:03 and I took so much time to render this forest

01:44:05 and it was so laborious to render

01:44:07 my hard drive with all my editing files

01:44:09 my 3D program and all that

01:44:11 to render it at home

01:44:13 that I thought "I might not make a sea floor"

01:44:15 "maybe it will add another week or three

01:44:17 of production to the video"

01:44:19 because we never know, there are problems

01:44:21 that can happen with this thing

01:44:23 so I thought "let's go easy"

01:44:25 regarding the challenges involved

01:44:27 I made one virtual background

01:44:29 I might not need to make two

01:44:31 and once the video is in one part

01:44:33 all my guilt and the idea

01:44:35 of not working hard enough

01:44:37 disappeared because I thought

01:44:39 it would have been bad in a single video

01:44:41 to have two different backgrounds

01:44:43 after a transition

01:44:45 I would have needed to make a transition

01:44:47 in motion, the camera would have to go through the branches

01:44:49 and I would have to create a whole side

01:44:51 and then I would have to go into the water

01:44:53 at least if I was doing it

01:44:55 I wouldn't have just made a cut

01:44:57 I wouldn't have accepted to make a transition

01:44:59 of a virtual background

01:45:01 or I would have added a sketch that would talk about it

01:45:03 it wasn't the kind of thing I wanted to do

01:45:05 with a video in one block

01:45:07 and it would have been less good

01:45:09 but at the beginning, there should have been two different backgrounds

01:45:11 and that's why

01:45:13 there was such a change

01:45:15 in the rhythm

01:45:17 that I went back to an exposition

01:45:19 of problems, talking about the parallel

01:45:21 between Terminator 2 and Avatar 1 and 2

01:45:23 because it should have been the beginning

01:45:25 of the second part of the video

01:45:27 and so as a second part of the video, I thought

01:45:29 we can allow ourselves to do the thing

01:45:31 and it didn't go so bad, it made a little pause

01:45:33 on the progression of the movie

01:45:35 on which we already spent 45 minutes

01:45:37 you go back to something more analytical

01:45:39 more based on the staging codes

01:45:41 and the way of making the movies

01:45:43 more than on the analysis of a specific scene

01:45:45 so finally, I thought it didn't go so bad

01:45:47 Sorry for my wife, I don't know how she got out of the kitchen

01:45:49 Sorry for my wife, I don't know how she got out of the kitchen

01:45:51 for once, it's one of those

01:45:53 that I was often mentioned

01:45:55 when I did a test broadcast with Arkady

01:45:57 he had to stop the video

01:45:59 because he was dying of laughter

01:46:01 I'm very happy that this joke

01:46:03 managed to go through because it sums up

01:46:05 the problem of the movie in a very funny way

01:46:07 it's really the thing

01:46:09 it's even an incomprehensible story

01:46:11 of the kitchen, I was very happy

01:46:13 to have found this joke

01:46:15 knowing that, if you look closely

01:46:17 in the movie, Nettiri

01:46:19 is in the kitchen during the whole movie

01:46:21 she almost never leaves

01:46:23 their house

01:46:25 the rare times you see her in it

01:46:27 she's cutting, folding the banana leaves

01:46:29 so she's cooking with Kiri

01:46:31 and the other time she's cutting

01:46:33 figs or dragon fruits

01:46:35 so she's cooking again

01:46:37 when the kids come back

01:46:39 she's doing nothing special

01:46:41 she's cleaning the house

01:46:43 she's cleaning her bow

01:46:45 literally, she's doing that during the movie

01:46:47 is she behind the vacuum cleaner

01:46:49 or in the kitchen?

