Damien Valentine 00:47
Hello and welcome to another instalment of And Now For Something Completely Machinima. I'm Damien Valentine. I'm joined by Phil Rice. Hello, Tracy Harwood. Hello. And of course, Ricky Grove.

Ricky Grove 01:03
Happy to see you,

Damien Valentine 01:05
Likewise. And this week we're talking about your film and and I believe he got some news for us as well.

Ricky Grove 01:13
I'd like to say Merry Christmas to everybody. And happy new year, this is the time for that. At our home, we don't celebrate Christmas, we celebrate Christoween, which is our Halloween version of Christmas. We have a black tree and on it as all sorts of monsters and vampires hanging and Jackal landers and a big Krampus at the top of the tree. And then we put all of our presents in black, wrapping paper. So Happy Christoween to those people who want to celebrate that. I tubing. Yes, two bits of news. There's a half life, retrospective documentary for 25 years of Half Life celebrating its anniversary covers the background of Valve, its formation, its creation, its idea for Half Life. There's also going to be a remastered version of it, which I'm eager to play myself. It's terrific. We'll have a link on the show notes. So you can watch it. Half Life has probably, its certainly in the top five games I've ever played. Half Life 2 is better mostly for the graphics and story. But the extraordinary game hours and hours and hours of fun and imagination and dreaming about it. It's just great. And then the second piece piece of news I have is an I believe it's Grand Theft Auto 6, which is being advertised as coming out. There was a brief sort of trailer kind of thing that was posted. I don't know all the details of what's going to be included. But it certainly looked good. You know, Rockstar Games, they do wonderful detail. So it's going to be a fascinating thing. We'll put a link on our show notes too. If there's a trailer, we'll definitely put that out there and you can keep your eyes I don't know exactly when the game is going to be released. But we'll keep you updated here because it's a very important game for machinima. And then lastly, I wanted to mention that the downloadable content for Elden Ring looks like it's going to be out in February at some point and From Software does a wonderful job on their their DLCs so this will be a very large I think there'll be new characters. New game play area. It's very exciting. So keep keep your eyes tuned and we'll let you know.

Phil Rice 03:50
Alright, I'm sorry to interrupt or do what what was the name of that? That game that's coming out. You said gramme, gramme,

Ricky Grove 03:59
grammes. grammes. Cracker Otto.

Phil Rice 04:06
I've heard of that. That sounds interesting.

Ricky Grove 04:11
Yeah.

Phil Rice 04:12
Just a coincidence, but that's actually the name of a crime, Grand Theft Auto.

Ricky Grove 04:18
You're pulling my knee. You're pulling my leg fell off to

Phil Rice 04:21
check that out.

Damien Valentine 04:22
I'm surprised that she's sick.

Phil Rice 04:25
I'll start with the first one.

Damien Valentine 04:27
Yeah,

Phil Rice 04:27
it's worked my way forward. So I know the whole story. I

Damien Valentine 04:31
was surprised surprise. I've actually managed to make six of them and we've not noticed befor.

Phil Rice 04:35
Never heard of it.

Tracy Harwood 04:37
I've been Miss anyway. I thought they'd been on about that for about 40 years. I don't know.

Ricky Grove 04:42
Yes. Well, certainly it will be a favourite of all of the avant garde. Oh, cinema filmmakers for the Italian Machinima Festival.

Tracy Harwood 04:51
Those that are left.

Ricky Grove 04:53
Yeah, those that are left. Well, anyway, let's get on to my film.

Tracy Harwood 04:58
Before we do that, I've got one bit Have you used to share? If you don't mind?

Ricky Grove 05:02
I do mine. All right, go ahead. All right.

