Rhesus Monkeys at the NIH

I've been working on a project at NIH for about half a year now. The short version is that I am creating a digital Rhesus Macaque monkey to show to infants to aid in cognitive responses to facial expressions. With that intro I'm now going to just mirror a post I am making on my science blog (there is no real science, but it's all science related. I'm just an artist, not a scientist). Quickly though, I want to note that this amount of time is due to me redoing the entire project about halfway through, that and spending way too much time trying to figure out fur in maya.

The project is finally starting to really come together on my end. Rebuilding the model was a very good move, I'm much happier with the model now and I'm having more success with how it moves. I decided to attack fur one more time since I have a new improved model. The hope was that the fur would magically look better on a better model that had a better texture underneath. This turned out to not be true, however in my new attempt I learned a few new things, the most important of which was that UV size determines the density of the fur. This was the reason I was getting awful looking patches of fur around the shoulders. So I redid the UV map which was kind of a pain because it also forced me to bring the new map into Zbrush so I could export new normal maps and texture maps.

Texture:



Normal:



So, with those fixes the fur started to look better. I'm still facing the issue of a weird hairline around the face. I am considering adding one more fur description just to define the hairline and make that work a bit better but for now I've moved on to blendshapes and rigging to test how the fur looks and moves when I animate the model. I have yet to rig but my blendshapes are finished. Another problem that I faced was with fur jittering but I found out this can be fixed by simply creating a texture reference object. This freezes the placement of the fur so it all is more or less baked onto the model and doesn't float all over when the model moves.

So here is a clip of my first few blendshapes, I've added a few to the rig but they aren't seen here. This is still a preliminary render. Not the final:

A few more speedsculpts and a busted power supply.

To those of you who follow me on twitter, you no doubt know that I've been hit with a case of power supply (psu) death. Fortunately I bought a Corsair and they have a pretty generous 5 year warranty. They are replacing the piece for me for free, no questions asked. While it's wonderful that I dont' have to buy a new one, I still have to wait for a new one. Well, this has been tough because it turns out my Zbrush addiction is worse than I had realized . Really this is probably a good thing, but still, it's been a boring few days at home. Kinda nice I guess to be able to just play video games, and finish my Lost re-watch without the guilt of knowing I should really be working on characters, but I miss it and I'm really dying to get back to it.

This happened on Saturday so I was unable to complete the Ana model in time for the release of season 2 of Azure but I'm sure we'll figure out a way to use the model.

In the mean time I have been able to just do a few speedsculpt warm-ups at work. I'm nowhere near my goal of doing one a day, but at least I'm still getting a few done.

So, please enjoy, here are two new ones. Both were approximately 45 minutes.



and

How about another speed sculpt. This one could use more work. Perhaps I should switch to a 60 minute sculpt time but I tend to feel like I hit a point where, if I start to push it farther then I have to really double my time to make the rest consistent. Anyway, the point of these is to just generate concepts and loosen up. So perhaps I'm thinking about it too much. Please feel free to comment with any thoughts.

45 minutes:

Two More Speed Sculpts

Well, I've been doing well in keeping up with these. I think I'm at 4 of them in 5 days, and that includes a weekend where I was hardly home. So, here are the next two:

50 minutes












45 minutes (First speed sculpt from Zsketch instead of a sphere)

Speed Sculpts

Well, here's my first personal 3D related post. Something that I see a lot around sculpting forums and modeling forums is speed sculpts. Basically you allot yourself a relatively short amount of time, anywhere from 30 minutes to 4 hours, and you just do as much as you can in that time. And this means working from scratch or from primitive geometry (square, sphere, etc.).

To me this is really a great practice for multiple reasons, and to be clear I'm currently just beginning to try to actually make this a habit for myself. First of all, sculpting this way really forces you to get used to working without wasting too much time thinking too hard about a stroke or getting distracted by any other possible distraction (chat, facebook, twitter, tv, video games). I know that focus is one hurdle that I've yet to fully overcome so I'm hoping these sculpts help to increase my discipline. Secondly, it's really a fantastic way to just generate a ton of models and build up a huge concept library. The purpose of these is not to have a finished piece, but to just quickly put to screen a new character or bust. Outside of technical skill, I feel that concepting is about the most important thing for an artist. You have to be able to do more than just build models. I certainly don't mean to imply that there is no merit in working off of others concepts, I do this frequently, but I find it important to actively create concepts as well.

So anyway, I'll try to post a good number of these on here, always using the label "speedsculpt" so at anytime you can look through all that I have done.

For now, here are my first two:

Time: 45 minutes











Time: 50 minutes