01:46:51 she's only used for that in this movie

01:46:53 it's crazy, and I thought it was cool

01:46:55 to denounce it in a funny way

01:46:57 Kiri Kiri Kiri

01:46:59 the Kiri Kiri

01:47:01 I don't remember when it started

01:47:03 so yes, the Kiri Kiri Kiri

01:47:05 it made a lot of you laugh

01:47:07 it made a lot of others angry

01:47:09 because I put it everywhere

01:47:11 but for me it was the joke

01:47:13 it was the abuse of this sketch

01:47:15 to put it everywhere, to never pronounce

01:47:17 the name of Kiri again

01:47:19 I even had passages where I pronounced it

01:47:21 in the voice-over without realizing

01:47:23 I didn't realize it while recording

01:47:25 so in the editing I removed those Kiris

01:47:27 to replace them with the other Kiri Kiri Kiris

01:47:29 for me it's that

01:47:31 the running gag is funny because it's a running gag

01:47:33 because it comes back all the time

01:47:35 and it's even annoying

01:47:37 to do that I carefully chose

01:47:39 the Kiri Kiri Kiri

01:47:41 because I had one in mind

01:47:43 that's why it's so good!

01:47:45 Kiri Kiri Kiri!

01:47:47 and this one is

01:47:49 hard to find in good quality

01:47:51 it was very often

01:47:53 it was starting while the ad was not finished

01:47:55 so there were still noises

01:47:57 of the spoon in the bowl

01:47:59 or the Kiri's tart

01:48:01 on the thing

01:48:03 or the kid moving

01:48:05 in the fields, there were animal noises

01:48:07 and so the Kiri was not clean

01:48:09 I watched

01:48:11 like 12 different Kiri ads

01:48:13 on the internet

01:48:15 I can only focus on the end with the jingle

01:48:17 but I watched 12 different ones

01:48:19 until I found the best one

01:48:21 because they have a new jingle now

01:48:23 and I told myself I will use

01:48:25 the most used one

01:48:27 the one most people will have heard

01:48:29 but it's disgusting! They did it on purpose to have kids

01:48:31 who were not in sync with each other

01:48:33 so they sing the Kiri Kiri Kiri in every way

01:48:35 Kiri Kiri Kiri!

01:48:37 So it's a whole YouTube

01:48:39 to have done this Kiri Kiri Kiri sketch

01:48:41 and typically

01:48:43 in my old videos

01:48:45 I would have just made the logo appear

01:48:47 and I would have made it move

01:48:49 but I told myself no, everything has to be

01:48:51 in the face, so the Kiri logo

01:48:53 will appear by doing a "shboing shboing"

01:48:55 in the middle and then stand like this

01:48:57 and just this little

01:48:59 bounce on the side at the beginning

01:49:01 it's stupid but it makes the thing more impressive

01:49:03 and more professional, etc.

01:49:05 It was not easy to do because I'm not good at motion design

01:49:07 it's not the thing I was

01:49:09 trained for

01:49:11 and it's never the kind of thing I tried to do regularly in my videos

01:49:13 so all these little things

01:49:15 animations, logos, etc.

01:49:17 I'm not good at it, I don't have many ideas

01:49:19 but I told myself I had to find something

01:49:21 for the Kiri so I did it like this

01:49:23 I also made it smaller than I usually do

01:49:25 I usually do bigger things in my videos

01:49:27 because I'm used to

01:49:29 editing with a small editing window

01:49:31 so I put it big to see it well in my editing window

01:49:33 and I always forget that in the end it will be projected in 1080p

01:49:35 or on a 4K screen or something like that

01:49:37 so this time I told myself I have to be careful

01:49:39 not to do things that are too overwhelming on the screen

01:49:41 so I had a little fun with this level

01:49:43 I hope that thanks to this

01:49:45 it will make them a little more...

