OctaneBench® 4.00 - 4x Tesla P100-SXM2-16GB     2 results

Maximum 1,013.40 Average 1,008.19
Minimum 1,002.98 Median 1,002.98

GPU Model No. GPUs Compute
OctaneRender Support
v4 v3 v2
Tesla P100-SXM2-16GB4????

Kernel Score #2 Weight #3 Sub-total
Info Channels10920.10109.16
Direct Lighting10020.40400.88
Path Tracing9960.50498.15
Total Score #21008.19
Scene Kernel Ms/s #4 Score #2
Interior (by Julia Lynen)Info Channels594.781154
Interior (by Julia Lynen)Direct Lighting196.291103
Interior (by Julia Lynen)Path Tracing90.061055
Idea (by Julio Cayetaño)Info Channels720.43838
Idea (by Julio Cayetaño)Direct Lighting199.16946
Idea (by Julio Cayetaño)Path Tracing179.86928
ATV (by Jürgen Aleksejev)Info Channels390.811245
ATV (by Jürgen Aleksejev)Direct Lighting148.50976
ATV (by Jürgen Aleksejev)Path Tracing125.40971
Box (by Enrico Cerica)Info Channels742.511129
Box (by Enrico Cerica)Direct Lighting136.12984
Box (by Enrico Cerica)Path Tracing138.801032
These values are calculated from the averages of all submissions and may not be representative of actual performance.

2 results

#1 What score is recommended for Octane?
This depends on your scene complexity and time-frame, but we recommended a score no lower than 45 for good render performance.

Please note that cards must have a score of 20 or higher to meet Octane's minimal performance requirements. While cards below this level may still be compatible, Octane's performance will be significantly impacted.
#2 What does the score value mean?
The score is calculated from the measured speed (Ms/s or mega samples per second), relative to the speed we measured for a GTX 980. If the score is under 100, the GPU(s) is/are slower than the GTX 980 we used as reference, and if it's more the GPU(s) is/are faster.
#3 What does the weight value mean?
The weight determines how each kernel's score affects the final score, and kernels that have higher usage are weighted higher.
#4 What is Ms/s?
Ms/s is mega-samples per second, this value is the average of all the results uploaded to OctaneRender for this/these GPU(s).