I realize that hardware questions are off-topic here, but I think this question about documentation and optimization is not really hardware dependent - it's a question of being able to reproduce consistent results with different versions and installations of Blender.

It took me a while searching the web and this site to realize why I was getting such poor results rendering the classroom benchmark scene on my new system that came with an NVIDIA GeForge GTX 1070. The default settings were set to render with the CPU, and when I tried to switch it to CUDA-based GPU rendering, things only got slower. Searching the web I found others had gotten it to render in about 7 minutes on the same hardware (GTX 1070). Finally I found the answer: increase the tile size. It seems like a lot of effort to figure out some basic configurations if you're not intimately familiar with how rendering works. And I still haven't matched the other benchmark. I'm at 8:38, and the other benchmark was at 7:47.

So my questions are:

  1. Is there some way to know to what I should attribute this variation? My benchmark was on Blender 2.78c and the original was on 2.78a. So it could be that or something else.
  2. Is there any documentation covering optimal settings for the benchmark scenes? I realize optimization is very dependent on hardware and scene, but especially for benchmark scenes on the most typical hardware, it would be useful to know how to reproduce consistent results so we can, for example, compare the performance of different versions of Blender.

Edit: The classroom scene I downloaded from https://www.blender.org/download/demo-files/, and I just realized that it's not technically labeled as a benchmark.

  • $\begingroup$ @cegaton But some of the "standardized settings" you mention seem to be missing. Do all the blenchmark timings listed at blenchmark.com/gpu-benchmarks really run at a tile size of 16x16? That would be terrible if so because they would all run slower than CPUs! $\endgroup$
    – BlueMonkMN
    Commented Jun 3, 2017 at 15:02
  • $\begingroup$ The results listed on blenchmark.com/gpu-benchmarks are done with the settings for the BMW blenchmark so they should take into account the a larger tile size. $\endgroup$
    – user1853
    Commented Jun 4, 2017 at 17:04

1 Answer 1


Use the blenchmark Addon: http://blenchmark.com/article/benchmark-your-cpu-or-gpu. The istructions on how to install and use it are explained very clearly on the site

There are two basic settings: one for CPU and one for GPU, don't use the one for CPU to test a GPU.

The idea behind the blenchmark is to have a standardized set of parameters by which to compare systems. Just set the compute device, leave all other setting untouched and let it render.

Changing the settings for each machine would result in a comparison like watermelons and oranges.

Note that it is quite easy to render faster than the average results, as the blenchmark is designed to work on many different types of hardware and Is not optimized for any particular one and it is set to work on a very wide range of hardware.

  • $\begingroup$ This answers the question for blender-based benchmarking in general, but it applies only to the BMW benchmark. How is one supposed to use the classroom benchmark? $\endgroup$
    – BlueMonkMN
    Commented Jun 4, 2017 at 9:55
  • $\begingroup$ I don't know such benchmark, but if it doesn't have different settings for CPU and GPU I don't think that the test would be of any use. $\endgroup$
    – user1853
    Commented Jun 4, 2017 at 16:30
  • $\begingroup$ It's one of the standard benchmarks from blender's site: blender.org/download/demo-files I see now that not everything on that page is labeled "benchmark". I didn't notice that before. $\endgroup$
    – BlueMonkMN
    Commented Jun 4, 2017 at 16:31
  • $\begingroup$ The classroom file has no indication on what Tile size to use for GPU. Dealing with arbitrary values set by each user or using one set clearly for CPU makes no sense as a benchmark. $\endgroup$
    – user1853
    Commented Jun 4, 2017 at 17:09

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .