• Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
    Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.

L5639 a0

Patriot

[H]ard|DCer of the Month - March 2011/June 2013/De
Joined
Dec 15, 2010
Messages
2,502
hey guys I have 2 A0 samples of the L5639 2.13ghz hexacore

They fold smp just fine... They had not run linux bigadv stably but then again... linux bigadv was canceled because of that...

I am thinking about doubling them up in a server and running windows bigadv... anyone think it will work?
 
I don't see why it wouldn't work. Of course, you could always send one of those chips over here and I could test it for you ;).
 
It will work just fine. My single L5639 rocks on -bigadv in windows.

Buy an SR-2 to put those puppies in and OC.

Fold On!
 
I don't see why it wouldn't work. Of course, you could always send one of those chips over here and I could test it for you ;).

I did that for jebo... lol... I said merry Christmas when he asked when I wanted it back... he was a tad happy...

I think it was a b1 and started at 2.66 but you would have to ask him...
 
It will work just fine. My single L5639 rocks on -bigadv in windows.

Buy an SR-2 to put those puppies in and OC.

Fold On!

I much prefer to borrow power... and sr2s are hard to come by...
 
Can't you just run bigadv in wine? Alot cheaper than getting win server, and theyre about done on fixing the linux issue with bigadv anyway.
 
It will work fine under Windows. I ran 2 x L5640s stock and they were goof for a very consitant 73K ppd. If those A0s have the 2X turbo, you should see the same. Even if they don't, they are plenty fast for bigadv.
 
Can't you just run bigadv in wine? Alot cheaper than getting win server, and theyre about done on fixing the linux issue with bigadv anyway.

Or this - it does work.
 
Can't you just run bigadv in wine? Alot cheaper than getting win server, and theyre about done on fixing the linux issue with bigadv anyway.

lol... there is no cost involved... for I bring these not home...
I have not had luck getting wine to run on rhel either...

It will work fine under Windows. I ran 2 x L5640s stock and they were goof for a very consitant 73K ppd. If those A0s have the 2X turbo, you should see the same. Even if they don't, they are plenty fast for bigadv.

that much really... I have only been getting 48k out of two 2.8ghz quads... running bigadv... perhaps it is windows 8 that holds it back...
Might try 2k8r2 after the next bigadv finishes...
And no I did not mistype :)
 
Patriot gets all the cool toys. It's a fact of life....at least he shares :)

Bigadv loves cores. Stock hex cores beat overclocked quads.
 
teach me the ways of getting 2k8 !?

lol...not 2k8 client 8 :)

the backgound is black and in like 30font says "shhh... lets not leak our hard work"

Patriot gets all the cool toys. It's a fact of life....at least he shares :)

Bigadv loves cores. Stock hex cores beat overclocked quads.

Sweetness...
 
I have only been getting 48k out of two 2.8ghz quads... running bigadv... perhaps it is windows 8 that holds it back...
Might try 2k8r2 after the next bigadv finishes...
Are you talking about two Nehalem quads with Hyperthreading? If that's the case, then 24k PPD each sounds perfectly fine for -bigadv.
 
lol oops um I am getting like 40k regular a3 smp out of dual nehalem quads with ht...
those are the 2.8s

getting 48k ppd out of dual 2.93 nehalem quads running bigadv...

are the 32nm that much faster clock/clock ? or does bigadv like cores that much...
 
-bigadv likes extra cores, and the Sandy Bridge CPUs are much faster clock for clock compared to Nehalem CPUs. 32nm Nehalem chips are pretty much the same as the 45nm models clock-for-clock. A Nehalem quad with HT running at 4GHz gets about 32k PPD with nothing else running.
 
Dual X5550 (2.93) does 58K ppd. The same board with dual L5640s (2.4) did 73K ppd. So yes, bigadv likes the extra cores.
 
bigadv loves physical cores. so the more physical cores you can throw at it the better the performance. take a look at 10e's quad socket 1.9ghz 12 core magny cour's setup (48 cores). just beats out MIBW's 4.3Ghz dual hex core (24 threads).
 
I have 3 L5639's (Rev B1, 2.13GHz) currently folding.

Two of them are in an SR-2 (which is back in stock at Newegg) @ 3.6GHz (200x18). They are currently getting 134,505 PPD on a P6900.
The other is in an Asus P6X58D @ 3.9 GHz (216x18) getting 49,883 PPD also on a P6900.

My point is, if you don't want to get an SR-2, you may be better off splitting them up and overclocking them in an X58 board rather than putting them in a dually that does not overclock very well.
 
im just afraid that a0s wont overclock that well...

well... I think I need to get my p6tse rmad and find out lol...

I have them up and running now... in a server (so stock of course)
they grabbed a 2685 wu ... dunno if that one is bad for ppd... but they are getting 23min23s a frame or 53k ppd...
the other module of the unit has dual 2.26ghz quads... have a 6900wu 35.8k 30min 53s frame

just fired these up...so don't know if the ppd is stable or not...
Posted via Mobile Device
 
Those numbers sound about right to me. Of course, if you split all of those CPUs into separate machines and overclocked them, you could get a lot more PPD.
 
I am sure of that...also more much more expensive to setup than what I have now.... though I might consider taking the hexa cores out for a spin...

we shall see... need to make sure they are stable folding before I go through the trouble...
 
well... as soon as I passed jebo_4jc up on ocw he went silent... built up an arsenal and has been folding here... soo... now I must surpass him again :p
 
well... as soon as I passed jebo_4jc up on ocw he went silent... built up an arsenal and has been folding here... soo... now I must surpass him again :p

Hey Jebo, ever get the feeling your being followed? ;)
 
Hey Jebo, ever get the feeling your being followed? ;)
more like, I can hear the mower starting up behind me. Patriot was hitting 200k-300k ppd for a while, when I thought I was big stuff with around 100k. I knew I didn't stand a chance back then, so I rejoined team 33 :)
 
more like, I can hear the mower starting up behind me. Patriot was hitting 200k-300k ppd for a while, when I thought I was big stuff with around 100k. I knew I didn't stand a chance back then, so I rejoined team 33 :)

Yeah then they nuked bigadv on linux... really pissed me off lol...
Thats pretty much all I run at work... primarily fold on rhel5u5

soo... I have been scrap pile digging and assembling an arsenal to run jebo down :)

the protos we dont need for testing anymore are getting combined with the spoils of scrap... to make uber folders...

I need to start putting ubuntu or something on them or figure out the dependencies to get wine working on rhel...
currently I have 3 running bigadv on windows um 8 shhh...
(2x 2.16hexa, 2x 2.26 quads, 2x 2.93quads) I think they have 8-12gb each... what can I say... I know where to dig...

I have been tempted to grab a 2p g34board off newegg and see how those 1.7ghz 12cores I found a few weeks back overclock...
currently those guys are folding regular a3s on linux... pathetic ppd...
and have dual .2.8 i7 also just reg smp...

of note... I don't think its worth it splitting up the a0 hexacores... I have found an a0 2.8ghz before as well as a b0... the b0 is folding for jebo... and the a0 is with someone else and doesnt oc more than a turbo setting...
 
ouch that OC is painful.

My understanding is that overclocking G34 is painful. Ask 10e here about his experience. I assume you have ES chips right? There's some sort of software overclocking method that I wouldn't call straightforward.
 
Back
Top