Scene for Benchmarking

Here are 2 files one should fit in an 8GB GPU and the other in a 12GB GPU
For the 12 GB I used a:
Titan X with 12 GB and simmed in 14 m 48 secs
For the 8GB scene
1070 8GB simmed in 5 m 53
Titan X 12GB simmed in 7m 1 sec
Though the Titan has more cores and more mem the 1070 has a higher core and mem clock rate.

Some one on the cafe tried the 12GB scene on a 2080-Ti 11GB GPU and simmed in 6m 5 secs which is more than 2x faster than the Titan X
Though not a render test he got frame 75 of the larger file to render in 3 m 41 secs and I got 6m 31 where 5m 30 of that was preparing geometry. Using the 1070 and the Titan X.
Converting to vdb i get the frame to render in 30 seconds !!
As this is not a render test leave the C4D light on when rendering in redshift, I know it should not be on but i forgot to turn it of and the cafe chap tested with the scene as is so i am basing my times on an as is file. 8g.c4d (342.1 KB) 12g.c4d (315.1 KB)

bumping to see if we can get some more benchmarks…well any in fact…come on peeps chip in.

The sim times include cache storage, which often exceeds the net GPU sim time. The total sim time is therefore a system benchmark. The HW specs should thus include at least the SSD model but also the CPU to be sure.

1 Like

OK can’t edit that post by I did these tests on an I7 3930K at 4.0 Ghz with 64GB RAM and a standard SATA 7200rpm Hard Drive.