sprockets Shelton's new Char: Hans It's just donuts by ItsJustMe 3D Printing Free model: USS Midnight Rodger Reynolds' 1950s Street Car Madfox's Pink Floyd Video Tinkering Gnome's Elephant
sprockets
Recent Posts | Unread Content
Jump to content
Hash, Inc. - Animation:Master

jpappas

Hash Fellow
  • Posts

    476
  • Joined

  • Last visited

Posts posted by jpappas

  1. Yes, that's a nice improvement! It looks heavier and like something that must be very large in size. You got that body and heady movement going nicely. :-)

     

    The only thing I can say now is you may want to tweak the feet as they lift off the ground. They look like they're still flat as they come up but they probably should rotate a bit like a human foot, with heal up and toe down as they lift, and then swing back to normal in the air as they get ready to come down again.

     

    -Jim

  2. Al,

     

    I didn't compare the .mov to your still image until just now, but the difference is that in the still image your camera background is blue, and in your animation the camera background is black. The black background color is giving you the higher contrast in the animation but otherwise the scene is exactly the same. My guess is when you render the animation you have something additional turned on in the render settings, like the Alpha channel?

     

    -Jim

  3. Al,

     

    That looks really good! I would say, you might want to add more movement in the body and head as each leg moves, to try and convey how big and heavy each leg is. You did a nice job in making each leg look mechanical like how I remember them in the movie.

     

    For the low contrast render, we were having a discussion on the "low contrast blues" a few weeks back. Check out the four steps here to see if it helps:

     

    http://www.hash.com/forums/index.php?s=&am...st&p=197637

     

    -Jim

  4. Colin,

     

    Just a stab in the dark, but in your TWO folder are there any conflicted files? Is the tinman.mdl conflicted?

    I ran into problems early on like this and it happened when A:M tried to load a conflicted file, I think SVN sets conflicted files to unwritable. If there are any, you would need to Revert the files.

     

    -Jim

  5. George,

     

    Nice! And I like how Tinman is now showing up with Nancy's texturing, it really makes him stand out.

     

    My critique might be at how the original take was done, but it's that Tinman does too much of the same symmetrical hand/arm gesture, that is, both hands moving out then in again as he talks.

     

    As I watched it a few times, I thought of one spot for your consideration where you could break that up. In the beginning when he says "since I was made of tin" that's where he does the first double hand gesture, out then in. But wouldn't it be great if he just used his right hand here to knock on his head or body? So he's saying "since I was made of tin" and he knocks his chest two times to show it. Knocking on his chest would be more heartfelt, knocking twice on the forehead would be more comical. If you do this we could point it out to our sound foley person so he could add the knock sounds. :)

     

    The other thing that caught my eye was when his right hand goes to his forehead it slides on the forehead. I think it would be more appropriate if the hand made contact there and stuck there until he removed it, like a person touching their head when they have a slight headache. The way it slides now is catching my eye as somehow wrong, I guess that's something that would happen only if someone were wiping sweat from their brow.

     

    Good job overall!

     

    -Jim

  6. Kuep,

     

    This looks great! The only one tiny thing I can see are his left hand fingers, although his hand on the axe moves in the shot which is great, those fingers never move. I would add just a tiny clench or two. But nice job overall! :)

     

    -Jim

  7. Michel,

     

    I see lot's of improvement in the second version, very cool! One thing I would suggest, right before he says "I'm sorry" he looks up, presumably at Woot who is in front of him. That's good but he does this very quickly and it's almost lost. I would try to give that a little pause, so he will look up at them and say "I'm sorry" then puts his head down in his hands and say the rest.

     

    The rig is amazing isn't it? I've been doing this for what, four weeks now, and I still have so much to learn about it.

     

    -Jim

  8. Hi,

     

    This is my first revision scene. The original scene was by Jason Simonds and I made sure I kept his original intent, which was scarecrow leaning in with his hands on the table to listen to Woot. My goal was to give him non-symmetrical facial poses, hand poses, and some slight hip rocking motion for his lean forward with head lag. I also attempted to have Scarecrow do an 'eye-dart' to glance at Tinman before looking back and leaning in to hear Woot, I hope that is noticable.

