Re: OT: Max 2009 features...

Date : Mon, 7 Apr 2008 18:24:16 +0200
To : XSI(at)Softimage.COM
From : "Ahmidou Lyazidi" <ahmidou.xsi(at)gmail.com>
Subject : Re: OT: Max 2009 features...
Hi Stefan
you have a spider web because you didn't pick the vertices, or you picked just one vertice, or many times the same vertice.

2008/4/7, Stefan Kubicek <xsi(at)tidbit-images.com>:
Hi Alan,

What I get looks more like a spider web rather than the regular UV grid you're getting in that screenshot, and I swear I chose the exact same settings.
Also, when I go back to the primitive's geometry settings and change the Subdivisions the beautyful new
UV layout goes bye bye.

Your example illustrates one thing very well: You still need to click a lot to create a dataset
(in this case UV's) that should be a standard part of the geometry right from the moment of it's creation.
In Max, you click a check box to generate UV's on the primitives rollout and voilla. One click!

As for PhysX, I remember browsing the SDK docs about rigid bodies some time ago, and there was a line saying that
the Physics environment supports only rigid bodies _"yet"_. To me that sounds very much like something bigger was/is
planned on the long run.

Your cloth demo is another fine example of PhysX. And here's the link to that guy who hooked it up to Particle Flow in
Max for fluid simulation, I still think it looks pretty good and I'm sure it solves faster than the XSI built in fluid solver
does: http://dimo3d.blogspot.com/

Stefan



Hi Stefan,

Some things to add to what you said: in Houdini, afaik, all solvers
speak to everything and also, for those that don't know, the UI is
actually made ALL in Python and thus any programmers/master-scripters
out there can customize the hell out of it if they want, through
custom or modified code.

On another note:
Primitives don't create a default UV layout. Ok, that's no big deal with
most, but you tell me how to parametrically map a torus with the given
toolset. Yeah, you can create another nurbs torus and GATOR the hell out of
it, but...you have to do it.

I don't like it either, but uhmmm...
http://img86.imageshack.us/img86/4318/walkonmeshcanunwraptubukn8.jpg
This simple torus took about 5 or 8 seconds to unwrap with XSI's
built-in tools. You were saying? ;)

GATOR is overkill for a torus. It's much easier to do what I
illustrated above. I encourage that you try it yourself:

1. Press I and select your (edge) seams and rightclick-->Disconnect
components, like so:
http://img241.imageshack.us/img241/7519/seamsanddisconnectcompoyw6.jpg
2. Give the object (in this case, the torus) a texture projection.
(Any simple one will do. We just need UVs to exist to begin with.)
3. Press I again, and do Ctrl+A to select all edges from the torus.
4. Do "Walk on mesh". It's a little button with an icon of a scroll
being unwrapped:
http://img241.imageshack.us/img241/6931/walkonmeshgg6.jpg
5. Then it's gonna go in a picking session. For this case, we just let
it pick itself by going ahead and right clicking.
6. A little PPG pops up. (Play with the "Interior Smooth" to
straighten your UVs if you like.) Done! :D
7. When you're happy with it, select all edges again,
rightclick-->Weld Boundary Points/Edges, then Freeze M.

This method works great for unwrapping any tube-like geometry, too.


Back to XSI simulations... how come we have PhysX but we can't do this??
http://www.youtube.com/watch?v=zgWur7HaIks
I was playing with that demo the other day. It ran in realtime, super
smooth with dynamic tearing! :o (....and it's not even "new".)

Hope to see it someday,

  -- Alan


On Thu, Apr 3, 2008 at 7:40 AM, Stefan Kubicek <xsi(at)tidbit-images.com> wrote:
I think it comes down to the fact that it's simply cheaper to implement
third party software improvements
(e.g. a new mental ray version or shaders) than developing everything from
scratch.
XSI benefits from this as well extensively through Shave, Mental Ray and
Syflex.

