PrimeGrid
Please visit donation page to help the project cover running costs for this month

Toggle Menu

Join PrimeGrid

Returning Participants

Community

Leader Boards

Results

Other

drummers-lowrise

Advanced search

Message boards : 321 Prime Search : Computation times

Author Message
Profile dannyridel
Volunteer tester
Avatar
Send message
Joined: 3 Feb 19
Posts: 962
ID: 1097922
Credit: 32,467,903
RAC: 35,594
Discovered 1 mega prime321 LLR Gold: Earned 500,000 credits (652,718)Cullen LLR Silver: Earned 100,000 credits (221,150)ESP LLR Gold: Earned 500,000 credits (517,207)Generalized Cullen/Woodall LLR Amethyst: Earned 1,000,000 credits (1,021,028)PPS LLR Ruby: Earned 2,000,000 credits (3,046,515)PSP LLR Gold: Earned 500,000 credits (616,940)SoB LLR Gold: Earned 500,000 credits (533,625)SR5 LLR Silver: Earned 100,000 credits (357,669)SGS LLR Gold: Earned 500,000 credits (726,147)TRP LLR Silver: Earned 100,000 credits (247,743)Woodall LLR Silver: Earned 100,000 credits (144,022)321 Sieve (suspended) Gold: Earned 500,000 credits (506,814)Generalized Cullen/Woodall Sieve (suspended) Bronze: Earned 10,000 credits (30,033)PPS Sieve Ruby: Earned 2,000,000 credits (3,566,518)AP 26/27 Ruby: Earned 2,000,000 credits (2,029,586)GFN Turquoise: Earned 5,000,000 credits (5,885,152)WW Jade: Earned 10,000,000 credits (11,992,000)PSA Silver: Earned 100,000 credits (373,034)
Message 146441 - Posted: 9 Dec 2020 | 14:52:52 UTC
Last modified: 9 Dec 2020 | 14:53:34 UTC

I just got my new r7 3800xt and was overclocking it.
I reached 4.5GHz with 1.2V, and was running 321 LLR on all 8 physical cores. The task is going to finish with a time of ~3.5hrs!
How fast is that? How does it compare to other processors? afaik my R5 2500U requires ~24hrs for one 321 llr task.
____________
My lucky number is 6219*2^3374198+1


Profile petnekProject donor
Send message
Joined: 22 Apr 10
Posts: 5
ID: 59182
Credit: 57,906,363
RAC: 0
321 LLR Gold: Earned 500,000 credits (549,801)Cullen LLR Gold: Earned 500,000 credits (571,857)ESP LLR Gold: Earned 500,000 credits (705,769)PPS LLR Amethyst: Earned 1,000,000 credits (1,201,679)PSP LLR Amethyst: Earned 1,000,000 credits (1,517,183)SR5 LLR Gold: Earned 500,000 credits (510,656)SGS LLR Silver: Earned 100,000 credits (115,411)TRP LLR Gold: Earned 500,000 credits (527,012)321 Sieve (suspended) Gold: Earned 500,000 credits (515,894)Cullen/Woodall Sieve (suspended) Silver: Earned 100,000 credits (444,656)Generalized Cullen/Woodall Sieve (suspended) Silver: Earned 100,000 credits (259,965)PPS Sieve Sapphire: Earned 20,000,000 credits (20,504,077)Sierpinski (ESP/PSP/SoB) Sieve (suspended) Bronze: Earned 10,000 credits (20,038)TRP Sieve (suspended) Bronze: Earned 10,000 credits (47,932)AP 26/27 Ruby: Earned 2,000,000 credits (2,106,403)GFN Ruby: Earned 2,000,000 credits (3,668,831)WW Sapphire: Earned 20,000,000 credits (20,136,000)PSA Ruby: Earned 2,000,000 credits (4,500,550)
Message 150470 - Posted: 23 May 2021 | 11:02:59 UTC

I'm running 321 LLR on R5 2600 and from testing FFT length on Prime95 is best to run 1WU on 1 thread. My times are in this way 36 hours per WU.
____________

Profile [AF>Amis des Lapins] Jean-Luc
Send message
Joined: 5 Feb 10
Posts: 3
ID: 54689
Credit: 2,228,974,648
RAC: 19,859
321 LLR Gold: Earned 500,000 credits (821,096)Cullen LLR Bronze: Earned 10,000 credits (18,509)PPS LLR Bronze: Earned 10,000 credits (18,748)PSP LLR Bronze: Earned 10,000 credits (68,231)SoB LLR Silver: Earned 100,000 credits (120,209)SR5 LLR Bronze: Earned 10,000 credits (26,889)TRP LLR Bronze: Earned 10,000 credits (20,729)Woodall LLR Bronze: Earned 10,000 credits (30,574)Cullen/Woodall Sieve (suspended) Ruby: Earned 2,000,000 credits (2,309,114)PPS Sieve Double Amethyst: Earned 1,000,000,000 credits (1,621,075,444)TRP Sieve (suspended) Bronze: Earned 10,000 credits (18,050)AP 26/27 Emerald: Earned 50,000,000 credits (94,440,543)GFN Jade: Earned 10,000,000 credits (16,983,356)WW Double Silver: Earned 200,000,000 credits (493,008,000)
Message 153772 - Posted: 2 Feb 2022 | 15:17:46 UTC

Please, I have a computer with an AMD threadripper 3990X 64C/128T CPU.
For the 321 LLR project, I saw that it was catastrophic to use 64 threads per task.
I tried 8 threads per task, it is much better.
Then I tried with 6 threads per task, it seems even better !
Does anyone know what is more efficient : should I put 1 thread per task ?
Should I put 3 threads per task, as I think I understand from reading another topic in this forum ?
Thanks to you.

Profile NickProject donor
Avatar
Send message
Joined: 11 Jul 11
Posts: 1855
ID: 105020
Credit: 4,928,333,469
RAC: 15,512,727
Discovered 9 mega primesFound 5 primes in the 2020 Tour de PrimesFound 2 mega primes in the 2020 Tour de PrimesFound 4 primes in the 2021 Tour de PrimesFound 2 mega primes in the 2021 Tour de PrimesFound 3 primes in the 2022 Tour de Primes321 LLR Sapphire: Earned 20,000,000 credits (23,436,311)Cullen LLR Emerald: Earned 50,000,000 credits (50,743,293)ESP LLR Emerald: Earned 50,000,000 credits (50,849,415)Generalized Cullen/Woodall LLR Emerald: Earned 50,000,000 credits (50,473,530)PPS LLR Emerald: Earned 50,000,000 credits (75,402,242)PSP LLR Emerald: Earned 50,000,000 credits (51,069,305)SoB LLR Emerald: Earned 50,000,000 credits (51,885,310)SR5 LLR Sapphire: Earned 20,000,000 credits (41,856,372)SGS LLR Sapphire: Earned 20,000,000 credits (20,482,069)TRP LLR Sapphire: Earned 20,000,000 credits (20,307,419)Woodall LLR Emerald: Earned 50,000,000 credits (51,058,101)321 Sieve (suspended) Sapphire: Earned 20,000,000 credits (20,380,527)Cullen/Woodall Sieve (suspended) Gold: Earned 500,000 credits (744,531)Generalized Cullen/Woodall Sieve (suspended) Turquoise: Earned 5,000,000 credits (5,007,004)PPS Sieve Double Gold: Earned 500,000,000 credits (631,169,403)TRP Sieve (suspended) Bronze: Earned 10,000 credits (21,181)AP 26/27 Emerald: Earned 50,000,000 credits (50,339,393)GFN Double Gold: Earned 500,000,000 credits (591,721,141)WW Double Ruby: Earned 2,000,000,000 credits (3,142,104,000)
Message 153773 - Posted: 2 Feb 2022 | 15:46:30 UTC - in response to Message 153772.
Last modified: 2 Feb 2022 | 16:20:46 UTC

Please, I have a computer with an AMD threadripper 3990X 64C/128T CPU.
For the 321 LLR project, I saw that it was catastrophic to use 64 threads per task.
I tried 8 threads per task, it is much better.
Then I tried with 6 threads per task, it seems even better !
Does anyone know what is more efficient : should I put 1 thread per task ?
Should I put 3 threads per task, as I think I understand from reading another topic in this forum ?
Thanks to you.

