What kind of temps are you getting with the Asus A7V266E?

[UsandThem]

What are other A7V266E users with Athlon CPU's 1.3 Ghz and higher getting for CPU temps?

I have a 1600+ with a Taisol cooler and I get 50c/122f idle, and it goes up 54c/129f at full load. I know in the past some Asus boards reported CPU temps higher than they actually were, but 54c/129f makes me feel a little nervous as I have already killed two Athlons in the past (one died because my MSI board was supplying .10v more to the CPU than needed and slowly killed it and the other when my old heatsink/fan clip loosened just enough to allow the CPU to get damaged ).

My cooler is rated to cool Athlons 1.4 Ghz and higher and is AMD approved. I use Artic Silver II and have a XP version copper spacer. I have a front 80mm fan and a rear 80mm fan plus a 80mm fan in my power supply.

My CPU is locked and not overclocked.

 

[UsandThem]

Well, since I was worried, I went and bought a Zalman CNPS5005-CU just to be safe. I still get the exact sames temps.

When I removed the old heatsink it was making perfect contact with the CPU die.

I wonder if the A7V266e temps are inaccurate, or if it just my board.

 

[pay]

Well, ASUS is known for about a reading +5-10C compared to other boards, but im not sure about the A7V266-E

 

[Mikewarrior2]

antiAbit,

If you've got the thermal jumper set to DUron/T-bird, then you are fine. Asus used to, and from early indications still do, compensate the reading upward to give a better representation of actual die temp. Yes, they are higher than other MBs, but that is not equal to "higher than actual". Better than actual, yes .



Mike

 

[UsandThem]

<< If you've got the thermal jumper set to DUron/T-bird >>



Yeah, that is what I have it set to.

I am having no heat related problems or system hang-ups, so I will just leave all as-is.

I just can't afford to buy a new CPU right now, and was a little worried when others are reporting 40c at idle.


At any means, this new Zalman cooler is very nice and very QUIET If anybody is in need of a new cooler, check this one out and see if it fits your needs.

Zalman

 

[Mikewarrior2]

that's hte problem.... You can't effectively (or even partially effectively) compare against other MBs. Different MB's read different temps, and in addition no one ever gives their ambient case temp. If you have a case temp of 30, adn someone else has one of 20, then the temps are going to be different?



Mike

 

[UsandThem]

<< If you have a case temp of 30, adn someone else has one of 20, then the temps are going to be different? >>



True. My case temp is always at 25c, so that is perfect. I was just really thrown off with the CPU temps.

I know one thing though, when I build myself a little entry server in the future, I will use a (sorry AMD) Intel PIII or PIV. Intel CPU's are expensive but "bomb proof".

Thanks for your help and your site on cooling is very interesting.

 

[Rastini]

I too am getting strange temp readings. 54C idle and 60C under load. My a7v266-e is running solid, so I sure hope that my temp gauge is off.

 

[Bglad]

A new Asus board and again complaints about high temps. If the new board fits the Asus mold, it will run higher voltage than most, run hotter than most, and read hotter than it really is! LOL

Not really a complaint though, I've had several and am completely satisfied. I just wish I knew what my real temps are.

 

[Mikewarrior2]

Bglad,

Explain and define REAL temps...

Do you like under-read inaccurate readings? Asus and other manufacturers (no, they aren't the only one) compensate the reading upwards to give the end user a better "reading" of what the actual CPU temp is. I'm tired of hearing people claim the reading is "higher" than actual when in fact the readings on 99.9999% of motherboards is "lower than actual". Its time to end hte crap about how asus' boards read "higher than actual", because it is a farce. Higher than what? other boards, yes. Higher than actual CPU temp? no.

And remember that XP hcips further render socket-thermistors even more inaccurate. The newer PCB is great at blocking heat from the backside thermistors (BTW, An XP1600 and T-bird1400 both pump about the same amount of heat into the heatsink, which heavily indicates that the t-bird has much more severe heat loss).



Mike

 

[LocutusX]

<< (BTW, An XP1600 and T-bird1400 both pump about the same amount of heat into the heatsink, which heavily indicates that the t-bird has much more severe heat loss).
>>



what do you mean? is that good or bad?

 

[Mikewarrior2]

what I mean by that is:

The 72W T-bird 1.4 pushes about the same amount of heat into a heatsink as a 62.8W XP1600. Same mhz, but same amount of heat into the heatsink.

