 Thank you. I don't know. Is the mic working? All right. Cool. So, unfortunately, there's a problem with one video, but hey, that's life. We're just going to move on and get to it. So that shouldn't be the next slide. What's going on? Okay. That's a different slide, but fair enough. I'll go back to it in just a second. So first, I'm going to quickly say who I am. So my name's Midge. I'm a motion designer and 3D animator. I love education as well, so I do a lot of teaching, too. And I'm mostly interested in free and open-source software, IT stuff, pipeline stuff, a huge hardware nerd as well. I love to build everything myself. And I love watching movies and TV, so a lot of my inspiration comes from movies and TV generally, and got me into sort of being a motion designer. So what do I do? I make a bunch of weird art, basically. I love that motion design has become this thing where it's like, if it doesn't really fit in any category, you can just call yourself a motion designer, and it's like, looks weird, it looks abstract, it looks kind of techie, awesome. So yeah, I make all kinds of stuff. This is, I like to abuse all the tools in Blender as well. For example, this is all just one scene with different camera angles and a bit of lighting changed. And it's just one object with a bunch of shading on it, and that's it, and a bunch of micro displacement. And it's one of the images I did when it got first sort of put into Blender. And my way of working is I really like sort of stream of consciousness working. I hate having to switch between multiple applications or having to render out multiple sequences or something, and then bringing them back in and doing things. And a lot of that is where my work comes from. And this is a really good example. This is done with a hair system. It looks really big and huge, but really it's just a hair system, and you can abuse all these tools to make really interesting looking things. Fun fact, this was rendered before the hair patch, and I took a really, really long time, and then it got patched, and all of a sudden it rendered super fast. So, but yeah, let's see. Oh, gotta go back first. For some reason, the slides are in the wrong order and everything. I blame Apple. All right, so the thing I'm going to talk about today is days. Now, my birthday is about two weeks, and my brother's birthday is about a week before me, and about six months ago, we were talking to each other and we were like, hey, we've never really done a birthday party together. Wouldn't that be cool? And we were like, okay, we can definitely do something, and he does a lot of work in a local venue, and we managed to book the venue, and then we started thinking, well, what if we actually turn it into a party, and then, you know, we started thinking even more. I was like, well, you know, I'd love to, I've been moving more into the VJ scene and doing more, a lot of work with musicians, and like, maybe it's a good way to showcase some of my work. So, days turned into how I ended up talking to myself into creating 60 VJ loops. Yeah, because I started talking to people, and now we're going to have to go through all those slides again to get at the right one. Come on. You can do it. He really doesn't like it. All right, so I went from zero to 60 in four months, and I'm really glad there's going to be some render calculations and stuff in there as well, and I'm really glad I did all of this for the presentation after the work was done. Because if I would have seen a lot of these numbers, I would have gone, ooh, I'm not getting anywhere near that, because this is just like a giant pain in the butt, because it's all personal work. It's not paid, so I have to do other work in between as well to get it all done. And my reasoning here is like, I wanted 60, and I wanted them to be out 20 seconds each, so if you loop them three times, you get a minute. And that way, they're all at 2K resolution, so I can move around a little bit, and obviously I went for the cinematic 24 frames a second, mainly because I have to render less frames. And even if you calculate it out, it's 28,800 frames. And yeah, like I said, if I would have known this in advance, wouldn't have even started. But I'm really happy to announce that I've managed to finish it, not even before the party, but actually, like now, they're already done. So I'm going to give you a sneak preview of all of them. So that's the first video, I don't know. We need to do some juggling. And there's sound as well. No sound. No. Just pause it. This all worked at home on a Linux machine, all right? Look at all that professional gear failing. Am I the annoying guy with the videos? No sound. Oh, come on. I can try something. Ah, there we are. So that was basically the last four months of my life. Music was done by my brother, so I just wanted to give him a shout out as well. So, yeah. So how does a project like this get started, right? I talk myself into it, so it starts with one guy, and the last time I did this presentation, I was sort of halfway, and that still said one idiot because I was in the middle of it, and that's how I felt. And, well, you know, getting back to those 28,800 frames, you need to get it rendered. So I've got two workstations, Zeus and Rinsler. You can guess what my favorite movie is, most likely. And between them, I've got the power of five GPUs, and for some reason that doesn't feel like showing up, but there you go. So there's one workstation with three GPUs and one workstation with two GPUs, and everything, all those 28,000 frames, were all rendered by me on my machines over the last six months. And again, this is another one of those slides. It's really interesting when you do it afterwards because one loop is 480 frames. If I render them at one minute a frame, it takes about eight hours to render one loop. Now, the average was not one minute. I can tell you that right now. It was more like three, and I think towards the end it might have even started peaking towards four minutes. So the total render time, and this is just on one workstation, the main workstation of three GPUs, would be 60 full days of actual rendering, like 24-hour days. Now, I use the second machine as all a little bit, so it probably got cut down to about 45 to 50 days even. And I don't know how I managed to