What I like in 2009 though is the Help Center search function, which ties
many sources of help together.
Also, I like the concept of relaxing the UV's by keeping the button pressed
until it looks right, instead of having to crank some spinner up and down
for iterations and hitting a "Relax" button several times, and then still
not quite hitting the desired end result. Eliminates the guesswork.

Commercial spline mapping solutions exist for quite some time, e.g. Texture
Layers from www.mankua.com for 3dsmax.

As far as open ends are concerned: Max is definitely the worst example
overall (It started stagnating with Version 3 imho), but even XSI shows some
signs of this.

e.g. Cloth and Hair solvers "don't know each other", the same goes for
fluids and rigid bodies. Now this is surely
more a technical problem rather than a design problem on which both Houdini
and Maya have their own take, and apparently available middleware that's
capable of that (PhysX) doesn't seem to cut for high end packages..

We have two cloth solvers in XSI. The standard version, and Syflex.
There are 2 ways to achieve soft bodies: Syflex and the old built in
version.
None of them interacts with any of the other dynamics stuff like rigid
bodies or fluids. (see above)

There is no central facility for handling rc or pulldown menu contents and
layout. Plugins tell the registrar at load time in which menu they want to
appear and that's it. No easy/user friendly way of reorganising those that I
know of.
In Max you can completely customize the UI, pull down and quad menu entries
with a simple drag and drop UI tool, which beats the crap out of XSI & Maya
in this regard (don't know about Houdini and C4D).

Small UI inconsistencies: The left command panel has bitmap buttons to
switch between Weight Paint tools, Main Tools, Palette and Scripting tools.
The MCP has tabs with text instead, which looks much tidier. (ok, that's
probably a bit pedantic, but it simply doesn't look consistent, and the
Icons (especially the Palette icon) seem to suffer from JPG artefacts).

We're in version 6.5 now and the sculpting tools ("painting" deformations)
are a joke.
Primitives don't create a default UV layout. Ok, that's no big deal with
most, but you tell me how to parametrically map a torus with the given
toolset. Yeah, you can create another nurbs torus and GATOR the hell out of
it, but...you have to do it.

Other than that: I Love it :-)

Stef





> Raffaele Fragapane wrote:
>

> Looking into the release, there´s lots of mR functionality
> implemented. This seems to become the biggest chunk of updates
> with any of the major 3Dapp licensees. This is either because
> there is nothing else to be developed (I doubt that) or because
> implementing the latest mR version and keeping track of it´s
> changes takes more and more time in the development cycle,
> leaving not much room to actually implement other new features.
>
> The splineUV thing is also something a smart guy released for
> Maya (using a NURBS surfaces as input) a while back, might
> even have transfered over from the Maya Bonus tools, concept-wise.
>
> *The IES lightprofile stuff and viewport display is still great.
>
> *Softselections are about time, the implementation seems nice.
>
>
> In general, regardless of the 3D app I´m looking at, I´d find
> it about time the developers find the strenght to consolidate
> existing functionality, update and weeding out where neccessary.
>
> Otherwise, with age, there comes an increased risk of bogging
> down and falling flat on your face...especially if there´s all
> sorts of open ends, nuissances and whatnot building up slowly...
>
> Cheers
>
> tim
>
>
>
>
> ---
> Unsubscribe? Mail Majordomo(at)Softimage.COM with the following text in body:
> unsubscribe xsi
>



--
Using Opera's revolutionary e-mail client: http://www.opera.com/mail/



---
Unsubscribe? Mail Majordomo(at)Softimage.COM with the following text in body:
unsubscribe xsi


---
Unsubscribe? Mail Majordomo(at)Softimage.COM with the following text in body:
unsubscribe xsi



--
Using Opera's revolutionary e-mail client: http://www.opera.com/mail/

---
Unsubscribe? Mail Majordomo(at)Softimage.COM with the following text in body:
unsubscribe xsi


Search the XSI List archives here or use the advanced search form to search across mailing lists. Searching help is available.
This site supposedly brought to you by Benjamin Grosser and the Imaging Technology Group.