Text only | Text with Images

Crunchers Anonymous

Cafe => General Discussion => Topic started by: Caprio on September 27, 2012, 04:08:25 AM

Title: Difference in memory availability for GPU 0 vs GPU 1
Post by: Caprio on September 27, 2012, 04:08:25 AM
Hello,

Why is the memory availability so huge between cards in different slots?

27/09/2012 20:39:54 |  | NVIDIA GPU 0: GeForce GTX 680 (driver version 306.23, CUDA version 5.0, compute capability 3.0, 2048MB, 8382279MB available, 3252 GFLOPS peak)
27/09/2012 20:39:54 |  | NVIDIA GPU 1: GeForce GTX 670 (driver version 306.23, CUDA version 5.0, compute capability 3.0, 2048MB, 1961MB available, 2915 GFLOPS peak)
27/09/2012 20:39:54 |  | OpenCL: NVIDIA GPU 0: GeForce GTX 680 (driver version 306.23, device version OpenCL 1.1 CUDA, 2048MB, 8382279MB available)
27/09/2012 20:39:54 |  | OpenCL: NVIDIA GPU 1: GeForce GTX 670 (driver version 306.23, device version OpenCL 1.1 CUDA, 2048MB, 1961MB available)

I.e. why is there 8TB memory available for device 0 while device 1 has 1,9GB available? Is this a bug?
Title: Re: Difference in memory availability for GPU 0 vs GPU 1
Post by: arkayn on September 27, 2012, 08:56:48 AM
Yes, it is supposed to be fixed in the BOINC 7.0.36 build.
Text only | Text with Images