01:49:47 funny

01:49:49 every time it comes back

01:49:51 it didn't work for some of them

01:49:53 but for most of you it worked so I'm really happy about it

01:49:55 Wait, it's Disney now Avatar

01:49:57 so it's banned

01:49:59 Under the ocean

01:50:01 Under the ocean

01:50:03 So typically, this thing

01:50:05 I never managed

01:50:07 to sing this song again like I did

01:50:09 when I recorded the voice-over the first time

01:50:11 I don't know how I was, I must have been in the mood

01:50:13 when I recorded my voice-over

01:50:15 It's my voice-over but I didn't even cut between the two

01:50:17 Under the ocean

01:50:19 I even had the right timing

01:50:21 even if I didn't have the music behind to do the thing

01:50:23 I recorded the thing perfectly the first time

01:50:25 and I never managed to do it again

01:50:27 Sometimes there are some luck

01:50:29 and sometimes my lips look a little weird

01:50:31 especially at the end when I say "Under the ocean"

01:50:33 I should have my mouth a little more open

01:50:35 Under the ocean

01:50:37 Under the ocean

01:50:39 I closed it more when I recorded

01:50:41 because I had done the accent differently

01:50:43 and I took more time

01:50:45 to put the two chorus together

01:50:47 and so

01:50:49 it didn't work

01:50:51 it was ugly, it wasn't beautiful, it didn't match the music

01:50:53 and I had to reuse

01:50:55 the very first version

01:50:57 I had recorded for the video

01:50:59 and luckily I kept it

01:51:01 That's all the editing work

01:51:03 You try to correct things sometimes and you realize you can't do better

01:51:05 But of course Norman

01:51:07 has nothing to do with the fact that we can access the world of the dead

01:51:09 by connecting his hair to trees

01:51:11 or the fact that Ewa is a tangible entity

01:51:13 and verified by the fact that Jack's prayer was heard

01:51:15 and that all ecosystems can modify themselves

01:51:17 according to the will of these entities

01:51:19 That for example

01:51:21 that all ecosystems can modify themselves

01:51:23 according to these entities

01:51:25 That passage

01:51:27 I was sure that everyone would quote it

01:51:29 saying "I don't understand what you're saying, you're talking too fast

01:51:31 we don't understand what you're saying"

01:51:33 Nobody told me about it

01:51:35 I thought it could be hard to understand

01:51:37 People have no problem

01:51:39 and the passages I thought were perfectly understandable

01:51:41 people tell me "No, we don't understand what you're saying"

01:51:43 It doesn't make sense

01:51:45 And then the

01:51:47 most viewed passage of this video

01:51:49 because it's the best sketch I've found

01:51:51 in 10 years of career

01:51:53 "Philippe, I know where you're hiding"

01:51:55 "Come here, I'll kill you, asshole"

01:51:57 "Shut up"

01:51:59 "Come here, asshole"

01:52:01 "Bastard"

01:52:03 I was talking about Avatar 2

01:52:05 with Arkady

01:52:07 and at some point

01:52:09 Arkady was much more dry

01:52:11 than me on Avatar 2

01:52:13 and he said "Avatar 2 is a fucking dork"

01:52:15 and as he told me "Avatar 2 is a fucking dork"

01:52:19 the first thing that comes to my mind

01:52:21 is "Philippe, I know where you're hiding, asshole"

01:52:23 "Philippe, I know where you're hiding"

01:52:25 "Come here, I'll kill you, asshole"

01:52:27 and he told me

01:52:29 "It's a fucking dork"

01:52:31 "Dude, you just gave me a sketch"

01:52:33 "I'm not gonna tell you what I'm gonna do yet"

01:52:35 "but you're gonna love it"

01:52:37 and I spent a whole time

01:52:39 to find a good audio source

01:52:41 for this passage

01:52:43 I always try to have the best quality sources

01:52:45 so a good audio source for this passage

01:52:47 and the passages that would allow

01:52:49 a sort of lip sync

01:52:51 of the Avatar 2 movie

01:52:53 on the movie called

01:52:55 Hitman the Cobra

01:52:57 It was a bit laborious because it was complicated

01:52:59 but in the end I managed to...

