OdinsEye2k

Understood - I had tried to do some of these things myself (adapting the topologies I've seen the Mayans use on CGtalk and a couple of techniques from Stahlburg's foot tutorial), but having someone who's practiced in that style of working apply it to Hash patches is fantastic for learning. 'Maintain curvature,' eh... I guess now that the animation features use biasing correctly (no 'popping' when crossing the 0 line) that also changes the game by a fair amount. Can't wait to see what you would do with smartskin on this guy with these insanely large muscles. Talk about flexin'

A great looking model so far. It's a type of sculpting that I've seen few people post around here, the way it jumps from one topology to another (i.e., the thick muscle bands on the upper leg, moving from a squared groin/waist region up to the ellipsoid abdominals). How are all of those three-pointers you've got behaving so far? Animation-wise, I'm looking at a couple of concerns (which you may already have ways around). One is in the deformation area between the leg and pelvis only having one control loop, in addition to a high aspect-ratio 5-pointer in that immediate area. Are you planning to control this via biasing once the leg starts lifting? Either way, this thing is solid - and a great inspiration for some tricks to try on my own particular model.

Yeah, I guess that's a way to do it as well. I think part of the reason was that I also had a bizzilion objects that I didn't want to activate/deactivate all of the time. Of course now, I think you have an on/off pose that activates/deactivates the entire set - so problem again solved So nice to have all this options.

Funny, Parlo, I feel the same inner conflict when I see NASA-rendered footage for various things. On the one hand, I know that they look a lot different from Hollywood stuff because they are somewhat based on real data. But they just never seem to convey all of the awesome forces going on in some events. They're actually pretty darned sterile, which isn't right either. Back to the thread at hand, a fair amount of camera shake and shudder would be pretty nifty for the rocket, Javier. Most rocket launcher vibrations are a mix of sine waves that occur in the 10-5000 Hz range.

If you want to speed up your render and animation times in a complex scene (say the warehouse in Killer Bean 2 or a forest) - first make yourself a much simpler version with key locations highlighted. Then animate your characters to heart's content. Once finished, bake the resulting choreo work into a single action for each character, and then slap that on your characters where they are in the far more complex scene. That allows you to animate without destroying your processor in the process for complex sets.

Actually, aside from shaking the engine a lot, you probably wouldn't see a whole lot of volume change in the plume (unless, I guess you *really* changed the throttle). You tend to design these things to be as steady as possible - every variation in pressure with time or position in the flow is more efficiency lost to your engine. Not to mention tearing the hell out of your combustor and being even louder than it would be normally. Hotspots would tend to show up as really bright streaks compared to the flame (see the Shuttle Main Engines at like T-2 seconds when they go from 60-100% throttle) A space rocket would be even more placid than a jet engine, I would think, since you don't have so many design points to try and satisfy (take-off in thick air, cruise in thin) or inlet flow distortions. You basically just have a bonfire with a nozzle attached. Or, in the case of a plasma rocket - a microwave with a magnet attached. Even though a fighter jet engine is super-powerful, you actually rarely see the flame change volume. Heck, even the beads that show up in the flame don't move that much IIRC.

Pretty slick - don't suppose you'd care to share the method with us (although I will admit a vested interest for the space animation contest, so I'm more than understanding if you want to keep your edge if you are entering as well) ? Also, I have to give technical nitpick. That looks like you have a diffuser/fan on the front of your rocket. Plasma rockets usually don't have those - but they do tend to have big magnetic rings to form a magnetic nozzle, which are even cooler IMHO. Not to mention a microwave generator that would do a potato in miliseconds

Thanks for the comments so far, everyone. I've tried to incorporate all of what I have heard so far. There are new updates - plus a current revision of the model itself on my site. It is currently undergoing some basic rigging to check out how well it will animate. I must say, it's looking pretty good - little smartskin will be needed thanks to the rigging setup. So, please, check this out: http://www.prism.gatech.edu/~gtg724n/common.html And pass me back some feedback. I really would like to finish nailing down this "base mesh" with well-defined muscle structure so that I can start building the deformer poses. More or less shrink/scale the various components and proportions at the bone level, and then another set at the mesh level to deal with flabbiness/skinnier/bulkier limbs. Oh - almost forgot to mention - the face itself is not a point of this mesh (unless it is lacking in the spline definitions needed for altering) - it is mostly a blank slate from which folks can do their own deformations. I will leave it unposed and undeformed to allow for more unique characters. Thanks all - I'm hoping the actual model will provoke a little bit more criticism/critique.