Personally, I think its a great thing. Assuming Best Case Scenario, the T-bird has a heat loss of 9+W through PCB, Pins, and other secondary heat pathways. Since, there's no Best Case Scenario (even the XP will lose W of heat elsewhere), the T-bird is probably losing 20% of its heat output through various other pathways, and not directing it to the heatsink. In this manner, the XP is much more efficient when it comes to cooling.

However, what this does is change (and change significantly) how we treat the XP versus T-bird heat drop (~12-17% depending on the chip). Even though there is a heat drop, the HEATSINk, in equivelant MHZ t-birds and XPs, will recieve the same amount of heat. Therefore, cooling remains the same as with T-birds.



Mike

 

[Bglad]

Well Mike,
You are the first person I've heard claim that Asus reads more accurate than most other boards. The fact is I don't know what is accurate. However, every other board I've used reads significantly lower temps than Asus. Now, I'm almost positive that Asus pushes their voltages more than most other boards for stability. What I don't know is whose temperatures are more accurate. Obviously higher voltages would result in higher temps, but I don't believe that the few extra 1/10's volt Asus pushes would result in the drastic temp increases I see on Asus boards. So the question remains, are Asis temp readings accurate (meaning they run hotter than other boards) or do they read hotter (meaning either they over compensate or all other brands lie). The answer I don't know but the fact remains, Asus is the black sheep. It is possible that Asus simply runs hotter and the temps are accurate. With all the talk about Asus temps, I've heard few complaints about heat related stability problems. I'm clearly saying I don't know the cause of the issue but I've talked with dozens of people with similar results to mine. You need some evidence to back your opinion that Asus temps are more accurate than others, but then we are left with the issue that they simply run hotter than most other boards.

 

[Mikewarrior2]

Well Bglad,

I've posted this thousands of times. Socket-thermistor Temp readings are not accurate. NO socket-thermistor setup is accurate. I never said that Asus boards are more Accurate than other motherboards. Maybe you should try reading my posts. I said that Asus' read "higher' not more accurately. Higher due to different compensation techniques employed by Asus. And yes, other manfucturers do this. This is why each and every Socket-a mb reads temps totally differently.

Evidence: Take a chip running in an Epox 8kha+, and transfer it to an A7V266e with the same exact case and heatsink and grease. Result: Instant 8-10C "hotter". Before you claim it is the voltage, the Epox boards also increase voltage over the "set" number. These increases are the same voltage/mhz/etc. Why? The Asus compensates the reading, while EPOX does not. Does that mean the Asus "socket-thermistor" is more accurate? No. This does mean that Asus gives a better representation of actual die temp. Meanwhile, the CPU is outputting exactly the same wattage and heat as with the Epox.

At best, a socket-thermistor only shows 3/4th of the core temp rise over ambient. At worst, they show 1/4th or less of the core temp rise over ambient.

Want to see what T-birds actually run at? Check this review:

A7VTroubleshooting.com Heatsink Tests w Internal Diode

That's right, 34 to 50+C over ambient case temp.

Cross comparing to Overclockers.com socket-thermistor tests, where an Abit KT7 shows a hotter processor running the [bDynatron DC1206BM-O @ ~18C over ambient. That is underreading ACTUAL core temps by 16C. Which shows me that, if anything, other motherboards are under-reporting temps due to how they are measured. Asus, by virtue of compensation, are better at showing temps closer to actual DIE temps.

Once again, define actual. Actual backside temps, yes the Asus is "hotter" reading. Actual die temps, even the Asus boards are probably under-reading it.


Mike

 

[rkpindia]

using athlon 1700 xp and a7v266-e with retail heatsink fan (from amd). have set therm moniotr on jumper switch to reserve (allows on die temp monitor i understand). runs at 50C under no load and 58c under divx (heavy load) i have two 80mm case fans in yeongyang cube. MB runs at 25c

 

[tazdevl]

A7V266E with XP 1800.

Mobo 22/25C
CPU 36/40C

See my rig for cooling setup.

Jumpers set to Palomino and Tbird (reserved (2-3) results in no variation of CPU temps)

 

[Paladin]

SK6 w/30cfm Delta:
CPU 42/48

Fong Kai 320 case with 2 92mm fans
25/28

*edit* that's with my XP 1800 O/C to 1.62 GHz.

 

[Fuddam]

FWIW

my XP 1600 locked and not clocked

Idle 32C+/-
Load 50C+/-

Globalwin WBK38 - LOUD :frown:

have had a few problems, but more Win XP related than anything to do with temps. I think....