Join PrimeGrid
Returning Participants
Community
Leader Boards
Results
Other
drummers-lowrise
|
1)
Message boards :
Number crunching :
Better multi-threading
(Message 132272)
Posted 1383 days ago by dh1saj
Same here.
Guess thats OK for the time being.
22.08.2019 20:29:34 | | Re-reading cc_config.xml
22.08.2019 20:29:34 | | cc_config.xml not found - using defaults
22.08.2019 20:29:34 | | log flags: file_xfer, sched_ops, task
22.08.2019 20:29:34 | PrimeGrid | Found app_config.xml
22.08.2019 20:29:34 | PrimeGrid | Entry in app_config.xml for app 'llr321', plan class 'mt' doesn't match any app versions
22.08.2019 20:29:34 | PrimeGrid | Entry in app_config.xml for app 'llrPPS', plan class 'mt' doesn't match any app versions
22.08.2019 20:29:34 | PrimeGrid | Entry in app_config.xml for app 'llrESP', plan class 'mt' doesn't match any app versions
22.08.2019 20:29:34 | PrimeGrid | Entry in app_config.xml for app 'llrMEGA', plan class 'mt' doesn't match any app versions
22.08.2019 20:29:34 | PrimeGrid | Entry in app_config.xml for app 'llrCUL', plan class 'mt' doesn't match any app versions
22.08.2019 20:29:34 | PrimeGrid | Entry in app_config.xml for app 'llrSOB', plan class 'mt' doesn't match any app versions
22.08.2019 20:29:36 | PrimeGrid | General prefs: from PrimeGrid (last modified 08-Dec-2016 20:02:03)
22.08.2019 20:29:36 | PrimeGrid | Computer location: home
22.08.2019 20:29:36 | | General prefs: using separate prefs for home
22.08.2019 20:29:36 | | Reading preferences override file
|
2)
Message boards :
Sieving :
Primorial and Factorial sieving are reopened
(Message 115102)
Posted 1934 days ago by dh1saj
Just trying to run Factorial on nVidia GPU hosted on Win10 machine with fsievecl64.exe
Running "-l" returns:
List of available platforms and devices
Platform 0 is a Intel(R) Corporation Intel(R) OpenCL, version OpenCL 1.2
Device 0 is a Intel(R) Corporation Intel(R) HD Graphics 4600
Platform 1 is a NVIDIA Corporation NVIDIA CUDA, version OpenCL 1.2 CUDA 9.1.75
Device 0 is a NVIDIA Corporation GeForce GTX 980
OK so far.
But then I use "-f1" or "-f1 -d0" to start a sieve
but whatever I try, once started a sieve, all I get is:
Platform 0 is an Intel(R) Corporation Intel(R) OpenCL, version OpenCL 1.2
Device 0 is an Intel(R) Corporation Intel(R) HD Graphics 4600
The sieve seems running well, but just on the HD4600 (Platform==0).
So it obviously does not like my GTX980 (Platform==1) but uses the onboard GPU instead.
What am I doing wrong?
Any hint how to get it run on the nVidia?
Jochen
|
3)
Message boards :
Number crunching :
PrimeGrid is Moving!!!
(Message 109946)
Posted 2100 days ago by dh1saj
Do we run into a deadline for the move?
(Other words: Do we have to leave Rackspace until xxxxx?)
|
4)
Message boards :
Project Staging Area :
PRPNet Help
(Message 109253)
Posted 2128 days ago by dh1saj
http://pgllr.mine.nu/PRPNet/ is down since quite a while.
This is why http://prpnet.primegrid.com exists.....
Some DOC might be outdated in details.....
|
5)
Message boards :
Number crunching :
TRP-Sieve HAS ENDED!!!
(Message 106492)
Posted 2259 days ago by dh1saj
...
And last how many antiquidated supercomputers you know:
...
Well, many. Actually all that existing.....
Technology still proceeds that fast, that once a new supercomputer, whatever kind of, has been installed, it can be considered outdated. Newer/better/faster/cheaper/.... already available.
And NO, I do not want to take part in speculations. I'm just saying that even supercomputers getting "antiquidated" much quicker than we may imagine....
|
6)
Message boards :
Seventeen or Bust :
The SoB Double Check is DONE!!!
(Message 106384)
Posted 2263 days ago by dh1saj
...
On the other hand, it's not clear whether SoB's double check compared residues. They may have simply run everything twice. If that's what they did, then it's entirely possible for there to have been two faulty calculations on the same number. It doesn't happen often, but it does happen. So it's still possible that there's missed primes at lower n values as well.
...
If I understand that http://www.primegrid.com/workunit.php?wuid=519659035 right, it should confirm the assumption that SoB did not compare residues.
We see 3 imported results and obviously, none of them was correct.
Anything misunderstood on my side, or is it really like that....?
If so, strange, at least.
|
7)
Message boards :
Number crunching :
LLR 3.8.20 Going Live!
(Message 106373)
Posted 2264 days ago by dh1saj
I also think I got confused when boinc client appeared to report an error, now I suspect it only does so if a subproject hasn't been run on a particular system before.
Right.
Had the same here, but once downloaded the first WU of SoB (in that particular case) all ran fine.....
|
8)
Message boards :
Number crunching :
Isaac Newton's Birthday Challenge
(Message 104057)
Posted 2324 days ago by dh1saj
...
Opening GCW-LLR we chose to mean "sieved to optimal depth for a 4 million digit candidate". The largest candidates in the sieve file are 15 million digits. The sieve will therefore end when we're sufficiently sieved for 15 million digits. I therefore expect that sieving to 15M will take 14 times as long as sieving to 4M (15^2 / 4^2). That means a total sieve time of 3500 days, of which we've already done about 200 days of sieving. So I was wrong; it's not closer to 5 years than it is to 1 year. It's closer to 10 years than it is to 1 year.
...
Doesn't this loudly call for a GPU sieving app?
I, technically, have no clue how to realise a most efficient GPU application, but I stupidly think: We do have a GPU app for PPS sieve. Even if GCW sieve is somewhat different, the core algorithm should be similar. Sort of.
So, one being able to code GPU binaries should be able to adopt the GCW sieve algo onto the PPS sieve framework.......
Easy said, admittedly.
But wouldn't it be worth to dig into this direction?
Even an not-so-super-efficient GCW GPU sieve might be able to reduce Mikes 3500 days to something arround 1 year. Or maybe some 3 years with much deeper sieve, which would result in a much higher hitrate on the LLR side. And this, finally, would surely draw attention onto the project.....
Just my 0.02$
|
9)
Message boards :
Number crunching :
Isaac Newton's Birthday Challenge
(Message 102897)
Posted 2347 days ago by dh1saj
Will there be new SIEVEFILES prior to the challenge?
Just asking as I want to be sure to have all 14 files on all joining machines before the challenge starts.... downloading them now and seeing new ones tomorrow might be... well, non efficient ;-)
Is there an estimation what the benefit of brand new files would be (compared to the current ones, dated Oct/Nov) - if any?
|
10)
Message boards :
Number crunching :
Winter Solstice Challenge
(Message 102472)
Posted 2359 days ago by dh1saj
By the way, I have not a clue what is meant by "base clock" anymore. This non-overclocked GPU has a "base clock" of 1506 MHz, a "boost clock" of 1708 MHz, and happily goes off and crunches at 1885 MHz without my having touched anything.
Noted the same and did initially not understand what's going on here....
But after a bit of digging.....
The BIOS of (some of?) the Pascal-cards seems to be quite flexible.
It doesn't care the clock very much, it mainly maintains two figures:
1.) Power limit
2.) Temperature limit
As long as either is not touched, it happily increases the boost-clock up to the sky.
This as well works opposite - if your cooling struggels, the GPU is clocked down to base clock. But I don't know what happens when this lower limit is once touched.
Hmmmmm.... knowing this...... one should think about liquid N2 cooling of the GPU..... :D
|
Next 10 posts
|
