Firstly, thanks to Claymore for the improvements here! On most of my rx 470 4GB cards, I am seeing about a 3-4% increase in hash-rate!
However, I was hoping v11 would also alleviate an issue I've been seeing since starting to use v10.2 a while ago. For some reason, GPU 0 on all 4 of my rigs acts very strangely... most of the time hashing well below the others... fluctuating between 0 and 450, while the other 5 are right at the 700 mark. Now, if this were only on 1 rig I would think the GPU is going bad, but since it is the exact same GPU slot 0 on all 4 rigs, it is more likely some kind of setting that i need to tweak.
Has anyone else noticed this on their rigs? If so, any suggestions on how to fix it?
FYI: after just upgrading all 4 rigs to v11.0, GPUs 1-5 are hashing better by about 3-4% but GPU 0 is still bad on all 4. Although, it doesn't seem to be fluctuating as much... pretty much stuck at 232 on 3 rigs and weirdly, 120 on the 4th.
Any help resolving this would be much appreciated!
However, I was hoping v11 would also alleviate an issue I've been seeing since starting to use v10.2 a while ago. For some reason, GPU 0 on all 4 of my rigs acts very strangely... most of the time hashing well below the others... fluctuating between 0 and 450, while the other 5 are right at the 700 mark. Now, if this were only on 1 rig I would think the GPU is going bad, but since it is the exact same GPU slot 0 on all 4 rigs, it is more likely some kind of setting that i need to tweak.
Has anyone else noticed this on their rigs? If so, any suggestions on how to fix it?
FYI: after just upgrading all 4 rigs to v11.0, GPUs 1-5 are hashing better by about 3-4% but GPU 0 is still bad on all 4. Although, it doesn't seem to be fluctuating as much... pretty much stuck at 232 on 3 rigs and weirdly, 120 on the 4th.
Any help resolving this would be much appreciated!
When Claymore miner starts, you will see GPU0 has a few less MB than other same cards.
Code:
GPU #0: Pitcairn, 1975 MB available, 20 compute units
GPU #0 recognized as Radeon 270/270X
GPU #1: Pitcairn, 1984 MB available, 20 compute units
GPU #1 recognized as Radeon 270/270X
GPU #2: Pitcairn, 1984 MB available, 20 compute units
GPU #2 recognized as Radeon 270/270X
GPU #3: Pitcairn, 1984 MB available, 20 compute units
GPU #3 recognized as Radeon 270/270X
GPU #0 recognized as Radeon 270/270X
GPU #1: Pitcairn, 1984 MB available, 20 compute units
GPU #1 recognized as Radeon 270/270X
GPU #2: Pitcairn, 1984 MB available, 20 compute units
GPU #2 recognized as Radeon 270/270X
GPU #3: Pitcairn, 1984 MB available, 20 compute units
GPU #3 recognized as Radeon 270/270X
Try to lower the -h parameter for GPU0
Thanks a ton for that tip! even though my GPU 0 was showing exactly the same MB available as other 5 GPUs I still gave your method a shot. on startup, all were at -h value auto of 892, so I lowered GPU 0 to 812 as a test and it worked immediately! Got that card up to 672 (from 120)! I may play with it and inch up a bit but that's already a huge improvement! Now i will make that same change to my other rigs! Thanks again! Really appreciate it!
I'm glad to help !
Testing again and again when a card doesn't work like you think it should
But sometimes a card is really lower than others : I have 2 Sph R9 290 ref board, one running at 700 H/s no matter how I overclock, the other at 810 H/s @1040/1500 . Flashing bios didn't help so far...
Anyway, if you feel like sending me some ETN, feel free to do so !
Code:
f4VR7QEKoHSNCosFG21EReQ3nS8TPxTUzHCzmFJ6trNe3KHG9rAKhq3F8AYWdLyCe7bLrPbKAX9seMhUmbRcCsxz9SBnNYMgfFk5JkJ9kqCst
Everyone have a lovely last 2017 day !
#Share #Data #Love