Gday
just a question is there anyway to limit "generated but not accepted".



Regards Dan

Someone found good settings for NVIDIA 970 cards?
Autotune makes my rig crash

I use the data from configgen and have this setup for all my rigs:
C:\Users\Mining1\Desktop\Yacminer-3.5.0-yac2-x86\yacminer.exe --scrypt-chacha -o stratum+tcp://stratum2.tumblingblock.com:5555 -u worker.1 -p password -R 2560 -g 1 -w 128 --buffer-size 2590 --lookup-gap 8 --temp-cutoff 85 --nfmin 4 --nfmax 30 --starttime 1388361600

From that I get different results for every rig:

5 x Sapphire Dual-x R9 280x 3GB: 998h/s, 998h/s, 490h/s ,139h/s & one SICK
5 x Sapphire Dual-x R9 280x 3GB: 998h/s, 998h/s, 998h/s, 362h/s & one DEAD
5 x Sapphire Dual-x R9 280x 3GB: 998h/s, 998h/s, 490h/s ,139h/s & one SICK
4 x Sapphire Dual-x R9 280x 3GB: 998h/s, 998h/s, 998h/s, & one DEAD

5 x ASUS Radeon R9 280x 3GB: 1,047Kh/s, 1,047Kh/s, 1,047Kh/s, 1,047Kh/s & one DEAD
5 x ASUS Radeon R9 280x 3GB: 1,047Kh/s, 1,047Kh/s, 1,047Kh/s, 1,047Kh/s & one DEAD
4 x ASUS Radeon R9 280x 3GB: 1,047Kh/s, 1,047Kh/s, 1,047Kh/s & one DEAD

5 x MSI Radeon R9 3 GB: 998h/s, 998h/s, 998h/s, 998h/s, 998h/s

Anyone got any idea why one card is always dead in each rig from Sapphire/ASUS but not from the MSI-cards?
Before mininig at N14 they didn't go DEAD/Sick...?

Ok, I'll try that. If it says GPU 0 is sick would that be automatically the first engine to slow down, in your example: --gpu-engine 1040,1040,1040,1000,1000 and slow that from 1040 to lets say 1000?

So tried --gpu-engine 1031, 1031, 1031, 1031, 1031 and get : 0 (Sick/dead), 1,009 Kh/s, 1,009Kh/s, 490h/s, 134h/s.
Then I try --gpu-engine 900, 1031, 1031, 900, 900 and get: 0 (Sick/dead), 881h/s, 881h/s, 489h/s, 130h/s
Why is GPU 1 and 2 affected by the GPU 0 engine set?

You are having memory issues, to much engine speed not enough memory. You can see that your gpu 0, 1 and 2 are affecting each other and 3 and 4 are proly sharing. Depending on the design of your MOB your GPU may be sharing a data bus. Example, my 6 gpu rigs are the same and  gpu1 and 4 share a bus and changing anything on 1 affects 4, but only when there are 6 gpu's running. try gpu-engine 1000 and gpu-memclock of 1250 for start.

So this is the batfile:
C:\Users\Mining1\Desktop\Yacminer-3.5.0-yac2-x86\yacminer.exe --scrypt-chacha -o stratum+tcp://stratum2.tumblingblock.com:5555 -u worker.1 -p password -R 2560 -g 1 -w 128 --buffer-size 2590 --lookup-gap 8 --gpu-engine 1000 --gpu-memclock 1250 --temp-cutoff 85 --nfmin 4 --nfmax 30 --starttime 1388361600

gives me: 0 (Sick), 978,978, 450, 130

Reboot the rigs. Do not just restart the miners when you make these changes.

Gettin` 1188 h/s on 280X @ 1200/1500
Now 4 (cards) x 20-25 accepts without HW.
Config from configgen.