Welcome to MilkyWay@home

Posts by sesef

1) Message boards : Number crunching : 79XX Dont Run (Message 52464)
Posted 14 Jan 2012 by sesef
Post:
with these drivers http://forums.guru3d.com/showthread.php?t=356972 i have significant speedup, but there is bad news :( After drivers update I can't control card with Afterburner :(
2) Message boards : Number crunching : HD7970 on the horizon .. (Message 52454)
Posted 13 Jan 2012 by sesef
Post:
At the moment its running NVidia OpenCL app at around 52-53 seconds settings @1115/300. Which is a good time for a WU written to (principly) run on NVidia hardware.



Try to run 2 wus in parallel, you should get few extra seconds :). Now I have about 43 sec/wu @1200/685
3) Message boards : Number crunching : HD7970 on the horizon .. (Message 52433)
Posted 12 Jan 2012 by sesef
Post:
Some fun with 7970

best clock without voltage change


best clock


I think OC over 1200 Mhz is worthless. You can get 1200 Mhz without voltage change/with small change and get extra 3 sec. For 1300 Mhz I had to set almost 1,3V to speed up only 3 more seconds :(.
4) Message boards : Number crunching : 79XX Dont Run (Message 52392)
Posted 11 Jan 2012 by sesef
Post:
I've made some tests with NV OpenCL app on Obix007's 7970.

input
milkyway_separation_0.82_windows_x86_64__cuda_opencl.exe -np 14 -p 0.3818788877972418000000000 30.0000000000000000000000000 -20.0000000000000000000000000 76.0000000000000000000000000 38.4090067002612400000000000 0.0000000000000000000000000 0.0000000000000000000000000 1.0000000000000000000000000 5.7985829621085900000000000 40.3449848062901640000000000 45.7000000000000000000000000 0.3401773264955795000000000 0.0000000000000000000000000 1.0000000000000000000000000


output
21:33:27 (3956): Can't open init data file - running in standalone mode
Error loading Lua script 'astronomy_parameters.txt': [string "number_parameters: 4..."]:1: '<name>' expected near '4'
Error reading astronomy parameters from file 'astronomy_parameters.txt'
Trying old parameters file
Using SSE3 path
Found 1 platform(s)
Platform 0 information:
Name: AMD Accelerated Parallel Processing
Version: OpenCL 1.1 AMD-APP (851.6)
Vendor: Advanced Micro Devices, Inc.
Extensions: cl_khr_icd cl_amd_event_callback cl_amd_offline_devices cl_khr_d3d10_sharing
Profile: FULL_PROFILE
Didn't find preferred platform
Using device 0 on platform 0
Found 2 CL device(s)
Trying to show unknown MWDoubleExts 6
Device Tahiti (Advanced Micro Devices, Inc.:0x1002) (CL_DEVICE_TYPE_GPU)
Driver version: CAL 1.4.1658 (VM)
Version: OpenCL 1.1 AMD-APP (851.6)
Compute capability: 0.0
Image support: CL_TRUE
Max compute units: 32
Clock frequency: 1125 Mhz
Global mem size: 3221225472
Local mem size: 32768
Max const buf size: 65536
Double extension: Unhandled MWDoubleExts

Compiler flags:
-cl-mad-enable -cl-no-signed-zeros -cl-strict-aliasing -cl-finite-math-only -DDOUBLEPREC=1 -DFAST_H_PROB=1 -DAUX_BG_PROFILE=0 -DNSTREAM=2 -DCONVOLVE=120 -DR0=30.000000000000000 -DSUN_R0=8.500000000000000 -DQ_INV_SQR=6.857229770206179 -DUSE_IMAGES=0

Build status: CL_BUILD_SUCCESS
Build log:
C:\Users\597x2-2\AppData\Local\Temp\OCL35C8.tmp.cl(236): warning: attributes
ignored here
__global const __read_only RPoints* r_pts,
^

C:\Users\597x2-2\AppData\Local\Temp\OCL35C8.tmp.cl(238): warning: attributes
ignored here
__global const __read_only LBTrig* lbts,
^

LOOP UNROLL: pragma unroll (line 302)
Unrolled as requested!
LOOP UNROLL: pragma unroll (line 279)
Unrolled as requested!

