BOINC Benchmarks

[Insidious]

Originally posted by: mrwizer

Originally posted by: Insidious

In MY OPINION (yes, that is only my opinion... I do not claim to be omnipotent.. or even correct) It isn't cheating to use the 5.3.1 boinc client and I will respect no Rosey any more or any less as a result of their personal choice in this matter.

-Sid

Is it just the standard build client, or an optimized build that has different files for P4's, AMD's, etc.? And 5.3, is that a development version?

a standard boinc client has been optimized for different instruction sets used by modern PCs. (SSE, SSE2, etc...) and so there are more than one 5.3.1 file. You only use the one that applies to your computer.

I don't know any more about it than is written at the web site from which it originates. I do know that the Rosetta client and it's processing of work units is absolutely UNaffected by these optimizations. They only change the behavior of the boinc infrastructure. These clients in no way Jeopardize the science of Rosetta.

 

[Rattledagger]

Optimized clients and applications has different builds for different cpu-type, while standard uses the same.


There will ocassionally be 2 users with artificially inflated benchmark crunching the same wu, but as long as majority of users isn't doing this these spikes in granted credit will be averaged-away. Also, if someone is having much too high claimed credits, it's easily spotted...

 

[Insidious]

Originally posted by: Rattledagger
.... Also, if someone is having much too high claimed credits, it's easily spotted...

If they cared, wouldn't they just look at the client version which is reported with every completed work unit? No need for detective work and point comparisons.

 

[Rattledagger]

Originally posted by: Insidious
It would be so incredibly simiple for BOINC to disallow any version of the BOINC client they chose to. (EVERY result you send in also indicates what client version was used.)
I have made a personall choice to assume if they are going to allow it, then how silly would I be to intentionally crunch for less credit than was available.

Since BOINC is open-source anyone compiling it can also set whatever version-number they wants, so this will not work...


As for the optimized v5.3.1, if not mis-remembering this client is older than v5.2.2, so it's really time to upgrade...
Both v5.2.13 and v5.3.3 comes in variants with artificially inflated benchmark-score, so this shouldn't be a reason for not upgrading.

 

[Insidious]

I didn't know there was a 5.3.3...

/me heads off to find it now

Thanks RD! :beer:

edit: ah yes, the open source debacle.... it does have an opportunity for synergy and it is quite fasionable, but boy oh boy does it create difficulties.

I will be very happy when a system is in place that makes all of these questions moot.

 

[Freewolf]

Two post under the other one I posted the link to ( Which is sticky to the top of their forum hiding in plain view) is this one from Doctor Baker thanking them. I pretty sure he isn't blind and is as able to see the sticky Optimized Clients post as I am and yet he said nothin about it.
So I'ld say it doesn't seem to brother him any.


http://www.free-dc.org/forum/showthread.php?t=10420

 

[mrwizer]

Originally posted by: Freewolf
Two post under the other one I posted the link to ( Which is sticky to the top of their forum hiding in plain view) is this one from Doctor Baker thanking them. I pretty sure he isn't blind and is as able to see the sticky Optimized Clients post as I am and yet he said nothin about it.
So I'ld say it doesn't seem to brother him any.


http://www.free-dc.org/forum/showthread.php?t=10420

Not to get this thread off topic, but have you read the latest posts to that thread???? :Q

 

[Freewolf]

yeah just now and that's some crazy numbers Poorboy talks about if it's true. I mean from those number 1 guy with one machine could do as much as all of the RA with the 40 something machines we have crunching.

 

[Insidious]

I won't cry when they fix all of this.

Thanks Freewolf

-Sid

 

[rise]

Originally posted by: Freewolf
yeah just now and that's some crazy numbers Poorboy talks about if it's true. I mean from those number 1 guy with one machine could do as much as all of the RA with the 40 something machines we have crunching.

:Q wow thats nuts. that was posted after i read the thread originally.

 

[Freewolf]

Looking at the top hosts I don't see the kind of numbers PoorBoy was throwing about

http://www.boincsynergy.com/stats/hosts.php?project=rah

 

[Freewolf]

bty if anyone wants to see what it takes for the Rebel Alliance to produce the numbers we are just look here

http://www.boincsynergy.com/stats/boinc-stats.php?id=9030&project=rah

 

[Insidious]

Originally posted by: Freewolf
bty if anyone wants to see what it takes for the Rebel Alliance to produce the numbers we are just look here

http://www.boincsynergy.com/stats/boinc-stats.php?id=9030&project=rah

That's a LOT of 'pooters :shocked:

did you see that one that is in the top 100 worldwide for RAC...

 

[mrwizer]

PoorBoy? I thought it was laurenU2...

 

[Insidious]

LOL

That's ME! (and only 1/2 an X2 at that)

LOL

 

[rise]

Originally posted by: Freewolf
Looking at the top hosts I don't see the kind of numbers PoorBoy was throwing about

http://www.boincsynergy.com/stats/hosts.php?project=rah

i wonder if what he was seeing was merged results, like it looks like this guy maybe merged some comps/results??

i've been running the optimized as well but i don't see anything like that. i'll have to go back over my early days when we ran it with TAS and see what my average was back then.