Welcome to MilkyWay@home

The GPU ripple

Message boards : Application Code Discussion : The GPU ripple
Message board moderation

To post messages, you must log in.

Previous · 1 · 2 · 3 · 4 · Next

AuthorMessage
Profile banditwolf
Avatar

Send message
Joined: 12 Nov 07
Posts: 2425
Credit: 524,164
RAC: 0
Message 13330 - Posted: 28 Feb 2009, 15:37:12 UTC - in response to Message 13329.  

Yes!

CPU: Intel(R) Pentium(R) D CPU 2.80GHz (2 cores/threads) 2.79298 GHz (362ms)

CAL Runtime: 1.3.145
Found 1 CAL device

Device 0: ATI Radeon HD 4800 (RV770) 512 MB local RAM (remote 28 MB cached + 512 MB uncached)
GPU core clock: 680 MHz, memory clock: 750 MHz
800 shader units organized in 10 SIMDs with 16 VLIW units (5-issue)
supporting double precision

0 WUs already running on GPU 0
Starting WU on GPU 0
Calculated about 1.85078e+012 floatingpoint ops on GPU, 6.18221e+007 on FPU.
Calculated about 8.03964e+008 floatingpoint ops on FPU (stars).
WU completed. It took 1 seconds CPU time and 25.726 seconds wall clock time @ 2.79307 GHz.


NO!

Doesn't expecting the unexpected make the unexpected the expected?
If it makes sense, DON'T do it.
ID: 13330 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile GalaxyIce
Avatar

Send message
Joined: 6 Apr 08
Posts: 2018
Credit: 100,142,856
RAC: 0
Message 13332 - Posted: 28 Feb 2009, 15:41:00 UTC - in response to Message 13330.  
Last modified: 28 Feb 2009, 15:56:47 UTC

Yes!

CPU: Intel(R) Pentium(R) D CPU 2.80GHz (2 cores/threads) 2.79298 GHz (362ms)

CAL Runtime: 1.3.145
Found 1 CAL device

Device 0: ATI Radeon HD 4800 (RV770) 512 MB local RAM (remote 28 MB cached + 512 MB uncached)
GPU core clock: 680 MHz, memory clock: 750 MHz
800 shader units organized in 10 SIMDs with 16 VLIW units (5-issue)
supporting double precision

0 WUs already running on GPU 0
Starting WU on GPU 0
Calculated about 1.85078e+012 floatingpoint ops on GPU, 6.18221e+007 on FPU.
Calculated about 8.03964e+008 floatingpoint ops on FPU (stars).
WU completed. It took 1 seconds CPU time and 25.726 seconds wall clock time @ 2.79307 GHz.


NO!

Amazing. And I got 12.961071 credits for it.

ID: 13332 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile GalaxyIce
Avatar

Send message
Joined: 6 Apr 08
Posts: 2018
Credit: 100,142,856
RAC: 0
Message 13335 - Posted: 28 Feb 2009, 15:49:26 UTC

It had to happen sooner or later. The first sub-one second ;)


CPU: Intel(R) Pentium(R) D CPU 2.80GHz (2 cores/threads) 2.793 GHz (364ms)

CAL Runtime: 1.3.145
Found 1 CAL device

Device 0: ATI Radeon HD 4800 (RV770) 512 MB local RAM (remote 28 MB cached + 512 MB uncached)
GPU core clock: 680 MHz, memory clock: 750 MHz
800 shader units organized in 10 SIMDs with 16 VLIW units (5-issue)
supporting double precision

0 WUs already running on GPU 0
Starting WU on GPU 0
Calculated about 1.85078e+012 floatingpoint ops on GPU, 6.18221e+007 on FPU.
Calculated about 8.03964e+008 floatingpoint ops on FPU (stars).
WU completed. It took 0.96875 seconds CPU time and 25.627 seconds wall clock time @ 2.79307 GHz.



ID: 13335 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Misfit
Avatar

Send message
Joined: 27 Aug 07
Posts: 915
Credit: 1,503,319
RAC: 0
Message 13375 - Posted: 28 Feb 2009, 19:07:01 UTC - in response to Message 13335.  

And people thought my suggestion of 1 credit per second was too much.
me@rescam.org
ID: 13375 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile GalaxyIce
Avatar

Send message
Joined: 6 Apr 08
Posts: 2018
Credit: 100,142,856
RAC: 0
Message 13462 - Posted: 1 Mar 2009, 12:14:33 UTC - in response to Message 13375.  

And people thought my suggestion of 1 credit per second was too much.

