 Excellent. Thank you for coming, especially so late in the day. All right, my cryptic title. You can lead the designer to water. Well, let me first start with my name. My name's Gareth Coley. And I'm a motion graphic designer. And why am I here? Well, basically, I'm going to talk about moving the studio from a proprietary to an open source platform. It was an attempt that I did a few years ago. And everything that went on about it. But before we do that, let me just talk a bit about myself. Who am I? Some of you in the audience might recognise that. I've got my start on motion graphics with the Commodore Amiga way back in the 90s. One of the best computers ever made. I used deluxe paint, imagine. Anyway, here's a picture of something that I did way back then with a friend of mine, Mark September. You can actually see the pixels. I think it was 300 pixels high, which was high raise back then. Anyway, since then, I've got some examples of some of the stuff I've done. I've worked in a variety of industries, broadcast, promos, commercials. I call them VR sets. It's not VR in the traditional sense. It's VR like you make a set and then it's all linked up via cameras. Technical animations, all sorts of stuff that I've done over the years. I first started playing with Blender. I think slightly before it was open sourced. It's version 1.5 of the manual. But throughout the years, I played with a bit and then left it. I only really started to use this package in 2008. This was one of the first commercials that I used this package for. I'll play it now. I must click play. Why aren't you playing? Space. It was working. Anyway, it's a very nice metable animation. I'll show it to you guys if I have time at the end. Back to the talk. I worked for a company called ETV in Cape Town, South Africa. It's the new building against the Table Mountain. We were a studio that had initially one channel and it expanded to seven channels. It's free to air. It was not only terrestrial but also satellite based. There was a motion graphics team of ten people, which I ran. Another member of the team there, Wesley Matia. He's now living in Holland. Let's hope this one plays. How do I play the hell? When I press play, does it? I'm just going to see you guys get an idea. Have I got multiple versions of this? Let me just close that. Did I just kill the whole thing? There we go. I just want to show you some of the work that we did. I think it's probably important to... Here's the metable animation. Trust me. Catch Arnold Schwarzenegger as the ultimate killing machine. The Terminator double feature, Sundays at 8 on E. So that was... That's my clips. KidPrimer, ETV... Anyway, I'll leave that one. Let me get back to the presentation. Or else we're going to run out of time. So why make the move? One of the big things that we wanted to do is get some of the features that are in Blender that we didn't have in Cinema 4D in our tool chain. Things like cloth, smoke, water. Those are usually plugins that you have to buy for Cinema 4D. Our industry was changing. We had the print team, motion graphics team, a production team. We started to do more and more of each other's jobs as the industry changed. So suddenly we were working on digital billboards and that's usually part of the print team. But now it's coming to us and we're working on motion graphics for digital billboards. So we wanted to make sure that as the industry changed the teams that were involved would be able to start using the tools. And then the investment involved in doing that wouldn't be exorbitant as it would be with a proprietary piece of software. It's a cost saving initiative. So the process. The first thing I did was talk to the team, see if they thought it was a good idea. I didn't want to do it without everyone thinking, what is he doing? He's making some cracks. So I spoke to them. Everyone thought it was a good idea. I think one of the big things was they could see the investment in themselves because they could then use this at home either for freelance jobs or when they moved on to possibly their own company. It wouldn't be necessarily them having to invest in another piece of proprietary software for them to work at home. After I spoke to them, spoke to my line manager, she thought it was a good idea. I then had to write this big sort of document with numbers and everything to senior management. It wasn't a quick throw it in and use it. They wanted to know why. There was a cost implication. So we had basically a fifth of what we were spending would be spent on blend as opposed to Cinema 4D, which obviously management likes. They did have a slight issue management, the finance and the IT guys were a bit worried about support. I had to do a little bit of education around the open-source world and how support tends to come from the community and it's a symbiotic relationship. Whereas with a commercial piece of software your relationship comes from the money that you give them. It doesn't necessarily come back to you, even though people like to think it does. It's a bit. Anyway, they all said, yeah, let's do it. So this was around 2015. We started to use Blender. Mainly we started to use it in projects that didn't require any part to touch any other part of the tool chain that we were using. So people could use it for individual promos or individual parts that wasn't going to have to link into big brand channel imaging or anything like that. So here's a couple of examples. We used it in a promo here and we did some tracking in Blender. This was done by Wesley. He's sitting over there. And there's the result of the tracking. We also used it. For a cloth simulation