This may be not quite correct as I don't have AMD and I haven't looked into this for a while.
I think that CPU has CCXs of size 8 cores - so 8 x 8 core CCXs.
Communication between CCXs is slower.
Communication within a CCX is much faster.
I can't remember the FFT size of 321 (it isn't one of the projects with large FFTs), but I could guess running 4 core tasks might be more optimal (if the CPU does have CCXs of size 8 cores) as that CPU has a bucket load of L3 cache - so there should be no issue to run tasks entirely within cache.
I can't remember if AMD needs managing to make sure tasks don't cross CCXs.

Someone will give a better answer.

Profile dannyridel
Volunteer tester
Avatar
Send message
Joined: 3 Feb 19
Posts: 962
ID: 1097922
Credit: 32,467,903
RAC: 35,594
Discovered 1 mega prime321 LLR Gold: Earned 500,000 credits (652,718)Cullen LLR Silver: Earned 100,000 credits (221,150)ESP LLR Gold: Earned 500,000 credits (517,207)Generalized Cullen/Woodall LLR Amethyst: Earned 1,000,000 credits (1,021,028)PPS LLR Ruby: Earned 2,000,000 credits (3,046,515)PSP LLR Gold: Earned 500,000 credits (616,940)SoB LLR Gold: Earned 500,000 credits (533,625)SR5 LLR Silver: Earned 100,000 credits (357,669)SGS LLR Gold: Earned 500,000 credits (726,147)TRP LLR Silver: Earned 100,000 credits (247,743)Woodall LLR Silver: Earned 100,000 credits (144,022)321 Sieve (suspended) Gold: Earned 500,000 credits (506,814)Generalized Cullen/Woodall Sieve (suspended) Bronze: Earned 10,000 credits (30,033)PPS Sieve Ruby: Earned 2,000,000 credits (3,566,518)AP 26/27 Ruby: Earned 2,000,000 credits (2,029,586)GFN Turquoise: Earned 5,000,000 credits (5,885,152)WW Jade: Earned 10,000,000 credits (11,992,000)PSA Silver: Earned 100,000 credits (373,034)
Message 153791 - Posted: 3 Feb 2022 | 0:07:43 UTC - in response to Message 153773.

Please, I have a computer with an AMD threadripper 3990X 64C/128T CPU.
For the 321 LLR project, I saw that it was catastrophic to use 64 threads per task.
I tried 8 threads per task, it is much better.
Then I tried with 6 threads per task, it seems even better !
Does anyone know what is more efficient : should I put 1 thread per task ?
Should I put 3 threads per task, as I think I understand from reading another topic in this forum ?
Thanks to you.

This may be not quite correct as I don't have AMD and I haven't looked into this for a while.
I think that CPU has CCXs of size 8 cores - so 8 x 8 core CCXs.
Communication between CCXs is slower.
Communication within a CCX is much faster.
I can't remember the FFT size of 321 (it isn't one of the projects with large FFTs), but I could guess running 4 core tasks might be more optimal (if the CPU does have CCXs of size 8 cores) as that CPU has a bucket load of L3 cache - so there should be no issue to run tasks entirely within cache.
I can't remember if AMD needs managing to make sure tasks don't cross CCXs.

Someone will give a better answer.


By looking at the tasks he completed, the FFT size looks like it's around 1M for 321 as of right now. That indicates 1*8=8MB of cache needed for each 321 task. Total L3 cache on the 3990x is 256MB, which means that you can fit at most 32 tasks within the cache. However, as Nick did say, each CCX has 8 cores and their own cache, making it not a very wise decision to run tasks on more than 8 cores. The rest is for you to try and test out! Whether is 4 cores / task or 6 cores / task or less, it depends on your CPU.
____________
My lucky number is 6219*2^3374198+1


