Hopefully, just an OOPS!! moment ....

Mike Hewson
Mike Hewson
Moderator
Joined: 1 Dec 05
Posts: 6,125
Credit: 127,318,678
RAC: 18,705

RE: ... rate of climb is so

Message 76759 in response to message 76758

Quote:
... rate of climb is so strong ...

Ah, Akos .... you are the Chuck Yeager of E@H! :-)

Cheers, Mike.

I have made this letter longer than usual because I lack the time to make it shorter. Blaise Pascal

peanut
peanut
Joined: 4 May 07
Posts: 162
Credit: 9,644,812
RAC: 0

It is sort of a shame that

It is sort of a shame that the "OOPS" happened on the top computers page. Akosf would have a well deserved top spot. I've only been participating in E@H for a short time, I don't know if akosf has been on top before or not. The only thing that can unseat him now is a >8 cpu beast or an "OOPS".

Brian Silvers
Brian Silvers
Joined: 26 Aug 05
Posts: 772
Credit: 282,700
RAC: 0

RE: It is sort of a shame

Message 76761 in response to message 76760

Quote:
It is sort of a shame that the "OOPS" happened on the top computers page. Akosf would have a well deserved top spot. I've only been participating in E@H for a short time, I don't know if akosf has been on top before or not. The only thing that can unseat him now is a >8 cpu beast or an "OOPS".

...or your system with the tweaks that he has done...

Me, personally, I'd just be happy to see a Windows app with the SSE optimization that you and other Intel Mac OSX users have been able to enjoy...

Jim Bailey
Jim Bailey
Joined: 31 Aug 05
Posts: 91
Credit: 1,452,829
RAC: 0

RE: . Me, personally, I'd

Message 76762 in response to message 76761

Quote:

.

Me, personally, I'd just be happy to see a Windows app with the SSE optimization that you and other Intel Mac OSX users have been able to enjoy...

You mean an optimized app for the other 85% of the user base? Wouldn't want anyone to waste any time on something as trivial as that, would you?

rbpeake
rbpeake
Joined: 18 Jan 05
Posts: 247
Credit: 219,266,664
RAC: 0

RE: RE: . Me,

Message 76763 in response to message 76762

Quote:
Quote:

.

Me, personally, I'd just be happy to see a Windows app with the SSE optimization that you and other Intel Mac OSX users have been able to enjoy...

You mean an optimized app for the other 85% of the user base? Wouldn't want anyone to waste any time on something as trivial as that, would you?


Ha, ha, good point! However, I would believe that is known (it is not high level gravitational physics, after all! ;) I am assuming that the project scientists have a reason for being so cautious.....

Although it would be nice, imho, if the project scientists would share their thoughts about this with us, if for no other reason than that we would stop bugging them about it! (OK, maybe not completely stop bugging them, but maybe reduce the volume of bugging! ;)

Jim Bailey
Jim Bailey
Joined: 31 Aug 05
Posts: 91
Credit: 1,452,829
RAC: 0

As run times increase it

As run times increase it costs me more to do a given number of WU's. It's a simple matter of being cost effective. When it costs me more than I figure it's worth, and no one has made an attempt to help with and improved science app, then it's time to move on. I shut down 4 computers last week and moved 3 to another project at the same time. That leaves 2 on Einstein and they have been set to "No New Work".

Mike Hewson
Mike Hewson
Moderator
Joined: 1 Dec 05
Posts: 6,125
Credit: 127,318,678
RAC: 18,705

RE: Ha, ha, good point!

Message 76765 in response to message 76763

Quote:

Ha, ha, good point! However, I would believe that is known (it is not high level gravitational physics, after all! ;) I am assuming that the project scientists have a reason for being so cautious.....

Although it would be nice, imho, if the project scientists would share their thoughts about this with us, if for no other reason than that we would stop bugging them about it! (OK, maybe not completely stop bugging them, but maybe reduce the volume of bugging! ;)


This is an old issue, settled I think in early-mid '06 ( ? ). E@H is the data analysis pipeline for an international scientific collaboration. This entails considerations beyond mere statements of program output equivalence. Perhaps you could best understand it as a highly conservative approach in an unprecedented area of science investigation. Being right slowly is far more important than being wrong quickly. Given the very large body of investigators who are entrusting their data to us, this binds the pace of such things. It can seem unfair that one's enthusiasm is being 'throttled back', particularly with the substantial material contributions collectively being donated by participants. But there is, and has to be, a process of verification of new components with extant ones. This is largely invisible to us and for software this is quite non-trivial. Plus the scientists are incredibly busy. Just ask Bernd how many 'spare' moments he has. Was it three? Or four? :-)

I continue to be amazed [ what am I, a sheep? :-) ] that E@H allows us to really advance such cutting edge science. I liken it to standing beside the likes of Einstein with a pencil & paper jotting down notes, making calculations, and running out for coffee & donuts! While we lack a confirmed wave detection to date, E@H has already made significant contributions to astronomical thought, see here for instance.

Cheers, Mike.

I have made this letter longer than usual because I lack the time to make it shorter. Blaise Pascal

rbpeake
rbpeake
Joined: 18 Jan 05
Posts: 247
Credit: 219,266,664
RAC: 0

RE: ..E@H has already made

Message 76766 in response to message 76765

Quote:

..E@H has already made significant contributions to astronomical thought, see here for instance.

Cheers, Mike.


Thanks for the link, that is very exciting news indeed! Nice to realize we already have an observatory that is capable of making meaningful astronomical contributions! :)

DanNeely
DanNeely
Joined: 4 Sep 05
Posts: 1,314
Credit: 1,750,616,091
RAC: 924,184

RE: Ha, ha, good point!

Message 76767 in response to message 76763

Quote:


Ha, ha, good point! However, I would believe that is known (it is not high level gravitational physics, after all! ;) I am assuming that the project scientists have a reason for being so cautious.....

For some unknown reason the SSE enabled code that's working in the *nix apps is blowing up in win32. ATM the rare crashes have a higher priority.

Depending on how radical a rework is it, the optimization plan once the base app is declared bug free is just to prove akos's apps are functionally identical and to just drop the new hot loops in.

rbpeake
rbpeake
Joined: 18 Jan 05
Posts: 247
Credit: 219,266,664
RAC: 0

RE: ATM the rare crashes

Message 76768 in response to message 76767

Quote:

ATM the rare crashes have a higher priority.

Depending on how radical a rework is it, the optimization plan once the base app is declared bug free is just to prove akos's apps are functionally identical and to just drop the new hot loops in.


Thanks, Dan!

Comment viewing options

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