Tracy Harwood 05:06
Well, I just wanted to say really, I don't know if you saw that John McInnis Studios spoke at the Unreal Engine Festival in New Orleans in October about the upcoming film that he's been developing with his studio called Baad Viibes. And he spoke with folks from Nextera and also AWS, which I think is quite interesting. bad vibes is a is a really cool project that's adopting both a crowdsource and a cloud based approach to the production process. So far, really, I think there haven't been that many tools or platforms that have professionalised that kind of that part of the creative process. I think it's something that the machinima community did back in those very early days, when things were shared through the dot com community for, but it's not something that was ever Well, folks were never able to commercialise the kind of work that they were doing in terms of the way that they were sharing it or anything that kind of came out of it. And the big difference here is that John's approach is obviously being done as a means to create exploitable content. So I think this is one that we really need to be paying attention to, because I think what it will do is enable the development of this kind of customizable pipeline, the front end part of it, the creative part of it, which will be done to an industry recognisable standard. And that ultimately means this particular project code could become quite influential and become, I think, probably the link between some of the very early machinima practices in terms of the way the community was harnessed. And some of these kind of new virtual production tools that these big studios are attempting to work with, because it's trying to link those two things together. So I'll say that John's doing updates through his channel, and hopefully will at some point, get to have another chat with him about how far he's got with it, and what the reactions to the project will be when it comes out, which I think will be next year at some point. That's my news.

Damien Valentine 07:36
Good news. Very interesting. Alright, so Ricky, you've got a film called Backflip? Yep. I

Ricky Grove 07:43
came across this film in an unusual way because I was actually looking for something about motion capture and AI. And that led me to another place. And that led me to another place. And that led me to another place. And that led me to Vimeo. And while I was at Vimeo, I almost always go to the Editor's Choice collection of videos, because those are, they're almost always super high quality. And I was just going through looking for interesting stuff. And I saw this Backflip. And what intrigued me was the screen capture that was used to be used to identify it. It looks that's so strange. It was such an odd mixture of live action and sort of glitched game capture, you know, and so I watched it that was just knocked out. And now you know me by now, and listeners probably know that I like things that are very creative, and that are stylish. They, for some reason that that appeals to me a great deal. And so this, this clicked all the boxes for me, the idea of it of this guy trying, essentially, he's afraid to try a backflip in real life, because he thinks he's going to hurt himself. Which is probably true. Unless you're very athletic and gymnastic. Don't try to do a backflip at home because you're probably going to break your neck. So he thought, well, what if I could do it in this digital virtual world that I would create. And so he goes through the process of trying to train his, his character, which is him, he took a picture of himself, and then made a 3d model out of it and left it sort of rough, especially around the face and everything which gave it a kind of abstract thing. And then he tried to train it to do backflips Well, initially it wasn't very successful. Until eventually he could get it to where could do backflips? Now this journey to create this virtual him to do that backflips has all sorts of meandering plots where he does many other things. And you get to see what's intriguing is the style is you get to see his real life video. But also in the back, you get to see the game elements, the, the vertices, and all of that stuff, which I thought was just fascinating and very funny. But funny in a strange way, because you keep wondering, are you poor guy, you can't go out and maybe try to do something else now. But he was bound and determined to use his skills as a game creator, to make himself to, in fact, backflip so you can experience it's a creative hybrid of the game, 2d and 3d in a very unique way. And it also shows how you can use AI to create entertaining and thought provoking films, with all of the snoot snooting a cock at AI and its legality. This is an example of how to use AI in a way that is interesting and unusual. And also fits into the documentary machinima movement. As an essay film, there was a subsection of documentary in real life that was called the essay film, where a person would explore a personal obsession, or a personal idea or a personal event. And they would document that. This is exactly that. But it's done using machinima. And for that, that's a third level of creativity that I found fascinating. I just love this film. It was thought provoking. It was funny. It was strange. It was weird. What did y'all think about it? Well,

Damien Valentine 11:59
I started watching it. Sorry, I gotta go first. Because I don't want to have to follow the rest of you. I watched started watching and I thought, What is this? It was not what the you know, the first minute or so we're trying to figure out what's going on here. And as was Ricky chosen this. But then I really caught on, this is actually a lot more than you think it's gonna be. One of the things that really stood out to me is the amount of work that actually went into this. Because it's made to look like it wasn't, it makes it look like this is just like it was thrown together. But when you watch it properly, you think I know he is that's the trick, he's put a lot of work into this. And is when he cuts between several different locations. And there's one where he's in his room, and he falls against some furniture and the furniture is actually it's, you know, it's just a photo of room. And you think he's just got this virtual character gets a photo, but he doesn't because the furniture moves when he impacts it and the stuff on top of the table that moves as well. But it matches the photo perfectly from the original location. So you don't expect it to suddenly jump forward. And so that's a huge amount of work to find, or to create 3d models of the environment that matches the photo perfectly. And to position it so well, that you don't expect it to move until it is character hits it. And then to actually make it move, and then he's like, Oh, there's more to this going on. And that's where it rested up just how good this film was. Because you can see this is he's put a lot of effort into this. And obviously, the 3d he's crazy, that 3d model is his character. But he's done it from different angles as well. So he's done a side one and a back one. And it cuts between them, but badly. But it's done on purpose. Yes, exactly. And I can really appreciate this film for that. And that's really funny. And it's definitely worth this is. It's not a long film, but it's worth sticking to all the way through because you get so much out of it if you just give up on it after the first minute because you have to stick through it to really appreciate it. But it's worth it. So yeah, Ricky, this is a really great choice. And I'm glad you chose it because I would probably not have watched it just from the thumbnail. Right? Right. We have to say this way each other we pick a film, which we wouldn't have normally watched. But then we have seen ourselves and we can really appreciate it and we introduced something completely different. So yeah, thank you Ricky for choosing this.

