Message boards :
Number crunching :
Q6600 Failure TEMP....
Message board moderation
Author | Message |
---|---|
Mike O Send message Joined: 1 Sep 07 Posts: 428 Credit: 6,670,998 RAC: 0 |
I just got a Q6600 system and its NOT doing so good. Running the 64bit seti 6.0 app, it crashed hard at temps of around 70c. I have a P4 that is over clocked 10 percent to 2.2ghz and its temps hit 75c and runs like a champ. I have a better fan for the Q coming to replace the Intel Heating fan. ;) That will drop it down to a max of 50c I hope. Anyway, I switched over the the AK_V8 64bit SSSES app and the thing has not stopped yet. (Knocks on wood) Im thinking of replacing the heatsink on the nvidia 680i chip.. but Im going to wait till the masscool deheater gets here for the Q6600. I got one heck of a deal on this system new on ebay $449.00 but it has a very cheap ECS mobo. Other than that, it has 4gig pc6400 ram, 250gig 7200rpm sata drive, 20x dvd burner and XP Pro 64bit. I priced out the componets at newegg and I can't buy the same things for $449. Anyway.. im sure this has been asked before.. It sounds like a standard problem but... what is concidered a safe max temp for a Q6600?? Not Ready Reading BRAIN. Abort/Retry/Fail? |
Jason A. Countryman Send message Joined: 29 Aug 03 Posts: 139 Credit: 50,172,873 RAC: 2 |
I am running mine at around 65 degrees. I think its a little high but not dangerously so. |
BMaytum Send message Joined: 3 Apr 99 Posts: 104 Credit: 4,382,041 RAC: 2 |
Mike O. From here Intel® Core™2 Quad Processor, you can download Intel's Quad Core Thermal and Mechanical Design Guidelines (PDF), which are a bit complicated but informative. See also this quick spec summary Intel® Core™2 Quad Processor Q6600 which lists Thermal Spec= 62.2C. You'll need to read the PDF to know what that means..... I also use Alex Kan's V8 seti client optimized for my C2D E8400 (Win32 SSE4.1). AKv8 does load both cores tp ~100%, and I see ~145F on days when its hot (>80F) in the house. AKv8 crunches workunits much faster than the standard v5.18 seti client (I haven't tried v6.x, since AKv8 is so good). Sabertooth Z77, i7-3770K@4.2GHz, GTX680, W8.1Pro x64 P5N32-E SLI, C2D E8400@3Ghz, GTX580, Win7SP1Pro x64 & PCLinuxOS2015 x64 |
archae86 Send message Joined: 31 Aug 99 Posts: 909 Credit: 1,582,816 RAC: 0 |
You have not mentioned what clock speed and voltage you are running at. Without that the temperature does not mean much regarding performance comparisons, though higher temperatures always raise the probability of hard failure sooner or later. |
john deneer Send message Joined: 16 Nov 06 Posts: 331 Credit: 20,996,606 RAC: 0 |
I just got a Q6600 system and its NOT doing so good. Hello Mike, In my experience a b0 stepping can be run quite comfortably when kept in the low 60's, whereas a g3 stepping can endure a little more heat and can be run without trouble in the high 60's. That's Centigrades, of course. You don't mention the stepping you have, so maybe you don't know. You can find out by using cpu-z, which will give you information on the exact type of processor you have in there. Regards, John. |
zoom3+1=4 Send message Joined: 30 Nov 03 Posts: 65763 Credit: 55,293,173 RAC: 49 |
I just got a Q6600 system and its NOT doing so good. I've heard of the B3 stepping and the G0 stepping, But the B0 and G3? When did those come out? As I have a B3 Q6600 cpu paired with a motherboard not designed for a quad, I'll be fixing that in December If possible(that and some more Patriot PC2-8500 Viper ram). Mine is forced to run @ 2.61GHz due to the motherboard and some ram problems too. The T1 Trust, PRR T1 Class 4-4-4-4 #5550, 1 of America's First HST's |
Mike O Send message Joined: 1 Sep 07 Posts: 428 Credit: 6,670,998 RAC: 0 |
thanks folks for the input.. OK.. here is the specs for the Quad.. clocked at 2.4ghz. stepping B revison G0 Vcore is at 1.28 800mhz ram fsb 1066 I moved the computer in to the basement where my farm used to be. Temps are now bouncing around 50c plus/minus 5. I also turned the 120mm fan around so it blows INTO the case and put 1/2 of a cd jewel case in front of it to divert the air flow towards the Quad. I think a lot of the issues im having have to do with the size of the case.. its VERY SMALL and the microATX puts the cpu LGA775 very close to the side of the case. Also, the power supplys voltages are not that great. the +5 buss is reading at 3.83 to 4.10.. the +12 is at 13.2. 3.3v is holding. I know that TTL chips wont run below 3.8 volts but I have NO CLUE about these VLSI bridge chips. This is a little off topic but.. almost all the WUs the quad is getting are processed in 30 to 40 minutes/core. Im getting maybe 8 to 12 credits for each. I averaged the numbers and its looking like.. If this stays the same.. the best I will get for an rac is around 2k/day. hardly what I was hoping this thing could do. I know its at stock freq but right now, im just trying to keep it going until the better fan/heatsink gets here. Next question.. what kinda RAC are you guys getting and at what clock speeds? Not Ready Reading BRAIN. Abort/Retry/Fail? |
dnolan Send message Joined: 30 Aug 01 Posts: 1228 Credit: 47,779,411 RAC: 32 |
