Imagined Worlds Project: Presentation feedback, and end of project summary

Okay, so lets stop avoiding the elephant in the room, it is time I talked about the feedback given by the lecturers about the project. My last post was before post-production, so I will put the links to the slides and the animation video itself at the bottom of this post.

We spent a lot of time trying to render out our animation on the day of the presentation, which didn’t really look good for us, but it was genuinely the best we could do. We then presented what we had, and overall I feel that the presentation could have gone better, because I was anxious and Ebonny ended up doing most of the talking, and I feel like I put too much responsibility on her there.

 

~ The Feedback ~

When it came to the feedback itself, I hate to say it, but I’m not surprised that we got the feedback that we received. It hurt a lot to hear, but honest feedback rarely means nice feedback. The only part of the feedback that got to me was the fact that we got a satisfactory for effort, it actually really hurts because we both put a tremendous amount of time into making it the best we could, it just unfortunately doesn’t show. Getting a pass feels fair, any higher grade would have been nice, but if I’m completely honest I am grateful for the pass. To look at our animation from an outside perspective, it doesn’t move great, and I get it.

The more personalised feedback is definitely going to help. Gary said that he didn’t know where the last two weeks of progress went, and that I completely understand, because the final animation doesn’t look too different to the blocking. I did work to my best ability to add as many frames in as I could, but in honesty, life got in the way a lot, and I got burned out after trying to rush everything to get things done in time, so by the time it came to making the animation look better, I was exhausted. This sounds like a terrible excuse, because there was plenty of class time to work on it, but this is just how it turned out. We also received some feedback on the lighting, which to be honest wasn’t rendered quite as we had hoped. The lighting did look nice on the test renders, but I guess we didn’t think about how it would obscure the animation and environment. We maybe should have just stuck with daytime lighting instead of making it dusk for the final render, because the dark lighting not only obscures the animation, but it also takes away from a lot of Ebonny’s hard work on the texturing and environment.

 

~ What I have learned ~

It is now time for some positivity, because I have learned an awful lot from this project. One of the biggest things that I have learned is probably to work in a bigger team, because in a small team there is still the same amount of jobs, its just that each member has more jobs. This sounds manageable in theory, but in practice it means that everyone gets burned out very quickly, and it makes it hard to recover. I’m probably the worst example of this, because I took on way too many jobs for someone who doesn’t really know how to do any of them, and ended up burning myself out on tasks that I wouldn’t even have chosen to otherwise do / be graded on. Another important thing that I have learned is that it is not as easy as it sounds to catch up on missed work time at home. If I’m honest, I occasionally wasted time at college, and it became hard to catch it up at home, and I ended up pulling a lot of all nighters. Not only is this kinda unhealthy, it also meant that my productivity became worse at college because it made a viscous cycle of having no energy or motivation, then having to catch up fast and burn more energy. This is a selfish work ethic, and it ended up being unfair on Ebonny, and she didn’t deserve that.

 

~ What I would do differently next time ~

For the synoptic project, I definitely want to work in a bigger team. Working with Ebonny was a pleasure, but if we had more members, there would be a more diverse skill range, and that is really what we were lacking. It meant that we both had to spend a lot of time trying to do things that we weren’t good at, and if someone with more knowledge and skill had done those things, it might not have taken so long. This sounds a tad defeatist, but I think for the synoptic project I might chose not to do animation, because I’m not sure its where my talent lies after all, and I think that its currently the wrong time to push myself at something that I don’t enjoy. It seems like a better idea to focus more on areas that I want to do, and I’d rather change my specialism to concept art, because currently I just find that way more enjoyable, and I’m less likely to get annoyed or stressed/burned out doing that. Lastly, this sounds bad, but next time I definitely need to spend more of my in college time on work. I got easily distracted during this project, and that contributed to things taking longer than they needed to, so next time, I will definitely be more focused in college time, because pulling all nighters to catch up on the missed time created a terrible cycle of making more work for everyone, and was ultimately kinda selfish. I think I was a bad team mate.