Ricky Grove 14:51
Oh, you're very welcome.

Damien Valentine 14:54
Sorry, Tracy.

Tracy Harwood 14:54
You will need to go next. So do you want to go next, Phil?

Phil Rice 14:57
No, go ahead.

Tracy Harwood 14:58
Okay, well, all right. If so this guy has created this kind of virtual asset in his own minute image, I think use photogrammetry to do that, what sort of data to do that?

Ricky Grove 15:09
That's right. That's right.

Tracy Harwood 15:10
I suspect that's what he's done. And then he's animated it using an algorithm to explore how it learns to do this backflip. And he's taken this process, this Nikita Diakur, has taken the process from a published research article. And then he's tried to apply the technique, and I'm guessing he's probably done it in something like Unreal, or maybe he's created another kind of bespoke environment to do it. Actually, I think it's a work of creative genius in many ways, it's actually using this kind of research based approach to animate, where where the virtual character or the agent is given, basically, something to mimic as a kind of reference, and it has to learn how to do that. But without a reference for all the interim stages, so the agent learns, basically learns some really awkward kind of stages on its sort of journey to create the end result. Now, this is basically I think, you know, as you said, it's kind of a sort of a documentary of the learning process, including all these kind of awkward interim stages of the virtual agent learning to do this kind of backflip, which apparently took him three years to do, three years, which is astonishing. But actually, what I love about this is that it's it's a, it's a creative way to show a research process that demonstrates the algorithmic approach. And, and what it's doing is actually putting emphasis on a specific type of game mechanic. You know, which, which, let's face it, many, many of us now really have kind of avoided this. Because the advancement in mo cap techniques means that these kinds of learned motions can be stitched together from actual recorded behaviour, or at least, what you can do is add more of the interim stages into the process so that the AI has kind of like a path through which it can learn to mimic in a much quicker and more accurate process of learning. So with that in mind, I then thought, right, what what do we know about this guy? What's his background? What's What is he doing here? So I had a little bit of a dig, as I do. And I would suggest that this guy is what I would call a found media artist. Because his ouvre seems to be one of exploring the glitch in dynamic simulation by computers. And it's, it's glitch only in the sense that the behaviour is not a fully functional representation of human mechanical movement, but an abstraction of it with with clearly some key missing information, such as the way in which a set of bones can move at the same time. So animation, in looking at what he's done is really only one of the techniques that I think he's using here. So, so what he has then done as a creator is actually for me, the really interesting part of this. And the process really, because in this film, what he's done is he's woven over this learning process of kind of imagined sentient mind state for the avatar, in order to devise a story. He's given it a rule, and the rule is about not hurting itself. Or at least he's justifying the glitch, by presenting us with that additional rule based on his human understanding of why he's not himself, been able to throw a flip with his own physical body. So he's commenting on all kinds of human like emotions such as ambition, fear, uncertainty. I picked up achievement failure determination. But interestingly, what I didn't pick up on was these other really important human states such as exhaustion, frustration, and you know, energy replenishment, let's say, so he's only picking up on a part of it but not the whole thing. And then what you've got really is this kind of well, some of call it gross but I'm call it kind of voyeuristic style of documentary that's kind of delightfully sadistic at one level and deeply disturbing another very much in the mould of what I would call experimental machinima which exploits the glitch in the game primarily That's what a lot of experimental machinery is doing. It's playing with that, that glitch, which basically ends up them being a satire. And in this case, what we're talking about is a satire of computer simulation processes. In some ways, it reminded me a