we did a new channel reveal and again this was done by Wesley himself. Unfortunately all these videos are not playing. This is very frustrating. Okay, so the area where it was a little bit more sticky was around the imaging of the channel. We had an existing look that was defined by the brand and senior management and we had executed this successfully in Cinema 4D. So we had to ensure that Blender, whatever we did inside Blender would gel with what was already in Cinema 4D. Didn't mean that we had to have the exact same reflection with the exact same sort of thing but it had to be close. It didn't help that everything that we had bar one channel was gloss. So everything that we did was expensive in terms of time. So I did some tests. I took the logo from Cinema 4D into Blender and started to play with cycles and its material system. Got a really nice look out of it, I must be honest. When we showed it to brand and creative director and everyone, they liked it. We tended to have a bit of a more real feel to it than what we were getting out of Cinema 4D. Here's another example. We've got the Cinema 4D interface and this was a whole series of glasses, glass plasters and each channel would then we would just basically have these plasters moving around and with the glass and the reflection. That's a really nice interesting look. So that's the Cinema 4D render. Looks good, nothing wrong with it. And this was the Blender render. Now both of these haven't been touched up massively just a little bit of levels. If I just go back again, you can see there's a Cinema 4D. So the reflections and the refractions are quite different but everyone actually preferred the Blender when I gave them a blind test. There's something more real about the cycles renderer. So we were about to start refreshing this channel, U Movies Plus and I was going to drop everyone into the deep end and force ourselves to use Blender for this one rather than rely on Cinema 4D. Then I got a really nice job offer to go work for the Discovery channel in the UK and so this didn't happen. And then what happened is Wesley left and he moved to Amsterdam. So basically this is where the story of Blender being used in ETV ends because when I spoke to my replacement, Jamie, asked them whether they were using Cinema 4D or Blender and they're using Cinema 4D again. So it was a very interesting experiment and so I've got some lessons that I've learnt that I thought would be useful for you guys. So what was the first lesson I learnt was you need to have more than one champion to the cause. In this case it was me. Wesley was a close second. He really liked the idea of Blender. He's a 2D animator by trade and when Grease Pencil got added he was like, oh, this is looking good. So when I left and then Wesley left, no one in my team was using it anymore. My line manager left six months or three months after me and the head of channels had left six months before me. So after that there was no one in the company that knew about the project at all and when times are tight and everything, people go back to what they know. So that was a big lesson I learnt. Next thing is time. There's never enough time. I'm sure all of you know when you're working commercial production it's one job after another and trying to learn new things while working on a job is very difficult and often when things are very tight, as it often is, people will go to the tools that they know rather than the tools that they want to learn. So you need to make time and one of the ways I thought of doing this was to actually have a conversion team. So you're part of the studio working on day to day and the other part, maybe a much smaller part, working on just converting the workflows across to Blender. And you could do this in a number of ways. One of the ones was get the unit system sorted and the templates that you're going to use so that they exist in both packages so that when a new project starts and you need the logo, it's there. Same with the assets. You saw I brought the e-logo across but there's many more assets especially when you've got seven channels to look after. So all of those assets have to come across and materials have to be aligned. All projects. All projects are things that we would often... You have your end board straps and your now, next, later and all those things that you go back to often. Those will need to be converted to be able to be used in either package. And lastly, file formats. Decide on a set of formats that both packages can use and stick to them. You can actually document everything and too many companies have worked at nothing is documented. And that relates back to having a champion to the cause. Often when those people leave, you end up with these black boxes that no one knows how to use. All right, training wasn't a problem. That was actually quite easy so people weren't familiar with Blender but we subscribed to Blender Cloud and with all the awesome tutorials people got up and running with the interface and how to use it and people were often quite pleasantly surprised at how it worked. We started to have these sort of sessions, hour long sessions called playtime where you could just down your tools on your existing job and just play around. Build a coffee mug, look at this feature. Anything, it didn't really matter. Like a child, you just have playtime educational time. We also had weekly challenges so either weekly or early challenges same thing down tools for half a day, you'd say right, take a logo, animate it or model a character or anything, just to get familiar with the software and the interfaces. Other things which we didn't get to do and