Index  | Recent Threads  | Unanswered Threads  | Who's Active  | Guidelines  | Search
 

Quick Go ยป
No member browsing this thread
Thread Status: Active
Total posts in this thread: 5
[ Jump to Last Post ]
Post new Thread
Author
Previous Thread This topic has been viewed 1848 times and has 4 replies Next Thread
Former Member
Cruncher
Joined: May 22, 2018
Post Count: 0
Status: Offline
Reply to this Post  Reply with Quote 
CUDA-vs-OpenCL

A paper from IEEE.org

Product Type: Conference Publications

TITLE:
A Comprehensive Performance Comparison of CUDA and OpenCL

ABSTRACT:
This paper presents a comprehensive performance comparison between CUDA and OpenCL. We have selected 16 benchmarks ranging from synthetic applications to real-world ones. We make an extensive analysis of the performance gaps taking into account programming models, optimization strategies, architectural details, and underlying compilers. Our results show that, for most applications, CUDA performs at most 30% better than OpenCL. We also show that this difference is due to unfair comparisons: in fact, OpenCL can achieve similar performance to CUDA under a fair comparison. Therefore, we define a fair comparison of the two types of applications, providing guidelines for more potential analyses. We also investigate OpenCL's portability by running the benchmarks on other prevailing platforms with minor modifications. Overall, we conclude that OpenCL's portability does not fundamentally affect its performance, and OpenCL can be a good alternative to CUDA.

Available Format: PDF
Member: US$10.00
Non-member: US$31.00

Interested?
http://ieeexplore.ieee.org/xpl/freeabs_all.js...rnumber=6047190&tag=1

Note:
The "o" in "optimization strategies,..." is missing in the source webpage as seen 2012.04.22Su.1230.utc. I supplied that "o" in the abstract written above.
;
[Apr 22, 2012 12:53:56 PM]   Link   Report threatening or abusive post: please login first  Go to top 
Former Member
Cruncher
Joined: May 22, 2018
Post Count: 0
Status: Offline
Reply to this Post  Reply with Quote 
Re: CUDA-vs-OpenCL

I thought the main difference though, had to deal with the fact that NVIDIA actually helps its devs. Customer support, right?
[Apr 22, 2012 2:38:28 PM]   Link   Report threatening or abusive post: please login first  Go to top 
Jim1348
Veteran Cruncher
USA
Joined: Jul 13, 2009
Post Count: 1066
Status: Offline
Project Badges:
Reply to this Post  Reply with Quote 
Re: CUDA-vs-OpenCL

Having spent several years Folding on Nvidia cards after using ATI/AMD, the difference is real enough there, though that was before OpenCL days. I strongly suspect that they define "fair" the same way that educators define "fair" when they want to achieve certain test results.

But the difference can go the other way. Will they say that AMD's increased performance over NVIDIA in certain OpenCL projects is unfair too?
----------------------------------------
[Edit 2 times, last edit by Jim1348 at Apr 26, 2012 2:28:50 PM]
[Apr 26, 2012 1:33:36 PM]   Link   Report threatening or abusive post: please login first  Go to top 
JEklund2
Advanced Cruncher
Finland
Joined: Aug 10, 2006
Post Count: 119
Status: Offline
Project Badges:
Reply to this Post  Reply with Quote 
Re: CUDA-vs-OpenCL

laughing When talking about anything related to sw in my list the "fair" can be placed anywhere in the next "value chain" :

truth --> lie --> statistics --> benchmark --> propaganda --> ???

( maybe the benchmark can be even "right side" than the propaganda .. tongue )
----------------------------------------

[Apr 26, 2012 6:47:34 PM]   Link   Report threatening or abusive post: please login first  Go to top 
mikey
Veteran Cruncher
Joined: May 10, 2009
Post Count: 824
Status: Offline
Project Badges:
Reply to this Post  Reply with Quote 
Re: CUDA-vs-OpenCL

laughing When talking about anything related to sw in my list the "fair" can be placed anywhere in the next "value chain" :

truth --> lie --> statistics --> benchmark --> propaganda --> ???

( maybe the benchmark can be even "right side" than the propaganda .. tongue )


I think in the end the only "fair" thing they are considering is there is now only ONE piece of software to write that covers BOTH types of gpu cards, not one program for EACH type of card. This makes it easier on the projects but not as efficient for the crunching. Time should bring the tweaks back and we should be closer to where we were again.
----------------------------------------


[Apr 27, 2012 1:53:14 PM]   Link   Report threatening or abusive post: please login first  Go to top 
[ Jump to Last Post ]
Post new Thread