Very nice. I think I forsee more tomatoes getting perfectly sliced than during a knife demonstration at the county fair.

Well, it's no java_bot, but I did compress the image load down a bit by offering links from the overall pictures to the close-ups. And I can't do HTML in my sleep - that space is reserved for the non-constrained optimizer I have to write in Matlab this week

At first, I thought that was kind of crazy, Iham, but thinking on it for a while, I guess there really is nothing mystical about that 90-degree turn. I also like the Imagemap idea to pop up various parts - I think I'll try and use it for current and future postings. It'll probably be a while before I can get this going though - I'm at lab all day today

Hey everyone, A while ago, I mentioned that I was working on a pair of base meshes for my own projects, and that I also would be willing to share with the community at large. I have a full body finished now, but there is still a lot of tweakage to be done. I've been looking at this for a while now, so I was hoping to get some fresh perspective other than the random references I've cobbled together. Please visit the page and let me know what you think and what features you think would be useful in a basic mesh of this type (in terms of being able to stretch/shrink parts of the body, change levels of musculature, whatever). -Edit: Still graphics heavy, but the initial load has been reduced by a factor of 2 or 3. http://www.prism.gatech.edu/~gtg724n/common.html Thanks in advance, Bjorn

A beautiful render, indeed. You say that the GI helps to warm up the room - does that mean that GI now includes color with the photon bounces? In other words, the lighting from GI should now automatically match the 'temperature' of the room depending on the dominant color, yes? I.e., an orange room would have 'warm' lighting and a blue-grey room would be 'cool.'

Ichi looks like he's trying to be in the stealing position - if you're really going for this, take a look at some baseball pics. Base runners tend to carry their weight fairly low and on their toes. I'm not sure about his stance in real life, but for most players, they tend to hunch over, almost like a baseman or tennis player, because they are really doing the same thing. The goal of the stance is to be able to quickly direct their motion in either direction, to avoid the pickoff or to start their stealing run. Of course, Ichiro doesn't obey a lot of the mechanics of baseball (heck, watching his swing, it barely obeys the laws of physics) so hard to say.

Now with hair and a bit more glint in the eyes... I'm currently doing a bit of smart-skinning on the upper half before moving on to model the legs. My model-and-smartskin parts in separate files scheme kind of backfired on me - the smartskin's went looking for the original CP numbers rather than the new ones the mesh attained when imported. Luckily, the fix is doable in WordPad, which, while tedious, is far superior than trying to remember how I pulled things off the first time. Anyhow, the new pic with a new hairstyle.

Are you refering to the lip color or the splines themselves? I did the color map following a cgtalk tutorial posted there - a couple of tricks that came in fairly handy: Face tutorial by 'cANT' As for the splinage, I just drew out the larger shape and then extruded inwards. And of course had to chose where the 5-pointers would go around them, but that's more of a orbis/depressor (that junction where wrinkles show up in old folk - also known as 'laugh lines') issue. The only real trick is to get that bump on them right, which is a fair amount of tweaking and looking at your own face. I found that it's a little sharper than you might think, and the bump usually is more toward the outside of the lip.

Thought I'd move Maggie over to the WIP so I don't spend so much list time. Warning: graphics overload (4 shots). She's become very soft, almost cartoony now. I'd kind of like to try and push her back to a little more realism. I've tried using specular highlight to get the glint in her eyes to become alive, but it's not really working. I'm going to try a clear overlayer tomorrow to see how that works. Any other hints would be welcome. I did move the decal a bit and it does improve the look. She's got an extra ring on the lips now, and the upper half of the jumpsuit is modelled as you can see. The arm is attrociously dense, but I'm going to need it when I start wrinking the fabric around the arms and the shoulder region. I'll probably use displacement maps for the smaller wrinkles on the body. Still haven't decided on a good hair-do - long or short, I wonder? I could see it both ways, but I think the long would end up in a pony-tail, probably with a few good strays up around her eyes. Anyways, thanks everyone for the comments so far. That and adding bones as I go (I found once a model's done, the motivation to rig up goes away) will help get this one complete. Then to her partner-in-crime. Too bad the 'reflections' contest is already over... coulda had fun with that one. I'm going to try and figure out how to make the artifacts on the collar go away - dunno if it's porcelain trying to make too steep a turn or not yet. If you want to see any more views to be able to better crit, please let me know. Bjorn