Good animations-How often?

Zenitor, where can i find the speedy render?

Here is the link where AN][ARES has put it’s marvelous Blender version:

The latest version also works for AMD processors.

The Blender Official renderer has been enhanced (quality)in V 2.34,but it is slower than the V 2.33a.

The Antares version I’m using is the version made from the V 2.33a,because it is faster than the latest one. I am not sure that this is still true because I have seen a topic written by Antares,saying that he had recompiled the soft and encreased the speed. But I am not sure that he was speaking about the very latest version…

You have to check and compare…


Thanks Roubal

Although this post is a bit late I wanted to point out that 2.34 has a bug that slows the rendering time down when raytrace is used. I’m currently using a CVS build and it proves to be much faster as it does not feature the same bug (instead, it has others).

HOWITZER LIVES!!! yeah i use cvs too.

Hi ! A stupid question : What is CVS ?

Is it an OS ? Is it a patch or a script ?

I use Win XP.


the latest version in delevelopmant is often reffered to as cvs version


Short answer: it’s how a lot of open source software is developed between people who rarely, if ever, physically meet.

Roubal, it seems to me that those render-times could be vastly improved by making very aggressive use of compositing and the Sequence Editor. And maybe by making much less use of techniques like Ambient Occlusion.

I do not wish to challenge your approach nor your experience but simply to point out to the Gentle Readers who are also reading this, some possibilities to consider… “Getting a nice image” and “having a life” should not be an either/or choice! :wink:

Notez vous! In the following, I do not mean “layer” to refer to the "twenty-odd buttons on the render screen. I mean, “layers or strips of film, sandwiched together in the Sequence Editor.”

Backgrounds: If the camera does not move, then anything that does not move within the camera field need only be rendered once, as one frame which is then composited into as many output-frames as may be required.

Ambient occlusion: Nice for creating translucent shadows in dark places, but what if we instead created a layer that we could composite into the partially-shadowed area? What if we take advantage of the “Layer” option for lights… enabling some lights to “shine right through” opaque objects in order to provide useful light in the otherwise-shadowed area? What if we build up what appears to be a single solid object as several layers, just in order to achieve an “AO-like” effect?

Flares and glows: Easily added in post-production, using layers that have nothing but the necessary visual effect. All kinds of “visual spice” can be added in this way, in so-called “beauty passes.”

Individual layers per character: You are very likely going to make some “tweak” to some part of your scene. If you put each part onto a layer, only what changes needs to be re-rendered.

Motion blur: If the object is moving, then it needs motion-blur. If the background against which it is moving does not also move, it does not require motion-blur! The shadows cast by the moving object upon said background can be part of the moving-object layer.

Depth of field: A sharply in-focus layer is composited over one that is more blurred. A depth-mask is used to control an output-filter that more or less blurs the content of a particular layer. The same “controlled distortion” techniques can produce a pretty darned realistic fake of the “caustics” in water and glass.

“Fake?” The word carries such a negative connotation when in this context it really refers to “saving yourself from 10 day renders.” The simpler you can make the task you set in front of your computer, the faster you get results.

Alfred Hitchcock’s acting advice to Ingrid Bergman:
“Ingrid, fake it!”

Also, if you are looking into an online rendering farm . . . check out this FREE one. Warning. . . it is still in pre-alpha stage!

Personally, I find it much easier to just distribute the rendering process among the computers manually.

IE: Computer 1 has frames 1 to 100, C2 has frames 101 to 200, C3 has frames 201 to 300, etc, etc, etc.
And if you skip any frames it’s your fault (seriously, one frame it not that big of a deal unless you are rendering fast scenes or rendering them at <15fps)

Interesting program though.