juggle this. When I look at that number, I'm like, how I was doing other stuff, I was doing client work, I had to get a whole lot of other things done, and somehow two full months, 24-hour-a-day months got taken out of those four months just to render. A lot of people might say, well, why didn't you use a render farm? Well, I tried, but the way I work, I don't know, there's something about it that render farms really don't like. And I tried render street one, for example, I was really excited about it, and then when you're used to rendering on a whole bunch of GPUs and then you have to fit your frame into half an hour on a CPU, it becomes near impossible. It's absolutely crazy. And even with the high power build of the last few months, it still came out cheaper to just render it myself. So it's been a very hot summer, not just outside, but it's been a lot hotter even in my office, so very interesting. But I'm really happy that I got it done. So how do you keep, how do you stay inspired through something like this? Because doing other stuff the whole time, I'm trying to push myself, trying to get better at it, and at the same time, not trying to get completely depressed and want to lock myself up and go completely nuts. So after I made a few of them, I started thinking about my methodology and I should really put some time into it to figure out a way to keep myself going. And the first thing I thought about was like, what's my target audience? Well, it's drunk people, right? It's going to be all projected behind the DJs and all over the place and I'm still figuring that out a little bit, but most people are going to be doing this and they might stare at it and other people are going to be doing this and staring at it. So it really releases you. It gives you some kind of freedom because you're like, as long as it's weird, people will probably be into it, which is really nice. And another thing that I found after just doing a few of them is I had to let my ideas go. I couldn't go in with a preconceived notion of what I wanted to make because I would just get disappointed if I didn't get there in time because you do feel the pressure. I can act really cool about it now, but I felt the pressure the last couple of months. So I found that letting my ideas just letting go of them and not trying to go in with, I'm going to make this or I'm going to make this was really, really important to me and really important just in general to stay motivated. And the way I started working around that is just by picking out tools in Blender that I hadn't used much and just going, all right, now I'm going to make a loop with this tool. I'm going to try and figure out how to loop it and just hope it works. And yeah, I did a lot of really fun stuff. So this is where the second video comes in. So this image is actually from a contest that was done recently on Remington Graphics. It's a YouTube channel. Grant Wilk there hosts it. He's a really cool guy. And the point of the challenge was to create something from a sphere without modeling or sculpting. When I read that, I was like, yes, I hate modeling and sculpting and not because it's not useful. It's super useful, but I've never been like a really good modeler. So modeling for me is too slow to get my idea out. And I don't like the fact that you can't iterate very quickly, unless you're like a super pro modeler and you've got all your hotkeys set up. And it's just not me, I'm afraid. So I do everything through modifiers. So the contrarian that I am, I was like, all right, sphere challenge, I'm going to make a cube. That was like the first thing that came to me. And this just gives you an idea of the modifier stack that I use. And you see me sort of run through it. And this is my favorite way of working because I can generate ideas like that. I can change one parameter and all of a sudden I get a completely different result. You know, especially if you can combine it with stuff like particles and change seeds and start messing with things and really just iterate really, really quickly. And then you see the end result there. To get to that kind of workflow, I now know which modifiers are very prone to crashing and which ones aren't. So there's some of them that are really slow, some of them that are really fast. And it's been interesting to see some of the development happening lately on them. So what's the other part of that is you have to dare to fail. This is just a handful of stuff that didn't make it in the video and won't ever see the light of day because I didn't like them or they were rendering too slow or there was something about them that I just wasn't into. And this is just a very small selection. So I didn't create 60 loops. I probably created almost 120 at this point. But that comes back to what I was saying earlier about sort of stream of consciousness. I want to be able to work as quickly and as focused as possible. So if I'm working on something and I do this all the time, if I have something that's like it takes an hour maybe or a half an hour, I'm working in Blender. I'm not liking what I'm seeing. File, new, crash. Just throw it out, slash, remember and keep going, keep going. I don't have a very creative background. So mine is like creativity through brute force most of the time. And it does work. And some of them, maybe I'll revisit them at some point, but I think it's a really important part of being an artist. It's just daring to mess it up. There's nothing wrong with it. You don't have to show this to anybody. It's really important that you keep going for yourself and you become more critical of your own work. So there was going to be another interesting video. I don't know if it's this one. All right. So first, another thing that obviously pops up is you have to loop everything. And you have to get pretty creative at some point because after you've done use the same technique like a few times in a few loops. You want to start doing new stuff. You want to start looping new tools. And you have to really start thinking about how stuff works and how to loop stuff. And yeah, it's a kind of a brain melter sometimes, but some of the stuff that I've been able to loop is geometry. Very simple. It's a displaced plane and the way I approach that is I displace it. I then mirror it and then I can