Author Topic: Lower performance of OpenCL demo program on Intel Xeon compared to Intel Core  (Read 2866 times)



0 Members and 1 Guest are viewing this topic.

gpu_user1

  • Newbie
  • *
  • Posts: 1
    • View Profile
Hello,
I tested "CL CPU - Post FX" demo program on my laptop and a Xeon based workstation. The systems' info shown below:
===================================[ System / CPU ]
- CPU Name: Intel(R) Xeon(R) CPU E5-1620 0 @ 3.60GHz
- CPU Core Speed: 3591 MHz
- CPU logical cores: 8
- Family: 6 - Model: 13 - Stepping: 7
- Physical Memory Size: 8192 MB
- Operating System: Windows 7 64-bit build 7601 [Service Pack 1]
- PhysX Version: drivers not installed


===================================[ System / CPU ]
- CPU Name: Intel(R) Core(TM) i7-2670QM CPU @ 2.20GHz
- CPU Core Speed: 2195 MHz
- CPU logical cores: 8
- Family: 6 - Model: 10 - Stepping: 7
- Physical Memory Size: 8192 MB
- Operating System: Windows 7 64-bit build 7601 [Service Pack 1]
- PhysX Version: drivers not installed

To my surprise the Xeon computer performed worse than the laptop. The laptop showed 19 frames per second as oppose to 9 frame per second shown by Xeon computer.
Is there any explanation on this? Do I have any setup problems?
I have the following 2 platforms on the systems (no difference in performance):
  • amd-catalyst-omega-14.12-with-dotnet45-win7-64bit
  • Intel OpenCL RunTime 15.1

Your comments will be appreciated.