01:53:01 "Come here, I'll kill you, asshole"

01:53:03 It's a particular replica

01:53:05 of Quaritch

01:53:07 that I isolated some parts

01:53:09 and I accelerated it a bit

01:53:11 so that the lip sync would fit better

01:53:13 but once you accelerated it,

01:53:15 it fit perfectly

01:53:17 and after a while I got a final version

01:53:19 I sent it to Arkady, he was laughing his ass off

01:53:21 and I think it's one of the best ideas

01:53:23 I've ever had in my life

01:53:25 and I got it because Arkady told me

01:53:27 that I could have this idea

01:53:29 without Arkady telling me this

01:53:31 so that's why I think it's really important

01:53:33 to be able to talk with people,

01:53:35 to have a time to be calm

01:53:37 and to be able to discuss certain things

01:53:39 because if you're always in the lead

01:53:41 to make videos, there are ideas that don't come to mind

01:53:43 and here, it's one of the best ideas I've ever had

01:53:45 because it fits perfectly

01:53:47 to the lip sync because I had fun doing it

01:53:49 but it's still the situation

01:53:51 "Philip, I know where you're hiding,

01:53:53 come here, I'll kill you, asshole"

01:53:55 and he wants to kill him

01:53:57 so that's exactly it, but I would never have had

01:53:59 this idea if I hadn't had my discussion with Arkady

01:54:01 so you can thank Arkady, he didn't find the sketch

01:54:03 but without having discussed with him

01:54:05 it wouldn't have existed, so you can thank Arkady

01:54:07 for this sketch

01:54:09 and I think it works even better

01:54:11 because it's completely unexpected

01:54:13 no one expects there to be an extract of

01:54:15 Hitman the Cobra in the middle of this review

01:54:17 and at the same time it fits so perfectly

01:54:19 that the funny effect is there right away

01:54:21 so yeah, I think it's

01:54:23 one of the best sketches I've found

01:54:25 in my entire career

01:54:27 The passage on

01:54:29 Spider's hostage take is constantly

01:54:31 written over and over again

01:54:33 in the development team, I ended up separating

01:54:35 the moral question

01:54:37 about Quaritch

01:54:39 from the moral question about Nettiri

01:54:41 because otherwise it would have been too complex

01:54:43 to deal with everything in one thing, I was doing too many back and forth

01:54:45 it was too laborious, so I said to myself, we're going to separate it in two

01:54:47 we're going to do Nettiri first, we're going to do the end of the movie

01:54:49 and then when we get to the moment

01:54:51 where we're going to save Quaritch, we're going to get back to this thing

01:54:53 but otherwise

01:54:55 it was way too laborious to do it

01:54:57 in this sequence, and then we finish the movie

01:54:59 because there are still a lot of scenes to deal with

01:55:01 so yeah, it was very very very laborious

01:55:03 to write this passage, and I hope it went well for you

01:55:05 because I struggled on this passage

01:55:07 Hitler's son

01:55:09 who during all his childhood

01:55:11 heard that his father was Hitler

01:55:13 and during the last months had the proof that his father was Hitler

01:55:15 saves

01:55:17 Hitler!

01:55:19 I almost did something here

01:55:21 I wanted to do a counter

01:55:23 with Godwin's points

01:55:25 but I didn't know how to illustrate the Godwin's point

01:55:27 there was no one who theorized this thing

01:55:29 that any argument

01:55:31 will end up

01:55:33 coming at some point in a comparison

01:55:35 with the Third Reich

01:55:37 I think people wouldn't have recognized it

01:55:39 I would have put it in a room like in Mario

01:55:41 with points next to it, and I'm not sure people would have understood

01:55:43 that it was Godwin, so I would have accumulated the Godwin's points

01:55:45 so in the end I didn't do this sketch

01:55:47 I didn't think about it for a long time

01:55:49 I don't know if you liked it

01:55:51 but I didn't plan to put the Hitler's quote at the end

01:55:53 which was like "Nein, nein, nein, nein"

01:55:55 it wasn't in my text, I didn't plan to put it

01:55:57 but it was at the end, when I saw the thing

01:55:59 after all these Hitlers, I said to myself

01:56:01 I really need to put an angry quote

01:56:03 which would be good to finish this thing

01:56:05 and suddenly I had in mind the Hitler who comes and says "Nein, nein, nein, nein"