array it because it's got a perfect seamless shape, which is really nice. Another thing is a displacement modifier. If anyone here has ever seen my YouTube videos, I don't know what I would do without a displacement modifier. It's like the base of all my work. And you can set a modifier to use a custom object as a coordinate. So you just make a curve, have that follow the curve, and then when it loops around back to the beginning, you get perfect movement, which is really interesting. Next one is dynamic paint, which is fun too because it's like displaced with a little extra edge with some fade in it. You can do all kinds of things. So there's a plane under it that's being displaced and then it's being all pushed into a vertex group and then that vertex group is driving another displaced modifier. But it gives you really interesting looking organic feeling things and the trick to that is just having a lead in. So it's basically the same trick as the displaced, but you move your rendered portion over a little bit and have that lead in perfectly. And then I discovered this last one fairly late, but I have a feeling it's not going to load. They're not coming up, are they? So there's three more that unfortunately don't load. One of them is oceans. Ocean modifier is very loopable as well and the waves, which is a lot of fun. And then I've already forgotten what the middle one was. Wow. Oh yeah, cloth simulations. You can actually loop cloth simulations by exporting two separate point caches with a different seed and then mixing them together, which is really interesting. And the last one is particles. When I figure that one out, I was super happy. And basically it's a complete and utter hack. I'll use the particle instance modifier to get geometry and then I'll export that geometry as an alembic and then bring in the alembic twice and offset one to the other and have it sort of work together. If you're interested in this kind of stuff, I recently did a YouTube video on all these techniques so you can hear me babble about them for like 45 minutes or something. But I go through all of them and set them up from start to finish. And then there was going to be like a really interesting video where I actually had a time lapse of me making a loop. So just imagine me, like a blender screen and things going really fast. But basically what I was going to explain for this is that I never leave blender. I do everything from start to finish. So I start with all the modifiers and stuff and then I jump between shading, lighting, compositing and modeling because of those modifiers all the time and going back and forth, back and forth, back and forth. And because I start compositing really early, I keep my sort of end frame in mind as I'm doing stuff. And especially when I started using the Blender Cloud add-on to bring in HDRIs so I don't even have to leave blender. Yeah, it's awesome. I really love it. I don't have to leave blender anymore. So I've got this whole little toolbox that allows me to just create, create, create, create, create. And to me, that was like a really, really awesome thing. Now I've got a bunch of images overlaid here that hopefully... Oh, there we go. Hopefully we'll illustrate this. And this is a very simple example, but being able to go back and forth, you can figure out where you're going to fix just in shading and rendering and what you can fix by doing compositing. So I just have a couple of compositing layers. They're not going to show, are they? Oh, there we go. So I abuse the mist pass all the time. Yeah, what is it? This one is sunbeams, I think. That should really be able to go inward. So any developers listening, please. There we go. More glow, more crazy stuff. And then I glitch it out and use the filter filter. I love the name of that one. You can really highlight your edges and do weird stuff with it. I know it's probably meant for other things, but I really don't care. Coming from a background of using a lot of proprietary software, it was really refreshing to see that Blender approaches the problem in a different way. It gives you all the tools, doesn't tell you what they do, which at the beginning is kind of weird, but something that the person that was speaking before me also said, not Eddie, but the other guy at Brent, I think it was his name, is I noticed when I'm teaching as well, students approach it very differently than like these streamlined tools that you're not supposed to use for anything else. So I really like that. And I just keep building on and then I put a scan line effect there. And the reason I'm able to do this is because I can just go back from forth to the pipeline without ever leaving it. And that to me is just absolutely crazy. I also come from kind of a VFX background, so I don't even render EXRs anymore. I just render my final composited frames to PNG sequences. Yeah, I know. But it works for me, so I don't mind. And doing everything on GPU allows you to make it go fast. So this is the final frame. So what I really wanted to say to everybody is just mess around and have fun. Like just break everything and then see what happens when you try to put it back together again. That's how a lot of my stuff kind of sort of ends up being. I let the tool guide me a little bit and then I take over when I feel like I actually have a concept that I can take out of it. Also, you know, use whatever works. I've almost completely converted to open source tools, which is really crazy. It's also very interesting when you talk to like an accountant and they're like, so you must have a lot of software you can like, you know, put in money-wise and then get money back from. I'm like, no, I actually don't have any software costs right now. And they're like, oh, that sucks. You're going to have to give everything to the tax man. I'm like, all right, that's what I get for going open source, I guess. So yeah, it's unfortunate we couldn't see the time lapse video, but yeah, I want to thank everybody. I've got a website. I've got Instagram, Twitter, all that stuff. There's a bunch of YouTube tutorials as well that I do. So if you're interested in this kind of work, give them a look or just say hi. It's all good. I've been trying to stream as well, which has been a very interesting experience. But yeah, that's it for me.