 


Important links:

The Google Slides Presentation

The Animated Short on Youtube

Advertisements

Creative and Technical Model Submission

After many weeks working on my Creative and Technical model, it is finally time to submit the final product.

Lilith Yunalesca – Creative and technical model

If the Powerpoint isn’t working, you can view the Google Slides here

Overall, a lot of things went well. The process of designing the character was perhaps my favourite part, because I really enjoy designing characters, and collecting research images was really fun once I had settled on a theme. I feel that mostly making the low and high poly models worked out well too, but at the time it didn’t feel like it because it was a super slow process, however in hindsight it was one of the more enjoyable parts of this project, and I didn’t run into any technical issues with this. A more recent thing that went well was baking ambient occlusion, I’m not sure exactly how successful this was, but it feels like a victory because it’s something that I sought out knowledge for.

There were many obstacles throughout this project, most of which I had never encountered before. One big example was baking the textures from high poly to low poly. The big issue was that the textures just wouldn’t bake right because the way that I constructed the models made it so that it the silhouette of the high poly model was quite different to the shape of the low poly one after smoothing. This meant that the textures baked super weird, and the model looked awful. The only way that I thought to overcome this was to just use the high poly model as my final model, and add the textures onto it, however, this linked into my next big problem, which was that the textures didn’t export right because I was missing the necessary maps. However, I overcame both obstacles with the same solution! In the bake settings for my model, I selected the “use low poly as high poly” option, and baked the missing maps. It was then just a case of plugging in the right models in the right places, and boom, it was fixed! This is technically my first success at baking with my own model, so overcoming those problems felt amazing, and it meant that the baking part of the pipeline wasn’t missed out after all! The last big obstacle felt the worst, because it was that I had to re-do most of my renders after failing to realise the textures hadn’t exported properly. This probably could have been easily avoided had I just paid more attention, but I didn’t realise that it hadn’t worked right until someone pointed out that the detail in my substance file look different to my renders. However, I overcame this problem by firstly, overcoming the first two obstacles mentioned above, then secondly, re-rendering and replacing what didn’t look right. I ended up learning how to render directly through substance painter, and this makes the renders look way nicer than rendering in Maya did, so this felt like a win win! I then just applied the fixed texture maps to my Sketchfab model and turnaround model in Maya, then batch rendered the file again and re-compiled it in Adobe Premiere, and everything looked much better than it did before.

If I was to do this again, I think that I would do everything differently, because way too much went wrong, and it ended up making everything behind schedule and harder to get done. One of the biggest problems was taking too long to make the low poly model, and not devoting enough time to the rest of the pipeline, because had I given everything an equal amount of time, the last tasks wouldn’t have been so rushed, and I probably would have payed more attention to the small little things that I did wrong, for example, the high poly model having a pipeline-breakingly different silhouette, or the textures having been exported wrong and wasting an important day’s worth of rendering. I had planned to get a nice set of action renders of the model in all its glory, however with everything else that had gone wrong, I couldn’t chance rigging the high poly model to the existing character rig, and so the posed renders that I got were all of the low poly model. It would have been nice to compile them together as I had planned, but I did not leave enough time, so if I was to repeat this project, I would definitely like to at least change that!


Useful links to things from the powerpoint:

The Google Slides document

The model turnaround Youtube video

The model in Sketchfab

Creative and Technical Model: Rendering!

It has been a long journey, but today I finally got my renders sorted! I spent all of today making some renders of my model, trying to get some cool artsy renders. Since she is a rigged character model that moves, I decided to try some poses out so that if I had the time, I could maybe make some professional renders with action poses!

 

I then put them together in some professional looking renders! I wanted to make a set of action poses, and an artsy set with textures maps and aesthetic render zoom ins, but unfortunately I had to scrap the action pose set because I ran out of time! Anyway, here is what I have!

Lilith professional render + text

I also took some time to render out my model spinning around as a model turnaround. I keyframed her rotating around so that she would do a full spin in 4 seconds, then looped it for 20 seconds and added a fade out effect in Adobe Premiere. I also uploaded it to Youtube, so you can check it out here!

