SI 2013 sightings

[Eugen]

As there seem to be some "wind of change" at Softimage, I'd like to place a little reminder:

If there's one "feature" focus should be set on, it's the one critical problem that makes Softimage fall behind M&M (max & Maya). It's what Luc-Eric mentioned at an earlier post in this thread:
SDK.
Let's use that here as a general term for all the options users/3rd-party developers have to shape Softimage to their own ideas/needs, including every aspect - Operators, User Interface extensibility, ICE SDK, Rendering API...

Generally, the SDK still is not open enough, comparatively. Too much of a blackbox.
I do not mean that it is awkward, but incomplete.
Interestingly, you can use more scripting/programming languages in SI than in any other package I know, but still, the SDK is the limit.
I believe this is one important aspect that makes people stick to M&M (strange marketing aside).

In M&M, you can define new base-classes, something completely new that can show up in the viewport.
Examples are T-Splines in Maya, PowerNURBS in max, and many more. In SI, writing plugins like that is simply impossible.

It's also not possible to write new topology operators consistently, meaning equal to factory operators. UVs cannot be created on-the-fly in the operator callback, because of SDK restrictions with Clusters. Too much is still unexposed here.

ICE does not fully rehabilitate SI in that regard. It's a new playground you can do countless things with it, but it is still incomplete when it comes to modeling (curves and surfaces creation still unsupported), and it also cannot override basic design philosophies.
There even had to be "workarounds" introduced, to make on-the-fly UV creation and material assignment possible.

UI: In Maya, almost every aspect of it is a MEL script, and can be reconfigured, if needed. In SI, too much of the UI is hardwired C++. The initiative to open up to Qt did not come from the devteam itself, but from other people (Steven Caron, and others).

Rendering API: it took a very long time until it was mature enough to make it possible for Chaosgroup, Solidangle, Nextlimit etc. to "marry" their renderers with SI. An ongoing process still, one that should be fully supported further by the SI team.

It's part of the success story of M&M that it was possible for users/3rd-party devs to "rip it apart" and shape it to their needs, without having to crawl to the king's throne first and beg for new features.
Development needs to become even more "democratic", I believe. Provide people with a good and complete framework, attract the developer-crowd, and it can only be for the good of SI.

Hope the new heads at SI with their fresh perspective will have that in mind.

[gfxman]

I said countless times, anything to do with future direction, developments and roadmaps is off limits. Due to Autodesk public listed status, we can't openly discuss these subjects. I know its frustrating, but its the same on my side. There's alot I would like to say, but can't.

That's a shame, but i understand it's not your fault.

As for guys moving projects, its odd that people always see this as a negative thing and not in the least bit positive. When guys like Ronald and Phil left Softimage, people spelled doom and gloom for ICE, and yet, years later it's still here and in even better shape. Essentially all Autodesk are doing is moving and re-focusing resource. Softimage isn't being singled out - Maya and Max have also done the same. As for guys like Luc-Eric and Guillaume moving to more Maya focused projects, then personally I think its a good thing as their experience will be useful.
As for replacements in the Softimage dev team, I don't manage or work in the dev team so I can't provide a full picture, though I do know there are new people in place, but it would be wholly inappropriate of me to name them and essentially 'out' them on this type of forum without their knowledge or input.

I can't see what is the positive thing, when key roles people leave softimage to Maya, the "enemy"
We see people going, but we don't see who is coming and from where, and to do what. So, the idea we have here, is that

- softimage is being kept in darkness (no mention in the autodesk front page, and it's being like that from day one)
- a lot of people already left at the time of the acquisition, and are leaving now, after the "shiny" 2013 release, at least 2 are going to Maya
- no news about who is coming to softimage, we just see people leaving, we don't see people coming

Where is the good thing for softimage development? I think the Autodesk acquisition has been the worst thing it could ever appen to this software ( the worst would be behing dismissed at all...)
I can say this after 3 years in the making, and the 2013 release, a release in past would have been a 0.1 release.

I never was on the negative train about the acquisition, in the first time i thought, it was a good thing....now i changed my mind....

[Hirazi Blue]

I can't see what is the positive thing, when key roles people leave softimage to Maya, the "enemy"

People, people, Maya isn't the enemy... Autodesk is!!!

On the Mailing List Luc-Eric Rousseau mentioned the following:

yes.. we've been interviewing, hiring and training many new developers
and QA for the last two years. there are more than a dozen new
softimage people compared to three years ago.

(Quoted from here)

That sounds a little better, now doesn't it?

[Daniel Brassard]

[Clagman]

Haha!

[bottleofram]

I was about to go on a lengthy rant, but that... that just made me smile.

Whatever i wanted to say - it doesnt matter. We're all sick of autodesk and they couldnt care less. Frankly, if there was a viable alternative to Softimage out there, i wouldn't think twice about starting a slow (and probably painful) process of migration. Unfortunately, there isnt one. I demand a quality animation toolset and (ironically) i can only fall on to maya for better job opportunities at this point. If only the torture could be less painful, that'd be swell.

[bottleofram]

Not sure whether people here have found their way to the Mailing List but heres a quote from Luc-Eric Skywa-- Rousseau ( ;) ) which may (or may not) clear a thing or two:

Actually, you're getting a lot of official information right here in
this thread.

None of us moving to the new Maya FX Montreal team, which is led by
me, are really doing it to quit Softimage, it's just a natural
evolution of thing. As written previously, this was planned and we've
been hiring and training a whole lot of new group of great people to
work on Softimage.