Profile [AF>Amis des Lapins] Jean-Luc
Send message
Joined: 5 Feb 10
Posts: 3
ID: 54689
Credit: 2,228,974,648
RAC: 19,859
321 LLR Gold: Earned 500,000 credits (821,096)Cullen LLR Bronze: Earned 10,000 credits (18,509)PPS LLR Bronze: Earned 10,000 credits (18,748)PSP LLR Bronze: Earned 10,000 credits (68,231)SoB LLR Silver: Earned 100,000 credits (120,209)SR5 LLR Bronze: Earned 10,000 credits (26,889)TRP LLR Bronze: Earned 10,000 credits (20,729)Woodall LLR Bronze: Earned 10,000 credits (30,574)Cullen/Woodall Sieve (suspended) Ruby: Earned 2,000,000 credits (2,309,114)PPS Sieve Double Amethyst: Earned 1,000,000,000 credits (1,621,075,444)TRP Sieve (suspended) Bronze: Earned 10,000 credits (18,050)AP 26/27 Emerald: Earned 50,000,000 credits (94,440,543)GFN Jade: Earned 10,000,000 credits (16,983,356)WW Double Silver: Earned 200,000,000 credits (493,008,000)
Message 153829 - Posted: 3 Feb 2022 | 18:08:18 UTC

Thanks so much for your help, I'm in the process of testing !

Profile robishProject donor
Volunteer moderator
Volunteer tester
Avatar
Send message
Joined: 7 Jan 12
Posts: 2134
ID: 126266
Credit: 6,680,677,919
RAC: 2,856,486
Discovered the World's First AP27!!!Discovered 11 mega primesDiscovered 1 AP272018 Tour de Primes largest primeFound 4 primes in the 2018 Tour de PrimesFound 1 mega prime in the 2018 Tour de PrimesFound 1 prime in the 2019 Tour de PrimesFound 1 prime in the 2020 Tour de PrimesFound 1 prime in the 2021 Tour de PrimesFound 1 prime in the 2022 Tour de PrimesFound 1 mega prime in the 2022 Tour de Primes321 LLR Turquoise: Earned 5,000,000 credits (9,681,253)Cullen LLR Emerald: Earned 50,000,000 credits (50,204,972)ESP LLR Turquoise: Earned 5,000,000 credits (8,577,288)Generalized Cullen/Woodall LLR Sapphire: Earned 20,000,000 credits (20,294,046)PPS LLR Emerald: Earned 50,000,000 credits (86,803,321)PSP LLR Turquoise: Earned 5,000,000 credits (9,600,575)SoB LLR Sapphire: Earned 20,000,000 credits (36,846,875)SR5 LLR Sapphire: Earned 20,000,000 credits (35,513,347)SGS LLR Turquoise: Earned 5,000,000 credits (5,741,242)TRP LLR Sapphire: Earned 20,000,000 credits (28,958,280)Woodall LLR Turquoise: Earned 5,000,000 credits (5,062,771)321 Sieve (suspended) Turquoise: Earned 5,000,000 credits (7,141,753)Cullen/Woodall Sieve (suspended) Turquoise: Earned 5,000,000 credits (7,892,369)Generalized Cullen/Woodall Sieve (suspended) Turquoise: Earned 5,000,000 credits (5,515,338)PPS Sieve Double Gold: Earned 500,000,000 credits (842,736,516)TRP Sieve (suspended) Silver: Earned 100,000 credits (121,416)AP 26/27 Double Bronze: Earned 100,000,000 credits (110,369,857)GFN Double Turquoise: Earned 5,000,000,000 credits (5,300,444,699)WW Double Bronze: Earned 100,000,000 credits (109,336,000)
Message 153839 - Posted: 3 Feb 2022 | 23:46:21 UTC

Vaguely remember stream saying 3.5 cores (ie 3 or 4) for llr2 is most efficient. Don't qoute me on that. Just stuck in my head 🙃

