NVIDIA Maxwell (GTX 970) - OpenCL vs CUDA - 1:0

\n studio-striking\n

Message boards : Number crunching : NVIDIA Maxwell (GTX 970) - OpenCL vs CUDA - 1:0
Message board moderation

To post messages, you must log in.

AuthorMessage
Microcruncher*
Avatar

Send message
Joined: 8 May 11
Posts: 11
Credit: 1,075,941
RAC: 0
Message 6270 - Posted: 29 Oct 2014, 18:13:10 UTC
Last modified: 29 Oct 2014, 18:18:47 UTC

Driver Version 344.48 (344.16 is buggy and serial crashes on machines with 32 GB RAM), Windows 8.1 - 32 bit CUDA/OpenCL clients from distributed.net:

CUDA Client from Distributed.net (v2.9110.519b):

[Oct 29 17:56:28 UTC] RC5-72: using core #6 (CUDA 4-pipe 64-thd).
[Oct 29 17:56:33 UTC] RC5-72: Benchmark for core #6 (CUDA 4-pipe 64-thd)
0.00:00:02.56 [1,705,648,911 keys/sec]


OpenCL Client from Distributed.net (v2.9111.520):

[Oct 29 17:49:05 UTC] RC5-72: using core #0 (CL ANSI 1-pipe).
[Oct 29 17:49:09 UTC] RC5-72: Benchmark for core #0 (CL ANSI 1-pipe)
0.00:00:01.93 [2,289,978,422 keys/sec]
[Oct 29 17:49:09 UTC] RC5-72: using core #1 (CL 1-pipe).
[Oct 29 17:49:12 UTC] RC5-72: Benchmark for core #1 (CL 1-pipe)
0.00:00:01.65 [2,682,184,538 keys/sec]
[Oct 29 17:49:12 UTC] RC5-72: using core #2 (CL 2-pipe).
[Oct 29 17:49:16 UTC] RC5-72: Benchmark for core #2 (CL 2-pipe)
0.00:00:01.54 [2,855,800,503 keys/sec]
[Oct 29 17:49:16 UTC] RC5-72: using core #3 (CL 4-pipe).
[Oct 29 17:49:19 UTC] RC5-72: Benchmark for core #3 (CL 4-pipe)
0.00:00:01.81 [2,432,125,750 keys/sec]
[Oct 29 17:49:19 UTC] RC5-72 benchmark summary :
Default core : #-1 (undefined) 0 keys/sec
Fastest core : #2 (CL 2-pipe) 2,855,800,503 keys/sec


Can someone with a 780ti post OpenCL results?
ID: 6270 · Rating: 0 · rate: Rate + / Rate - Report as offensive

Message boards : Number crunching : NVIDIA Maxwell (GTX 970) - OpenCL vs CUDA - 1:0


 
Copyright © 2011-2024 Moo! Wrapper Project