| Author |
Message |
|---|
 Michael Goetz Volunteer moderator
Project administrator
Send message
Joined: 21 Jan 10
Posts: 14011
ID: 53948
Credit: 435,681,308
RAC: 871,866
                               
|
|
I just installed the new Windows and Linux (but not Mac) CPU versions for the following GFN projects:
GFN-16 Yes, you can run GFN16 on CPUs again!
GFN-17-LOW
GFN-17-MEGA Yes, you can run GFN-17-MEGA on CPUs again!
GFN-18
GFN-19
GFN-20
Genefer 3.3.5 provides substantial speed boosts for CPU on those projects.
This applies only to the CPU apps. GPU apps are not affected.
GFN-21 has not been updated because it's still capable of running the FP64 transforms, which are the fastest.
I have removed the CPU GFN-22 app. It was a mistake.
Mac versions of the new app will be made available when testing is completed, but this has been delayed by a lack of Mac testers.
If you want to help out testing the Mac version, head over to the 3.3.5 testing thread.
____________
My lucky number is 75898524288+1 |
|
|
|
|
|
awesome stuff many thanks ! |
|
|
|
|
|
The GFN-22 worked fine when set to run with multiple cores via the app_config.
____________
Werinbert is not prime... or PRPnet keeps telling me so.
Badge score: 2x3 + 5x4 + 5x5 + 5x7 + 1x9 + 3x10 = 125 |
|
|
Michael GoetzVolunteer moderator
Project administrator
Send message
Joined: 21 Jan 10
Posts: 14011
ID: 53948
Credit: 435,681,308
RAC: 871,866
|
The GFN-22 worked fine when set to run with multiple cores via the app_config.
Um..... no, it didn't. You may have thought it worked, but the software only actually used multi-threading on GFN-21. In reality it was only using a single thread on GFN-22.
I really, really, wish multithreading worked on GFN-22, but Yves has to rewrite the transforms individually for each n.
____________
My lucky number is 75898524288+1 |
|
|
|
|
|
Thank you for the new software, and thanks to all the testers. I appreciated your work. |
|
|
|
|
|
I am a bit on the crazy side of doing these long tasks, so getting rid of GNF-22 CPU probably only affected me... I like to think my old i7-3770k can do serious work. :-)
On the plus side, not having to do GFN-16/17-mega on the GPU frees up my GPU cards for other tasks.
____________
Werinbert is not prime... or PRPnet keeps telling me so.
Badge score: 2x3 + 5x4 + 5x5 + 5x7 + 1x9 + 3x10 = 125 |
|
|
|
|
|
Just got an email to say one of my GFN16 CPU tasks is a prime !
But when I look at http://www.primegrid.com/workunit.php?wuid=687966829 it says the other person's task isnt finished yet can you check all is OK ?
edit : turns out its fine https://primes.utm.edu/primes/page.php?id=131438 :) |
|
|
|
|
|
rud how you getting so many points? on the xeon? |
|
|
|
|
|
Threadripper 3990X 64-Core - works fine up to GFN 17 mega CPU but no good past that. get a great first rate with 16 but bad with 17 mega so will do those again in distant future when multi threading available |
|
|
Yves GallotVolunteer developer
Project scientist
Send message
Joined: 19 Aug 12
Posts: 820
ID: 164101
Credit: 305,989,513
RAC: 1,728
|
Threadripper 3990X 64-Core - works fine up to GFN 17 mega CPU but no good past that. get a great first rate with 16 but bad with 17 mega so will do those again in distant future when multi threading available
The rate is about 64 tasks in 1 hour => 1 task per minute... this is twice as fast as the RTX 2080 Ti! But of course, GPUs are first.
Multithreading will not be efficient with GFN-17: these numbers are too small. Each square operation of a one million digit number takes about 1.5 ms. |
|
|
|
|
|
GFN-16 CPU tasks, my timings.
I9 10980XE 18 tasks concurrent 710 secs, equivalent 39 secs / task
3950X 16 tasks concurrent 611 secs, equivalent 38 secs / task
3090 GPU single task 48 secs running in 3950X with Pcie 4 SSD timing.
Timings vary of course, these are example results.
____________
|
|
|
Michael GoetzVolunteer moderator
Project administrator
Send message
Joined: 21 Jan 10
Posts: 14011
ID: 53948
Credit: 435,681,308
RAC: 871,866
|
|
The Mac version of 3.3.5 is now live.
____________
My lucky number is 75898524288+1 |
|
|