01:56:07 and I said to myself "yeah, it fits well"

01:56:09 and the last thing I want to talk about at the end

01:56:11 I don't know if you noticed

01:56:13 but there is a progression in my way of naming

01:56:15 Avatar 2 from the beginning to the end of the story

01:56:17 it's called "Avatar 2, the waterway"

01:56:21 and I found a way to reduce it each time

01:56:23 Let's go for Avatar 2 and the well hydrated road

01:56:25 We have 2h13 left

01:56:29 from Avatar and the wet road

01:56:31 because I guarantee you that when you start an Avatar

01:56:33 and the wet passage

01:56:35 there is a decrescendo of the size

01:56:37 of the words used to talk about it

01:56:39 from the road, the way, the passage, etc

01:56:43 I had a lot of fun with it

01:56:45 there is also some running gag

01:56:47 in some things, just in the voice-over

01:56:49 in the basic texts

01:56:51 and in some terms I use

01:56:53 So there you go

01:56:55 I hope you enjoyed this video

01:56:57 and this ending passage

01:56:59 where I thank everyone

01:57:01 I had planned it from the beginning

01:57:03 but I received an advice from Arcadis

01:57:05 I had recorded it in the same tone

01:57:07 as the rest of the voice-over

01:57:09 and I looked angry

01:57:11 I almost looked like I was yelling at you

01:57:13 and Arcadis told me

01:57:15 "Maybe we should make a small difference

01:57:17 on the ending text when you say "thank you"

01:57:19 because you say it dry

01:57:21 "Thank you everyone, goodbye"

01:57:23 That's how I recorded it

01:57:25 and he said "maybe it's a bit dry

01:57:27 so take a breath before doing it"

01:57:29 and then I changed the tone

01:57:31 to say "I got out of my text and now I'm going for real"

01:57:33 and that's what I did

01:57:35 and it's not what he advised me

01:57:37 I recorded the end of the text

01:57:39 I didn't know what I wanted to say

01:57:41 but I created the sentence live

01:57:43 I didn't have a text under my eyes

01:57:45 I just wanted to thank this person

01:57:47 and say this thing

01:57:49 but that's all I did

01:57:51 So I improvised this ending passage

01:57:53 so it would be easier

01:57:55 in another tone

01:57:57 and we could feel a difference

01:57:59 between the angry Dudu and the happy Dudu

01:58:01 at the end of his video

01:58:03 I don't know how long this video will be

01:58:05 I think I recorded it between 5 and 6 hours

01:58:07 I'm so angry

01:58:09 it's so monstrous

01:58:11 that's why I lost my voice at the end

01:58:13 and I'm having trouble articulating

01:58:15 I talked a lot

01:58:17 so I don't know how long this video will be

01:58:19 maybe in several parts at the end

01:58:21 I don't know what it will be

01:58:23 I thought it would be faster

01:58:25 I didn't think I would have so much to say

01:58:27 but there were a lot of complex things

01:58:29 I hope you liked it

01:58:31 I might have crammed a lot

01:58:33 because it was too long

01:58:35 I'm so happy about the welcome you gave him

01:58:37 because you gave me a big present

01:58:39 to reward the choice I made

01:58:41 and the risk

01:58:43 of not being able to release

01:58:45 a lot of videos in the meantime

01:58:47 So thank you so much

01:58:49 for making this video a big success

01:58:51 thank you for watching it

01:58:53 no matter how long it is

01:58:55 I don't know how it will feel

01:58:57 and I hope you will continue to follow me

01:58:59 for other adventures

01:59:01 Take care, go watch movies, ciao!

01:59:03 [Music]

01:59:05 [Music]

01:59:07 [Music]

01:59:09 [Music]

01:59:11 [Music]

01:59:13 [Music]

Anecdotes - PJREVAT Avatar 2

Category

Transcription

Recommandations