this is a really nice one is you create old work. So now you don't have to think about design you know what you need to execute because you've done it before, you just have to do it in a different program so it's very quick for you to now think about just the specifics of Blender and not necessarily anything else. The plan was to create bespoke training not just for the studio, we clearly have stuff that you would have to do that's related to ETV but also just for the wider process and the aim was then to release that into the community and of course the old deep end that's always the best way to learn. I find through experience throw yourself into a real project and you force yourself to learn because you've got a deadline so you often can then work at home especially with Blender because you don't have to pay for it. One area that I thought would be nice for us to take something from Cinema 4D is it's manual it's got a context sensitive manual so if you look here you can see there's the interface of Cinema 4D you could hover your mouse over any object or menu item or feature or tool tip or anywhere and you press control F1 or Apple F1 and a manual would pop up with the exact thing that you need to know about. This was hugely useful for people that were wanting to get up and running quickly so you could learn about something that you need to know now and get to know it. There are areas that we battled with the material system the cycles render engine it's a new fangled way of lighting and rendering scenes. Cinema 4D is still using its old system great flexible system but it is using an old methodology so I know I battled with it and some of the team did because you're going from that type of very typical and even Blender internal uses a very similar system to something like that so that was for concrete, painted concrete and it's very different and so it takes a bit of time to learn this stuff. The other area that we needed some work done or focus was on network rendering. At the time, network rendering is not as easy as in Cinema 4D. Cinema 4D if you look up close you have a queuing system that you could do on your machine or you could use a client server approach but both of them use the thing called team render so you would just have an ID and a password which you would then, the software would connect all the machines together automatically and whether you're using a client server or a queuing system you could just render so it was very artist friendly something that we'd like to see in Blender. Spline tools as well are a very important part of a motion graphic designer's job being able to manipulate splines very easily and this includes text which I think was touched on in an earlier talk. Text is very important and Blender has some spline tools and it's text tools but it would be lovely to see these being developed a little bit further and dare I say a NURBS implementation would be wonderful. From a user interface point of view I know some people there's always talk about the user interface of Blender but if you look between Blender and Cinema 4D they're actually quite similar. You've got a big 3D viewport top right here you've got your object tree in Cinema 4D and you've got a similar thing in Blender, same with the properties so that I didn't actually have an issue with. This was rendering was a big learning curve for us. ETV at that time we had Mac Pros which were great render machines but not when it comes to graphics. The graphics cards are terrible. Great for CPU rendering and then when I started using Cycles which just loves GPUs it would take forever. So some of those renders I made up there I would have to leave it over the weekend and stuff like that. This laptop which I got just recently showed me the power of a GPU so those renders that I had when I rendered them again were extremely quick. So where at ETV we would be doing stuff to reconfigure everything to a GPU based system. Motion graphics designers largely live in an Adobe world and in an After Effects world. After Effects is uniquely placed to be almost like an advanced video editor. There isn't an open source equivalent like this. Well none that I've found so far. So my big thing would be to get Blender to play nicely with After Effects. Maxon and Adobe were very clever in that they started to work together so they've built this very tight integration of the two packages. So in Cinema 4D when you render you can say you want an After Effects file and if you're rendering render passes it puts everything together packages it and creates pre-comps and then you open up an After Effects file with everything ready for you. And then similarly the other way if you're designing in After Effects you can then export your scene to Cinema 4D to do sort of things like 3D type and stuff like that. You could even create a Cinema 4D file from inside After Effects. And lastly my big thing which I had to learn was how to learn things. So things like cycles and some of the other new render systems have forced a completely different way of lighting and rendering and then your material systems change. So I'm along in the two 3D and motion graphics designer who's very used to the old way of working and it took me in my forties now I have to start to learn things a little bit sticky so I had to learn and I came up with that phrase but I quickly found out that someone else actually more illustrious came up with that phrase, Alvin Toffler who was a futurist and he very eloquently said that and I think that's sort of the take home that I would like to say is that we've got to learn, unlearn and re-learn things. That's my talk, thank you very much.