NEW: WINDOWS TEST APPLICATION FOR EINSTEIN@HOME

Bjoern Henke
Bjoern Henke
Joined: 4 Jan 05
Posts: 5
Credit: 115,825,452
RAC: 180,558

Einstein 0.03 is running fine

Einstein 0.03 is running fine here without any errors.
Graphics are working even on the system with an ATI graphics card (but they always worked on that system).
Processing times have increased around 8-9 percent. It really looks like the increased time is related to the boinc stuff in it, as the new SETI application 4.18 is affected as well.

Here are my processing times:

Mobile Athlon 4 - 1.2 GHz

Einstein v4.79: 10.8 hours
Einstein v0.03: 11.7 hours (+8,33%)

SETI v4.08/v4.09: 5.3 hours
SETI v4.18: 5.7 hours (+7,55%)

Athlon XP 2400+

v4.79: 6.6 hours
v0.03: 7.3 hours (+10,61%)

Athlon XP 3200+

v4.79: 5.6 hours
v0.03: 6.1 hours (+8,93%)

SETI v4.08/v4.09: 2.2 hours
SETI v4.18: 2.4 hours (+9,10%)

Chipper Q
Chipper Q
Joined: 20 Feb 05
Posts: 1,540
Credit: 708,571
RAC: 0

The transition from 4.79 to

The transition from 4.79 to 0.03 worked as advertised on an HP Media Center PC (P4 @ 3.2GHz, 840MHz FSB, running XP Pro – I don't think it's 'dual core', but BOINC has been making good use of a 2nd CPU from somewhere(?) - I thought from one of the 2 factory installed (Radeon X300 Series) video cards).

Doesn't the Scheduler program monitor and resolve performance issues regarding individual machines and user preferences? (I was impressed that BOINC runs benchmarks on initialization and at periodic intervals.)

Heffed
Heffed
Joined: 18 Jan 05
Posts: 257
Credit: 12,368
RAC: 0

RE: The transition from

Message 12994 in response to message 12993

Quote:
The transition from 4.79 to 0.03 worked as advertised on an HP Media Center PC (P4 @ 3.2GHz, 840MHz FSB, running XP Pro – I don't think it's 'dual core', but BOINC has been making good use of a 2nd CPU from somewhere(?) - I thought from one of the 2 factory installed (Radeon X300 Series) video cards).


No, that wouldn't be the video cards. That's the hyperthreading feature of your processor. A single core, but two 'logical' CPUs.

Shaktai
Shaktai
Joined: 8 Nov 04
Posts: 183
Credit: 426,451
RAC: 0

Pentium D 840 since 21 Jun

Pentium D 840 since 21 Jun 2005 0:53:58 UTC
46 units total
1 download error (h1_ unit)
1 computing error
4 pending
8 to be processed
32 successfully completed and validated.

A pretty good beta so far.

Chipper Q
Chipper Q
Joined: 20 Feb 05
Posts: 1,540
Credit: 708,571
RAC: 0

RE: RE: The transition

Message 12996 in response to message 12994

Quote:
Quote:
The transition from 4.79 to 0.03 worked as advertised on an HP Media Center PC (P4 @ 3.2GHz, 840MHz FSB, running XP Pro – I don't think it's 'dual core', but BOINC has been making good use of a 2nd CPU from somewhere(?) - I thought from one of the 2 factory installed (Radeon X300 Series) video cards).

No, that wouldn't be the video cards. That's the hyperthreading feature of your processor. A single core, but two 'logical' CPUs.


Thanks, Heffed! That explains why some of portions of my code seemed to be already executed before even being called (in an unrelated prime number crunching routine).

Divide Overflow
Divide Overflow
Joined: 9 Feb 05
Posts: 91
Credit: 183,220
RAC: 0

So, the graphics work for ATI

So, the graphics work for ATI video cards once again. A nice plus for those with that hardware who want the eye candy. The application still seems to be running slower on a majority of hosts, a big minus for almost everyone out there. It was mentioned that updates to the BOINC library code and a new application version might change this. Is there a v004 beta application in the works?

hm
hm
Joined: 15 May 05
Posts: 10
Credit: 16,825,087
RAC: 0

RE: So, the graphics work

Message 12998 in response to message 12997

Quote:
So, the graphics work for ATI video cards once again. A nice plus for those with that hardware who want the eye candy. The application still seems to be running slower on a majority of hosts, a big minus for almost everyone out there. It was mentioned that updates to the BOINC library code and a new application version might change this. Is there a v004 beta application in the works?

I am having a problem with the beta with ATI...when I go to wake up the screen saver, it oftne bombs the client with an unrecoverable error as in: 7/11/2005 8:58:06 AM|Einstein@Home|Unrecoverable error for result l1_0249.0__0249.4_0.1_T08_S4lA_0 ( - exit code -1073741819 (0xc0000005)

Heffed
Heffed
Joined: 18 Jan 05
Posts: 257
Credit: 12,368
RAC: 0

hm, please alter your sig

hm, please alter your sig line. It stretches the page.

Thank you.

Jord
Joined: 26 Jan 05
Posts: 2,952
Credit: 5,779,100
RAC: 0

hm, look in the Boinc Wiki

hm, look in the Boinc Wiki for an explanation.

Hans Sveen
Hans Sveen
Joined: 18 Jan 05
Posts: 11
Credit: 13,164,229
RAC: 0

Hello! It seems to be some

Hello!
It seems to be some problem with the validitor, or am I missing something; seems to be delivered in time as the 4th result, I know 3 is enough!
So why no credit on this wu?
Wu in question is http://einsteinathome.org/task/6128119

Thank for some light on this issue!

PS.
Are You sure the problem with the Ati card are gone, take a look at this result, http://einsteinathome.org/task/5938274

Comment viewing options

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