Heat vaues Duron/Athlon

[jaydee116]

I need to know what the proper heat leavels for the Duron.I am getting mixed information. Some say don't get above 120f and some say 160f+ is ok. Please give me your opinions.

A7V
Duron600@800

Thanks

 

[jsbush]

Anything under 50C is good. Don't know what that is in ferenhight. I don't even think I spelt that right

 

[johneetrash]

46c = 114f

i forgot the exact formula though, i cheated using mbm somethign with 1/9 i think..or something...bleh

 

[Mule]

Well as the speed of the processor increases so the the heat disapated. And I think that Athlons produce more heat than Durons. I don't remember exactly how much heat but I think a 1.2 Ghz Athlon is somewhere near 60watts while a 900 Mhz Athlon is near 40-45watts.

It is not uncommon to find 1.2 Ghz systems which run IDLE in the 50's and upwards to the 60's under full load.

My current system which is a 1.1 Ghz Athlon ran 46 IDLE 49 Max, but with me quieter case fans it runs 49 IDLE 53 Max. It runs without a hiccup so I'm really not too worried about the heat factor. Never has it stalled on me yet. I think you start to worry if it nears 64C which is the low threshold for the Athlon processor.

However motherboard readings aren't the most precise, but their not too bad either.

The rule of thumb for me is that if it boots up well, you have decent cooling, and it doesn't ever crash on you (even while running some CPU extensive stuff) then you're alright.

 

[Mikewarrior2]

use Radiate

Durons use a bit less than corresponding t-birds.

As far as temps being remotely accurate, I have to completely disagree. The reason you should stay below 55C reading temps on most mbs, is the higher you get the lower and lower the "change" in temp you'll see.

The readings are Approximately accurate around 1ghz, and after that the changes don't really show up anymore. results of compression due to socket-thermsitor measurement. a 1.2ghz chip could be reading 52C, with a 25C ambient case temp. Taking into account that it is the kT7 UL bios, so you can chop 10 off the reading temp... so you're "reading 17C over ambient". What if you're particular setup is getting a 2X temp change compression. The 17C over ambient is actually a 34C core temp change over ambient?

Sound far fetched? Sure it does. But it clearly explains why systems at 60C "read" temp on most boards are inaccurate. Because the CPU probably isn't running at 60, more like 70-80C.

This doesn't apply as much to the A7V. The A7V "Compensation" tends on the side of being slighly too much compensation. a 58-60C A7v reading usually isn't unstable.


Mike
P.S. This is where i'm seeing MB problems pop up: On a Kt7-readings of 60+C tend to be unstable, on an A7V its around 65C, and on a K7T Pro2a the problems appear to start when a cpu is beign read at 50-55C(this mb has no compensation).