bit of Phil's pick earlier than yes, yeah, I thought so do well, but, but that wasn't exploring glitch necessarily. But but the codified rules in the game. And it got me thinking it really got me thinking this one because because if if glitch is what what we're about, this is a very distinctive style of animation that we've talked about at some length over over some time. And we've seen it many times over the years really, where the where the focus is a very specific glitch in a game mechanic. And the artist is specifically focusing on that. So I was thinking, right, what else have we seen that does that and I remembered, as I was sort of casting my mind back a film from many years ago, which I'm sure you'll all remember really well, which was called the 1k Project II buy Blackshark outline, it was about 2000, early 2005, I think it was released. And what that was doing was using something called Ghost, ghost, ghost mode in the game, TrackMania sunrise. The glitch that the guy was playing with, is basically the mechanic in the game that allowed allowed them to overlay the playthroughs of the game, endlessly, you could put 1000s, of overlaid run throughs through it, and it would just continue to do it or just basically, depending on how good your and computer graphic card was, I suppose and then back in the day. And it was a glitch that was kind of discovered by the community. But it's actually one that the players themselves took full advantage of. There were there were many, I think, as I recall, examples of this 1k project, where it wasn't called product to that I remember there being a project 3 a project 4 and I don't know how many different types of variations of it, there was, there were many hundreds of examples of of machinimas, that came out around the same time that we're practising doing the same kind of thing overlaying the content. But I think what I've kind of lost my train of thought really well, I guess really well is going with it is it's not quite the same thing as the as the example that that you've picked up on here, Ricky, because what what what the glitches here has is kind of been stitched together over over a kind of, over kind of time period. And in order to kind of produce a sort of narrative arc. But what the glitch in that other one, that other example was was, was a kind of simultaneously shown. So it wasn't a kind of a linear narrative at all, but it's sort of a simultaneous thing. But nonetheless, the focus was just on the one aspect of the of the algorithm in the game, which I, which I kind of thought was really interesting. And super interestingly, as well, what I did, because I thought, you know, I remember this is so long ago, I thought, not sure I really fully remember all of what I was told about it that way back when. So I thought I'll ask this guy and see if, you know, see if he will at least acknowledge, you know, a little message about you know, can you tell us more about the backstory to it? Because I was intrigued by the this glitch idea. And can you believe it? 20 years on 20 years on, he's got back to us with a whole story about it. And his role in developing it, and it's truly fascinating. But what I'm going to do with that is a separate blog, because it's it's not my story as what what we're talking about. But it's, I think one of the very first examples, I remember that that kind of pushed that boundary and ended up with a really interesting and very successful output from it. In fact, Blackshark 1K II project made it to the Guinness records. And it was it made it because it was the first example I believe of a French viral machinima so it made it in French version of it. So this this story of glitch I think is one that's really really interesting. I have to say, I always thought it would disappear over time, this, this focus on this, because I assume that, you know, as as things got patched, and you know, more things were perfected through a kind of a longer dev time and whatnot, I just thought it would go away. But I think somehow, it's it's neither gone away nor diminished in its impact. And it's allowed, I think, artists like Nikita Duikur, to make a really unique contribution. Because what he's able to do is focus on the aesthetic of that very specific imperfection. And in Diakur's case, what we're talking about is the way the avatar learns to mimic movement. So it's fascinating. Thank you really enjoyed it.

Ricky Grove 25:51
You've got enough there for a good article, you should write it. Possibly.

Phil Rice 25:57
Yeah, this is it's a work of genius. I mean, I don't know how else to say it. I don't think it can be overstated. It's just extraordinary. I mean, there are great documentaries, which are informative, and entertaining, typically. And then there are great mockumentaries, which are, they're not really trying to be informative at all. It's all about the joke. So you know, This Is Spinal Tap, for example, back packaged as a documentary, but it's just a big spoof or a joke, right? It's rare that's something is both at the same time. And I think this is this is it's like you can, if you you can watch this as a documentary of a process of, of basically a scientist, a computer scientist, trying to accomplish a task and documenting the process of that, and all the rough edges of it and everything. And that is exactly what this is. And it's also performative. Like it is an extraordinary comic performance. And the thing that really just blows my mind is that it's both Yeah, it's not a mix. It's actually both simultaneously it is two things well put. And I just loved it. You know, I mean, the the nerdy part of my brain just loved what bit of the computer science aspect of it that I could follow, which is most of it, but not all of it. But, you know, I loved that. But then, you know, the kid in me, sees him bonk his head and, and oh, my goodness, the foley, the sound work in this is just wizard. Like, I mean, it's real. I don't know how. There's a lot about this film that I confess, I don't know how he did this, like, Damien, you caught something about the photo overlay and stuff like that. And that was more specific than I was able to get to on my own watching it. I'm not sure how much of that was that I was too busy laughing and being entertained by it, to really analyse. I watched it twice. And both times, it was I was just yuck it away. It was so funny, and absurd and fascinating. So the only thing that I'll I'll say I'm not going to dig into the how it is made, because I'm not going to pretend that I am confident that I know. But one thing it made me think about I don't know if this is part of his intended message of the film or not. But you know, there's a lot of talk about, you know, how close AI is getting to, you know, be competitive with human thinking and all that stuff. And this is, you know, I've heard some people in the either in this profession or clinical psychologists or people who study neurologists, and they study the brain. And, and many of them insist, we are really under estimating just how much computation goes into things that we think of is very simple when it comes to manipulating the body. We've spoken on previous episodes about what seeing actually is when you try to try to deal with it computationally. It's not just like passing into a thing. There's identification and prioritisation and all these things happening for what we call seeing and yeah, a backflip here as hard as it is, for a human being to learn a backflip, it's probably easier assuming that you've you've got the the working musculature and are in appropriate shape and aren't too overweight or whatever. It's actually probably a really relatively seemingly simple thing for a human to learn to do this with maybe some instruction, the instruction to avoid the injury, which is why he embarked on this in the first place, right? If I just go and just decide to go do a backflip out in the parking lot, you know, call 911. Right? I mean, it's not going to go good. He said that when he tried it, he ended up with a broken toe. It's like, no, that would, that would be career ending for me. Indeed, it would be a neck or something far worse. And yet, I know that if somebody who was a professional gymnast, and if I were in the appropriate level of conditioning, I could probably learn how to do a backflip and a half in an afternoon. From a from a professional. Because what I lack is the understanding of what first of all, what does it feel like? You know, there's the motor memory part of it, what does it feel like, but also just what physically is involved in making that happen? You know, it's not just pushing down on your feet in a certain way. It's their something else, what do you do exactly. And once you're taught that, virtually anybody, again, if you have the appropriate, you know, if your body's in the right shape for it, anybody could do that. And yet, for a machine to break that down computationally and understand it and reproduce it even virtually. This this is this, this documentary is a perfect example of oh my goodness, how extraordinarily things you don't think about. Oh, just amazing. So that's what got me thinking about was, you know, not some, okay, it's easy to have some kind of a pat response to go. Yeah. You know, I like to joke sometimes they whenever I worry about AI taking over the world, I just have an interaction with Siri on my phone. And then I don't worry anymore. You know, Siri, wife goes call my husband Siri says which husband? Siri? I say how much longer is it going to keep raining? Siri says it's not raining. It's raining. So like, I'm not worried about AI. And it's easy to be flippant. And go see this as a reason why we're not Yeah, but I'm talking like seriously like, what a monumental task. And then that's amazing enough. But he gets success. Yes, he actually does it. Oh, like somebody he gets it to happen. Once you reach the end of this about 10-12 minute video, is it what took him three years but I mean, for us we can digest in 12 minutes. I mean, I was just it was like at the end of some some clocky sports movie or rocky or something? Where when when he wins, you're just literally yet. I mean, I was cheering. It was just amazing. Because I could then I had felt like I had even the smallest appreciation for how difficult that was. So what a wonderful experience watching this and being made to think about all of that. It just Yeah. I mean, just just wonderful. There's How often does a video do all of that? Yeah, not that often something and oh needy crashed the car. Oh, hey, funny joke. You know, bathroom humour again from Phil. Great. Thanks, Phil. This, it's like, Man, I felt like parts of my brain that don't light up on a regular basis were lit up from watching. Oh, that's excellent. I'm so glad like it just Yes. Just stimulating in a totally different and elevated way. And then of course, talking to you guys about it. And Tracy I mean, you just had a whole insight on this that just none of that even occurred to me like it's so now I've got more to think about I enjoy watching it again. So yeah, I mean, just Bravo and this Nikita a very interesting person. Sure. Whoever made this is it's just like somebody I'd love to just talk to. What a mind going on, you know behind this. I'd

