Temperature measurements
We test the Peter in an installed state in a Corsair Graphite 600T and torture the AMD proband with Furmark and GPGPU applications. The picture shows our combined heat and power plant:
So that the “Turbo” (i.e. the actual TDP brake) does not take effect, we work with the older Catalyst 12.4, which recognizes the card but prevents the downclocking.
If you look at the final value at the respective fan current, then up to 18 Kelvin temperature difference is already a really fat house number. The increase in the temperature curves is also quite meaningful, because you can also infer the results with a more moderate gaming load. The flatter the gradient, the better the cooling in the partial load range. The maximum value is reached after a maximum of 4 minutes (240 seconds) and the GPU does not heat up any further:
They see, they hear: nothing
Low temperatures alone are certainly nice, but worth nothing in the end if your ears thank you for the good cooling performance with a well-groomed tinnitus. What’s most impressive is the cooling performance at 2 x 5 volts or 5 + 7 volts, when the associated sound levels are just over 32 dB(A) and 34 dB(A), respectively. There is no card here that offers similar values in the direct combination.
Even an Asus card with the very good DirectCUII cooler doesn’t reach 62°C at about 41 dB(A) in the installed(!) state. Peter can confidently invite to an alpine pasture party and buy a round of GPU grilled sausages for everyone, because he is the coolest lamella freak we have had in the computer so far.
31 Antworten
Kommentar
Lade neue Kommentare
Urgestein
Urgestein
Urgestein
Urgestein
Urgestein
Urgestein
1
Urgestein
Urgestein
Veteran
1
Mitglied
Urgestein
Urgestein
Veteran
Urgestein
Mitglied
Urgestein
Mitglied
Alle Kommentare lesen unter igor´sLAB Community →