Kernel work group info:
Work group size = 256
Kernel local mem size = 0
Compile work group size = { 0, 0, 0 }
Keeping chunk boundaries as multiples of 2048
Range: { nu_steps = 640, mu_steps = 1600, r_steps = 1400 }
Iteration area: 2240000
Chunk estimate: 1
Num chunks: 1
Added area: 512
Effective area: 2240512
Integration time: 47.281927 s. Average time per iteration = 73.878011 ms
Integral 0 time = 48.013559 s
Likelihood time = 2.081959 s
<background_integral> 0.000076204271302 </background_integral>
<stream_integral> 0.000000007248892 19.757344732480210 </stream_integral>
<background_likelihood> -5.610462983778151 </background_likelihood>
<stream_only_likelihood> -238.985842705200840 -188.059457448391440 </stream_only_likelihood>
<search_likelihood> -5.289947178575774 </search_likelihood>
<search_application> milkywayathome_client separation 0.82 Windows x86_64 double OpenCL </search_application>
21:34:18 (3956): called boinc_finish
5) Message boards : Number crunching : ATI 4770 work ... or not? (Message 30294)
Posted 6 Sep 2009 by sesef
Post:
I have Asus 4770 Formula, and it work fine on cat 9.8 with win XP SP3 x86 and Win7 x64
6) Message boards : Number crunching : BOINC 6.10 support with ATI (Message 29477)
Posted 18 Aug 2009 by sesef
Post:
When milky admins will upgrade server to newest version with ATI applications support?
7) Message boards : Number crunching : Small WUs is doing it again! (Message 29174)
Posted 10 Aug 2009 by sesef
Post:
Hello MW@Home,

I took over the BOINC server this past weekend (Saturday is when these "small" WUS should have started showing up).

For the prior 2-3 weeks Travis had been running convergence tests on 5 degree wide stripes-- I am however working on 2.5 degree wide stripes which translates to half as many nu steps and thus half as large WUs. These WUs should be running just as fast as they usually ran before I introduced the 5 degree stripe size about 2 months ago-- is this so?

I can put some more 5 degree tests up later today when I get a chance.

Thanks for your patience,
John Vickers


I think mix with 2,5 degree and 5 degree wu's will be the best.
8) Message boards : Number crunching : Question for you Win 7 guys (Message 28996)
Posted 7 Aug 2009 by sesef
Post:
9.7 with 4770 and win7 x64 run stable, and wu crunch faster than 9.5
9) Message boards : Number crunching : Settings while running 2x 4870x2's (Message 27545)
Posted 12 Jul 2009 by sesef
Post:
I see you can chrunch 4 WU on each GPU. What had you set run like that

About commandline OxyOne have n2 w0.0 f20
10) Message boards : Number crunching : HD 4730 (Message 27219)
Posted 7 Jul 2009 by sesef
Post:
Someone have this card?

It have double precision support and work with milky?
11) Message boards : Number crunching : Power Usage (Message 24946)
Posted 11 Jun 2009 by sesef
Post:
Not mine but 1.65 kWh on 2x Xeon 3.6GHz + 2x4870x2
12) Message boards : Number crunching : New faster application? (Message 8392)
Posted 15 Jan 2009 by sesef
Post:
Beau

I do not think we can from our computers. As Travis posted here about the only method to verify that good results are being churned out.



And how would one do that? There were no sample inputs given with the code. Just the same link and app that others downloaded from arizmoon.com


Yes there are, if you go to http://milkyway.cs.rpi.edu/milkyway/download/code_release/
all of these have sample input files.



Ok, I downloaded the Milkyway release .7 (I am assuming that I was supposed to download the latest one?) Now what do I do with them? Sorry--I dont mean to be a pain...

Thanks





You should be able to put the milkyway executable and the other executable in the same directory with the sample inputs and run them. This would be (from where you uncompressed the files) ./milkyway/bin/test_files/



Ok, copied them into the same folder and ran them, now what am I looking at?




in this folder you'll find file "out", last line of this file is your result.
13) Message boards : Number crunching : New faster application? (Message 8375)
Posted 15 Jan 2009 by sesef
Post:
Faster applications returns incorrect results.

Original app:
fitness: -3.046272445362277

"Faster app":
fitness: -3.046289346362010

sesef, could you also check your faster app you are using please?


which applications do you mean?

sorry i thought in your team there where also optimized apps used made by some of your teammade.
i had the user mindc in my mind.
nevermind if im wrong.


Ahh you mean about this app, it returns correct results (same as oryginal app -3.046272445362277)
14) Message boards : Number crunching : New faster application? (Message 8363)
Posted 15 Jan 2009 by sesef
Post:
Faster applications returns incorrect results.

Original app:
fitness: -3.046272445362277

"Faster app":
fitness: -3.046289346362010

sesef, could you also check your faster app you are using please?


which applications do you mean?
15) Message boards : Number crunching : New faster application? (Message 8355)
Posted 15 Jan 2009 by sesef
Post:
Faster applications returns incorrect results.

Original app:
fitness: -3.046272445362277

"Faster app":
fitness: -3.046289346362010


Which of the faster apps gives out garbage results?
SSE 2 or SSE3? Or all? *scratches head*


MW-update18

SSE:
fitness: -3.046289346362010

SSE2:
fitness: -3.046289346362010

SSE3:
fitness: -3.046289346362010

My CPU dont have SSSE3 and SSE4.1 instructions, so I cant test it.
16) Message boards : Number crunching : New faster application? (Message 8348)
Posted 15 Jan 2009 by sesef
Post:
Faster applications returns incorrect results.

Original app:
fitness: -3.046272445362277

"Faster app":
fitness: -3.046289346362010




©2024 Astroinformatics Group