Autodesk, Pixar and Open Subdivs

[Mathaeus]

As the "Cult of Roosendaal" implements everything they can get their grubby hands on, I'm not all that surprised.
I wonder, however how long it will take them this time
and if this will be deemed important enough to actually change their roadmap drastically?

Disclaimer: Still not a big fan of Blender, in case you hadn't noticed...

So you still don't have a big picture of him. At least, small plastic figure, of course printed by new 3d printing toolset in Blender.
Now seriously, implementation of Bmesh (which allowed polygons with more than four sides, Catmull Clark subdivs...), took about 1-2 years, depends on counting. Together with slow rewiring of related modifiers, modeling ops, so on. Finished last year, so probably they will have smarter business to do, today.

But this time, I have impression they got "the offer they can't refuse", so maybe everything will be done much faster.

[Lord Futzi Voldemort]

As the "Cult of Roosendaal" implements everything they can get their grubby hands on, I'm not all that surprised.
I wonder, however how long it will take them this time
and if this will be deemed important enough to actually change their roadmap drastically?

Disclaimer: Still not a big fan of Blender, in case you hadn't noticed...

Considering what they achieve with only a small number of volunteers who work mostly part time one or two on one project each, they can be considered as lightning fast, compared to companies with millions of budget and a staff of full time workers. Every one in the commercial world should bow before them.
And not to forget: they really listen to their users ;)

[Nizar]

Bmesh was a very tedious and long project. Due open source nature and the programmer behind the first code did a "bit" of mess and was very slow. Other projects came to life in more short time, i.e. compositor improvement or cycles.

Now blender foundation duplicated the Google found for developing blender, also, gain a lot of interest. Actually I think is the best choice for many freelancers. Also, around the entire software, there is a great consistency, all work with some logic. I like it (and I was a detractor) and work with blender is very fun: very fast, no frills and nice organization (you need to understood logic of the software), also is really stable, actually didn't see a single crash in months. And yes, I like also the interface

And yes, Lord Futzi, agree with you. Yesterday I found a bug, reported it, and in hour Campbell answered me explaining is just in TODO list for the next release. Great.

Pixar subdivision implementation will add also ptex in blender, and paint in blender is just nice (it is not Mari o Mudbox, don't forget, but doing a decent job

[Hirazi Blue]

@Lord Futzi Voldemort - If this were an arms race, Blender would undoubtedly rule the world, but adding feature upon feature does not necessarily make a very useful application. Their hard work is no doubt commendable, BTW, and they are doing very interesting stuff, albeit in the high-profile open source community, where your work is so much more visible than in the closed shops of proprietary software..
It's like comparing oranges for free with apples at the hefty price tag. Sadly, I just don't care too much for oranges, at any price, I'm afraid.
;)

[Nizar]

Blender is good tools, independently by the price. I using now more and more blender and not because is free.

[Lord Futzi Voldemort]

@Lord Futzi Voldemort - If this were an arms race, Blender would undoubtedly rule the world, but adding feature upon feature does not necessarily make a very useful application. Their hard work is no doubt commendable, BTW, and they are doing very interesting stuff, albeit in the high-profile open source community, where your work is so much more visible than in the closed shops of proprietary software..
It's like comparing oranges for free with apples at the hefty price tag. Sadly, I just don't care too much for oranges, at any price, I'm afraid.
;)

What are you doing with your oranges?

[Mathaeus]

@Lord Futzi Voldemort - If this were an arms race, Blender would undoubtedly rule the world, but adding feature upon feature does not necessarily make a very useful application. Their hard work is no doubt commendable, BTW, and they are doing very interesting stuff, albeit in the high-profile open source community, where your work is so much more visible than in the closed shops of proprietary software..
It's like comparing oranges for free with apples at the hefty price tag. Sadly, I just don't care too much for oranges, at any price, I'm afraid.
;)

Well to be fair, they already planned to do something with viewports, this summer.
And this one is not just a feature, could be more a basic thing. Also, if it will go in the same way as bmesh, OpenSubdiv will be replacement of existing subdivs.
Even more (if you're still able to read Ton's words )
"In Cycles OpenSubdiv will replace the incomplete subdivision surface implementation, to enable faster BVH builds and lower memory usage for subdivision surfaces, as well as Ptex rendering support".
If this will work, Blender will be the first 3d app, having the same engine for viewports and renderer, which then is strategic advantage.
I don't see the open source nature, as problem. Remember the evolution from Unix/Irix/whatever, to Linux. Does people wants to use this app, that's important.
As far as I know, a lot people wants, and that's nice.

[Hirazi Blue]

Nope, that was not what I meant by my open source remark. Open Source enables developers to be way more open concerning their roadmaps and the progress of implementations, if you compare it to proprietary software, especially certain kinds of "Sarbanes-Oxley" related development, where you'd have to keep everything you do a secret until you can release. In that sense Blender's development is much more high-profile...
;)

[Mathaeus]

well, 'if you're still able to read Ton's words'. meant something I can only understand as your.. dislike... of this guy - to extent where you don't want to read his words.
'open source is no good', well it seems I've talked with myself ( as usually). For example, sad story, that one of well known Blender artist has to show a collection of works which aren't accepted in some popular internet gallery, where only bad thing that's repeat, is 'created in Blender'. These galleries are probably irrelevant for those who already has working experience, but for new guys... I think the best way is what already happening, that Blender Artist site has almost the same traffic as mentioned gallery (together with all sub-forums).

Now I'll try to stop in this thread.

[Hirazi Blue]

As you continually and probably quite deliberately misunderstand/misrepresent everything I say about your dear Blender, I have nothing more to add!

[luceric]

blender is still mostly opengl 1.x and their roadmap has a re-write and re-design the viewport drawing, to name one re-write. It's kind of ironic indeed that they keep shoving new things in on top of the old stuff. I would have expected an open source app to be always on top of things, or near that, since they shouldn't be dragged down by legacy support, supporting clients, etc, and there should be constant enhancement in all areas since the open source volunteer should be doing "the right thing" instead of working on features driven by marketing. That's kind of the point of open source, people would go in an contribute fixes and enhancements across the board, there isn't supposed to be forgotten areas. Compare to other open source projects like apache, linux, firefox, vlc, etc, these are doing much better at what they do, there are no apologies required. Blender is five years older than XSI, and just as old as 3DSMax but except for UI, it seemed to be immuned to criticism on the internet, because it's free.

Also, the code base should be much smaller than the 12 mil lines that are max/maya/softimage, so it should be easier/faster to change it. But it doesn't look so different to a commercial app in practice. They've figuring out that maybe they should not have worked so much on the game engine and want to scale that down. that's time they could have put into something else instead. My point is that is that the roadmap doesn't look like something you'd expect from an open source that's been in development since 1995. It's very similar the one of any old application that is due for rewrite and renovation, like lightwave or something. except it's free.

Even more (if you're still able to read Ton's words )
"In Cycles OpenSubdiv will replace the incomplete subdivision surface implementation, to enable faster BVH builds and lower memory usage for subdivision surfaces, as well as Ptex rendering support".
If this will work, Blender will be the first 3d app, having the same engine for viewports and renderer, which then is strategic advantage.

In my opinion, this has no meaning whatsoeve. Max has been using using the mental ray subdiv engine in the viewport for a decade and therefore already does this. XSI installs its own subdivision engine in mental ray and also does this, and it shares geometry data structures through geometry callback with mental ray. And every app that has a built-in renderer shares the engine. Ho were we talking about doing displacement in the viewport with the same code that is used in the renderer? Not sure that's a strategic advantage (artist doesn't care about code, just the result being the same, and not everyone wants to sue Cycles), but does anyone really think opensubdiv will be fully implemented in blender before Maya? OpenSubdiv is being developed for Maya in the first pace.

[Nizar]

Blender development, if excluding the project developed directly by Blender Foundation (composed by three guys) is erratic: a volunteer see the Dynamic Topology in 3dcoat and try to develop it inside Blender, another one like Pie Menu, so doing it. They announcing the new feature, but no one are sure they will finish it in time or finish at all. So Blender is subjected by commercial rules, or, more exactly, by fashion rules.

Blender guarantee a legacy compatibility (from 2.43? don't remember).

Some projects was ready very fast, Cycles is one of these, Bmesh was a single case where the feature, in high demand by community, was so delayed.

Blender was old like 3dsm but the found they have, and number of developers employed in the project is a "bit" different, and comparing Blender to other open source project is not fair: linux is supported by IBM and other big names (MS), FireFox has big revenue from google etc. Blender has only the Google summer of code, some (poor) CEE founds, donation by users (about 3000 euro per month) and Blender shop.
Independently by founds Blender give you a capable toolset, you can make money with this and live with only blender in your arsenal. This is great.

As Luceric said, an artist don't need to know what exactly is under the hood, indeed I don't see particular issue in Blender: opengl 1.x or 2, I don't know, I know can run my project and many others, some doing shorts, some realizing nice archiviz and other simple animation.

Blender will not substitute softimage , but is nice it exist and the agility and effort by Blender Foundation deserves respect, especially when comparing with big company not particular loved by its own customers.

[Lord Futzi Voldemort]

It's kind of ironic indeed that they keep shoving new things in on top of the old stuff. I would have expected an open source app to be always on top of things....

Well we know that procedure well from your employer don't we? And I expect commercial software to be on top of things even more and it doesn't happen....

[bergj]

It's kind of ironic indeed that they keep shoving new things in on top of the old stuff. I would have expected an open source app to be always on top of things....

Well we know that procedure well from your employer don't we? And I expect commercial software to be on top of things even more and it doesn't happen....

Maybe you should jump ship to houdini or something if you hate AD that much.

[Lord Futzi Voldemort]

It's kind of ironic indeed that they keep shoving new things in on top of the old stuff. I would have expected an open source app to be always on top of things....

Well we know that procedure well from your employer don't we? And I expect commercial software to be on top of things even more and it doesn't happen....

Maybe you should jump ship to houdini or something if you hate AD that much.

So I cannot make any comment anymore that has a hint of ad in it?
I was reacting to a statement of another member and I think it was completely valid to point that out. It is something everyone here knows. I didn't bring it up.

If you don't like what I have to say, feel free to ignore it, but don't patronize me!

[bergj]

So I cannot make any comment anymore that has a hint of ad in it?
I was reacting to a statement of another member and I think it was completely valid to point that out. It is something everyone here knows. I didn't bring it up.

If you don't like what I have to say, feel free to ignore it, but don't patronize me!

Dont troll the other members of the forum then if you dont want to be "patronized". Your signature - a good day is a day without AD, your attitude over and over pretty much saying that AD is shit - it gets OLD. Like I said, try out houdini and /or something else like Blender.