Bah, I seem to have found some kind of sweetspot.
Using TC@8193 & Gpu@1105/1500 I get 770Kh/s on all 6 cards, pretty stable.
Might actually try 8193 and slightly higher clocks on the watercooled one as well just to see what happens...
Topic: Inconsistent results with Zuikkis kernel and 280X (Read 837 times)
no idea?
Hi
I'm just trying out the zuikkis kernel on my rig.
Basically it's a 6x Asus DirectCU II Top R9 280X rig, running BAMT and SGminer 4.0.0.
On the default kernel, and 1085/1500 clocks (and the usual settings i.e. G2, I13, TC@8192 etc) I get 755Kh/s on all 6 cards.
With zuikkis I manage to meerely squeeze 5 cards up to (yet, seemingly more unstable) 765Kh/s, using 1095/1500. The sixth card is actually going slower (lower 700's).
Any logical explanation for that?
Does the 4.1.0. version add stability/speed?
I have a watercooled 280X (same brand and model) in my main PC which is running much higher clocks, and has profited from the zuikkis kernel much more (from 785 to ~825). It's running SGminer 4.1.0. All other parameters are equal.
Any input?
I'm just trying out the zuikkis kernel on my rig.
Basically it's a 6x Asus DirectCU II Top R9 280X rig, running BAMT and SGminer 4.0.0.
On the default kernel, and 1085/1500 clocks (and the usual settings i.e. G2, I13, TC@8192 etc) I get 755Kh/s on all 6 cards.
With zuikkis I manage to meerely squeeze 5 cards up to (yet, seemingly more unstable) 765Kh/s, using 1095/1500. The sixth card is actually going slower (lower 700's).
Any logical explanation for that?
Does the 4.1.0. version add stability/speed?
I have a watercooled 280X (same brand and model) in my main PC which is running much higher clocks, and has profited from the zuikkis kernel much more (from 785 to ~825). It's running SGminer 4.1.0. All other parameters are equal.
Any input?
Jump to: