Message boards :
Number crunching :
corona_NSP16_v1, why does this require more units than the others?
Message board moderation
Author | Message |
---|---|
Send message Joined: 10 Feb 22 Posts: 31 Credit: 476,135,569 RAC: 119,565 |
corona_NSP16_v1, why does this require more units than the others? I have noticed the earlier Covid iterations were crunched within 1 week. This has been crunching for over 1 month. I am curious why that is, is it a more complex database file it is comparing the work unit key against or is it the same. More keys? Forgive me, my terminology is wrong but I think you know what I mean. |
Send message Joined: 11 Oct 20 Posts: 337 Credit: 25,678,026 RAC: 9,200 |
Hello! A little explanation can be found here. |
Send message Joined: 10 Feb 22 Posts: 31 Credit: 476,135,569 RAC: 119,565 |
Thank you that confirms more iterations. I hadn't noticed longer compute times but that's probably because I'm used to Rosetta. Do you know if the client is optimized for SIMD instruction sets or is it computing with straight x86? You could get a lot more throughput/energy efficiency if you run a check to see what instruction sets are supported by the computer and run the appropriate executable that could be developed. |
Send message Joined: 11 Oct 20 Posts: 337 Credit: 25,678,026 RAC: 9,200 |
Yes, application will use SIMD and SSE4 instruction compatibility of CPU. Thank you for participation! |
©2024 SiDock@home Team