Bruce, a question about An Optimized Application

Alinator
Alinator
Joined: 8 May 05
Posts: 927
Credit: 9352143
RAC: 0

Well, I can tell you it

Well, I can tell you it doesn't work for a K6-2 running 9x.

Moral of story:

1.) Make sure you exercise some caution before you slipstream it to all your hosts. Related to this, since this opti is essentially an RE hack, that might explain why Dr. Allen wanted to run it through the testing loop before releasing it as an "official" EAH app, because of problems like that.

2.) To announce to the EAH community with a potential CAYS link, with no explanation or even a link to a site to allow anyone to see what's up about it is irresponsible, regardless of how clever the work is.

3.) My gut feeling is Dr. Allen is not going to be too thrilled about this development.

Alinator

rbpeake
rbpeake
Joined: 18 Jan 05
Posts: 266
Credit: 967717074
RAC: 1217988

RE: Moral of

Message 23602 in response to message 23601

Quote:

Moral of story:.....
2.) To announce to the EAH community with a potential CAYS link, with no explanation or even a link to a site to allow anyone to see what's up about it is irresponsible, regardless of how clever the work is.

Alinator


Perhaps a low level link was intentional to limit access to only the most "savvy" users who would be able to deal with any problems, but nonetheless, I am just as irresponsible in using the application. No one is forcing me to use it....

UBT - Halifax--lad
UBT - Halifax--lad
Joined: 10 Nov 04
Posts: 20
Credit: 3162
RAC: 0

RE: RE: Moral of

Message 23603 in response to message 23602

Quote:
Quote:

Moral of story:.....
2.) To announce to the EAH community with a potential CAYS link, with no explanation or even a link to a site to allow anyone to see what's up about it is irresponsible, regardless of how clever the work is.

Alinator


Perhaps a low level link was intententional to limit access to only the most "savvy" users who would be able to deal with any problems, but nonetheless, I am just as irresponsible in using the application. No one is forcing me to use it....

exactly as with any optimised client there is always risk and most users know that

Join us in Chat (see the forum) Click the Sig

Join UBT

Jordan Wilberding
Jordan Wilberding
Joined: 19 Feb 05
Posts: 162
Credit: 715454
RAC: 0

RE: I would like to know

Message 23604 in response to message 23597

Quote:

I would like to know how Dr. Bruce feels about the optimized app before installing. I know it is validating and things, but getting a better official word makes me feel better.

So, is it good for us to use?

I don't know about you, but it feels damn good to me :)

such things just should not be writ so please destroy this if you wish to live 'tis better in ignorance to dwell than to go screaming into the abyss worse than hell

cmds
cmds
Joined: 1 Aug 05
Posts: 62
Credit: 348073
RAC: 0

RE: A36 I tested the

Message 23605 in response to message 23579

Quote:
A36

I tested the very fast application on my Intel P4 3.2 GHz with active HT.
Crunching times nearly the half of the original Albert.

But, the temperature (copper-core oft the cooling unit, checked with Digital Voltmeter and VCR ) will be nearly 10 Degrees lower!

Many Thanks to You. You have done an excellent work!

Chris
---
SpecialOffTopic-Team - FileBase
http://www.cmds.tk

*Die Signatur befindet sich aus technischen Gründen auf der Rückseite dieses Beitrages!*

DanNeely
DanNeely
Joined: 4 Sep 05
Posts: 1364
Credit: 3562358667
RAC: 1580

Bruce's looked over the code

Bruce's looked over the code changes for the optimization, praised them and promised new apps for all platforms using them, so I think any objection would be pro forma at worst.

The drop in cpu temp's probably related to the fact that the biggest algorithm change removed most of the slow division ops, so you're not using that part of your FPU much.

