Silly question

[coffeefoot]

:huh:Can someone explain to me what is the purpose of "Delta" temp? I tried to use the same method that TTL uses and came up with a:huh: figure of 48.5 c. But what does this tell me ? Please give me an answer as if you were talking to a puppy !

 

[jamesriley94]

Doesn't tell you anything.

It's just taking the temperature of the room away to compare coolers.

For example - NH-D14 with a 30 degrees C ambient vs BeQuiet Dark Rock with a 15 degrees C ambient temp - of course the bequiet is going to be cooler in that test. But in real life you can't really compare them like that.

It's just to keep testing fair. You don't really need to worry about it tbh.

 

[airdeano]

delta is the mean or difference in temperature minus any influences (room
temperature)
so taking the loaded high, subtracting the room temp gives the delta. doing this
from now on will give a comparison of the temperatures with the room temp not
a contributing factor.
so comparing winter temps and summer temps the deltas should be close to the
same.

airdeano

 

[coffeefoot]

Ok. Thanks for the quick response. So, if I understand your answers, it is a way of removing all temp variables to give a fair comparison.
So given my setup ( i5 3570k, z77 ,asus mobo, 16 gs ram, corsair H60-push/pull)Stress with Aida64 100% for an hour, a "D" temp of 48.5c is ok?

 

[jamesriley94]

Ignore delta temps on your own system.
It's only really beneficial for comparing coolers.

Stick to actual temps. You could stick your PC in the oven and have a delta temp of 48.5C, whilst your CPU was like 200 and something..... (ignoring the fact that the system would shut down at that temp)

But assuming that's actually around 68.5 degrees - that's fine.

 

[coffeefoot]

OK. Got it. Thanks for the help. There will be more !!

 

[airdeano]

usually, when i state temps its a loaded read-out , followed by ambient
temperature. this way it removes doubt and questions on validity or status.

airdeano

 

[coffeefoot]

thanks !

 

[233Kosta]

Well delta temps are great assuming the relationship is linear. It's not however. The higher the absolute ambient temperature - the higher the delta is going to be. I'm not sure if the relationship is square, square root, exponential or whatever else. It's definitely not linear though. If anybody feels like running this sort of measurement - be my guest, I'd happily process the data, however I don't have the time (or air conditioning) to gather so much data.

For the purposes of comparing coolers however - if the ambient temperatures during testing don't differ too much - delta temperatures are fine for performance comparison between coolers and I'd expect accuracy within 1 or 2 degrees.