“Integrating GPU support is going to be a mandatory task for any open source rendering engine. Luxrender team want to share his effort with other open source developers and decided to develop LuxRays has a stand-alone library. This library will be developed with the main task to accelerate Luxrender but it will be available for any other project that don’t want to waste its time developing a duplicate of our work. Indeed, as in any open source project, any contribution will be highly welcome.”
Yeah, and just think about it. Once not too long ago the clearly stated goal was to get lux integrated into blender. Radiance even visited Ton for a meeting (unofficial?) in Amsterdam. I have no idea what happened in that meeting, but soon after radiance left lux, and now the whole blender integration idea is nowhere to be found.
That’s because it involved 2.49 which was on it’s way out and didn’t have the flexibility to support such a large change.
With 2.5 an integration is no longer needed, however something like OpenCL accelerated rendering is something which will directly benefit blenders rendering engine, meaning it’s something worth while to do.
I don’t know what radiance’s plan was, but if it only involved integration on the level of yafray in the old days, you are of course right.
But if the plan was to make lux a fully official integrated part of Blender, then I don’t agree. On paper you might be able to access everything with the new API, but in reality there are many more variables. An external render will always be one step behind in both user base and feature integration - integrating the features of Blender that is (like particles/volumetrics and so on), not overall render features of course!
An external render will always be the ugly step child
so! what should we do now! build a serverpark with SLi/Crossfire comps:D and run network luxrenders! that would be awesome. that kind of renders for animation works
Regarding Blender integration, just hold your breath, you might have a nice surprise coming but probably not before Blender 2.6 release, the Blender API is still a bit too much in flux right now…
Would this be the 2.5 Lux script? I’ve seen some shots of it and I’m very excited about it. Having all of the available features in corresponding panels (materials, lights, environment and rendering) will be a huge (Read: huge huge huge) step forward, not only for blender for but for the lux community.
I can see both communities growing rapidly from these developments, they rival and in some cases out class commercial offerings.
I think many people would be very interested in hearing more about your thoughts and hopes for the future even if they are still fuzzy. Please don’t hold yourself back if you have more to add :yes:
Let’s just say that Blender and LuxRender API are both maturing and that this opens very interesting doors (hint: LuxRender API now has python bindings).
its not always the case mental-ray, v-ray, prman, maxwell… are external renders made by different companies from the ones making the 3d software yet are prefered renderers by most companies out there.
Even take the case of yafaray it is in front of BI on some place but it lags in others why that is the case I wouldn’t really know.
I just wanted to point out the fact that an external render does n’t always lag the internal offering. But this library is pretty cool I hope the yafaray guys and the Bl guys can make use of it.
that than counts also for blender because it hardly offers what VRay offers.
Whats the point of having all the fancy tools and compositor when the
important part, the render engine is kinda weak - duh.
Yes VRay might not have access to Blenders compositor, however I rather
have basic render to VRay but have access to a state of the art render engine.
I can always build and render passes done in different engines and compose it
back later.
While BI might not be the, uhm, most up to date render engine around, there are still not to my knowledge any external render engine around that supports all the blender features. There is always something you have to manually convert to geometry or bake or something. And that is of course to be expected when the devs of those engines have to take the features of multiple modelling packages into account. An internal render engine on the other hand is of course “forced” to support all the new features from day one.
External render engines are on the other hand often on the forefront when it comes to overall rendering optimisation and features.
Is there anything official going on between you guys and the blender devs? If I may ask.
Just imagine the power of an official bundling of the two.
True. My point is just, an external engine will (of course) never be as tightly integrated as an internal. You always have to go through loopholes to get everything to fit together. That is especially frustrating for newcomers I could imagine.
:eek: The only “Gasp!” here is that all these external renderers are very demanding in all sort of knowledges. Hard to approach and use if not completely skilled (and even especially technically schooled). Too many parametres to tweak and values to manage.
Therefore, a simple, built in, adapted and compact way to render is mostly needed… Yet, there should still be a way to optain stunning results… of course! :D:D
