Tobit Wants YOU

Tobit said:
Thanks dmolter! There will be some love coming to uniprocessor clients soon in the form of the A4 core. I can't comment much at this time but A4 is the uniprocessor version of the A3 core.
Click to expand...
I'm hoping for some multi-core support but I have a feeling it's going to be a bonus credit.
 
so i loaded up a couple uniprocessor clients on my laptop's 2.5ghz c2d.

I got an 11284 which is apparently only worth 145 points....about 300 ppd

I also got an 11278 which is worth 209 points.....about 231 ppd

edit: compare 531 ppd to the roughly 1200 ppd I got with the smp client and you can see why nobody folds these WUs, lol
 
Yeah and single core projects keep coming as well despite older systems becoming less and less attractive to run for a number of reasons.

Thanks though man.
 
Out of curiosity, I can't seem to get the classic uniprocessor client working on my laptop anymore. I fire it up and it keeps giving me an error and is unable to download the b4 Protomol core. I havn't folded on my laptop in a very long time, but thought I would help out with the overflow of classic work units since these have a very long deadline and I can just fold them whenever my laptop happens to be powered on.

Would I be able to run the uniprocessor work units using the SMP client by omitting the SMP flag by any chance? I don't know if the SMP client can be used that way.
 
Last edited:
amdgamer said:
Would I be able to run the uniprocessor work units using the SMP client by omitting the SMP flag by any chance? I don't know if the SMP client can be used that way.
Click to expand...
Definitely. The most recent SMP clients are considered a "unified" client in that they can run both uniprocessor and SMP work. Just leave off the -smp flag.
 
Tobit said:
Definitely. The most recent SMP clients are considered a "unified" client in that they can run both uniprocessor and SMP work. Just leave off the -smp flag.
Click to expand...

I was confused since the SMP setup and installation guide over at Stanford said to revert back to the classic client or something like that if you cannot meet SMP deadlines. I have it running right now, and boy is it slow. Surprisingly, I got a 10055 which has a deadline of like 4.59 days which is a lot shorter of a deadline than I thought. I only intend to fold these classic work units when my laptop is up and running since I don't want to kill my lappy.
 
I'm running the classic client 24/7 on 4 systems right now, old machines that won't run the newer stuff anyway. I was doing it because it was all I had to work with - didn't know I was doing a "good thing".
 
Well you can count me in for a couple of clients - at least for a couple of days. my e6600 doesn't seem to be making the bonuses so it can run 2 classics until it is decomissioned
 
jebo_4jc said:
overclock that E6600!
Click to expand...

Its stuck on a crappy 680i M/B that is on its last legs that i want to sell on so i'd rather not, As i said its being decomissioned in the next few days anyway so i would rather devote the time to getting more out my x6:D
 
Nathan_P said:
Well you can count me in for a couple of clients - at least for a couple of days. my e6600 doesn't seem to be making the bonuses so it can run 2 classics until it is decomissioned
Click to expand...

Hm...I suppose that means I should go check if my e6400 is making deadlines. I might be in for 2 as well.
 
I have my 2.5ghz T9300 core 2 duo laptop folding uniprocessor units

It would make SMP deadlines if I let it run 24/7 but obviously it's a laptop, soo.....

Actually I lied it's currently BOINCing but once the Christmas challenge is over it'll be back to uni CPU fah.
 
Some nice thread resurrection :).

I'm running two classic clients on my P8700 right now because I wouldn't consistently make SMP preferred deadline anyways because it's not a dedicated folder.
 
I'll have to fire up the unicore client on my Latitude XT, the ULV 1.2ghz conroe based C2D doesn't stand a chance of completing SMP WU's anyway.
 
Since starting folding years ago, it's been classic client (and about 1 year of PS3 client), and 1 SMP client. Lots of WUs, but not so hot on points.
 
Celerator said:
Since starting folding years ago, it's been classic client (and about 1 year of PS3 client), and 1 SMP client. Lots of WUs, but not so hot on points.
Click to expand...
Thanks Celerator! :cool:
 
I make it my point to at least squish a few uniprocessor units with a few different rigs. I have my Celeron laptop (1.3Ghz Celeron 743 ULV. I can get a little more than 1.3 from it.) and a P3 that ends up more often than not with my feet on the case (Desktop type, gets warm enough to be comfortable for footwarming during folding) both working on uniproc. If I ever get a new 775-based board, I'll drop an old E6600 in (slightly damaged, won't run at 266 and requires underclocking to 250 before it'll boot Windows) and some borrowed DDR2. Boot and run Ubuntu and fold 2 more uniproc.

While we may be of different teams, our competition means science wins.

Now, if you guys don't mind, I'm gonna go back to my 6701. I hate to say it, but I kinda like them.
 
Celerator said:
Since starting folding years ago, it's been classic client (and about 1 year of PS3 client), and 1 SMP client. Lots of WUs, but not so hot on points.
Click to expand...
That's the old school way to do it! Just throw a bunch of CPUs at the job and let them crunch. Nice work by the old vet.