So what does that mean? Well first it means that Autodesk is committed
to continuing the development of Softimage - otherwise we wouldn't be
spending so much effort building a new team. We worked a lot on this!
I interviewed every single one of these guys - and some we rejected.
The new guys have backgrounds in game production, real time shader,
physics, etc. They're bring new ideas and skills.

And it also means that Autodesk is renewing its effort on the Maya
FX's toolset, which is not super interesting to you guys obviously,
but is that a thing that makes sense.

It's hard to leave Softimage, but it's also hard to not get excited
with the new projects. We're going to Digital Domain and ILM this
week with Duncan. New experiences! If we can just stop hitting the S
key to orbit the camera, everything will be fine.

[Kerro Perro]

Not sure whether people here have found their way to the Mailing List but heres a quote from Luc-Eric Skywa-- Rousseau ( ;) ) which may (or may not) clear a thing or two:

Actually, you're getting a lot of official information right here in
this thread.

None of us moving to the new Maya FX Montreal team, which is led by
me, are really doing it to quit Softimage, it's just a natural
evolution of thing. As written previously, this was planned and we've
been hiring and training a whole lot of new group of great people to
work on Softimage.

So what does that mean? Well first it means that Autodesk is committed
to continuing the development of Softimage - otherwise we wouldn't be
spending so much effort building a new team. We worked a lot on this!
I interviewed every single one of these guys - and some we rejected.
The new guys have backgrounds in game production, real time shader,
physics, etc. They're bring new ideas and skills.

And it also means that Autodesk is renewing its effort on the Maya
FX's toolset, which is not super interesting to you guys obviously,
but is that a thing that makes sense.

It's hard to leave Softimage, but it's also hard to not get excited
with the new projects. We're going to Digital Domain and ILM this
week with Duncan. New experiences! If we can just stop hitting the S
key to orbit the camera, everything will be fine.

It somehow makes me feel good to know that even super-experienced guys like them have to say "D'oh!" when they open Maya ;)

I know i'm a glass half-full kinda guy but i look forward to what those new guys can bring - if they can wow me with anything before june i might even renew my sub!

Also i'm looking forward to the Refractive guys integrating Octane - i think it's better that IRay and they assure that it will still be very affordable a usual. I just hope they get it doen a little faster than it usually takes a 3d party renderer...

(Btw thanks Nizar and Maximus for your 2 cents on Modo, i guess i'll try it out if they ever release a demo though my gut tells me it won't temp me away from SI (yet..))

[woodengun]

Hey, I like using the "s" key to orbit!!

alt sucks!!

[Nizar]

(Btw thanks Nizar and Maximus for your 2 cents on Modo, i guess i'll try it out if they ever release a demo though my gut tells me it won't temp me away from SI (yet..))

The important thing when a XSI user look at modo tools is: modo is not XSI I strongly prefer XSI for modelling, rendering, rigging and animation (I doing very little animation, very basic), but the plus in modo is really nice and I use it for paint texture, very basic sculpt (for the rest I use extensively Zbrush) uvmap and test my texture in the fast render preview (believe or not, on my machine modo is more fast than octane). Now retopology in modo is pretty good more strong compared to XSI (where combination of the polygon constraint ans skrinkwrap is nice but not as new topology pen in modo) , so I using it sometime to retopo my mesh from zbrush. The workflow is the main issue, a bit clunky if compared to XSI.

I would like some basic sculpt brush in XSI, modern UVmap set of tools (ala UVLayout) and a 3d paint set... the base are just here and FX is really nice and can match with Bodypaint layers. I'm sure a day, when Softimage will turn back to AVID, ( ) will have this tools directly in softimage. Or tomorrow AD will bring Mudbox inside Softimage? Who know... are mad these developers


Luk-eric Skywalker make my

[Daniel Brassard]

Interesting little tidbits from the mailing list today. Some of the team members working on Softimage presented themselves.

http://www.mail-archive.com/softimage@l ... 00255.html

[iamVFX]

http://www.mail-archive.com/softimage@l ... 00255.html

Yury Khmel - Core, ICE, FaceRobot. 12+ years, last 5 as an architect in games
development.

Tovarisch!

[Fer]

Three info I got from this mail: Guillaume Laferriere is gone also , Autodesk have gathered all the veterans for their maya project, and Softimage is now a low cost development project. Sorry if I sound harsh for the new recruits, but just don't like how this is turning out.

just who gonna fix mental ray in Softimage right know ?

[ActionArt]

I think it's good in a way though, like Luc-Eric said, these are new talents. Maybe the viewport, and UV tools will get a good lift.

As for MR, don't they have "Master Zap" for that? He's probably too busy on other packages though. Maybe he'll get to SI next time around

[Nizar]

So old softimage montreal team is now Maya VFX labour and the new softimage team is Singapore based? I'm a bit confusing about.
The fact XSI is more strong in Asian/Japan studio influenced this decision or is only a coincidence related to tax and bureaucratic facility?

[brudney]

That's really interesting whats going on in AD right now.

When I read about oldies leaving SI team my heart stopped for a sec, but then I thought that maybe it's not the end of the world and hopefully the new guys will pump some fresh blood into SI. I guess time will tell. It would be a pity if SI slowly faded away... What would be even worse is that I'd have to start using maya again;)

Speaking of which, that's really interesting what they are planing to do with it. Hopefully they will do more than reanimate the corpse. What seems quite obvious to me is they're developing ICE-like platform and hopefully making the whole application take full advantage of its node based architecture and exposing them properly (houdini-sorta-thing?) Or maybe I'm just fantasising...

Anyway, who knows, maybe I'll take a look at maya in my spare time and try to remember what I tried to forget when I switched to SI;)