SPCR Folds Team Blog
Moderators: NeilBlanchard, Ralf Hutter, sthayashi, Lawrence Lee
-
- *Lifetime Patron*
- Posts: 4284
- Joined: Fri Apr 04, 2003 6:21 pm
- Location: Undisclosed but sober in US
Sorry, too many times I click on "quote" instead of "edit".
Anyway, my last utility bill for March and April, man, it was bad. It was twice what I expected. I haven't done this in a while, but I need to look at exactly what I was charged versus what I used. Yes it was cold, but it really sucked.
I feel sorry for those locally on oil, whose price is swinging way too fast to have it be used as a heating source.
At some point:
1. I will either shut down a folder OR
2. Purchase a 45nm quad and then shut down or slow some other PCs. The goal is less energy usage as opposed to more points.
I'll be sealing up every nook and cranny I can find in my home. I've been saying that for years but this time I mean it. Yes, many years time off is a choice between going on vacation or doing stuff at home, and I kept going on vacation. Well, time to pay the piper, because this energy crap is getting out of hand. I refuse to sit idly by and watch it become an ever growing piece of the budget. I'll go down kicking and screaming, and doing what I can.
Anyway, my last utility bill for March and April, man, it was bad. It was twice what I expected. I haven't done this in a while, but I need to look at exactly what I was charged versus what I used. Yes it was cold, but it really sucked.
I feel sorry for those locally on oil, whose price is swinging way too fast to have it be used as a heating source.
At some point:
1. I will either shut down a folder OR
2. Purchase a 45nm quad and then shut down or slow some other PCs. The goal is less energy usage as opposed to more points.
I'll be sealing up every nook and cranny I can find in my home. I've been saying that for years but this time I mean it. Yes, many years time off is a choice between going on vacation or doing stuff at home, and I kept going on vacation. Well, time to pay the piper, because this energy crap is getting out of hand. I refuse to sit idly by and watch it become an ever growing piece of the budget. I'll go down kicking and screaming, and doing what I can.
-
- Moderator
- Posts: 7681
- Joined: Mon Dec 09, 2002 7:11 pm
- Location: Maynard, MA, Eaarth
- Contact:
Hi,
I've settled for running two instances of the regular client while I'm in WinXP. Does anybody have any thoughts/guesses/hints on getting F@H SMP to work in Windows?
I added an E8400 to the Fold, btw. So, I now have four SMP machines (one Mac Pro, one E6750 @ 3.0gHz, one E8400, and my home Athlon 64 X2 4200+ machine when I'm booted to Ubuntu.
I managed to get the SMP client working in Ubuntu 8.04 -- I still having trouble getting the Windows SMP client working on this Athlon 64 X2 4200+ machine! I have uninstalled it, and I tried the newer beta. I also have WinXP x64 on this machine (a "triple" boot: either Ubuntu, or WinXP, or WinXP x64), and it "refuses" to work there either. All attempts in Windows, to no avail.VanWaGuy wrote:Neil, something just reminded me about a problem I saw regarding folding not starting. Something about the network config needing to be changed.
I do not remember all the details, but sometimes I had to go into system ->administration -> network then click hosts tab (other ways to do this too)
then, make sure there is a line 127.0.0.1 and your machine name,
and 127.0.1.1 and machinename.networkname.
I think sometimes I got 127.0.0.1 to be network name. Seems like I remember that messes up the MPI if it is not configured as above.
(Also, sounds like you got past needing to install the 32-bit libraries.)
I've settled for running two instances of the regular client while I'm in WinXP. Does anybody have any thoughts/guesses/hints on getting F@H SMP to work in Windows?
I added an E8400 to the Fold, btw. So, I now have four SMP machines (one Mac Pro, one E6750 @ 3.0gHz, one E8400, and my home Athlon 64 X2 4200+ machine when I'm booted to Ubuntu.
