Jump to content

Maya skinning...


abvfx

Recommended Posts

First time Side Effects and I were introduced to the capture pill idea was way back in 1995-1996 at R&H with Nick Ilyin and his crew working on a 3d theme park ride with dinosaurs in 3d (I believe). Boy that was a long time ago...

He implemented capture pills in PRISMS. We followed up on that and implemented them in Houdini's capture system way back then as an alternative to proximity skinning (which we call capturing or binding which makes more sense to me :rolleyes: ).

A couple years later Max introduces capture pills. :)

Umpteen years after that Maya introduces capture pills. :lol:

Glad to see that in the Maya video they get the same results Nick got at R&H, we got, and subsequently Max got way back in the mid 90's.

Now when I show capture pills in the "skinning" process, the Maya audience won't go "what the heck are those" but will say "oh yeah, now Houdini has capture pills too". All good.

Link to comment
Share on other sites

A couple years later Max introduces capture pills. :)

Umpteen years after that Maya introduces capture pills.

haha yeah i recognize them from back ages ago when i used Max, i didnt know about prism around that time ether so its cool it came from there :D But like u say, funny stuff Maya introduces it now as their "super cool new feature". oh im glad i hardly have to touch Maya now a days!

Link to comment
Share on other sites

hehe, well it was bound to happen

Someone needs to track the history of every update from each product, Maya, XSI, Max, Houdini, Modo etc and see where each feature started originally. I think it would be interesting to see.

Go ahead internet people :P

Link to comment
Share on other sites

Agh! The world of Autodesk Empire. :)

It is amazing how people look at some cool fur stuff and ask: "How did you do it?!"

And some dude comes and posts: "Probably they used 'Shave and Haircut' and then they wrote a script to create some filter attributes, then they wrote a script to blend between different attributes and then they wrote a script to do that and to do this.. blah blah"

It is done in Houdini Man! Wake up!

:)

Link to comment
Share on other sites

One nice thing they have done though is include a DQ solver and a blended linear to DQ solver which is really the only useful way to use DQ, Sesi you gonna add that? it's been around as a method for ages and is almost trivial to add.

Edited by sibarrick
Link to comment
Share on other sites

One nice thing they have done though is include a DQ solver and a blended linear to DQ solver which is really the only useful way to use DQ, Sesi you gonna add that? it's been around as a method for ages and is almost trivial to add.

Please submit this Simon.

Link to comment
Share on other sites

One nice thing they have done though is include a DQ solver and a blended linear to DQ solver which is really the only useful way to use DQ, Sesi you gonna add that? it's been around as a method for ages and is almost trivial to add.

Ha, I've been playing recently with your DQ otl, Simon. How nicely it solves many problems with particles tricks :)

Link to comment
Share on other sites

One nice thing they have done though is include a DQ solver and a blended linear to DQ solver which is really the only useful way to use DQ, Sesi you gonna add that? it's been around as a method for ages and is almost trivial to add.

As Jeff said, please log it. And when you do, please specify HOW you see it being added. From everyone I've talked to, no one wants to see yet another XXXDeform SOP. How would the user specify where the DQ solver should be applied? What parameters should be added and where? Adding it is one thing, having it work well with everything else. and still have it work reasonably fast to be flexible is another. Besides, tell that to the other hundreds of "trivial to add" RFEs assigned on my list. :)

Link to comment
Share on other sites

I'll do that....

But just to note it here, I would suggest you add it straight into the existing deform sop, it uses all the same weights etc. as the normal linear capture system you have already, it just needs an extra painted attribute to allow the DQ solution to blend with the linear one.

I did say "almost trivial" :P .... what I meant was it doesn't need a complete capture system as that is already there, so it's probably about 10% the effort of writing a whole new deformer.

@SYmek, I have a DQ otl??? I know I hacked together some HDK code but i never finished it off properly, did I make something else I've forgotten about.....

Edited by sibarrick
Link to comment
Share on other sites

Oh that! yeah that has some similarities.

I'll post my HDK DQ experiments using the existing Houdini linear capture system tomorrow when I've dug it all back up and seen what I have. The only reason I didn't finish it was because I realised I needed to re-write the whole linear deform node that Sesi already has just so I could add this in which seemed like a lot of wasted effort on my part. However if anyone fancies doing it I'll post what I have and it should at least demonstrate nicely what is needed, which really isn't much once you have a linear deformer setup.

Link to comment
Share on other sites

That may now be possible as VEx now supports arrays, however I think the problem is the way that the capture attributes are stored in Houdini's linear capture node is quite long in the tooth and use a different way of storing capture attributes to how you might do it now thus I don't think all of them can be read into vex.

However if the whole system could be re-authored to work with vex then not only could users easily extend it but you'd get the benefit of SIMD multi proc support, that would be a much bigger job but would be very cool.

Link to comment
Share on other sites

Guest Swann

That may now be possible as VEx now supports arrays, however I think the problem is the way that the capture attributes are stored in Houdini's linear capture node is quite long in the tooth and use a different way of storing capture attributes to how you might do it now thus I don't think all of them can be read into vex.

However if the whole system could be re-authored to work with vex then not only could users easily extend it but you'd get the benefit of SIMD multi proc support, that would be a much bigger job but would be very cool.

Funny that you mention this guys. Since I started my journey with C I try to convert some things that I find to VEX. And since C version I'm learning is C#, XNA is also one of the things that I'm interested in. Some time ago I came to this XNA example and was thinking about rewriting it but I wasn't ready. With Sibarrick's example it will be much easier to understand.

Edited by SWANN
Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...