People are always wrong about you


ID: 13462 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile GalaxyIce
Avatar

Send message
Joined: 6 Apr 08
Posts: 2018
Credit: 100,142,856
RAC: 0
Message 13533 - Posted: 1 Mar 2009, 20:56:55 UTC
Last modified: 1 Mar 2009, 21:02:53 UTC

I've just updated my profile and added this;

This is one of the first recorded sub-one second MilkyWay WU's, crunched by me with a Sapphire HD4850;

Task ID 13212621
Name ps_s21_10_2105_1235836088_0
Workunit 12897484
Created 28 Feb 2009 15:48:11 UTC
Sent 28 Feb 2009 15:49:08 UTC
Received 28 Feb 2009 15:54:49 UTC
Server state Over
Outcome Success
Client state Done
Exit status 0 (0x0)
Computer ID 40322
Report deadline 3 Mar 2009 15:49:08 UTC
CPU time 0.59375
stderr out <core_client_version>5.10.45</core_client_version>
<![CDATA[
<stderr_txt>
Running Milkyway@home ATI GPU application version 0.19 by Gipsel
CPU: Intel(R) Pentium(R) D CPU 2.80GHz (2 cores/threads) 2.79297 GHz (347ms)

CAL Runtime: 1.3.145
Found 1 CAL device

Device 0: ATI Radeon HD 4800 (RV770) 512 MB local RAM (remote 28 MB cached + 512 MB uncached)
GPU core clock: 680 MHz, memory clock: 750 MHz
800 shader units organized in 10 SIMDs with 16 VLIW units (5-issue)
supporting double precision

0 WUs already running on GPU 0
Starting WU on GPU 0
Calculated about 1.85078e+012 floatingpoint ops on GPU, 6.18221e+007 on FPU.
Calculated about 8.03964e+008 floatingpoint ops on FPU (stars).
WU completed. It took 0.953125 seconds CPU time and 25.528 seconds wall clock time @ 2.79307 GHz.

</stderr_txt>
]]>

Validate state Valid
Claimed credit 0.00127383979304833
Granted credit 12.961071
application version 0.19



ID: 13533 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Carl Johnson[SETI.USA]

Send message
Joined: 8 Apr 09
Posts: 2
Credit: 32,148,082
RAC: 0
Message 18629 - Posted: 14 Apr 2009, 3:16:43 UTC

I followed these instructions and it says 'waiting to run' on the milky way tasks. I have a dummy plug in the card since it isn't my primary. Is there a way to check to see if it is active?
ID: 18629 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Misfit
Avatar

Send message
Joined: 27 Aug 07
Posts: 915
Credit: 1,503,319
RAC: 0
Message 18740 - Posted: 14 Apr 2009, 21:19:06 UTC - in response to Message 13462.  

And people thought my suggestion of 1 credit per second was too much.

People are always wrong about you

So what does it take to get 355K credits in just 4 days?
me@rescam.org
ID: 18740 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile Debs

Send message
Joined: 15 Jan 09
Posts: 169
Credit: 6,734,481
RAC: 0
Message 18743 - Posted: 14 Apr 2009, 21:26:22 UTC - in response to Message 18740.  

And people thought my suggestion of 1 credit per second was too much.

People are always wrong about you

So what does it take to get 355K credits in just 4 days?


A RAC of 88750 ?
ID: 18743 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile verstapp
Avatar

Send message
Joined: 26 Jan 09
Posts: 589
Credit: 497,834,261
RAC: 0
Message 18762 - Posted: 14 Apr 2009, 22:55:58 UTC

>A RAC of 88750
I could possibly do that if my Radeons didn't spend so much time waiting for work. :)
Cheers,

PeterV

.
ID: 18762 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
John Clark

Send message
Joined: 4 Oct 08
Posts: 1734
Credit: 64,228,409
RAC: 0
Message 18764 - Posted: 14 Apr 2009, 23:20:31 UTC - in response to Message 18629.  

I followed these instructions and it says 'waiting to run' on the milky way tasks. I have a dummy plug in the card since it isn't my primary. Is there a way to check to see if it is active?



Carl

If you Google for GPU-Z (version 0.3.2 latest), it can be found on the TechPowerUp web site. The middle tab will show the GPU temperature, the fan speed the the percentage of GPU capacity used.

So, that will show the GPU is working.
Go away, I was asleep


ID: 18764 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile verstapp
Avatar

Send message
Joined: 26 Jan 09
Posts: 589
Credit: 497,834,261
RAC: 0
Message 18768 - Posted: 14 Apr 2009, 23:31:20 UTC

Everest gives me GPU usage and temperatures on the OSD.
Cheers,

PeterV

.
ID: 18768 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Carl Johnson[SETI.USA]

Send message
Joined: 8 Apr 09
Posts: 2
Credit: 32,148,082
RAC: 0
Message 18800 - Posted: 15 Apr 2009, 12:50:01 UTC - in response to Message 18764.  

Well I had it running and everything installed and so on and so forth. Then it started crashing all of the tasks immediately.

<core_client_version>6.6.20</core_client_version>
<![CDATA[
<message>
- exit code -1073741515 (0xc0000135)
</message>
]]>


I don't know what that means, it's going to be a stand alone machine so I'll format and start over.
ID: 18800 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile borandi
Avatar

Send message
Joined: 21 Feb 09
Posts: 180
Credit: 27,806,824
RAC: 0
Message 18802 - Posted: 15 Apr 2009, 13:31:33 UTC - in response to Message 18800.  
Last modified: 15 Apr 2009, 13:31:47 UTC

Well I had it running and everything installed and so on and so forth. Then it started crashing all of the tasks immediately.

<core_client_version>6.6.20</core_client_version>
<![CDATA[
<message>
- exit code -1073741515 (0xc0000135)
</message>
]]>

I don't know what that means, it's going to be a stand alone machine so I'll format and start over.


I solved this by installing .Net Framework 3.5
ID: 18802 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
John Clark

Send message
Joined: 4 Oct 08
Posts: 1734
Credit: 64,228,409
RAC: 0
Message 18807 - Posted: 15 Apr 2009, 13:52:38 UTC - in response to Message 18802.  

Well I had it running and everything installed and so on and so forth. Then it started crashing all of the tasks immediately.

<core_client_version>6.6.20</core_client_version>
<![CDATA[
<message>
- exit code -1073741515 (0xc0000135)
</message>
]]>

I don't know what that means, it's going to be a stand alone machine so I'll format and start over.


I solved this by installing .Net Framework 3.5


The Microsoft update Tuesday is over, but the patches still need downloading.

I have just completed the updating (manually) on 2 PCs - 1 and old Win2K one which needed 35MB of update, and the other an XP pro rig which needed 27.8MB of updated. None of which required any .NET updates, which suggests it is present on these Win PCs.

My older Server, running XP Pro, is just downloading the same update patches (again manually) and this includes .NET. The download totals 283.5MB of patches, including a 248.0MB download of .NET Framework 3.5 SP1 and .NET Framework Family Update.

This updating is going to take over 1.5 hours on a 0.5Mbps broadband line.

I wonder if the GPU will be affected, ot what?
Go away, I was asleep


ID: 18807 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile borandi
Avatar

Send message
Joined: 21 Feb 09
Posts: 180
Credit: 27,806,824
RAC: 0
Message 18813 - Posted: 15 Apr 2009, 14:51:09 UTC

My windows updates (XP Pro 32-bit) never, ever, call for .Net Framework, thus I have to download the setup tool and do it that way. From older threads, for some reason .Net 3.5 just seemed to work, but it wasn't figured out why. I didn't go for 3.5 SP1 though, and it all works my end without 3.5 SP1.

You could download the standalone .Net 3.0 file, then copy it across your machines, then update to 3.5 on each.

On a different note, I didn't realise broadband in the UK was still sold at 512kbps :S Well, unless you're in Hull where you have to use the local provider Karoo as BT doesn't have lines there, but even they are slowly upgrading all their lines for free to 1MB...
ID: 18813 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
John Clark

Send message
Joined: 4 Oct 08
Posts: 1734
Credit: 64,228,409
RAC: 0
Message 18815 - Posted: 15 Apr 2009, 15:03:49 UTC - in response to Message 18813.  

[quote
I didn't realise broadband in the UK was still sold at 512kbps :S Well, unless you're in Hull where you have to use the local provider Karoo as BT doesn't have lines there, but even they are slowly upgrading all their lines for free to 1MB...[/quote]

Only for me, and BB users like myself.

I was offered ADSLMax, the line rate adaption system, but as I required stability I stuck with the plain 512kbps ADSL. The reason for this is the line length from the exchange (8 km/5 miles), meaning the attenuation and SNR is OK but marginal.
Go away, I was asleep


ID: 18815 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile borandi
Avatar

Send message
Joined: 21 Feb 09
Posts: 180
Credit: 27,806,824
RAC: 0
Message 18837 - Posted: 15 Apr 2009, 19:01:35 UTC
Last modified: 15 Apr 2009, 19:01:51 UTC

5 miles? Geez, out in the sticks? =P

But fair play - stability is an important factor. BT have started to cap my area at 1MB between 4pm and 12pm :(, but I can get my full 8MB in the early morning.
ID: 18837 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
John Clark

Send message
Joined: 4 Oct 08
Posts: 1734
Credit: 64,228,409
RAC: 0
Message 18843 - Posted: 15 Apr 2009, 19:45:24 UTC - in response to Message 18837.  

5 miles? Geez, out in the sticks? =P

But fair play - stability is an important factor. BT have started to cap my area at 1MB between 4pm and 12pm :(, but I can get my full 8MB in the early morning.


I am on the so called 8MB, but it drops with the length of the line.

My attenuation down is 63.5 db, and the SNR varies between 10 db and 14db.

I believe I could get 1=MB with ADSL Max, but I am not sure as the SNR would be in the area of BB desynchronization.
Go away, I was asleep


ID: 18843 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile verstapp
Avatar

Send message
Joined: 26 Jan 09
Posts: 589
Credit: 497,834,261
RAC: 0
Message 18891 - Posted: 15 Apr 2009, 22:46:44 UTC

>I didn't realise broadband in the UK was still sold at 512kbps
ISPs in Britain, and probably most other places, call anything faster than dialup 'broadband'. So you can have, here in Oz anyway, 128kbps 'broadband'.
Its advertising, therefore its lies. It might be the big lie, it might be a small lie, but its still... :)
Cheers,

PeterV

.
ID: 18891 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Previous · 1 · 2 · 3 · 4 · Next

Message boards : Application Code Discussion : The GPU ripple

©2024 Astroinformatics Group