-
- Moderator
- Posts: 7681
- Joined: Mon Dec 09, 2002 7:11 pm
- Location: Maynard, MA, Eaarth
- Contact:
Windows firewall?
Neil wrote:
You can check this by looking at the properties of the network adapter. I think they are under the Advanced button of the Security tab; or vice versa.
HTH
Mike
Have you looked to see whether Windows is configured to allow F@H to communicate with the Internet? At some point in XP M$Soft switched the security defaults to assume programs should not be allowed to communicate unless authorised. I have seen cases where the users accidentally said "No".It definitely has an Ethernet connection -- and I should say that I had the SMP client running on this machine in the past, for quite a while (about a year?). I don't know what happened
You can check this by looking at the properties of the network adapter. I think they are under the Advanced button of the Security tab; or vice versa.
HTH
Mike
-
- Moderator
- Posts: 7681
- Joined: Mon Dec 09, 2002 7:11 pm
- Location: Maynard, MA, Eaarth
- Contact:
U/Volt?
Aristide
We too are having a short period of warm weather (the Dutch call it "summer"), and the peak temps for my E6750 reached 52 centigrade. Despite the fact that I blagged my way out of having any science lessons at school, I do know that heat = fan speed = noise, so I decided to take a look at undervolting. Just taking the voltage control of my Gigabyte P35-DS4 off automatic, and reducing the default setting by 0.0625v, I got a Vcore reduction of about 0.13 volts, temp reduction of 8-10 centrigrade, and a concomitant reduction of around 175 to 250 rpm. It did seem to increase the time for a WU - perhaps around 40 minutes to one hour out of 19.5. It is hard to tell, because not all 2605s are exactly the same, and because I use this system for other work as well.
In any case, I decided to reduce the voltage by another 0.125v, and increase the FSB from 400 to 414, which gets me back to about the same WU time, and my temps and speed are still quite a lot lower than they were - and I think it was hotter yesterday than two weeks ago when I started this experiment.
Of course, it is entirely possible that you have already reduced your voltage, and that I am teaching my grandmother to suck eggs. Just in case, I have posted a graph of the results on my website.
HTH
Mike
We too are having a short period of warm weather (the Dutch call it "summer"), and the peak temps for my E6750 reached 52 centigrade. Despite the fact that I blagged my way out of having any science lessons at school, I do know that heat = fan speed = noise, so I decided to take a look at undervolting. Just taking the voltage control of my Gigabyte P35-DS4 off automatic, and reducing the default setting by 0.0625v, I got a Vcore reduction of about 0.13 volts, temp reduction of 8-10 centrigrade, and a concomitant reduction of around 175 to 250 rpm. It did seem to increase the time for a WU - perhaps around 40 minutes to one hour out of 19.5. It is hard to tell, because not all 2605s are exactly the same, and because I use this system for other work as well.
In any case, I decided to reduce the voltage by another 0.125v, and increase the FSB from 400 to 414, which gets me back to about the same WU time, and my temps and speed are still quite a lot lower than they were - and I think it was hotter yesterday than two weeks ago when I started this experiment.
Of course, it is entirely possible that you have already reduced your voltage, and that I am teaching my grandmother to suck eggs. Just in case, I have posted a graph of the results on my website.
HTH
Mike
-
- *Lifetime Patron*
- Posts: 4284
- Joined: Fri Apr 04, 2003 6:21 pm
- Location: Undisclosed but sober in US
First - Iganu came to a full stop the week of May 18th
http://folding.extremeoverclocking.com/ ... =&u=267697
OK, I was in the OC camp, until Iganu taught me otherwise. My E6400 at 3.2GHz was in the 60C range all winter, and it was cold in here. So now its like 20C hotter in here, where was the CPU suppose to go?