     

    These shorter 3 second scenes have unique challenges. I was concious of the fact that I can't just start off quickly because the scene itself is a quick camera cut, so how to get enough of a holding pause at the beginning so the audience can register the new scene they're looking at?

     

    I used about 20 frames at the start to try and accomplish that, so the audience can rest their eyes on the scene before it happens. That left me only 50 frames or so to get scarecrow into pose and try to make it not look too quick or too linear. :P

     

     

     

    -Jim

    1_03_30_take.mov

  9. Hey Michel,

     

    Martin's got the answer there, you can use SVN Revert. More specifically, on a WinXP machine, you would navigate to the .cho file on your hard drive, right click, choose TortoiseSVN, then Revert. And this pulls the last one from the SVN server and overwrites whatever you have.

     

    You'll likely run into using Revert soon anyway. When you work on an animation scene and save it, A:M will sometimes write to the models like the Scarecrow or Tinman models themselves, even though they are not embedded and you haven't actually saved them. What it writes is just a small piece of data that won't drastically change the models, it's a rounding type difference in a piece of data in the models. But SVN will catch this and show that these too need to be SVN updated. But you don't want to do this. It would cause too many useless updates to the models if we all did it. So when you see this happen, just track down the file (scarecrow.mdl, etc..) and Revert as described above.

     

    -Jim

  10. Hi Michel,

     

    Here's my take on answering these questions.

     

    1. ...I don't feel too comfortable overwriting the cho I have been assigned yet. So what would be the procedure? Saving a new cho, and overwriting only after the approval of the director(Rhett or Martin?), or just go ahead, overwrite and hope my changes will really improve the shot.

     

    Overwrite the CHO. Always keep in mind we're using SVN, software version control. That means all iterations before yours are 'in the system' and can be reverted back to at any time. Let's say I did a shot today and just demolished someone's great work from before. The previous version can be pulled from SVN and updated over mine.

     

    2. ...who decides when it is ready to upload on SVN? The director or myself?

     

    Uploading to SVN safeguards your work, so do it as often as you need if you're working on a shot. If however something bad has happened and you feel you might have, let's say deleted the Rig from a character or something... that's when you would skip uploading to SVN and contact Rhett or Marting for guidance.

     

    3. Can you just reload the project file, if you feel you have not started right?

     

    Of course. I do this alot too. Just now in fact I turned on the Master Eye Target for scarecrow and for some reason his right eyelid is now permanently closed and I can't see any reason why. Since I hadn't saved yet, I just closed the Project and reloaded it.

     

    4. I was due yesterday but will be late, since I got more time on weekend.

     

    I'm the poster child for this, no worries. It would be great if we all could post WIPs during the week so we can get comments and have time to incorporate them, but I tend to work late during the week at IBM and most of my free time is on the weekends. My 4-5 hours per week on my shot usually happens on Sat. and Sunday.

     

     

    -Jim

  11. If the bottom of the shovel were curved to have a "C" shape like a snow shovel, that might help with the illusion and provide a visual clue for the spring he's getting from it, and at the same time show why it doesn't cleave the ground. Just a thought. :)

     

    -Jim

  12. Hey Jason,

     

    I'll be doing a 2nd pass on this. I'm thinking I can add a more pronounced facial expression, make the left hand a little different than the right when they come together, lift his shoulders up a bit when he moves his hands forward, and even though he's sitting, move his hips a little forward too.

     

    If you have any thoughts or ideas, let me know. :)

     

    -Jim

  13. Kevin, Colin,

     

    I went back to look at it and you're right, his eyes can be hard to track. I think that's because he's got so much white specular type reflection on his face and his white eyes seem to become part of that.

     

    But also, when he closes his eyes the white specular on his eyelids can cause them to appear open! Kind of like you're looking at the whites of his eyes. This is more noticable when it's animating but here's a screencap with his eyes shut. Just my 2 cents, but probably lowering the amount of reflection or spec. intensity on those eyelids would help.

     

     

     

    -Jim

    eyes_shut.JPG

  14. Hey Rodney,

     

    You're absolutely right, my original idea was to have a box on the table, a poster on the wall, maybe even some extra lions on the table behind him, to setup a little fight scene, but alas, I ran out of time! I was finishing the Voltron model right up until the very last moment. Literally, kids were coming up for Trick or Treat, and there I am handing out candies with the laptop and Voltron on the side! :lol:

     

    The face itself was the most difficult part since it's the most organic. I followed the Cooper tutorial for the first time but applied it to Voltron's type of face. The result works for this toy model but you would never want to use it for facial animation!

