S39L Observation Thread

StoneLord
StoneLord
Joined: 15 Jun 05
Posts: 13
Credit: 50798
RAC: 0

RE: RE: 00109404: 8B 4C

Message 26423 in response to message 26422

Quote:
Quote:
00109404: 8B 4C DE 80 83 EB 10 75 F3 89 EA 90 ->
00109404: 0F 18 44 DE 80 83 EB 10 75 F2 89 EA
SETI_K78b

Ok, it seems I returned too late to download this update so I applied patch by hands. Patched version finished current WU successfully but all others finished with error immediately. So this version is not for Athlon XP.

Quote:
Well, I modified, but I'm not interested in SETI.

Very sad news for SETI crunchers :( Your SETI_K78 outperforms both Crunch3r's SSE2&3 versions on all Athlon 64/ Pentium 4 I have to install it.
It's amazing, thank you very much!
Maybe you would consider possibility of creation 3DNow! optimized client for SETI too? Your D-40 for Einstein is great....
Thank you again.

Hi. Where can i download optimized client for s@h - > seti_K78b.zip, because the previous link is not operational.

Akos Fekete
Akos Fekete
Joined: 13 Nov 05
Posts: 561
Credit: 4527270
RAC: 0

RE: Ok, it seems I returned

Message 26424 in response to message 26422

Quote:
Ok, it seems I returned too late to download this update so I applied patch by hands. Patched version finished current WU successfully but all others finished with error immediately. So this version is not for Athlon XP.

It was a fast modified code to my Duron.
But I tried out on my Durons and on one of my laptops (Dothan core) some days ago.
It worked fine.

Quote:
Quote:
Well, I modified, but I'm not interested in SETI.
Very sad news for SETI crunchers :( Your SETI_K78 outperforms both Crunch3r's SSE2&3 versions on all Athlon 64/ Pentium 4 I have to install it.
It's amazing, thank you very much!
Maybe you would consider possibility of creation 3DNow! optimized client for SETI too? Your D-40 for Einstein is great....
Thank you again.

Well... I think you don't know that I'm a big SETI fanatic.
Probably, I was the only guy who used an optimised SETI classic client. :-)
And, I tried to build a FPGA (Xilinx Spartan-3 ~ 10USD) based SETI accelerator card about a year ago, but the speed of the PCI bus wasn't enough to my algorithm. I was in trouble because I didn't have the source code so I couldn't select better functions to reduce the needed bandwidth.

Einstein. The test of S-40 is running. It has better prefetch.

Raistmer*
Raistmer*
Joined: 20 Feb 05
Posts: 208
Credit: 181428947
RAC: 6029

RE: It was a fast modified

Message 26425 in response to message 26424

Quote:
It was a fast modified code to my Duron.
But I tried out on my Durons and on one of my laptops (Dothan core) some days ago.
It worked fine.


Hm, strange. I tried on Athlon Thoroughbred 2400+. Will repeat on Barton...

Quote:
Well... I think you don't know that I'm a big SETI fanatic.
Probably, I was the only guy who used an optimised SETI classic client. :-)
And, I tried to build a FPGA (Xilinx Spartan-3 ~ 10USD) based SETI accelerator card about a year ago, but the speed of the PCI bus wasn't enough to my algorithm. I was in trouble because I didn't have the source code so I couldn't select better functions to reduce the needed bandwidth.


Wow :) It sounds wonderful. So I will not worry about SETI optimization future now :)

Quote:

Einstein. The test of S-40 is running. It has better prefetch.


