Bruce, a question about An Optimized Application

Bruce Allen
Bruce Allen
Moderator
Joined: 15 Oct 04
Posts: 1119
Credit: 172127663
RAC: 0

RE: RE: Any results on

Message 23561 in response to message 23557

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.

This is my fault -- I got caught up in some urgent things at this end. I've just written to you off-list.

Cheers,
Bruce

Director, Einstein@Home

Stef
Stef
Joined: 8 Mar 05
Posts: 206
Credit: 110568193
RAC: 0

Any News?

Any News?

Bruce Allen
Bruce Allen
Moderator
Joined: 15 Oct 04
Posts: 1119
Credit: 172127663
RAC: 0

RE: Any News? Yes.

Message 23563 in response to message 23562

Quote:
Any News?

Yes. Akosf has done three things to speed up our executable. One of these optimizations is very clever: it eliminates large numbers of (slow) divisions. We're in the process of building and testing new executables (for all platforms) that incorporate these changes. They should result in very substantial speed-ups.

Bruce

Director, Einstein@Home

FalconFly
FalconFly
Joined: 16 Feb 05
Posts: 191
Credit: 15650710
RAC: 0

Sounds excellent! That's

Message 23564 in response to message 23563

Sounds excellent!

That's the spirit and why I like this Project alot :o)

ExtraTerrestrial Apes
ExtraTerrestria...
Joined: 10 Nov 04
Posts: 770
Credit: 579306861
RAC: 196998

RE: One of these

Message 23565 in response to message 23563

Quote:
One of these optimizations is very clever: it eliminates large numbers of (slow) divisions. We're in the process of building and testing new executables (for all platforms) that incorporate these changes. They should result in very substantial speed-ups.

That's great news guys :)
I'll be running 100% Einstein for some time now, because I'm running out of network traffic (wtf) .. so this may be just in time.

MrS

Scanning for our furry friends since Jan 2002

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

RE: RE: Any News? Yes.

Message 23566 in response to message 23563

Quote:
Quote:
Any News?

Yes. Akosf has done three things to speed up our executable. One of these optimizations is very clever: it eliminates large numbers of (slow) divisions. We're in the process of building and testing new executables (for all platforms) that incorporate these changes. They should result in very substantial speed-ups.

Bruce

If you need someone to help test in Linux, just give a shout!

I am really looking forward to this.

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

BORG
BORG
Joined: 8 Nov 04
Posts: 27
Credit: 200263
RAC: 0

RE: RE: RE: Any

Message 23567 in response to message 23566

Quote:
Quote:
Quote:
Any News?

Yes. Akosf has done three things to speed up our executable. One of these optimizations is very clever: it eliminates large numbers of (slow) divisions. We're in the process of building and testing new executables (for all platforms) that incorporate these changes. They should result in very substantial speed-ups.

Bruce

If you need someone to help test in Linux, just give a shout!

I am really looking forward to this.

I'm also running Einstein exclusively in Linux and would love to help.

BORG

B52
B52
Joined: 19 Feb 05
Posts: 45
Credit: 273899
RAC: 0

RE: RE: Any News? Yes.

Message 23568 in response to message 23563

Quote:
Quote:
Any News?

Yes. Akosf has done three things to speed up our executable. One of these optimizations is very clever: it eliminates large numbers of (slow) divisions. We're in the process of building and testing new executables (for all platforms) that incorporate these changes. They should result in very substantial speed-ups.

Bruce

Sounds really, really great. Looking forward to the new bins

Zilli Samuel
Zilli Samuel
Joined: 2 Mar 06
Posts: 6
Credit: 21583
RAC: 0

RE: We're in the process

Message 23569 in response to message 23563

Quote:

We're in the process of building and testing new executables (for all platforms) that incorporate these changes. They should result in very substantial speed-ups.

Any news? Are new binaries arriving?

I'd like to suggest to make anyway an SSE2 optimized version of the albert application (expecially for Win).
For Seti, I saw that the best boost is done with SSE2 optimization, and a lot of people own an SSE2 enabled CPU.
It shouldn't be too time-expensive to build only one optimized version...

josep
josep
Joined: 9 Mar 05
Posts: 63
Credit: 1156542
RAC: 0

RE: I'd like to suggest to

Message 23570 in response to message 23569

Quote:

I'd like to suggest to make anyway an SSE2 optimized version of the albert application (expecially for Win).
For Seti, I saw that the best boost is done with SSE2 optimization, and a lot of people own an SSE2 enabled CPU.

Read below, in the same thread we are now, a post from Bernd Machenschalk. He has tested it, to find that SSE2 does not give a significant improvement over SSE in Linux compiling with gcc.

And for Windows, he says there is no significant improvement when using SSE or SSE2 over the default optimizations from MSVC compiler.

Comment viewing options

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