I have an assignment at work where I might be able to use my stickman rig, and I wanted to clean him up for some time now, so I started to redo the rig. It’s a lot cleaner now, and I added some more stretch controls. In the middle of the upper, lowerarms and legs, and 2 more in the spine.
These extra controls kinda require you to do a bit more tweaking though, to not get intersecting joints and such. It isn’t hard, you just rotate the stretch controls, just like the previous version, only now there are more of them.
I could go back to having less stretch controls, which will be a easier (read faster) to manage, but I’m kinda starting to like it as it is.
screens:
Showing the FK controls.
showing the new stretch controls.
What kind would of update would you guys prefer. one with all the previous stretch options, or one with the additional ones?
Okay, I have a very stable version, but the default deformations in FK/IK (meaning, without using stretch) makes the spine act not as “rubbery” or smooth as I’d like. I’m experimenting with some different setups, but each seems to have it advantages and it’s quircks. And the quircks make it a no-go for me.
I’m really considering going back to 2 controls for the spine, and lose some extra stretch there, but it should be easier to animate when you only use the non-stretchy poses, which is what a lot of people seem to do anyway.
I’m still doubting on this, because I so much like the idea of that extra stretch for the spine too, but if it doesn’t work intuitive/good enough…
I’d like some more of your opinions on this guys. I know there are a lot of people who have downloaded the previous version rigs, there’s the occasional mails from people asking questions about them or youtube movies or someone’s blog or something, but now I’d like to ask those people that download and play with it to give some feedback for this new release.
It’s a BIG improvement over the previous version, but if nobody gives me feedback on it…
I know what I like in a rig, but I’d like to know what you guys like in a rig. Particularly from actual animators.
I’m working on some movies for work, and by doing that I’m finding what works and what doesn’t.
(one of them can be found here: http://www.youtube.com/watch?v=qJvc_Q4NDgQ )
A screen from one of these other setups, with an advantage (smoother and easier animating at default FK/IK, less keyframes needed) and a big flaw (less control, unreliable behaviour) (and yes, there’s a bug in this screen, it’s fixed already)
Shot at 2008-12-10
I have some testing left to do, and the break is over so I have to wait till I’m home again, but I think I’ve tackled the smoothness issues when only using the “non-stretch” controls.
on a side note, does anyone know how to remove (references to) images that are in the blend file? like images you opened in the UV editor or just plain textures with and image type, where you unlinked from the image, removed all the texures etc.
I have a number of files where I have images with no more users (checked the data browser and outliner for this) but they still stay in the blend file. None of these images where ever packed, I think it’s just the reference to them that stays in the blend.
Does anyone know how to remove those images?
Nice rigs FreakyDude. FK-based stretch is something I still need to learn. Gah you and JiriH are wearing me out!
Regarding how many controls you need, etc. I’d honestly just keep animating with the rig yourself, and change whatever you feel would help you animate better. I don’t think there’s such thing as a perfect rig - just one that helps you animate more efficiently - which judging from the video, you seem to be doing.
Regarding your images - could you not just remove the image from that said directory?
hey guys, thanks for the replies. The rigs won’t be really soon I guess, I want to make a full package with manual, like for Audine, and I haven’t gotten that far yet… Sorry bouza…
@Mikahl
it’s not that complex really, the trick is to get as much done with as little complexity as needed. Nothing you couldn’t handle, judging from your rigs. In any case, your animations are much better than mine.
And unfortunately, removing the images from my hdd has no effect, did that ages ago…
stickmanV2 had 87 bones, V3 has 104, that’s with a shitload of new controls, “double-limb-sets” (for smoother IK/FK) and a whole bunch of sliders. If you disregard those, he’s actually a lot simpler this time.
updates. just a bug left in the hinge for the tail, and devilman is done too. That and some testing…
Shot at 2009-01-18
Devilman is done too. Apparently, blender is extremely finnicky (at least it has been for me) at HOW you add your driven IPO curve. I know about the shark icon, and you should disable it so the IPO is connected to the constraint, rather than the action. But you can disable it in a dozen ways… I had everything set correctly (No shark icon pressed and the ipo and constraint set up correct) but it STILL had refresh issues…
This happens in both 2.48 and 2.47…
I believe I ran into that issue once when I was rigging a character of mine a while back. Basically, this is caused as a consequence of all of the ugly hacks that the old system had…
I’ve ran into that issue a couple of times now, I always got around it somehow, but it sure was annoying. I thought I just did something wrong. Glad to read drivers and animation data are being separated now.
Thanks for the info on that you two. I’m really looking forward to your new system anyway Algorith, you’re doing a really great job at that.
ps, does this mean we’ll have to redo old rigs for the sliders to work in 2.5 or do driven curves from old files get separated automatically?
…
and there’s still that image block issue in the outliner that i can’t seem to get rid of.
It is my intention to have the version patching code to fix such situations to work in the best possible way in the new system, without requiring any manual tweaking. Driven curves are quite easy to identify and separate, so there shouldn’t be many problems there. The only problem is porting all of the ugly old defines over… ick!
That’s cool. It would be cool if the porting works, but just in case, I for one don’t mind redoing my sliders in 2.5…
Update… well a bit of good and bad news really. Good news, the manual went a lot faster than I thought. It’s almost done, should have something good finished this weekend if all’s well.
Bad news… I found a bug while doing the manual. I’ll have to redo part of the chest mechanism, this might have me scratching my head if things don’t go as planned. This might also mean some new IPO’s…
Algorith, another question if you don’t mind. Could these old “hacks” cause any unwanted users of IPO data? I had one bone with a constraint and a driven IPO channel, I removed the constraint from that bone. Then I added two new bones, gave one of those bones two constraints, and gave the new constraints the driver that was used in the old constraint on the original bone. I’m rather certain there’s no other constraint using that IPO block, yet there are 3 users listed, and it doesn’t have a fake user.
… It doesn’t really matters for the rigs usability, but it’s messy, and I don’t like it. I’m going nuts tracking the potential user down, and it’s really annoying the sh*t outta me…
updatewise:
I’m pretty sure I have the main bug squashed. Found another bug which is also precent in the older version with the head. Also pretty sure I fixed that one, minus the extra user…
I think it was just a case of closing and reopening blender, because the extra user is gone now…
Also did some testing to see if everything works nicely again, and all seems well. still some things to do before I release it, but not for the actual blenderfile.
A little offtopic. Here another small project I started for the heck of it: