Posts by Raistmer


log in
1) Message boards : Number crunching : @Pre-FERMI nVidia GPU users: Important warning (Message 1571811)
Posted 1 day ago by Profile Raistmer
nVidia decided to drop pre-FERMI GPU support.

We already found that 340.52 driver (though simulating work) generates incorrect results with AstroPulse and pre-FERMI hardware.

This is thread on nVidia dev forums regarding issue:

https://devtalk.nvidia.com/default/topic/774899/cuda-programming-and-performance/340-52-driver-breaks-opencl-support-for-pre-fermi-cc-1-x-cards/

So, if you want to use your pre-FERMI nVidia hardware for AstroPulse crunching stay with pre-340.xx drivers.
2) Questions and Answers : Windows : Hello everyone, I'm new here. (Message 1569198)
Posted 6 days ago by Profile Raistmer
Hello, I am a novice. Very glad to have so many friends. But the worst is here seem to communicate in English, and I will not English, only with the help of the translation software. Oh, really makes people depressed.

Привет :) Ну, для многих англоязычных русский не сильно от китайского отличается :) А вообще языков тут много разных :)
3) Questions and Answers : Windows : C-50 and AMD C-60 APU with Radeon(tm) HD Graphics [Family 20 Model 2 Stepping 0] owners (Message 1569196)
Posted 6 days ago by Profile Raistmer
C-60 is the father and mother device of current OpenCL ATi apps :D Cause it's my dev host (and profiling too).

To maximize its throughput try to use BrazosTweaker utility. Though it can't tweak GPU speed (on Acer netbook I use) it can slowdown and undervolt CPU part that in turn allows BIOS to rise GPU speed to 400MHz instead of usual 276MHz.
Running AP in such config would be most productive (though user response can degrade due to lower CPU freqs).
4) Questions and Answers : Windows : AMD A10-5700 APU with Radeon(tm) HD Graphics (Message 1569188)
Posted 6 days ago by Profile Raistmer
Do I understand right that you see partial CPU load though all CPU cores should be busy with CPU app instances?

And I would recommed to use mostly OpenCL AP GPU app on your device if it gets enough of AP work to keep it busy.

Also, on GPU use -cpu_lock param, it helps a lot with GPU load:
http://lunatics.kwsn.net/1-discussion-forum/astropulse-v7-performance-illustration.0.html

To avoid driver restarts with ATi MB app try to set higher value for -period_iterations_num and use -sbs 256 param.
iteration of 1 (default) time to time too high even for HD6950 I use, APU can be too laggy with defaults.
5) Message boards : Number crunching : All of a sudden: Errors on APs (Message 1568309)
Posted 8 days ago by Profile Raistmer
I hope that this fixed in latest AP v7 builds.
Soon 7.04 will be deployed and will see.
6) Message boards : Number crunching : Lunatics Windows Installer v0.42 Release Notes (Message 1568225)
Posted 8 days ago by Profile Raistmer
Thanks for report. Little we can do w/o task itself. But please continue to look for such cases. And try to catch task.



Here is what I think is pertinent & links to the output I saved for analysis.

Task ID: 3681602891
AstroPulse v6 Anonymous platform (NVIDIA GPU)
AstroPulse v6 Windows x86 rev 2399

single pulses: 2 repetitive pulses: 0 percent blanked: 3.18 Single pulse: peak_power=38.13 dm=5946 fft_num=29147136 peak_bin=29154000 scale=2 Single pulse: peak_power=130 dm=9335 fft_num=17809408 peak_bin=17820800 scale=6

Task ID: 3681602892
AstroPulse v6 Anonymous platform (CPU)
AstroPulse v6 Windows x64 rev 2163

single pulses: 2 repetitive pulses: 1 percent blanked: 3.18 Single pulse: peak_power=38.13 dm=5946 fft_num=29147136 peak_bin=29154000 scale=2 Single pulse: peak_power=130 dm=9335 fft_num=17809408 peak_bin=17820800 scale=6 Rep. pulse: peak_power=9469 dm=-9776 peak_bin=1280 scale=4 ffa_scale=1 period=236.14

Task ID: 3719572871
AstroPulse v6 v6.06 (opencl_ati_100)
Windows x86 rev 1832

single pulses: 2 repetitive pulses: 0 percent blanked: 3.18 Single pulse: peak_power=38.13; dm=5946; scale=2 Single pulse: peak_power=130; dm=9335; scale=6
7) Message boards : Number crunching : Panic Mode On (89) Server Problems? (Message 1567586)
Posted 9 days ago by Profile Raistmer

As far as I am aware no one in the lab does or has ever read these boards.


Time to time reading these boards I envy those who don't :P
8) Message boards : Number crunching : @NVidia-developers: Low background GPU load RISES crunching throughput significantly (Message 1566585)
Posted 11 days ago by Profile Raistmer
Really elegant way would be to add such ability to app_config.xml and not to relay on science apps.
app_config already can pass cmdline to app. All it needs is to distinguish compute devices it governs.
9) Message boards : Number crunching : @NVidia-developers: Low background GPU load RISES crunching throughput significantly (Message 1566509)
Posted 11 days ago by Profile Raistmer
The possibility to configure different GPUs dependent on their socket address is a real killer feature! *doublethumbsup* ;)

Thanks everyone!


Install second BOINc client and enjoy.
http://vyper.kafit.se/wp/index.php/2011/02/04/running-different-nvidia-architectures-most-optimal-at-setihome/
10) Message boards : Number crunching : All of a sudden: Errors on APs (Message 1566465)
Posted 11 days ago by Profile Raistmer
Check GPU clock for shader/memory..
Also, it can be not hardware effect but software one. Change driver priority of smth alike. Even it can be CPU effect - higher CPU (not GPU) power state allows quicker CPU response to driver needs. It's all connected...
11) Message boards : Number crunching : All of a sudden: Errors on APs (Message 1566134)
Posted 12 days ago by Profile Raistmer
As another possible fix try to add -cpu_lock option and see if there will be any crashes with this option active?

EDIT: also you could try to help in debugging by participation in beta:
http://setiweb.ssl.berkeley.edu/beta/forum_thread.php?id=2186&postid=52208
12) Message boards : Number crunching : All of a sudden: Errors on APs (Message 1566109)
Posted 13 days ago by Profile Raistmer
Yes, it's elusive "crash after finish" bug that plagues mostly Nv builds.
Unfortunately, it didn't eradicated so far and shows itself time to time in AP 7.03 NV too.

I experienced it on my host too with high number of crashes... and suddenly all crashes disappear for few weaks. Maybe host reboot helped, maybe some Windows updates....

I plan to take next attempt against this bug soon but for now we must just live with it. Try to reboot host.
13) Message boards : Number crunching : "Sticky" ATI card (Message 1565745)
Posted 13 days ago by Profile Raistmer
Sometime last year I had an issue whereby the mouse would "stick" or not move when commanded for a while, and it only happened whan the gpu was set to crunch. I turned off GPU crunching, but there was chatter on here about it at the time.
Since I upgraded the PC earlier this year (but kept the ATI 6950), I restarted GPU crunching, even to the point (after installing Lunatics) to doping 2 wu's on the gpu. To be fair, I get roughly the same amount of glitching with 2 wu's as I did with 1, but I'm getting fed up of it now. Replacing the gpu isn't an option.
Currently on Catalyst 13.12, later versions gave me issues so I swit6ch back.
Can anyone suggest anything i can try that might relieve the problem, or am I stuck between a rock and a hard place, ie crunch & glitch, or no crunch?

P.

You can try to add options specially designed to solve such issues. Consult with ReadMe file.
14) Message boards : Number crunching : Problems with PC. (Message 1564993)
Posted 15 days ago by Profile Raistmer
I use this on several Windows 7 machines, but think I used it on XP as well.

Not XP Home, I fear.

C:\test>ver

Microsoft Windows XP [Version 5.1.2600]

C:\test>timeout /?
'timeout' is not recognized as an internal or external command,
operable program or batch file.

Nor in Win XP Pro SP3.
But in Vista:

P:\bin\BoincData>timeout /?

TIMEOUT [/T] <таймаут> [/NOBREAK]

Описание:
Принимает значение таймаута, задающего фиксированный период времени
ожидания (в секундах), или ожидание до нажатия клавиши. Имеется также
параметр, зaдающий игнорирование нажатий клавиш.

Параметры:
/T <таймаут> Время ожидания в секундах.
Допустимый интервал: от -1 до 99999 секунд.

/NOBREAK Игнорировать нажатия клавиш, ждать указанное время.

/? Вывод справки по использованию.

Примечание: значение таймаута, равное -1 задает неограниченное время
ожидания до нажатия клавиши.

Примеры:
TIMEOUT /?
TIMEOUT /T 10
TIMEOUT /T 300 /NOBREAK
TIMEOUT /T -1
15) Message boards : Number crunching : Will the optimised apps still be faster with my old C2Q? (Message 1563771)
Posted 17 days ago by Profile Raistmer
Yea I know but it's a right pain looking at dozen+ like that.

you could use some script parsing client_state.xml file instead.
16) Message boards : Number crunching : blanked AP tasks (Message 1563764)
Posted 17 days ago by Profile Raistmer

I would guess that the change was only recently devised.


Not quite. Initial Joe's BLANKIT mod was introduced quite long time ago.
Need to keep telling that currently SETI@home is purely volunteered endeavor, I would say from Berkeley's side too, not only from our. What it means?
It means that despite of string on the bottom of beta forum pages SETI team has no funds from US government long ago already. Also that means that all team members can't dedicate long enough time quanta for SETI development, even day to day support severely deteriorated by non-SETI constant activity requirement.
All this incurs long delays in everything we do here. In each step.
Nobody can allow himself to live w/o money in this mad world...
17) Message boards : Number crunching : blanked AP tasks (Message 1563730)
Posted 17 days ago by Profile Raistmer
That's why I am asking the questions. Trying to figure if it's the way the new app handles things, or if the AP splitting errors shown on the SSP are related to the blanking issue.



I'm probably going to stir up a hornet's nest with this but.....
I actually don't mind too much (outside of the WOW challenge) when I get work units that eventually get highly blanked. Why, you ask? After all the reading the discussions on just what is going on when the data gets analyzed, I came to appreciate just what is going on. There is talk that v7 "will fix" the blanking problem. Well, not really. It actually just "replaces" the problem. I understand why it is being done. But sacrificing sensitivity for improve specificity leaves a lot of potential signals out of the mix. I know I'm late to the party on this and v7 is well underway in testing in Beta. I guess it just rubs me wrong way when someone says that v7 "fixes" the problem. It doesn't fix it, rather to try to find a signal in a mess of noise, it replaces that entire area with a blanket noise that easier to skip over. I really appreciate all the effort and time that has being put in over the years by everyone that has worked on this project. It's not easy to find people willing to donate their time and expertise in helping this project. I guess I posted this just so people are aware that in order to get a "Faster" AP we do so with the knowledge that might be "throwing out the baby with the bath water." Just my 2 pesos...lol

Zalster


I would just say "wrong approach".
Wait if Joe will describe it in more details maybe. AP v7 has better "science", not just better speed on GPU.
18) Message boards : Number crunching : Lunatics Windows Installer v0.42 Release Notes (Message 1563715)
Posted 17 days ago by Profile Raistmer
@those who reported issues in OpenCL AstroPulse v6 from this Lunatic installer release:

Please test OpenCL APv7.03 on beta site and check if it works OK with your setups.
Doing so allow to release AP v7 that will work "out of box" on your own hardware, w/o additional efforts.
19) Message boards : Number crunching : Lunatics Windows Installer v0.42 Release Notes (Message 1563713)
Posted 17 days ago by Profile Raistmer
So is the "problem" that there's no CUDA AP app?

Please see AP6_win_x86_SSE2_OpenCL_NV_r2399.exe from the v0.42 installer. While not CUDA it is the app that will give you the best use of NV cards for AP.

Yes I am using that, just wondered if there was an CUDA version if that would be better.


Perhaps. Lot of time both SETI apps do FFT. And proprietary cuFFT was significally better than open-source oclFFT on nVidia hardware when I tested them year or more ago.
I'm working to make oclFFT more flexible and tunable now, but expect cuFFT being faster still.
So, in my estimations, even direct porting of AstroPulse OpenCL code to CUDA would improve performance on nVidia cards. This porting requires time though.
Any volunteers?
20) Message boards : Number crunching : Lunatics Windows Installer v0.42 Release Notes (Message 1561454)
Posted 22 days ago by Profile Raistmer

Considering some have 4 top GPU`s running with a slow CPU and running at least 2 instances on each card would require at least a 8 core CPU to cope with high blanked AP`s.

And that's why we wanna AP v7 release ASAP.


Next 20

Copyright © 2014 University of California