____________
My lucky numbers 10590941048576+1 and 224584605939537911+81292139*23#*n for n=0..26

Profile composite
Volunteer tester
Send message
Joined: 16 Feb 10
Posts: 1003
ID: 55391
Credit: 880,109,081
RAC: 329,545
Discovered 2 mega primesFound 1 prime in the 2018 Tour de PrimesFound 1 prime in the 2022 Tour de Primes321 LLR Turquoise: Earned 5,000,000 credits (6,055,323)Cullen LLR Gold: Earned 500,000 credits (776,297)ESP LLR Ruby: Earned 2,000,000 credits (3,433,680)Generalized Cullen/Woodall LLR Ruby: Earned 2,000,000 credits (2,443,837)PPS LLR Sapphire: Earned 20,000,000 credits (33,804,782)PSP LLR Turquoise: Earned 5,000,000 credits (6,587,988)SoB LLR Sapphire: Earned 20,000,000 credits (44,987,733)SR5 LLR Turquoise: Earned 5,000,000 credits (6,205,694)SGS LLR Ruby: Earned 2,000,000 credits (3,627,819)TRP LLR Turquoise: Earned 5,000,000 credits (7,078,152)Woodall LLR Amethyst: Earned 1,000,000 credits (1,693,614)321 Sieve (suspended) Emerald: Earned 50,000,000 credits (50,256,050)Cullen/Woodall Sieve (suspended) Turquoise: Earned 5,000,000 credits (5,571,178)Generalized Cullen/Woodall Sieve (suspended) Emerald: Earned 50,000,000 credits (50,009,610)PPS Sieve Double Silver: Earned 200,000,000 credits (458,608,316)Sierpinski (ESP/PSP/SoB) Sieve (suspended) Jade: Earned 10,000,000 credits (10,165,888)TRP Sieve (suspended) Sapphire: Earned 20,000,000 credits (20,071,454)AP 26/27 Turquoise: Earned 5,000,000 credits (6,798,063)GFN Emerald: Earned 50,000,000 credits (57,113,430)WW Ruby: Earned 2,000,000 credits (2,072,000)PSA Double Bronze: Earned 100,000,000 credits (102,762,384)
Message 153854 - Posted: 4 Feb 2022 | 5:58:24 UTC - in response to Message 153839.

Vaguely remember stream saying 3.5 cores (ie 3 or 4) for llr2 is most efficient. Don't qoute me on that. Just stuck in my head 🙃

LLR2 main task will be first, almost certainly 100% of the time anyway, so why use more than one core per task and waste CPU cycles and energy for a non-existent race?

Profile Michael GoetzProject donor
Volunteer moderator
Project administrator
Avatar
Send message
Joined: 21 Jan 10
Posts: 13787
ID: 53948
Credit: 345,142,938
RAC: 12,033
The "Shut up already!" badge:  This loud mouth has mansplained on the forums over 10 thousand times!  Sheesh!!!Discovered the World's First GFN-19 prime!!!Discovered 2 mega primesFound 1 prime in the 2018 Tour de PrimesFound 1 prime in the 2019 Tour de PrimesFound 1 prime in the 2020 Tour de PrimesFound 2 primes in the 2021 Tour de PrimesFound 2 primes in the 2022 Tour de PrimesFound 1 mega prime in the 2022 Tour de PrimesFound 1 prime in the 2022 Tour de Primes Mountain Stage321 LLR Turquoise: Earned 5,000,000 credits (6,638,389)Cullen LLR Turquoise: Earned 5,000,000 credits (5,038,114)ESP LLR Turquoise: Earned 5,000,000 credits (6,177,890)Generalized Cullen/Woodall LLR Turquoise: Earned 5,000,000 credits (5,094,541)PPS LLR Sapphire: Earned 20,000,000 credits (23,416,635)PSP LLR Turquoise: Earned 5,000,000 credits (7,956,186)SoB LLR Sapphire: Earned 20,000,000 credits (36,067,618)SR5 LLR Jade: Earned 10,000,000 credits (12,645,567)SGS LLR Turquoise: Earned 5,000,000 credits (5,036,951)TRP LLR Turquoise: Earned 5,000,000 credits (5,084,329)Woodall LLR Turquoise: Earned 5,000,000 credits (5,032,821)321 Sieve (suspended) Jade: Earned 10,000,000 credits (10,061,196)Cullen/Woodall Sieve (suspended) Ruby: Earned 2,000,000 credits (4,170,256)Generalized Cullen/Woodall Sieve (suspended) Turquoise: Earned 5,000,000 credits (5,059,304)PPS Sieve Sapphire: Earned 20,000,000 credits (22,885,121)Sierpinski (ESP/PSP/SoB) Sieve (suspended) Amethyst: Earned 1,000,000 credits (1,035,522)TRP Sieve (suspended) Ruby: Earned 2,000,000 credits (2,051,121)AP 26/27 Jade: Earned 10,000,000 credits (10,902,645)GFN Emerald: Earned 50,000,000 credits (92,455,703)WW Emerald: Earned 50,000,000 credits (65,888,000)PSA Jade: Earned 10,000,000 credits (12,445,029)
Message 153860 - Posted: 4 Feb 2022 | 7:59:37 UTC - in response to Message 153854.

Vaguely remember stream saying 3.5 cores (ie 3 or 4) for llr2 is most efficient. Don't qoute me on that. Just stuck in my head 🙃

LLR2 main task will be first, almost certainly 100% of the time anyway, so why use more than one core per task and waste CPU cycles and energy for a non-existent race?


So that the tasks fit in cache.

With LLR2, the only thing that matters is "how many tasks can I complete each day?"

The rule of thumb is that you need 8x the FFT size. My Haswell i5-4670K is a great example. It's got 4 cores. What does the most tasks per day? -t1 x4, -t2 x2, or -t4 x1?

With PPS-MEGA, 8x the FFT size is about 2 MB. The i5-4670K has 6 MB of cache. Running 4 single core tasks requires 8 MB, which exceeds the cache size. Testing backs this up. 4 single-thread tasks is slower than 2 two-threaded tasks (because it doesn't fit in cache), and 1 4-threaded task is also slower because each thread adds a little synchronization overhead. So on this CPU, -t2 is best for PPS-MEGA. Real testing confirms the theory.

My Zen2 3700X has 8 cores and 32 MB of cache, so it can run 8 tasks at -t1 and have it all fit in cache. Testing also confirms that -t1 is best for PPS-MEGA on this CPU.

As for 321, the the +1 and -1 variants have slightly different FFT sizes, so let's use the larger FFT, which is 1 MB. These tasks therefore require 8 MB of cache. On a CPU with less than 16 MB, running a single task on all the cores should be best. With 32 MB of cache on the 3700X, running 4 -t2 tasks should be best.

Essentially, you want to run the lowest number of threads per task (and hence the most simultaneous tasks) provided the tasks fit in cache. Keeping the FFT data in cache is more important than reducing the multi-threading overhead.
____________
My lucky number is 75898524288+1

stream
Volunteer moderator
Project administrator
Volunteer developer
Volunteer tester
Send message
Joined: 1 Mar 14
Posts: 986
ID: 301928
Credit: 543,186,264
RAC: 434
Discovered 1 mega primeFound 1 prime in the 2018 Tour de PrimesFound 1 prime in the 2019 Tour de PrimesFound 1 prime in the 2020 Tour de PrimesFound 1 prime in the 2021 Tour de Primes321 LLR Jade: Earned 10,000,000 credits (10,011,570)Cullen LLR Jade: Earned 10,000,000 credits (10,009,374)ESP LLR Jade: Earned 10,000,000 credits (10,009,221)Generalized Cullen/Woodall LLR Jade: Earned 10,000,000 credits (10,012,217)PPS LLR Jade: Earned 10,000,000 credits (16,320,618)PSP LLR Jade: Earned 10,000,000 credits (10,044,081)SoB LLR Jade: Earned 10,000,000 credits (10,064,750)SR5 LLR Jade: Earned 10,000,000 credits (10,002,051)SGS LLR Jade: Earned 10,000,000 credits (10,001,295)TRP LLR Jade: Earned 10,000,000 credits (10,002,411)Woodall LLR Jade: Earned 10,000,000 credits (10,013,921)321 Sieve (suspended) Sapphire: Earned 20,000,000 credits (20,004,228)Generalized Cullen/Woodall Sieve (suspended) Sapphire: Earned 20,000,000 credits (20,047,667)PPS Sieve Sapphire: Earned 20,000,000 credits (20,866,490)Sierpinski (ESP/PSP/SoB) Sieve (suspended) Sapphire: Earned 20,000,000 credits (20,043,271)TRP Sieve (suspended) Sapphire: Earned 20,000,000 credits (20,015,177)AP 26/27 Sapphire: Earned 20,000,000 credits (20,045,194)GFN Emerald: Earned 50,000,000 credits (55,355,287)WW Emerald: Earned 50,000,000 credits (50,016,000)PSA Double Silver: Earned 200,000,000 credits (200,301,443)
Message 153869 - Posted: 4 Feb 2022 | 11:36:15 UTC - in response to Message 153839.

Vaguely remember stream saying 3.5 cores (ie 3 or 4) for llr2 is most efficient. Don't qoute me on that. Just stuck in my head 🙃

Running above 3-4 cores noticeable increases thread synchronization penalty, which must be compensated by speed gained by better cache usage. The penalty is roughly the same for each iteration, so it'll be less noticeable on slow tasks (SoB) but don't use too many cores on small FFT sizes.

Mike just wrote a great answer above.

Profile composite
Volunteer tester
Send message
Joined: 16 Feb 10
Posts: 1003
ID: 55391
Credit: 880,109,081
RAC: 329,545
Discovered 2 mega primesFound 1 prime in the 2018 Tour de PrimesFound 1 prime in the 2022 Tour de Primes321 LLR Turquoise: Earned 5,000,000 credits (6,055,323)Cullen LLR Gold: Earned 500,000 credits (776,297)ESP LLR Ruby: Earned 2,000,000 credits (3,433,680)Generalized Cullen/Woodall LLR Ruby: Earned 2,000,000 credits (2,443,837)PPS LLR Sapphire: Earned 20,000,000 credits (33,804,782)PSP LLR Turquoise: Earned 5,000,000 credits (6,587,988)SoB LLR Sapphire: Earned 20,000,000 credits (44,987,733)SR5 LLR Turquoise: Earned 5,000,000 credits (6,205,694)SGS LLR Ruby: Earned 2,000,000 credits (3,627,819)TRP LLR Turquoise: Earned 5,000,000 credits (7,078,152)Woodall LLR Amethyst: Earned 1,000,000 credits (1,693,614)321 Sieve (suspended) Emerald: Earned 50,000,000 credits (50,256,050)Cullen/Woodall Sieve (suspended) Turquoise: Earned 5,000,000 credits (5,571,178)Generalized Cullen/Woodall Sieve (suspended) Emerald: Earned 50,000,000 credits (50,009,610)PPS Sieve Double Silver: Earned 200,000,000 credits (458,608,316)Sierpinski (ESP/PSP/SoB) Sieve (suspended) Jade: Earned 10,000,000 credits (10,165,888)TRP Sieve (suspended) Sapphire: Earned 20,000,000 credits (20,071,454)AP 26/27 Turquoise: Earned 5,000,000 credits (6,798,063)GFN Emerald: Earned 50,000,000 credits (57,113,430)WW Ruby: Earned 2,000,000 credits (2,072,000)PSA Double Bronze: Earned 100,000,000 credits (102,762,384)
Message 153889 - Posted: 4 Feb 2022 | 17:03:18 UTC - in response to Message 153860.

Essentially, you want to run the lowest number of threads per task (and hence the most simultaneous tasks) provided the tasks fit in cache. Keeping the FFT data in cache is more important than reducing the multi-threading overhead.

Agreed. However there's no reason to run more threads than needed for the same number of tasks when there's no discernable speedup. In fact, the reduced thermal load from using fewer threads may allow some cores to run faster.

Post to thread

Message boards : 321 Prime Search : Computation times

[Return to PrimeGrid main page]
DNS Powered by DNSEXIT.COM
Copyright © 2005 - 2022 Rytis Slatkevičius (contact) and PrimeGrid community. Server load 0.68, 0.62, 0.88
Generated 6 Jul 2022 | 0:15:10 UTC