S5 patches - Proposal for a very small alpha test group

Gary Roberts
Gary Roberts
Moderator
Joined: 9 Feb 05
Posts: 5,593
Credit: 85,623,065,463
RAC: 67,044,852
Topic 191441

Please see the point posted in the READ ONLY thread and please make any comments or respond to discussion here.

If you have a fast machine with good turnaround time and you wish to be on the bleeding edge you can volunteer by registering yourself and your machine details here. Akos can then choose what he needs and his decision would be final.

EDIT: Please note that if you do volunteer you are exposing yourself to possible invalid results. You should only volunteer if you are prepared to stay the distance and can pay attention and report back in a timely manner.

Cheers,
Gary.

Pepperammi
Pepperammi
Joined: 20 Feb 05
Posts: 131
Credit: 437,943
RAC: 0

S5 patches - Proposal for a very small alpha test group

best machine;
11hours for a long unit
Pentium D 830 Running at 3.21Ghz 2MB(upto SSE3 able)
3GB Dual ChannelDDR2
Windows XP all updates

Second machine takes 15-18 hours for a long so maybe a little long;
Pentium 4 HT 3.4Ghz 1MB (upto SSE3 able)
1GB Dual Channel DDR
Windows XP all updates

Ulrich Metzner
Ulrich Metzner
Joined: 22 Jan 05
Posts: 113
Credit: 963,370
RAC: 0

Best machine (only available

Best machine (only available workdays):
Pentium 4 650 Prescott, 2 MB L2, 3.4 GHz, 3 GB 800 MHz Ram
Limited to only 1 Process for noise reasons ;)

All time available machine:
Athlon C Thunderbird, 256 KB L2, 1.4 GHz, 512 MB 266 MHz Ram

Momentary out of business but can be reactivated on demand:
Pentium III-M Tualatin, 512 KB L2, 1.0 GHz, 640 MB 133 MHz Ram

Aloha, Uli

Maarten Bos
Maarten Bos
Joined: 3 Mar 05
Posts: 4
Credit: 106,201
RAC: 0

Hello people Well i'm willing

Hello people
Well i'm willing to take a little risk and so offer my P4 northwood 2.8 GHz with HT to test a new app.
The pc runs 24 hours a day with a turnaround time with the new large units of i guess now 15 hours. (first units just started so i don't know precisely yet)
The system is running windows XP fully updated
Just let me know if my system would have the honor of testing an improved app.

Maarten Bos
Maarten Bos
Joined: 3 Mar 05
Posts: 4
Credit: 106,201
RAC: 0

maybe also handy to know is

maybe also handy to know is that i normally check my progress about 3 times a day so i should be able to spot invalid work units pretty soon

M. Schmitt
M. Schmitt
Joined: 27 Jun 05
Posts: 478
Credit: 15,872,262
RAC: 0

My X2 is SSE3 capable, does

My X2 is SSE3 capable, does two long WUs in about 7h and it is willing for testing. I have asked. ;)

cu,
Michael

Dave Burbank
Dave Burbank
Joined: 30 Jan 06
Posts: 275
Credit: 1,548,376
RAC: 0

I will gladly offer my rigs

I will gladly offer my rigs up for alpha testing.

My main rig A64 3700+ (SSE3) is about to start crunching S5 WU's this afternoon(EST), I will run a few standard app WU's to establish a baseline result, but after that she's all yours Akos. I also run a P4 2.0 ghz (SSE2) and a PII 450 mhz. These are also running the standard app to establish time baselines.

Glad to see you're at it again Akos
Dave

Edit: P4 2.0 not PIII 2.0

There are 10^11 stars in the galaxy. That used to be a huge number. But it's only a hundred billion. It's less than the national deficit! We used to call them astronomical numbers. Now we should call them economical numbers. - Richard Feynman

Akos Fekete
Akos Fekete
Joined: 13 Nov 05
Posts: 561
Credit: 4,527,270
RAC: 0

Hi people! Gary invited me

Hi people!

Gary invited me to here to give some answers.
So, what are the questions?

Gary Roberts
Gary Roberts
Moderator
Joined: 9 Feb 05
Posts: 5,593
Credit: 85,623,065,463
RAC: 67,044,852

Actually it shouldn't have

Actually it shouldn't have been this thread. I thought I'd fixed the link quickly but you nust have spotted it before I got the edit done. The link was supposed to take you to a post in the READ ONLY thread.

With your announcement that you were going to stick to SSE3 only, the problem basically goes away anyway and the people who are registering their interest will probably be disappointed if they don't have an SSE3 capable processor :).

You should, however, significantly boost the sales of SSE3 boxes so you should be asking for a commission. I've got a very large number of SSE capable machines. I wonder how much I could get for them as a trade-in on a small fleet of SSE3 boxes :).

Cheers,
Gary.

Akos Fekete
Akos Fekete
Joined: 13 Nov 05
Posts: 561
Credit: 4,527,270
RAC: 0

RE: Actually it shouldn't

Message 40246 in response to message 40245

Quote:

Actually it shouldn't have been this thread. I thought I'd fixed the link quickly but you nust have spotted it before I got the edit done. The link was supposed to take you to a post in the READ ONLY thread.

With your announcement that you were going to stick to SSE3 only, the problem basically goes away anyway and the people who are registering their interest will probably be disappointed if they don't have an SSE3 capable processor :).

You should, however, significantly boost the sales of SSE3 boxes so you should be asking for a commission. I've got a very large number of SSE capable machines. I wonder how much I could get for them as a trade-in on a small fleet of SSE3 boxes :).

Hm... I can exchange the SSE3 instructions to a slower FPU code any time to get an SSE compatible code, but these patches doesn't give to big performance improvement...

Can anybody do a time measurment with the same wu ( official / S5Txxxx ) ?
( I have only 10+ hours wus... )

Winterknight
Winterknight
Joined: 4 Jun 05
Posts: 614
Credit: 219,468,730
RAC: 40,413

RE: Can anybody do a time

Message 40247 in response to message 40246

Quote:

Can anybody do a time measurment with the same wu ( official / S5Txxxx ) ?
( I have only 10+ hours wus... )


Sorry haven't got time to do as you ask, but I have 45 short units that all get 20.46cr.
The first 42 were done on official app in an average of 9,480.95s
The last 3 were done with S5T0003 in an average of 9,336.7s

Thats 1.5% faster, it is a Dual P3 925MHz (coppermine), hostid=662822 running alongside Seti with cpu affinity enforced.

Andy
P.S. S5T0004 looks like it failed validation for one result.

Comment viewing options

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