Ziran
Ziran
Joined: 26 Nov 04
Posts: 194
Credit: 356403
RAC: 1626

Akosf wrote earlier in this

Message 23607 in response to message 23557

Akosf wrote earlier in this tread:

Quote:
Quote:
Any results on this so far? Any beta-testers needed? :)
I would like to help on speed optimization, but i don't know what is the way of it. Probably i can put my code on a webpage, but i think it would be not legal. I didn't get any e-mails in connection with legitimacy.

Something made him change his mind. We know from earlier in this thread that he has shown his optimized client to Bruce. His post only contained the link to the application. No installation instructions, so you need some familiarity with BOINC file structure to be wiling to install it. Look at the timestamp of his post.
Posted 7 Mar 2006 6:59:08 UTC
If you wanted to release a test application, isn’t just before work Tuesday morning an excellent time to do so, if you want to keep an eye on things. (just before 8 in Germany)
The sent back result contains this massage:
�Optimized by akosf (A-36) --> 'projects/einstein.phys.uwm.edu/albert_4.37_windows_intelx86.exe'�
If Bruce made a post wanting us for some reason to stop using this client, wouldn’t most of us reading this thread uninstall it so fast it would make a gravitational wave even SETI could detect. :)

Is that the X-files theme I hear in the background? :)

Then you're really interested in a subject, there is no way to avoid it. You have to read the Manual.

DanNeely
DanNeely
Joined: 4 Sep 05
Posts: 1364
Credit: 3562358667
RAC: 1580

PS has anyone tested it on an

PS has anyone tested it on an amd platform? I've got it running on my a64x2 and it hasn't crashed, but it'l lbe afew hours before I've got validatable results.

Alinator
Alinator
Joined: 8 May 05
Posts: 927
Credit: 9352143
RAC: 0

RE: Akosf wrote earlier in

Message 23609 in response to message 23607

Quote:
Akosf wrote earlier in this tread:
Quote:
Quote:
Any results on this so far? Any beta-testers needed? :)
I would like to help on speed optimization, but i don't know what is the way of it. Probably i can put my code on a webpage, but i think it would be not legal. I didn't get any e-mails in connection with legitimacy.

Something made him change his mind. We know from earlier in this thread that he has shown his optimized client to Bruce. His post only contained the link to the application. No installation instructions, so you need some familiarity with BOINC file structure to be wiling to install it. Look at the timestamp of his post.
Posted 7 Mar 2006 6:59:08 UTC
If you wanted to release a test application, isn’t just before work Tuesday morning an excellent time to do so, if you want to keep an eye on things. (just before 8 in Germany)
The sent back result contains this massage:
�Optimized by akosf (A-36) --> 'projects/einstein.phys.uwm.edu/albert_4.37_windows_intelx86.exe'�
If Bruce made a post wanting us for some reason to stop using this client, wouldn’t most of us reading this thread uninstall it so fast it would make a gravitational wave even SETI could detect. :)

Is that the X-files theme I hear in the background? :)

Be that as it may, and I am implying no malfeasance on Akosf's part in reverse engineering Albert, or malintent in his "obscure" test release. The fact remains that if EAH wanted to have app version variety in the wild they would have made the source code freely available and encouraged folks to have at it, like SAH does.

I took Dr. Allen's prompt and positive responses as an indicator he found the proof-of-concept app had merit, but we would have to wait until as many of the ramifications of full scale release could be evaluated and addressed.

Alinator

DanNeely
DanNeely
Joined: 4 Sep 05
Posts: 1364
Credit: 3562358667
RAC: 1580

RE: PS has anyone tested it

Message 23610 in response to message 23608

Quote:
PS has anyone tested it on an amd platform? I've got it running on my a64x2 and it hasn't crashed, but it'l lbe afew hours before I've got validatable results.

I've completed two and two halfs (units started under old finished under new). One and two halfs have validated successfully, the 4th WU is waiting on someone else, and won't be validated for annother day or two.

I've gone from 3:45+-15 per large WU to 1:40/WU.

CPU temp hasn't deltaed more than a degree or three at most (I didn't check beforehand), but a64s aren't blastfurnaces to begin with.

Comment viewing options

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