Hope it will be public available very soon to try it :)

PS Barton met the same situation. Current WU was finished fine, but all other ended with error after 2 sec of computing.
***UNHANDLED EXCEPTION****
Reason: Access Violation (0xc0000005) at address 0x00509400 read attempt to address 0x01F0FF80

Steve Cressman
Steve Cressman
Joined: 9 Feb 05
Posts: 104
Credit: 139654
RAC: 0

I tried both K78 and K78b on

I tried both K78 and K78b on my barton with no success. But the good news is that with the latest opp app, D-40, I am now doing work 3.7x faster than with the standerd app. Went from 260 min/result to 70 min/result. D-40 knocked off another 5 min :)

98SE XP2500+ @ 2.1 GHz Boinc v5.8.8

caesar1
caesar1
Joined: 6 Dec 05
Posts: 8
Credit: 1108550
RAC: 0

RE: Hi. Where can i

Message 26427 in response to message 26423

Quote:

Hi. Where can i download optimized client for s@h - > seti_K78b.zip, because the previous link is not operational.

akosf,

maybe you have overlooked this question...?

it would be very kind if you post another link

Akos Fekete
Akos Fekete
Joined: 13 Nov 05
Posts: 561
Credit: 4527270
RAC: 0

RE: RE: Hi. Where can i

Message 26428 in response to message 26427

Quote:
Quote:

Hi. Where can i download optimized client for s@h - > seti_K78b.zip, because the previous link is not operational.

akosf,

maybe you have overlooked this question...?

it would be very kind if you post another link

These files are removed, because some people found an illagel address problem. I said that this executable was written to myself, but it worked fine (on my computers). It did about 1500-2000 wus to me (Duron,AthlonXP,Northwood,Dothan) a little bit faster than other clients. I'm sure that some prefetch routines aren't perfect, but it wasn't problem to me. (I used Win2000 and BOINC 4.19/4.45.) Probably I will do a revision on this code.

Raistmer*
Raistmer*
Joined: 20 Feb 05
Posts: 208
Credit: 181428947
RAC: 6029

RE: These files are

Message 26429 in response to message 26428

Quote:
These files are removed, because some people found an illagel address problem. I said that this executable was written to myself, but it worked fine (on my computers). It did about 1500-2000 wus to me (Duron,AthlonXP,Northwood,Dothan) a little bit faster than other clients. I'm sure that some prefetch routines aren't perfect, but it wasn't problem to me. (I used Win2000 and BOINC 4.19/4.45.) Probably I will do a revision on this code.

Please, restore link for original seti_K78 with remark that it's work on Athlon 64/P4 SSE2/3 and may not work on SSE-only processors. This version works faster on P4/A64 than Crunch3r's SSE2/3 so it will be surely useful on these procs.

Akos Fekete
Akos Fekete
Joined: 13 Nov 05
Posts: 561
Credit: 4527270
RAC: 0

RE: Please, restore link

Message 26430 in response to message 26429

Quote:
Please, restore link for original seti_K78 with remark that it's work on Athlon 64/P4 SSE2/3 and may not work on SSE-only processors. This version works faster on P4/A64 than Crunch3r's SSE2/3 so it will be surely useful on these procs.

The link is live again. (SETI K7-08)
But I don't do any funny remarks and I don't promise anything.
Is okay?

Misfit
Misfit
Joined: 11 Feb 05
Posts: 470
Credit: 100000
RAC: 0

RE: RE: Please, restore

Message 26431 in response to message 26430

Quote:
Quote:
Please, restore link for original seti_K78 with remark that it's work on Athlon 64/P4 SSE2/3 and may not work on SSE-only processors. This version works faster on P4/A64 than Crunch3r's SSE2/3 so it will be surely useful on these procs.
The link is live again. (SETI K7-08)
But I don't do any funny remarks and I don't promise anything.
Is okay?


I did a few WUs with that app. They finished OK but took 7-10 min longer.

me-[at]-rescam.org

Reuben Gathright
Reuben Gathright
Joined: 15 Feb 06
Posts: 23
Credit: 6851215
RAC: 0

Thanks akosf for relinking

Thanks akosf for relinking SETI K708.

I have processed 12 wu's in the past few hours on my Dual Xeon 3.0Ghz 2mb. I am getting an average time of 21 min, my old average was 28 min using Cruncher's SSE3 app. I have a lot of 19 min results but sometimes they finish in 25 mins.

Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.