     

    Now that I've modelled him I'm thinking of ways to use him, maybe an attack on the floating city from the SciFi contest!

     

    Here's a larger quick render which shows more of the detail.

     

    -Jim

    side_view.jpg

  15. Hey hashers,

     

    Like the announcer in the show says "Voltron was needed once more..." so I had to model him. I learned so much about mechanical modelling by tackling this. Now I've got to learn organic modelling and I'll feel like I know what I'm doing. :P

     

    I had to keep it quiet for the last few weeks since he was in the Image Contest. I think it took me about 20 hours or so, working a little bit each night over a week. Each Lion has separate models for their own legs so the lions can easily come apart and stand on their own.

     

    Also, I've never rigged anything complicated so I gave it the most basic skeleton -- just so I can move it around, it's so basic it's pathetic, just a few bones on each limb and one for his hip, but its workable.

     

    -Jim

    robot.jpg

    wireframe.jpg

    voltron_rig.jpg

  16. Fab,

     

    There's a mountain range back there! Anyone who has struggled to make decent looking distant mountains (I'm speaking mainly of me of course) is going to be dying to know how you approached them. Modelled by hand, procedural materials or textures, you know, all the usual quesitons. :D

     

    And absolutely beautiful clouds you have there too, and also the bumped texture on the grass and dirt add so much.

     

    -Jim

  17. George,

     

    Using the Ease is not the only way to do Pose to Pose. I've been following the method from Anzovin's Animate! tutorials, which is simply, model each Pose in the Choreography action, and set the Interpolation of all these to Hold. That's your basic Pose to Pose. This allows you to quickly retime any pose by selecting all the keyframes (like we were talking about in your other post) and just moving them left or right.

     

    Now, when you're happy with the basic poses and timing, you select all the keyframes and reset the Interpolation to Spline and... yup, you get a floaty mess between the poses. But that's easy to fix too. Again, using the method described by Cristin McKee, you copy all the keyframes of the first pose and paste them a few frames after it, and do this for all the poses. This gets you closer to having 'holds' again, but now you are in Spline interpolation, and you can now do all the fine tuning and offsetting and inbetweens.

     

    So this method of animating happens just in the Choreography action and doesn't use Ease percentages or layering. Not that there's anything wrong with other methods, you just have to choose which is best for you. I think I understand the Ease method but have never used it so I can't comment on it. With A:M's large feature set, there's always 6 ways to solve a task!

     

    -Jim

  18. Xtaz,

     

    There's something unusual with how the AM:Stills Showcase page is showing this, it won't display but also seems to cause the webpage to constantly prompt for a username and password. Is anyone else seeing this? I'm using the Firefox browser on WinXP if that makes a difference.

     

    I'm not sure why, but it seems to only happening when I'm looking at a gallery page that contains either of these images, but the most recent one isn't displaying, it's just showing the name, peugeot~0.jpg.

     

    BTW, this is an awesome model and a great render. I was trying to look at it while browsing the Stills gallery when I ran across the web page problem.

     

    -Jim

  19. George,

     

    You've got alot of good advice so far. I agree the time frame of this shot is just not long enough for Woot to cover the distance without a camera cut, or him running. My vote is for a camera cut. If he's looking around in wonder, another closeup of that would be cool and allow for some facial expressions. Then when the camera returns to full view, he's arrived! (and Tinman will be impressed with Woot's teleporting abilities but he won't show it). :)

     

    -Jim

  20. I didn't get a refinement shot this week, this is a new shot that needed to be done. And this one was tough. It's not like I can look in a mirror to see how a person dismounts from balancing on an axe with his hands! In fact, given that the axe's wood base is pointy I don't think this is even physically possible! :D

     

    Tinman is supposed to dismount from his balancing trick and flourish with his hand to show how good he his. For this first pass I didn't address any facial expressions, but just tried to get the basic movement of falling from an axe look decent.

     

    -Jim

    1_03_38_take.mov

×
×
  • Create New...