Overall, this has been a pretty arduous days worth of rendering, however now that it is done and out of the way, it feels like a big weight off!

Creative and Technical Model: Schedule week 4 – Last remaining week!

Today marks the fourth and final week of model production in my last month until the deadline.

Lilith production schedule 5.12.17

So over the last week, I had planned to sculpt my model in Mudbox, however, Mudbox kept crashing, and I eventually lost patience with it and decided to keep my high poly model as is, because with having a small amount of time left, I didn’t want to invest lots of time into something that might crash again and waste limited time.

I decided to focus my time into baking and texturing. While baking my model, I encountered a problem. The smoothing on my high poly model changed the shape and silhouette of the model. It only made small shape changes, but the changes were too significant not to cause baking problems. It was a badly informed decision, but I decided to just use that high poly version of my model as my “low poly model”, and make a model that was slightly more smoothed as my “high poly”. I checked through my models, made sure the naming conventions were correct, made sure that everything had the correct materials assigned for texturing, and everything looked right. This too caused baking problems for some reason, so I decided to stick with my higher poly “low poly model” and just keep it as is.

I then started applying some textures to my model! I added some base colours to each necessary part, then added some more specialised textures, like specific material textures, to each piece of clothing.

lilith_tpose_render.png

Overall, I am very happy with how she looks. I’m annoyed that I couldn’t put much else into the high poly model sculpting wise, and that my baking didn’t really work, however I like how it looks now, and I’m happy with the texturing!

 

Imagined Worlds Project: Animation blocking

So it has taken a while, and it didn’t feel like any of the collective small amounts of progress were worth posting about, but I finally blocked out our animation!

 

 

It took quite a while, but by using Ebonny’s animatic as reference, I blocked out where the characters need to be for each part of the animation. Nothing went wrong while doing this, to be completely honest! There was some technically difficult things to figure out, like keyframing the visibility on Zach, and making sure everything was clamped and stepped (because for some reason those settings just wouldn’t stay right, and things kept switching back to splined).

The good news is, the models will definitely be moving for the MVP deadline check! The bad news are that we are not really where we should be, but hopefully it will at least look pretty good for the overall deadline, after all, it’s mostly about the world and sound design, but I’m determined not to let the team down!

Creative and Technical Model: Schedule week 3 – 1 weeks remaining!

Yesterday marked the third week of the model production schedule in this month before the deadline!

Lilith production schedule 29.11.17.png

This last week was supposed to mark the high to low bake stage of the pipeline, however circumstances have made it hard to meet this, so I’m slightly behind schedule!

Currently I am working on my high poly model. This has mostly been a case of adding in edgeloops and using the crease tool so that smoothing the model will work right. So far, I like how this has turned out, because she looks very smooth and nice! It has mostly been in preparation for taking her into mudbox, because it would be nice to sculpt the model.

My next steps are to take her into mudbox to get some nice sculpting on the model! I am hoping to do this tomorrow so that I can start baking and get back on schedule!

Imagined Worlds Project: Life imitates art!

Over the past couple of days, I have been continuing to skin the models for our animation! With Lilith complete two days ago, I set off with my newfound rigging skills to attempt to speedily and efficiently skin Zach.

To be honest, I didn’t use any new techniques on this. I duplicated the skeleton that I made for Lilith, because when modelling the characters, I used the same body with the intention of making rigging easier to save as much time as possible. With the duplicated rig, I then skinned Zach’s character model to it. I attempted to use the interactive skin bind with capsuling like I did for Lilith, but for some reason it didn’t want to work, so I manually painted the skin weights for all of the muscles myseslf. This took quite a while, but turned out rewarding, because I’m happy with how he moves! I then decided to pose the models and get some renders of them in a pose inspired by some concept art created by Ebonny.

Rowdy_Kids

 

Bros_dabbin'Bros_dabbin'_zoom

Overall, I am over the moon with the progress that I have made in this project. This experience has been invaluable, because rigging is really difficult, and now I can skin two models in the space of a week! Now that the models are skinned, I can make the control sets, then start animating! There isn’t much content in this post, but it felt worth blogging about because it’s progress!