Madrias said:
I make it my point to at least squish a few uniprocessor units with a few different rigs. I have my Celeron laptop (1.3Ghz Celeron 743 ULV. I can get a little more than 1.3 from it.) and a P3 that ends up more often than not with my feet on the case (Desktop type, gets warm enough to be comfortable for footwarming during folding) both working on uniproc. If I ever get a new 775-based board, I'll drop an old E6600 in (slightly damaged, won't run at 266 and requires underclocking to 250 before it'll boot Windows) and some borrowed DDR2. Boot and run Ubuntu and fold 2 more uniproc.

While we may be of different teams, our competition means science wins.

Now, if you guys don't mind, I'm gonna go back to my 6701. I hate to say it, but I kinda like them.
Click to expand...
Get outta here (flogs Madrias mercilessly) ;)
Are you saying you can overclock your laptop? How are you doing that? Clockgen?
 
jebo_4jc said:
That's the old school way to do it! Just throw a bunch of CPUs at the job and let them crunch. Nice work by the old vet.



Get outta here (flogs Madrias mercilessly) ;)
Are you saying you can overclock your laptop? How are you doing that? Clockgen?
Click to expand...

Friend of a friend had a non-standard (thereby warranty-killing) BIOS for the Acer Aspire 1410. I took the risk and loaded it, and can push it a little. My highest 'stable' OC is about 1.6 out of a 1.3 chip. Any more and I'm making the chip run way too hot. I like my laptops to stay under 50C. Usually I leave it stock, though.

Anyway, there's just something to be said about underclocking it a little (1.2 or 1.1) and running for hours on end while folding on battery. I can squeeze 4 hours running 1.3.

Granted, enough said on this silly notebook of mine. The big one is my Pentium 4 Northwood laptop. 3.0, Hyperthreaded, 17 inch screen, and literally 1 hour of battery life. I like my old computers. I run one uniproc. client on it, as otherwise the chip gets nasty hot. It idles around 50, folds one 'core' at about 80-90, and the one time I did a -oneunit for the second core, I shot temps up to 85+ for 3 stinking days. Had it sitting on a box fan to keep it running.
 
Madrias said:
Friend of a friend had a non-standard (thereby warranty-killing) BIOS for the Acer Aspire 1410. I took the risk and loaded it, and can push it a little. My highest 'stable' OC is about 1.6 out of a 1.3 chip. Any more and I'm making the chip run way too hot. I like my laptops to stay under 50C. Usually I leave it stock, though.

Anyway, there's just something to be said about underclocking it a little (1.2 or 1.1) and running for hours on end while folding on battery. I can squeeze 4 hours running 1.3.

Granted, enough said on this silly notebook of mine. The big one is my Pentium 4 Northwood laptop. 3.0, Hyperthreaded, 17 inch screen, and literally 1 hour of battery life. I like my old computers. I run one uniproc. client on it, as otherwise the chip gets nasty hot. It idles around 50, folds one 'core' at about 80-90, and the one time I did a -oneunit for the second core, I shot temps up to 85+ for 3 stinking days. Had it sitting on a box fan to keep it running.
Click to expand...

If you feel like tinkering sometime, try reapplying the TIM with some IC Diamond 7.

From working on my GPU's, it seems like AS5 really starts to lose effectiveness over 60-70C whereas IC Diamond can take any heat you throw at it. Now if you have never reapplied the TIM on these laptops, AS5 will be loads better and IC diamond better yet. I bet you could take 10C off.
 
R-Type said:
If you feel like tinkering sometime, try reapplying the TIM with some IC Diamond 7.

From working on my GPU's, it seems like AS5 really starts to lose effectiveness over 60-70C whereas IC Diamond can take any heat you throw at it. Now if you have never reapplied the TIM on these laptops, AS5 will be loads better and IC diamond better yet. I bet you could take 10C off.
Click to expand...

I don't have either of those, but I've got some OCZ stuff somewhere (can't remember what it's called, but it's pretty good, I think. Keeps an E8600 around 40C when folding flat out on stock air, stock speed). I've given the laptops 'names' to make referencing on forums easier, but I'm not using them here for now. I did reapply the TIM on the P4 laptop, as it wasn't involved to do (Gateway m675, it's one panel, a handful of screws, and the heatsink/fan array all comes up nicely), but it's just a bad design overall.

As for the Celeron, well, I don't feel like pulling it apart. Acer did build them to be near impossible to get into, and I don't feel like junking $450 worth of lightweight computing power on the off chance I pull 10C off the temps. Reducing 10C is as easy as setting it on a Delta-based fan cooler.
 
Tobit said:
As some of you know, I've always been a proponent of running as much variety of clients as you can to best benefit science. I was talking with sirmonkey this evening on our IRC channel and he seemed surprised to know that there is one work server at Stanford with over 2 million classic client work units available. That is just one server and there are several classic client servers!
Click to expand...

gee lets see whos fault that is... I maintained 1600 installs of the classic client many moons ago before the SMP client was worth a damn, for a whopping 166400 PPD (quite a feat in its day) then the GPU Client came along and people were getting the same PPD with 10 video cards... now I can get almost the same PPD with 1 SR-2 rig....

so which would you rather maintain, 800 computers each with 2 instances of the classic client?

or 1 SR-2?

they did this to themselves
 
You must log in or register to reply here.
Back
Top