Broken APEX49 Posted August 30, 2020 Share Posted August 30, 2020 Everyone, hello. While my score is calculation for some reason, the time moves slightly. When thinking a change in about 14 seconds, it's of more than 20 seconds. When the passage other persons are calculating is seen, there is no blurring, but 、、、. Are there any fatal problems? Please take care above. Quote Link to comment Share on other sites More sharing options...
Broken APEX49 Posted August 30, 2020 Author Share Posted August 30, 2020 By all means. Would you advise? Quote Link to comment Share on other sites More sharing options...
Broken APEX49 Posted August 30, 2020 Author Share Posted August 30, 2020 allocated device memory? Is the adjustment effective including an adjustment of bacth size? A way? Quote Link to comment Share on other sites More sharing options...
Crew Leeghoofd Posted August 30, 2020 Crew Share Posted August 30, 2020 Play with the batch sizes to improve your time.. And yes gpupi is not like superpi where the loops are more predictable. Loop after loop 1 Quote Link to comment Share on other sites More sharing options...
Broken APEX49 Posted August 30, 2020 Author Share Posted August 30, 2020 Mr. Leeghoofd Thank you very much for your information. Quote Link to comment Share on other sites More sharing options...
nick_name Posted August 30, 2020 Share Posted August 30, 2020 @GTI-R Can you let us know how it goes with your testing of batch sizes? I'm new and would like to learn more about running GPUPI also. 1 Quote Link to comment Share on other sites More sharing options...
jab383 Posted August 30, 2020 Share Posted August 30, 2020 nick_name: Please don't rely on someone else's tests of tweaks like GPUPI batch size. Your results will differ depending on many things. In this example, number of cores, core clock, process priority, memory size, memory clock and size of cache all make differences. I know it takes a bit of time to go through many combinations of batch and reduction sizes and more to have to do that for every CPU/memory combination you want to bench. That's the beauty of a complex, variable set of tweaks like in GPUPI -- it rewards effort. 2 Quote Link to comment Share on other sites More sharing options...
Crew Leeghoofd Posted August 30, 2020 Crew Share Posted August 30, 2020 GPUPI-CLI.exe -d 100M -c ... Calculate 100M on all CPUs in the system and test for the best combination of benchmark parameters (this will take a while) So probably replacing 100M with 1B will do the same for you ? Fastest to do is to check the other submissions in the competition... so for me I would try 10M 1024 first to see if it goes a bit faster, but as Jab383 mentions it might not jield the expected result.... 2 Quote Link to comment Share on other sites More sharing options...
nick_name Posted August 30, 2020 Share Posted August 30, 2020 2 hours ago, jab383 said: nick_name: Please don't rely on someone else's tests of tweaks like GPUPI batch size. Your results will differ depending on many things. In this example, number of cores, core clock, process priority, memory size, memory clock and size of cache all make differences. I know it takes a bit of time to go through many combinations of batch and reduction sizes and more to have to do that for every CPU/memory combination you want to bench. That's the beauty of a complex, variable set of tweaks like in GPUPI -- it rewards effort. Understood. I'm just trying to take in as much as I can where I can to learn. Quote Link to comment Share on other sites More sharing options...
Broken APEX49 Posted August 30, 2020 Author Share Posted August 30, 2020 意味がわからない Quote Link to comment Share on other sites More sharing options...
Broken APEX49 Posted September 1, 2020 Author Share Posted September 1, 2020 Batch size is the reduction tendency. The good way? Quote Link to comment Share on other sites More sharing options...
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.