I found that dropping from a 400MHz FSB to 333 has allowed me to continue to drop my volts. I was 1.475, now I am at 1.3625, and under full load it runs at 1.325. I will drop another .025 (1.30 under load) and see how it looks. Basically for every step down (.0125 volts) I usually see a 1C degree drop. And the heat wave here ended last night. Outside temp was 32C, then a thunderstorm, and then temps were 19C, all in a couple of hours. At that point I restarted the Intel and it's at 65/62C with a 25% OC.
(There's still plenty of hot air in the OT threads though. )
I have no idea if the current WU will be done in time.
So now I want to see how much my electrical usage will drop as well. I slowed down my AMD Black Edition a while back, and reduced electrical usage almost 40 watts on a 24/7 basis. I'd like to know your wattage reduction as well.
I appreciate your comments.
Thank you
http://folding.extremeoverclocking.com/ ... =&u=267697
OK, I was in the OC camp, until Iganu taught me otherwise. My E6400 at 3.2GHz was in the 60C range all winter, and it was cold in here. So now its like 20C hotter in here, where was the CPU suppose to go?
I found that dropping from a 400MHz FSB to 333 has allowed me to continue to drop my volts. I was 1.475, now I am at 1.3625, and under full load it runs at 1.325. I will drop another .025 (1.30 under load) and see how it looks. Basically for every step down (.0125 volts) I usually see a 1C degree drop. And the heat wave here ended last night. Outside temp was 32C, then a thunderstorm, and then temps were 19C, all in a couple of hours. At that point I restarted the Intel and it's at 65/62C with a 25% OC.
(There's still plenty of hot air in the OT threads though. )
I have no idea if the current WU will be done in time.
So now I want to see how much my electrical usage will drop as well. I slowed down my AMD Black Edition a while back, and reduced electrical usage almost 40 watts on a 24/7 basis. I'd like to know your wattage reduction as well.
I appreciate your comments.
Thank you
No killawatt = No statistics
Aristide
I haven't been able to find a source for a killawatt here in The Netherlands, so I can't answer ATM. Perhaps one of the native Dutch speakers who are following this thread could point me in the right direction?
But 62c still sounds high ... is that with a stock cooler?
Mike
I haven't been able to find a source for a killawatt here in The Netherlands, so I can't answer ATM. Perhaps one of the native Dutch speakers who are following this thread could point me in the right direction?
But 62c still sounds high ... is that with a stock cooler?
Mike
-
- Patron of SPCR
- Posts: 749
- Joined: Sat Nov 11, 2006 6:02 pm
I found this one: EnergyChecker (further below there are more products).
58 °C does seem high... With the hot weather here last week, my CPU reached 40 °C maximum.
58 °C does seem high... With the hot weather here last week, my CPU reached 40 °C maximum.
-
- SPCR Reviewer
- Posts: 561
- Joined: Tue May 30, 2006 8:22 pm
- Location: Vancouver, BC
-
- *Lifetime Patron*
- Posts: 4284
- Joined: Fri Apr 04, 2003 6:21 pm
- Location: Undisclosed but sober in US
I recall the Q6600, everybody wanted the G0 stepping because it ran cooler and OC'd better.
My processor is much older.
I'm considering a Q6700 for my next folder.
Does anybody have temps from an undervoltage or stock voltage 45nm folder?
I have the voltage down to 1.35 volts now, temps 59/61C and it's fairly warm up here. CPU-Z says around 1.29 volts under load.
My processor is much older.
I'm considering a Q6700 for my next folder.
Does anybody have temps from an undervoltage or stock voltage 45nm folder?
I have the voltage down to 1.35 volts now, temps 59/61C and it's fairly warm up here. CPU-Z says around 1.29 volts under load.
-
- Patron of SPCR
- Posts: 749
- Joined: Sat Nov 11, 2006 6:02 pm
I have recently got a Q9300 (45nm 6MB, 2.5GHz Quad), and it's running stock speed (I have no luck overclocking with the Gigabyte P31-DS3L), but at 1.15V (I think) rather than 1.25V stock. Temps are around 58 to 63°C for each core, measured with CoreTemp. I'm running it with 2 SMP clients and the affinity changer as well giving ~2500 PPD on WinXP Pro 32-bit. I'm slowly reducing the voltage to lower power consumption.
Temps are in my Antec Solo with standard 120mm exhaust on low, near the fanless Ninja (Rev A) and an older Seasonic PSU.
Temps are in my Antec Solo with standard 120mm exhaust on low, near the fanless Ninja (Rev A) and an older Seasonic PSU.
-
- *Lifetime Patron*
- Posts: 4284
- Joined: Fri Apr 04, 2003 6:21 pm
- Location: Undisclosed but sober in US
http://www.msnbc.msn.com/id/25317077/
Interesting story.
Well, for all my troubles. 400FSB down to 333 and 1.475Volts on the CPU down to 1.3 volts. Reduction in energy used? 150 to 130 watts. Probably a fair amount of watts used for my video card, but I still expected better than this.
Interesting story.
Well, for all my troubles. 400FSB down to 333 and 1.475Volts on the CPU down to 1.3 volts. Reduction in energy used? 150 to 130 watts. Probably a fair amount of watts used for my video card, but I still expected better than this.
interesting replies.
I am on a 90nm prescott, now at 13500hrs, also doing smp within tight time frame, (but mostly vid card now)on 3 cooling fans and using less than any pc I have owned.
I am not discouraged in relation to power bills and donate for 12hours or less daily.
Upon the past, I found fans have a means to do something HUGE with thier yoyos and a relation to hertz reaching out to null through them electrically.
I would reduce fans, make cooling intelligent, and maintain same power/speed of chips for less consumption.
I recently found a power suppply on earth called a silverstone fanless with a 300w and 450. My machine can still run a 300w. Sounds temping, with my reverse cooling...down to 2 fans for a very hot cpu on a 12x9.6 board, oc'd to 2995mhz and an 856fsb.
I am on a 90nm prescott, now at 13500hrs, also doing smp within tight time frame, (but mostly vid card now)on 3 cooling fans and using less than any pc I have owned.
I am not discouraged in relation to power bills and donate for 12hours or less daily.
Upon the past, I found fans have a means to do something HUGE with thier yoyos and a relation to hertz reaching out to null through them electrically.
I would reduce fans, make cooling intelligent, and maintain same power/speed of chips for less consumption.
I recently found a power suppply on earth called a silverstone fanless with a 300w and 450. My machine can still run a 300w. Sounds temping, with my reverse cooling...down to 2 fans for a very hot cpu on a 12x9.6 board, oc'd to 2995mhz and an 856fsb.
-
- Friend of SPCR
- Posts: 1439
- Joined: Tue Dec 14, 2004 4:06 pm
- Location: New Hampshire, US
- Contact:
Wow! Did I miss something????? 32nd place? In one month we've gone from 100,000 points daily to 70,000 ...in one month!? What gives? Should I abandon ship?
edit: I guess that was hasty...now that I think about it, there isn't really any other community I'd rather fold for/with. Overclock.net is the next place I use for info/news, but never for the level of advice that I find here.
edit: I guess that was hasty...now that I think about it, there isn't really any other community I'd rather fold for/with. Overclock.net is the next place I use for info/news, but never for the level of advice that I find here.
Last edited by ryboto on Wed Jul 23, 2008 7:40 pm, edited 1 time in total.
-
- Friend of SPCR
- Posts: 1439
- Joined: Tue Dec 14, 2004 4:06 pm
- Location: New Hampshire, US
- Contact:
That's unfortunate. Even if I didn't have AC, I'd be folding. Not on my video card, as it really would overheat, but I'd fold on the CPU. We're nosediving pretty hard though, unless we get some surge of users, I don't see the team returning to the former glory of 20th position.Plissken wrote:All of these silent folders get too hot in the summer?
That's why I stopped, but will resume in a month or so.