Originally Posted by karelbastos
Tested more here
And i really don't know why after WATER COOLED the two ZOTAC GTX 1080 FE i can't reach the same clock speed as before 2025 stable
And can't reach the same performance on benchmarks.
For me, lowe the temps will give me better results, but this not happen to me
Any of you guys, have a explanation ?
Mas i arrive nor stable is 2000mhz and 150 less points on benchmarks like HEAVEN.
BEFORE 67-70 C - NOW 45 - 50 C
I'm getting the same here.
I was able to run +152 for an final speed of ~2025-2050mhz when on air with the FE card running at about 65c and the EVGA SC at about 55c.
After 2xEK blocks, temps are much lower, both hovering around 40-50c at full load (I don't have a super efficient loop, I prefer silence and look, so-so performance (delta T) is fine for me) I haven't tried squeezing it yet, but from roughly adjusting settings in AB, I was able to get +100 to run, and it would clock to about 2000mhz.
I did notice though that the clocks didn't vary AT ALL, whereas when I was running the cards on air, even at stock clocks, the clocks were constantly bouncing all over the place during gaming or Firestrike Ultra. Once they were put on water, the clock went right to 2000mhz and sat until whatever I was running was 100% done (benchmark, gaming, etc.). I chalk it up to GPU Boost 3.0 working as intended. I think people (myself very much included) need to better understand how GPU Boost 3.0 works, how it works differently than prior versions, and what it means for squeezing performance out of the cards (not necessarily higher clocks).