pleavens

VERY COOL! Your displacement map is causing the gap, as you thought. You could fix this by having a band of RGB 128,128,128 that runs directly down the edge, and is feathered into the rest of the map. Phil

Oh goody! something to look forward to after work. Thanks, Aaver! Phil

What codec did you use? my Windows media 9 player was black from end to end. Phil

So do you have blueprints to work from, or are you interpreting this as much as possible from existing photos? Phil

We already have that with action objects.

You were correct in thinking that there is a problem in the program. There's no reason we should have to "work around" it. So I think you should still send it in to Hash to have them look into this offset key setting thats going on. Another thing I noticed, the option when right clicking the pose for "settings" seems to have departed. Phil

Please re-read my post. I never said that I did your tests in the constraint relationship(s). Everything I pointed out was done from testing in the actions you provided. Here for instance, are the keys I mentioned in the second objects constraints relationship. (this is from your latest project version) And I still don't understand why you are pushing the "force keyframe" button in this case, as the keys are set when you do the rotation itself. I've posted a modified version of your project with the things I found cleaned up. I DID find it very difficult to set the key values in the relationships properly. I had to delete the existing keys and manually assign the value in spline mode by changing the cp value directly. Otherwise AM was placing the value WAY down the timeline instead of at zero, as you can see in the above image. This is a definite issue. Modified project Phil

Action1 seems to work the way I would think it would. Rotate "Z" causes the second (hidden) bone to rotate "Z" at 50% of the amount applied. I used the "rotate manipulator" to do the rotation, to avoid any input errors. Action2 has a constraint applied to the model in its relationship which changes value over time, I believe if you were to remove the keys which cause the value to change, the issue you see will clear up. As keys are set by doing the rotation itself, I am confused by your statement about "hitting" the key button repeatedly. Phil

Another update with both eyes now having distortion driven poses. Frogger two eye "pop" movie For more fun right-click and "save as" to your desktop. Scale it double or larger for a fun view. I left it looping scaled to maximum, and my cat went nuts watching it. Phil

Ok, nobody liked that one I guess. Here's a new distort test with some additional spazz out added. New Frogger distort movie Tough crowd!

Thanks, here's a link to my first test of using a distort in a pose. So far only the right head/eye bone combination is done, but I hope you all get as much of a kick out of the test as I did. Frog distort test movie Action object distorts are cool! The eye is controllably scaled while still letting the underlying scaled bone movement occur. And it was easy to setup once I figured out the "two-boned" head layout. Ok, on to the other eye, and then the limbs and body. STRETCH me till it hurts, says the frog. C&C anyone? Phil

Select "k" Select "k" Select "k" Select "k" Select "k" Select "k" Select "k" It'll be done before ya know it. The other option is to select the side you want to keep. And hit "copy" Delete the whole thing, then paste or "paste with bones" the half back in. Warning, if paste offset is not "0" in OPTIONS/UNITS your pasted model will have moved from it's prior location. Warning two, Be sure "mirror mode and magnet mode" are OFF when you do the selection. Warning three, sometimes your 5 point patchs will need to be re-selected and applied. Warning four, sometimes if there are complex materials applied to a group within the selected group, your copy will "hang" All of these are much slower to talk about then to do. Once you've been modeling for a while you won't even think about it. Phil

Your absolutely right. I've since added a specular "shell" on the front of the eyes with "eye lights" to define the specular component. The original post image has been updated with the change. Thanks for catching that. Phil

Here's an update. Now boned with simple fan constraints and some texturing work. Still struggling with combining distortion boxes with the existing bone structure, but I'll post more when I get there. C & C all appreciated, I keep feeling that something is "just not right" (that should be my trademark) about it. <edit> Updated with speculer layer added to eyes </edit> Phil

It's about time I learned how to do some character animation with lot's of "squash and stretch" (some would probably say period) so.. here's my first attempt at a character that I think could be a natural choice for such abuse. My idea is to assign animated distortion boxes to all his main components. Arm and forearm bones, eyes and eyesockets etc. I have no idea how well this will work, but it's something I've been wanting to try for awhile. No bones, other then for distorted eye's, or final textures at this point. The Darktree on the main body is going to have to go as render times were ridiculous. Comments criticisms and whatnot all appreciated. Frog Spin Movie (sorenson 3) Phil

Hello everyone. Someone in HASH community chat asked me about a constraint system to allow both suspension and steering for an automotive project. I'm sorry but Hash Chat went down for me and I lost the connection. And I'm embarrassed to admit that I didn't really pay attention as to who it was that was asking. Anyhow, here is the project if anyone might need something like this. Zipped Project (bones only) Hope this helps someone

