thats too hot - your cpu wil lmost likely last for many years even at that temp
y should be at 50-55 idel in the bios
u should be at 65c tops under full load, maybe 70c
most likelu u need a better cpu cooler - did u use artic silver?
does your system have good air flow to cpu and is it oriented in the right direction? i.e u should have 12cm fan in and out and door fan or more
you rheat sink is not a highly rated one for this appication - start with a top of the line cooelr with 12cm or 9cm fan
85c??? " What is a safe temperature for this QX6700 if I am not bothered about it lasting more than 12-18 months."
Yeah 75C and you have convinced me on the cooler, I should have read more reviews before ordering the Eclipse (it looked impressive).
Will order a Zalman 9700 tomorrow which I gather is the current recommendation for air cooling?
One final question what temperature should I be getting under a full load, from the few tests I ran at 2.66Ghz the processor was still going over 70 fully loaded with the stock cooler (the Eclipse had not arrived then).
Is this normal or am I doing something wrong? I figured the QX6700 would be even with 4 cores these temps seem high.
All temps are with the Nvidia temperature monitor for xp.
Temps drop down to about 28-30 when ideling, room temp is about 22-25.
Unfortunately did not see your post in time so ordered and have installed the zalman.
Not very happy with the zalman had to angle so air blasts into PSU as if its mounted the correct way it actually touches the PSU, so I will order the recommended cooler.
On coretemp the idel temperatures are: 49, 49, 44, 42 (vs 34 according to Nvidia monitor).
Have a small problem with the intel thermal analysis tool, it only seems to support 2 cores not four. When I run it cores 1 and 2 quickly get upto 73/71 on Core Temp (Nvidia saying 60ish)... Stopped the test at that stage.
On Prime95 with all 4 threads running on small FFT's temperature stabilises around 69-72 after about 10mins.
Still not very happy with stability Prime95 is failing on core 3 after about 10mins. (its the coolest core)
Am very woried about the Nvidia Montior, clearly it is very wrong in its temperature analysis.
A quick change of cases to a full-tower so I can fit the fan so it is blowing through the rear and my temperatures have actually gone up a couple of degrees.
Is it better to blow the hot air from the CPU into the power supply rather than out through the rear ehaust fan?
Going back to the origional question given the quads run hot, how hot is safe? I dropped the processor back to 2.66Ghz and the temperatures are still hitting the 70's, is it safe having this processor run in mid 70's at full load?
Hey I wish I had your budget!
Recommend default 266 coreclock until you sort out your temps.
The Quad uses big watts and gives off major heat...
Sure it might OC as well as an E6700 but it's the heat and power that serve to hold it back...
I was joking (in another post) about needing a dedicated 15A AC circuit for that beast - but with an 8800GTX it's no joke!
The thing could be too much for even my 620W PSU.
Very good cooling, stock/default settings, and SpeedStep would all seem to be indicated
And make sure it's getting enough good pure AC power,
I must confess I played it safe on the power and bought a Galaxy 1000W, don't get me wrong I don't need it now but I'll probably put another 8800GTX in in Feb-March when prices drop and I figured with the way things have been going power wise I'd play it safe.
Being in the UK amps should be OK 13A * 240V = 3120W and the rest of the electronics on the ring don't draw that much (old PC is probably worst, 3.4Ghz P4, 6800GTX ~400W).
Temperatures are stabilising a bit, I applied more arthic silver and swapped the direction that the CPU block was on (more so pipes would fold better when I closed door).
Even though temps are a bit better I think I'll just leave it a 3Ghz, at 3.2Ghz with bus speed increase memory wasn't happy (and given my existing problems I don't want to raise voltage on RAM).
3.4Ghz was fine with a multipler increase but the temps were getting up to mid 70's again, the 200-400Mhz just doesn't seem worth the extra risk atm, might raise it again for a 3d Mark blast but other than that I'll keep it down.
Out of interest, how safe is it likely to be to run both CPU & GPU on the Tai-Chi default cooling rig, now I have the water-cooling I was thinking of swapping the 8800GTX for one of the water-block ones (More because the fan on the 8800GTX is doing my head in, I thought the old 6800GTX was bad but this is worse).
Also would I see much of a difference if I swapped the pump or CPU block?
Cooler master arogate displays work well for monitor of temp’s. I place the sensors under the water block as close to the CPU and GPU as possible. The bios temp’s reading is 5 degrees higher than the ambient temp’s that the sensors are displaying to the display. It has four extra 12 volt wiring for additional monitoring of fan rpm.
It’s a great way to monitor you system, its specially useful to see how well the system is cooling and will help anyone improve their own cooling efforts without worrying what is right, the software or the bios or whatever that person is using to monitor the cooling of the system.
Does not matter where the sensors are, the reading of the Gpu and the CPU temperatures displayed to the digital display will be consistent according to the temperatures of the sensor location, and the temperatures will remain the same on the digital display consistent with the bios information for temperatures . Cooler the heat sink or hotter the heat sink, the sensors will display the temperatures.
This is consistent to 5-degrees for CPU. I have personally tested this with a sensitive thermostat placed on the heat sink of the CPU, the thermostat is the same temperature reading as the sensor’s monitoring location .No matter what negative remark about the monitoring capabilities of this product, or other products of this kind, Is a selfish way to agree that this is one of the best ways to monitor the complete cooling of any system, including other areas in the case. This 5-degrees difference does not include any other areas in the case. The GPU sensor is exactly the same temperature as the GPU bios monitor, it is this bios monitor that sends all data of temperatures no matter what is used to read temperatures The digital display can do this task without the use of the bio’s temperature monitor and be very accurate. I do not use any temperature monitor other than my digital display. I used NVIDIA’s self installed temperature reader for the GPU ambient temperature for comparison. This also allows all system monitoring to be completely shut down in the bios of the GPU and CPU including fan speeds I have complete confidence that the digital display can handle all of the monitoring accurately without the use of system resources to run the software to monitor the temperatures.
The temperatures are accurate. It’s easy to see this is a win product to monitor system cooling. Picks to prove the reliability of this can be arranged and will be given so others considering the choice of this wonderful monitoring device can be used in confidence to monitor accurately.
Frostytech.com is the place for cooler reviews,,,that's all they do...
The other element that you might consider is getting the heat from your GPU [which is more that likely higher than the CPU's output??] out of your case,Look and see if the Vantec Iceberq 6 will do the job ,it exhausts out the back of youe case,pushing GPU heat all around inside the box ONLY heats up everything else,get rid of it ASAP.......