Ricky Grove 34:54
like to point out one last bit before we close out. And that's the narration Now, if you had a more talented narrator, somebody who was professional and creative, and who could do that thing with variety and drama and a beautiful rich voice, it would almost ruin. Thank ya, because his sort of inadequate, but honest and real narration makes a great deal of the film, of course, obviously the content and the visuals and all that but his narration is sort of lacklustre. Sometimes he's excited. Sometimes he's not. Sometimes he's, he's, you know, depressed about certain things, and excited about other things, makes the thing it's just so perfect. And that's why I think you should put it in the category of documentary.

Phil Rice 35:55
It's actually even more. There's actually something really interesting about that, Ricky, the narration is synthetic, is it? He created a voice clone of himself. A 15 minute voice sample was put through something along the lines,

Ricky Grove 36:17
right. I forgotten

Phil Rice 36:19
Elevan Labs. Yeah. Then it read it read his narration, how fast? Oh, yeah, that's part of how it pulled off the somewhat deadpan the minimum emotion that we've we've spoken about on prior films where Eleven Labs was used for the voice and talked about how the range of emotions kind of narrow what that thing can do. I've discovered something really interesting about that, by the way, that I think I'll share with you off air for now. But I figured out something new about how to coach emotion. Oh, well, thanks

Ricky Grove 36:55
for clarifying that. Yeah,

Phil Rice 36:58
yeah. But he didn't use that here. He actually used I think, the text to voice synthesis with his own cloned voice of himself. Because I think you're right, Ricky the deadpan that the the delivery that flat delivery for the most part, and kind of a random sense of peaks of emotion. Yeah. works. It makes this better. Yeah. Yeah. It does in a film where you're trying to be serious or dramatic, it wouldn't it would be distracting yet, but this it fits with that quirky close clipping through his knees virtual character that we give. Yeah, wonderful. Great.

Ricky Grove 37:38
I'm so glad you liked it all. I was worried that you might not it might be too strange for you. So I'm glad.

Tracy Harwood 37:45
Oh, it's great. It was really great picks probably my favourite this month, actually. Yeah.

Damien Valentine 37:49
That's great. The clothes tipping thing I see. That's something you don't typically see even something iClone is the closer static model attached to the character. But when he's doing those flips, and the jackets moving, yeah, that's simulating actual fabric on his body. It's not doing it well. But that's part of the joke.

Phil Rice 38:09
I see a way to do that. And iClone not that you would ever want to. But if you turn on soft cloth physics, and then paint the soft cloth physics map as to treat everything as flexible. And like silky. Yeah, it will do that.

Damien Valentine 38:27
But it'll do it as well as video. It will do. Bad

Phil Rice 38:31
as that video. Yes. Yeah. I bet because I was experimenting with it with a character that I brought in with AccuRig. And actually, yeah, it happened like the clothing split apart. And as soon as the physics was turned on, it's just weird stuff happens to be a lot of that. Yeah. So it's not something you'd ever want to do. But yeah, it's that's that's how that works is that basically, you saw with other objects in this film, there was a lot of physics being applied, the way the bottle wobbled, and things like that. I think that those are all within the simulation. And he must have turned on some kind of cloth physics simulation for certain parts of the outfit. And yeah, it. It really worked well, for it.

Ricky Grove 39:14
Sure did.

Damien Valentine 39:15
Yeah. All right. Well, I think that wraps things up, not only for this week or this month, but also for this year, because this is our last episode of the year. That's right. So if you have anything to say to us about this film are all about us. Send an email to talk at completely machinima.com. You can find our blog with our news and show notes at completely machinima.com. And that's it for now. We'll be back next year with a brand new episode. So we'll see you then. Take care everyone. Happy New Year.

Ricky Grove 39:48
Oh, hi, everyone. Happy New Year. Happy New Year.

Phil Rice 39:51
Happy New Year.