Tuesday, October 21, 2008

steve ly..3 hit combo



Ah, the famous 3 hit combo. Steve is clearly doing a piece targeting game companies, which is great...don't be afraid to do pieces specific to where you want to go. I wouldn't send this to ILM, but it's on it's way to being a really strong piece.

The thing with 'game reels' that you need to be careful of is catering them to the perceived look of a game animation. Game animation is tricky because it has a lot of restrictions on it because of the engine it's going into, and the ability of states to blend to each other. I can go into some of the restrictions if you guys are interested, but suffice it to say, when I do a reel of my game work, I tend to reanimate it as needed to either punch it up, or slow it down accordingly. The biggest culprit is the transition back to a resting ('idle') pose. I see a lot of student reels where the ending happens abruptly. Maybe that's how it is in a game, but the reason is so that the player can get control quickly. What companies of all types want to see is good motion and weight, so don't be afraid to have your character take an extra step, do a shoulder shrug, or anything to finish the action.

So the above action is good...the poses and general flow is correct. Steve did his homework and is doing a good job of taking some reference and starting to move beyond it. Using reference footage can be tricky...at a certain point you have to chuck it and just periodically glance back at it. Mocap and rotoscoping want your animations to be smooth, fluid, and a little sluggish.

Right now, the 2 things that will improve this piece are purely timing and spacing related..I did some cutting and pasting of frames in quicktime to both sharpen and pad the timing, accellerating into the punches, and holding after the fact...the knee kick definitely needs some holds. See the boxing capture below for slow motion footage of accelleration (link)

The other thing to watch (that i don't have too much time to get into) is how the feet travel. The offsetting is great...it's not too 'stompy', which can be a problem. On the other hand, they feel like they're in slow motion at times, so the jumps forward may need some frames trimmed to accellerate into each step...I'll look at it further, but it's very much improved from last week. You've arrived at what I think of as the polish stage..refining timing, pushing the elements to be stronger..





No comments: