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
1) Message boards : Number crunching : World Maths Day Challenge (Message 133724)
Posted 10 days ago by Profile DeleteNullProject donor
This is (may be) an error of the BOINC software. Even if you have a cache of zero, BOINC loads up to 1000 tasks if you switch fom AP27 to llrDIV (multi threaded). If the core count is higher than the count of WU that are processed in parallel this happens.

Sad once more we have 1000's of Tasks the have been aborted.
Maybe someone posting of how to run PG should be posted.
No sense in downloading a few hundred tasks and then turn around and abort them all and then downloading a few more hundred tasks only to abort them as well.
Maybe PG can set aborted tasks to be given back to the same host that has aborted them....

2) Message boards : Number crunching : Better multi-threading (Message 132213)
Posted 60 days ago by Profile DeleteNullProject donor
The BOINC client invokes the app -- the LLR wrapper -- with "--nthreads N". The new version of the wrapper will understand --nthreads as well as -t, and will pass that along to LLR.

Does that make sense?


Yes that is the reason for my question. --nthreads should be transformed into -t and into the parameter in the llr.ini (ThreadsPerTest=N) that is copied into the slots folder.
3) Message boards : Number crunching : Better multi-threading (Message 132209)
Posted 60 days ago by Profile DeleteNullProject donor
Must implement the LLR app the plan_class or is it the wrapper that has to implement this feature?

Will an app_config.xml still work normally? In other words, will it override any preferences set at the project?


Yes it will, but there's a wrinkle.

The LLR apps currently do not have a plan_class. The new LLR apps will have a plan class of "mt" (or something similar to that.) This means that although you can still use app_config to override the website preferences, you will need to modify your current app_config.xml file to add the <plan_class>mt</plan_class> line.

I am hopeful that I can create a transition whereby your existing, unmodified app_config.xml file (without <plan_class>) can continue to work. Undoubtedly, no matter how many forum posts I write, or news items I create, or how many warnings I put on the website, some people won't see it.

The best way to be prepared for the change is to include two <app_version> blocks in your app_config.xml, one each for the current apps (without <plan_class>) and for the new upcoming apps (with <plan_class>). I'll have more information on this once I work out all the details.


4) Message boards : Number crunching : What can we expect from Zen 2/Ryzen 3000? (Message 131060)
Posted 101 days ago by Profile DeleteNullProject donor
Both have hyperthreading, 7700K = 8 threads, 3700x = 2 x 8 threads.
Depending on the FFT length sometimes the 7700k has shorter times sometimes the 3700x.

But the 3700x calculates 2 WU in parallel when the 7700k is doing (only) one.......


This thing is pretty fast. A first test with LLR-PPS-MEGA shows (on average):
2 times faster than 7700k (memory 3000)


I checked your very interesting tests and found that Ryzen 7 3700X is twice as fast as Ryzen 7 1700X but is not faster than Intel i7-7700K...???

Intel i7-7700K, 800 sec
primegrid_llr -d -oDiskWriteTime=1 -oThreadsPerTest=4 llr.in
Using all-complex FMA3 FFT length 256K, Pass1=128, Pass2=2K, clm=2, 4 threads, a = 3
http://www.primegrid.com/result.php?resultid=1011442934

Ryzen 7 3700X, 982 sec
primegrid_llr -d -oDiskWriteTime=1 -oThreadsPerTest=8 llr.in
Using all-complex FMA3 FFT length 256K, Pass1=1K, Pass2=256, clm=1, 8 threads, a = 3
http://www.primegrid.com/result.php?resultid=1011450586

5) Message boards : Number crunching : What can we expect from Zen 2/Ryzen 3000? (Message 131050)
Posted 101 days ago by Profile DeleteNullProject donor
A new Ryzen 3700x (TDP 65W) arrived today together with a x570 board and DDR4 memory 3200.

This thing is pretty fast. A first test with LLR-PPS-MEGA shows (on average):
2 times faster than 7700k (memory 3000)
A bit faster (~ 10%) than 7820x (memory 3000 quad channel, AVX 512)

7nm vs. 14nm: 3 times more energy efficient in LLR.

It's also cheaper and you can use the (very good) fan that was included.
(default settings, memory with XMP profile #1)

6) Message boards : Number crunching : Badges - Full Sets! (Message 130873)
Posted 109 days ago by Profile DeleteNullProject donor
Impressive.
Did anyone already reach the next level of all bronze shields?


Thanks. And yes, e.g. Scott Brown.
7) Message boards : Number crunching : Badges - Full Sets! (Message 130860)
Posted 110 days ago by Profile DeleteNullProject donor
all active subprojects Emerald or better.
8) Message boards : Number crunching : Mega Prime finders by quantity found. (Message 130180)
Posted 137 days ago by Profile DeleteNullProject donor
Thank you Robish!

The only thing that is undone is the upgrade of my M-Badge.
But this must be another thread ;)
9) Message boards : Number crunching : Badges III (Message 129673)
Posted 153 days ago by Profile DeleteNullProject donor
This (3 days ago), and
that (today)

One more to complete this level.
10) Message boards : Generalized Fermat Prime Search : GFN-16 MEGA Prime Search starts March, 14, 2019 (Message 127995)
Posted 220 days ago by Profile DeleteNullProject donor
Hello Doctor,

I am using (8x8 threads, cc_config with 8 ncpus):
<app_config>
<app>
<name>gfn16_mega</name>
<fraction_done_exact/>
<max_concurrent>8</max_concurrent>
</app>
<app_version>
<app_name>gfn16_mega</app_name>
<cmdline>-t 8</cmdline>
<avg_ncpus>1</avg_ncpus>
<max_ncpus>1</max_ncpus>
</app_version>
</app_config>

This works fine for me.


Next 10 posts
[Return to PrimeGrid main page]
DNS Powered by DNSEXIT.COM
Copyright © 2005 - 2019 Rytis Slatkevičius (contact) and PrimeGrid community. Server load 1.72, 1.93, 1.86
Generated 21 Oct 2019 | 12:58:58 UTC