A good idea, but first I would like to see some success with hair collision. I like the shape you get from lower constraint values combined with collision being on. Not to mention that Gala looks kind of funny with hairs sticking through her face and ears. Unfortunately there seems to be an "undamped" quality to the hair dynamics with collision on, that totally ruins the intended effect. I am going to start some simple tests to see if I can isolate the problem areas. Has anyone had success in this area? - If so, I sure would like to know how you did it. Just had an idea... I will try a composite render with "shadow" on in the hair emitter, using the final render single-pass technique. There's a possibility that I can capture the shadow data somehow. (shadow only, or to the buffer) I'll keep you posted. Phil

I've done some more testing with interesting results. New Shaded Test I accidently discovered that lit shaded hair now seems to render. As it never would for me in the past, I am at a loss for an explanation. Hair render type comparison Here's a group shot showing the results of hair rendered using some of the various techniques as well as the total render time necessary to complete them. I put more time into her hair grooming, and used some force and collision for the first time in these tests. The next step is to see how well the collision dynamics work over time. Phil

Ross, you are absolutely correct, unfortunately the shaded render mode does not work with ANY lights in the choreography. So currently, there is no way around it that I am aware of. Of course for stills it's possible to wait overnight (or longer) for a final render, where this really pays off is for animations where the lighting issue will be far less noticeable anyway. Yes, as it's only being used on the black-out copy of the original model to do the shaded pass, there is no need for any of the "hair stuff" to be active on the original. Phil

Vern - It would be nice. But, if it's there I can't find it. Jim - I will try to do something in the next week on this. Here's a new still with the hair density increased to 300% and it's thickness reduced. Frame 13 of the animation, as it shows the hair detail better. Frame 13 full still 1024*768 Hair pass completed in about 2 minutes. It was about 30 sec at 100% density. Phil Leavens

Ok, the video version is now available. Total render time was about 30 min for the hair pass and 3hrs for the final VGA output at 9*9 multipass using a Jeff Lee animated skylight with one additional keylight. 1.4 gig machine with an OEM Radeon 9600 graphics card. 2 Second Shaded Hair Test Video And the still Phil

It's fairly complex, but if there's some interest I'll try and produce a tutorial on the process. Short answer is... Render hair shaded at as high a resolution as your system will allow, (in this case 2000*1500) with NO lights and nothing that will cause it to anti-alias. Use a copy of the model that you want for the final render, with any openings stitched closed, and all surfaces black with no falloff or specular component. Any models that will intersect (cover up) the hair will need to have this process applied. Set the camera to have a solid black background. The hair material cannot have transparency, or be tapered to zero at the ends. Also, using pure black for the hair color will create problems. Batch process the resulting sequence to remove the axis symbol from the left corner by painting it out black in your favorite image editor. Import the edited sequence as an "on top" camera rotoscope into a copy of the choreography. Make sure the "key color" is set to black and then switch models to the originals and add lighting and such to taste. Final render the composite. ------------------------------------------------------------------------------------------ The reductions in final render times are fantastic. The hair render for the animated test of this image took only 30 minutes for 60 frames to render at 2000*1500. Phil Phil

Hi all, I've been continuing my tests with compositing shaded hair and final renders directly in AM - no alpha channels or fancy software needed. Here's the result of my latest process, the TOTAL render time for this 1024*768 image, was 6:30 on my 1.4 gig machine. That's 6 minutes, not hours. This is frame 44 of the test animation which uses dynamics for the hair. I'll post a link to the final VGA animation later. EDIT - Link to VGA dynamic hair video 1.03 megs Quicktime UPDATED to high density hair. Link to full size version of still Phil

Do the tutorials from "The Art of Animation Master" - watching them on video first is even better. HASH online training videos Many new users skip these tutorials because they want to get straight to the "good stuff" and think it's just a bunch of boring exercises. This is especially true for users coming from other 3d applications. Don't make this mistake, their not perfect but there is a ton of detailed instruction on using the interface, usage of the modeling tools, particles, constraints and more in these seemingly simple tutorials. I would have killed for them back when I started with AM if I knew their value. In this same vein, AM provides you with a host of simple yet powerful tools that will enable you to tell your story. The easiest way to find yourself going nowhere with it, is to try to force AM to work the same as other packages you may be used to. Take the time to learn the usage of decals, smartskinning and the art of Hash patch modeling. Approach these with an open mind and a willingness to learn a different way of doing things and you may become one of the many who wonder why the other packages can't do things the AM way. Good luck, and feel free to ask for advice on this forum, there are a lot of people here who enjoy seeing others have success with our favorite animation package. Phil

Ask and ye shall etc. I notice that the particles are still rendering the toon line around their patch edge instead of their alpha channel edge. I wonder if this has ever been reported, anyone?