robcat2075

Its a nice shot!. How about moving the tall candle to the left or right so we can see the flame against the dark sky rather than the white window frame?

I think more R&D is needed. While they are more intimate in the new pose, the humanoid appearance isn't reading well. Perhaps something with two curves drawn together like two parentheses: )( or maybe () It's a tough problem. But I know you can do it.

It looks good, but I must be missing something in the Cog tutorial because when i move a shoulder up high like that the rib area comes way, way out to the side, even on the finished sample.

Yeah, it sure does seem to be bouncing fast, but I don't have tennis ball here to test. Maybe you based this on a video example? If you're going for VERY realistic, a tennis ball wouldn't elongate on the way down, however... Richard William's "Animator's Survival Kit" book has a spot where he talks about this very exercise and how you can "cheat" reality for a better effect. Great book.

Cute shot! If he zig-zagged instead of coming straight down , we'd get a better view of his pedaling action. Is he on a path constraint of a surface constraint?

Thanks for watching and making comments! Well, my hope was that he would appear to be listening during the pauses, then reacting to what he heard. But if it just looks like arbitrary poses strung together then that wasn't successful. You're right. The one thing I changed between the contest version and this post was to make him glance at the phone, but that was a bad idea. Indifference was better. Don't they say "If you blink, you lose"? The file was too small... or too big? Working toward my diploma from the "Make'em See It In The Back Row" school of acting.

Well, if you're just looking for "problems" here's what strikes me initially... -When he drops an arm down to his side, it seems to jump into place rather than ease in. -The transistion from shot 1 to shot 2 is an awkward jump cut because he's suddenly changes the phase of the walk cycle he's in (from "left foot forward" to "right foot forward") and because the change in camera view is too slight. -Likewise from shot 2 to shot 3 in the position of the right arm doesn't match up. These are things live-action filmakers go mad getting to match up because they really have to capture each shot separately and make them fit in editing. But for animation we can more easily do this as one performance in one chor with multiple cameras and get perfect "cuts". I think doing this in 3 chors is over-engineering it. -I don't see any action/reaction between the masses of the arms and the body, particularly when he tosses the object. The result is a rather stiff appearance to the action. My hand is small compared to my body but when I swing it out my body is still nudged in the opposite direction, and my body comes back when I swing my arm back in. His hands are huge in comparison to his body. The whole piece is really about the arms it seems, so you want to make them great. -many of the head motions have a linear quality to them, making for a jerky effect. There's also an odd jump right at the beginning of shot 2. -A head that huge would benefit from some anticipations on the big moves. That would take experimenting, it might mean a completely different performance for the head. This would be a difficult character to animate because the proportions are so extreme. The limbs are so short in comparison to the hands/feet that "successive breaking of joints" is going to be tough to fit in there, but that would be another thing to implement if you're going for broke on this shot. (I realize this is a character you were assigned). But I think this will be an effective sequence. I do miss the part about the loofa sponge, however.

From the four-hour Animation Showdown, my finest achievement in grotesque overacting so far: Fated Phone Call! (QT 217K) bonus animation: Dig (QT 85K) See if you can tell where my four hours ran out. Suggestions are welcome.

Is there any way to do that in plain-vanilla HTML?I'm not an HTML guru but I suppose the HTML work-around would be to display the icons in one HTML frame and have clicking on one load the related info into a nearby frame. DHTML, java or javascript would probably come closer to the behavior i was thinking of, but there are compatibilty problems across browsers. Flash is really the way to do stuff like that now. It's great for graphic interfaces. Everyone and their dog has the plug-in already. But then what would I put for "Star Wars 1.5"; "Officially Sued by Lucasfilm"? "Really Nice!" - Mom

That's a lot of great content on your site. Content is king, they say. My complaints : The text notes on each film don't seem visually connected with their picture. I think center justifying them is what's doing it. They're sort of saying "I'm by myself, pay no attention to the picture tagging along onthe left." Also the fact that all the text is the same size (the title is no bigger than the date of the film, for instance) prevents the eye from easily distinguishing one film's info from another. And to do varying text sizes, you'd have to ditch the Courier font. Courier is rather retro, but not in a good way. I think you should go with some san serif font that works with the one you've used for "Featured Films". Just from a web standpoint, I'd wanna redesign the page so that no scrolling is needed. Aside from the enabled scroll bar on the right of my browser window, there was nothing to make me think the page had anything below "Duck Sauce" on it. My first thought is to design an interface that just shows the pictures, and when they roll over the picture the pertinent info for that film is displayed in some comon area on the page. Clicking on the picture would take us to the actual film. There are lots of different visual metaphors that could embody that concept. Each film could be a position on an old-fashioned TV channel knob. The info could come up on something that looked like a "TV Guide" listing. I'm sure there are even cornier ideas than that. Whatever, the idea is to get the whole design visible in one window. Yes, web users know how to scroll, but don't make them scroll if you don't have to. Your films are so good, we don't want any of them to get missed. BTW, "Duck Sauce" is an "Official Selection, Portland Indy Animation Festival". Put that between two laurel leaves in the "awards" spot. Disclaimer: These opinions are coming from a guy who for two years did all his Flash work in gray tones, just to see if anyone would complain.