I averaged the numbers and its looking like.. If this stays the same.. the best I will get for an rac is around 2k/day. hardly what I was hoping this thing could do. I know its at stock freq but right now, im just trying to keep it going until the better fan/heatsink gets here. Well, I have a stock B3 Q6600 that was getting well over 3K before all this credit adjustment stuff started, it would probably still get over 3K, I think. but it's been doing Rosetta for a while now. It'll be going back to Seti in the next couple of days, so we'll see.... I have a G0 Q6600 that was doing well over 5K at 346 X 9 (3114), that one's also been mostly on Rosetta lately (trying to get that project to 1,000,000...). 2K seems low for a q6600 running the optimized app (which it looks like you are) unless you're not using all 4 cores or have it set to use less than 100% of cpu time. -Dave [edit] also, I took a look at the Q6600 in your computers list and it looked like it was doing 17.36 credit WUs in mid 1400 - mid 1600 seconds, so that's low - mid 20 min. for 17.36 credits, that's way off for the calc's you listed... [edit 2] And come to think of it, I don't recall ever seeing a WU in the 8 - 12 credit range (unless maybe a -9 error...). |
john deneer Send message Joined: 16 Nov 06 Posts: 331 Credit: 20,996,606 RAC: 0 |
I've heard of the B3 stepping and the G0 stepping, But the B0 and G3? When did those come out? As I have a B3 Q6600 cpu paired with a motherboard not designed for a quad, I'll be fixing that in December If possible(that and some more Patriot PC2-8500 Viper ram). Mine is forced to run @ 2.61GHz due to the motherboard and some ram problems too. Yes sorry, you are absolutely right. Must have been late :-) The g0 runs cooler and should be kept in the low 60's and the b3 can be run in the high sixties (but gets there at lower freqencies, needing more volts than the g0). Since you have a b3 you should try to keep it below 70C, preferably around 65. Regards, John. |
zoom3+1=4 Send message Joined: 30 Nov 03 Posts: 65763 Credit: 55,293,173 RAC: 49 |
I've heard of the B3 stepping and the G0 stepping, But the B0 and G3? When did those come out? As I have a B3 Q6600 cpu paired with a motherboard not designed for a quad, I'll be fixing that in December If possible(that and some more Patriot PC2-8500 Viper ram). Mine is forced to run @ 2.61GHz due to the motherboard and some ram problems too. In My B3's current setup It would be amazing If It got past 50C, Right now It's at 37C. The T1 Trust, PRR T1 Class 4-4-4-4 #5550, 1 of America's First HST's |
Ingleside Send message Joined: 4 Feb 03 Posts: 1546 Credit: 15,832,022 RAC: 13 |
Also, the power supplys voltages are not that great. the +5 buss is reading at 3.83 to 4.10.. the +12 is at 13.2. 3.3v is holding. I know that TTL chips wont run below 3.8 volts but I have NO CLUE about these VLSI bridge chips. It's doubtful the computer will run if those voltages is accurate, something they often aren't especially if not looked-on in bios but by some form of monitoring-software. I had a psu going bad, this was detected by hd stopping (and re-starting), and a closer look showed that +5V was at 4.5V. If you're even lower, it's surprising the system is still running. But granted, the hd was a power-hungry 10k-rpm SCSI, so maybe another hd is more forgiving... "I make so many mistakes. But then just think of all the mistakes I don't make, although I might." |
Mike O Send message Joined: 1 Sep 07 Posts: 428 Credit: 6,670,998 RAC: 0 |
I was going off what BoincView was telling me. I looked at my pending credits and I guess im doing better than I thought :) Im still confused here.. I guess I have been staring at this for to long. BoincView is now showing all up comming WUs in the range of 20.88 to 30.88 credits estimated. If its wrong about others, im sure these will produce the 50-60+ range im used to. I read all the specs on the Q6600 and its making better sense. My system would lock at around 67-70c. To be honest, Im supprised its running with the Vi at 3.83.. which is where it is now.. Maybe the monitoring couple is defective? Not sure but this is day 2 with no crash so im keeping my fingers crossed :) thanks all for the help :) I'll keep ya posted. I hope to OC this a bit after the new fan shows up. Its off to work... Peace.. Mike Not Ready Reading BRAIN. Abort/Retry/Fail? |
W-K 666 Send message Joined: 18 May 99 Posts: 19075 Credit: 40,757,560 RAC: 67 |
Strictly speaking TTL spec for 5V rail is 5V +/- 5% (+/- 0.25V) but there are now a lot of TTL compatible chips that will run at 3.8V, or so, that will accept the min 3V level of a '1' from a TTL device. |
john deneer Send message Joined: 16 Nov 06 Posts: 331 Credit: 20,996,606 RAC: 0 |
I was going off what BoincView was telling me. I looked at my pending credits and I guess im doing better than I thought :) Mike, the figures BoincView shows are lower than the credits you will actually be given for those wu's. I'm not sure how BoincView calculates its estimates, but it has probably little to do with the method that Seti uses (based on the number of floating point operations used in the calculations). I have 2 machines which I monitor with boincview and from a quick glimpse I think that the factor between credit given and credit estimated by boincview is not the same for those 2 machines. My guess is that BoincView estimates are based on benchmark scores in boinc and the time needed for finishing the wu. /edit/ As a rough estimate: the credit in BoincView is (approximately) only half of what you really get. Regards, John. |
©2024 University of California
SETI@home and Astropulse are funded by grants from the National Science Foundation, NASA, and donations from SETI@home volunteers. AstroPulse is funded in part by the NSF through grant AST-0307956.