S38 Observation thread

Erik
Erik
Joined: 14 Feb 06
Posts: 2815
Credit: 2645600
RAC: 0

Have to agree with Misfit.

Have to agree with Misfit. 3.0 p4 ht with S38 is slicing around 2000-2300 compared to C37. Even more impressive is 2.4 Celeron with roughly 6000 sec.

networkman
networkman
Joined: 22 Jan 05
Posts: 98
Credit: 7140649
RAC: 0

I must've missed the details

I must've missed the details on how C37 improves upon performance over A36, because the difference I was seeing was negligible - sorry, no numbers to post at the moment. But I'll try the S38 over the next day or two and let you know what I find. :)

Thanks for all of your efforts, akosf. :D

"Chance is irrelevant. We will succeed."
- Seven of Nine

Santas little helper
Santas little helper
Joined: 11 Feb 05
Posts: 36
Credit: 4699218
RAC: 0

3 valid S-38 / 3 valid C-37:

3 valid S-38 / 3 valid C-37: 0.76
on P4 Northwood 2.67 GHz 512KB L2 @ WinXP Pro SP2
:)

Greetings, Santas little helper

Zap
Zap
Joined: 12 Feb 06
Posts: 15
Credit: 3900434
RAC: 0

AMD64 xp 3000 Newcastle core

AMD64 xp 3000 Newcastle core 10% overclock.

Went from 14k plus secs with the original app through an average of some less then 6000 with A36 to now my first result with S38 in 4235 secs.
Quite impressive Akosf.
Since I met all my goals set for Seti I will dedicate more time to E@h and thanks to Akosf with more then three times the results a day.

RandyC
RandyC
Joined: 18 Jan 05
Posts: 6003
Credit: 111139797
RAC: 0

RE: Have to agree with

Message 26143 in response to message 26139

Quote:
Have to agree with Misfit. 3.0 p4 ht with S38 is slicing around 2000-2300 compared to C37. Even more impressive is 2.4 Celeron with roughly 6000 sec.

Based on 4 results, my AMD XP 1600+ system has dropped about 1700 secs per WU. About 20% decrease in processing time C37 to S38.

Seti Classic Final Total: 11446 WU.

Nothing But Idle Time
Nothing But Idl...
Joined: 24 Aug 05
Posts: 158
Credit: 289204
RAC: 0

RE: RE: Have to agree

Message 26144 in response to message 26143

Quote:
Quote:
Have to agree with Misfit. 3.0 p4 ht with S38 is slicing around 2000-2300 compared to C37. Even more impressive is 2.4 Celeron with roughly 6000 sec.

Based on 4 results, my AMD XP 1600+ system has dropped about 1700 secs per WU. About 20% decrease in processing time C37 to S38.

Same here with my P4 3.0 HT...S38 getting 20% decrease in completion time over C37. Project supplied Albert app ran 8 hours, S38 now running 2.75 hrs. Some old Einstein WU's used to take me between 11.5 and 13.5 hrs! Those were the good old days?

networkman
networkman
Joined: 22 Jan 05
Posts: 98
Credit: 7140649
RAC: 0

My very initial observations

My very initial observations with a few rigs that have completed workunits:

"APOPHIS" AMD XP2000+
8170 seconds A36
5972 seconds S38

"MARDUK" AMD XP2400+
6860 seconds A36
4981 seconds S38

"ANUBIS" AMD XP2500+
6800 seconds A36
4941 seconds S38

Again, a very small sampling, but the results are very encouraging. :) Thanks yet again to you, akosf!

"Chance is irrelevant. We will succeed."
- Seven of Nine

LiborA
LiborA
Joined: 8 Dec 05
Posts: 74
Credit: 337135
RAC: 0

I find very interesting fact.

I find very interesting fact. On my Athlon64 2800+ each other WU is crunched about 1% longer.

There is list of results all WU's crunched by S38:
Time reported ; CPU Time ; Ratio ; Optimalization
--------------------------------------------------------------
20 Mar 2006 21:55:22 ; 1340,32 ; ---- ; S38
20 Mar 2006 22:21:18 ; 1349,99 ; 0,99 ; S38
20 Mar 2006 22:47:03 ; 1360,04 ; 0,99 ; S38
20 Mar 2006 19:51:24 ; 1371,03 ; 0,99 ; S38/S37a
20 Mar 2006 23:12:48 ; 1371,10 ; 1,00 ; S38
20 Mar 2006 23:38:34 ; 1381,87 ; 0,99 ; S38
21 Mar 2006 00:04:19 ; 1393,44 ; 0,99 ; S38
21 Mar 2006 00:30:10 ; 1404,95 ; 0,99 ; S38
21 Mar 2006 00:55:51 ; 1415,35 ; 0,99 ; S38
21 Mar 2006 01:21:46 ; 1425,66 ; 0,99 ; S38

20 Mar 2006 21:29:37 ; 4617,36 ; ---- ; S38
21 Mar 2006 02:49:03 ; 4821,42 ; 0,96 ; S38
21 Mar 2006 04:21:04 ; 4872,18 ; 0,99 ; S38
21 Mar 2006 09:14:06 ; 4895,22 ; 1,00 ; S38
21 Mar 2006 12:42:22 ; 4923,44 ; 0,99 ; S38

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

I would like to see the

I would like to see the performance of the new sourcecode optimalised application. I think that will be much faster, because the optimalisation of sourcecode gives easier and more facilities.

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

RE: I find very interesting

Message 26148 in response to message 26146

Quote:
I find very interesting fact. On my Athlon64 2800+ each other WU is crunched about 1% longer.

...and from my "COBRA" (Dothan CPU):

result         time      claimed  granted
22191416 4,908.92 47.47 pending
22185063 4,908.88 47.47 pending
22178529 4,908.67 47.47 pending
22169717 4,908.81 47.47 pending
22166278 4,908.63 47.47 pending
22158199 4,908.91 47.47 pending
22152486 4,909.03 47.47 pending
22145696 4,908.94 47.47 pending
22140259 4,908.55 47.47 28.66
22133271 4,909.00 47.47 pending
22126189 4,908.97 47.47 pending

I think the different results would be interesting...
The cpu executes same amount of instructions, that needs same amount of time.

Comment viewing options

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