Thanks Yves! Are there any schemes on the horizon for shortening the render times for this sort of stuff? Also, it sure would be neat if we could do one render of a scene and bake the results onto the patches. I realize that wouldn't do for scenes with objects moving around inside them, but architectural fly-throughs could use it ok. Or is that something we can already do?

How about a sprite that looks like a solid particle instead of a fuzzy one? You might model and render a tiny crumpled cube and use that image as the sprite.

Thanks, Yves. I'm having some trouble with the spreadsheet, however. It opens up ok, but when I put new values in the white boxes, the "Suggested Sampling area" result always shows "#NAME?" which means excel "doesn't recognize text in a formula." The cell contains this formula... =MROUND(SQRT(D8/PI())*100,50) I'm on Excel 2000, which doesn't have MROUND listed as a function. If I subsitute ROUND the result will produce numbers again. Is that an appropriate substitution or is MROUND something completely different? thanks,

Cool lighting! Is there a sample Cornell Box project available anywhere with all the settings in it to get results similar to what is shown on Yve's discussion of Photon Mapping parameters? I've tried making my own but always extremely grainy results no matter what my settings.

That's a neat piece. As far as the walking motion of the fingers, true, it may not be anatomically accurate, but if this were one shot in a movie (not a clip that we can review a dozen times) you might get away with it. I think you've succeeded in giving the impression that they were doing their job, because you have the other hand doing it's thing and luring our attention away. If it were a longer scene you'd probably have to address the issue of balance: how does it keep from tipping over when one hand is in the air? This would really freak people out if it showed up at the Animation Showdown.

I think that looks quite good. The first time I watched it I thought he was reacting too soon to the 2nd light. After I've looked at it some more, I'm thinking maybe his arms could be loosened up some in the way they move. Overlappping motion. But overall, it's a successful shot.

That looks promising! It's so much more fun with your own character. Since he's almost all the same color you'll want to be careful to not make poses that leave the arms between the body and the camera. To animation... and beyond!

One investor, Steve Bing, provided half the money. I'm sure he felt it was a good investment based on the excellent prior track record of Robert Zemeckis and Tom Hanks. You might ring him up. He'll probably ask about your track record. I went to see it in true stereoscopic 3D! It's not like that money just vaporized. It shows up on the screen. If you took a dozen of the best A:M artists and put them to work for a year they might come up with something of the complexity of a typical shot in Polar Express. Great. Now what about the other 999 shots? There used to be an org that shovelled out money to small artists with no track record. It was called the National Endowment for the Arts. No blockbusters resulted. And how much did that gross in its US theatrical release? Why is this thread still in "Showcase"?

That's fabulous! Congratulations! That's a good short. Wish I had a Dish to see it!

No, I think feathering the transparency is a brilliant trick! That does alot to make the backing geometry appear to match the cutout. Hard edge cutouts always look weird. Will remember that one! thanks!

You can set the number of divisions for distort boxes in the Options>modeling tab. The number of divisions you choose will depend on when you need to distort your model. Once you go into distort box mode it puts what looks like a box of splines around whatever you had "selected". You can drag the CPs of that box around individually or select and group them or use any other tool just like CPs in a model. As you distort the proportions of the box the original modle mesh enclosed by the box will follow that distortion. Q:Why is this different from moving the original model's CPs? A:The box has far fewer CPs than your model so it is easier to make broad curves over a complex mesh than if you had to adjust the complex mesh directly. While distort boxes do a good job of moving the CPS they aren't smart enough to adjust the biases between distant CPS to match the new broad curvatures. That's why distort works best on meshes with fairly regular splinage.

Your treatment of the curved corners has several awkward spline intersections. Here's a take on your shape created with the mesh flat first, the using a 3x3x3 distortion box to bend it on X and Y. It still has a few lumps in it but Porcelain.mat would make it very smooth. A Vern noted above, the splinage should be very regular for good distortion box results. Mine should have been more regular.

It's so smooth now that I think you're losing "weight". Anyway, I think ANY walk cycle is going to look weird after it's repeated exactly several times. But I agree with the several above about the toe needing to push off more into the next step. And the shoulders would benefit from a different treatment and... I know you're wanting to do realistic and not cartoon, but Richard Williams "Animator's Survival Kit" has a great big huge section on walks and he addresses just about everything that's probably bothering you about what you have so far. Geez, only $20 used on Amazon... that's a steal considering he used to charge people $1000 to listen to him say the same stuff in person. Are black socks in now?

Hey filipmun, That's a pretty good mashing machine model or whatever they call it. The fact that you rounded all the edges does alot for it. I think the one thing that's probably bothering you is the crease at the edges near the ends of the the box shapes. Technically, the four point patches you have on the tops of the boxes are legal but a four sided patch made out of only two splines is always going to be a problem Take a look at this post http://www.hash.com/forums/index.php?showt...indpost&p=53091 for another way of doing boxes with rounded edges. Once you resplined the boxes, I don't think you'd need to add much more geometry. I think you could add the tiny details (like rivets and metal plate seams) with bump maps.

I'd love to see this but all I get is a white screen. Does anyone know where I can get the VP3 codec for Quicktime 6.5? The VP3 site only has it for v5 and the link they have for "theora.org" to get VP3 for v6 doesn't really have anything about VP# at all. ?