I have searched through the thread regarding invalid shares but I am still a bit confused.
When I look at a calculator, it seems I am getting about as much eth as I should expect. 146 mhs Day = .905135 $10.00
but I have a lot of invalid shares.
I am using ethOS
user ID:23168
Could you explain what a stale level is?
Do invalid shares affect the amount of eth I mine?
What can I do to reduce invalid shares?
In another thread someone suggest core clock could be an issue?
Another suggestion was a problem with DAG. I am using ethOS so each time I startup the DAG is regenerated right?
from MPH
Round Shares Submitted shares since last found block (ie. round shares)
Est. Shares 4,681,152 (done: 57.38%)
Your Valid 8,007.6900
20.9591 (1 level stale 0.11%)
Your Invalid 10,735.5726 (57.21%)
These are the GPU's and their clocks
Rig 1 (
http://paste.ethosdistro.com/aqiletafit)
miner_hashes:16.61 29.67 30.96 30.90
models:01: Advanced Micro Devices, Inc. [AMD/ATI] Tahiti XT [Radeon HD 7970/R9 280X]
03: Advanced Micro Devices, Inc. [AMD/ATI] Hawaii PRO [Radeon R9 290/R9 390] (rev 80)
04: Advanced Micro Devices, Inc. [AMD/ATI] Hawaii PRO [Radeon R9 290/R9 390] (rev 80)
06: Advanced Micro Devices, Inc. [AMD/ATI] Hawaii PRO [Radeon R9 290/R9 390] (rev 80)
bioses:113-C3865000-O77 MS-V30823-F5 MS-V30823-F5 MS-V30823-F5
default_core:1020 1040 1040 1040
default_mem:1500 1500 1500 1500
core:1000 1150 1175 1175
mem:1500 1500 1600 1600
Rig 2 (
http://paste.ethosdistro.com/funujicavi)
hash:37.7
miner_hashes:19.06 18.64
models:02: Advanced Micro Devices, Inc. [AMD/ATI] Tonga XT [Radeon R9 380X] (rev f1)
03: Advanced Micro Devices, Inc. [AMD/ATI] Tonga PRO [Radeon R9 285/380] (rev f1)
bioses:113-380X4DB71-W80 113-380PEDF71-W82
default_core:1040 990
default_mem:1450 1425
core:800 800
mem:1575 1575
Do you think I should change the clocks in some way? I worked at it for a few days to get the hash rate as high as I could and then really haven't messed with things.
TIA
Could you explain what a stale level is?
- > Ethereum's block can be generated not only from the highest height block. The biggest difficulty level block is chosen as main block among candidate blocks and others will be treated as "Uncle block".
Stale level means, your share is a bit late and new block is already found from pool. This late share is not useless, but should be valued lower by how far the block height difference is.
level 1 stale share means it's 1 block late, level 2 stale share means it's 2 block late. It can be upto 6 level according to Ethereum's uncle block spec.
Do invalid shares affect the amount of eth I mine?
-> It affects.
Uncle block's reward is little bit less than normal block.
Normal block's reward is 5ETH, Uncle block can be 4.375 ETH, 3.75 ETH, 3.125 ETH, 2.5 ETH, 1.875 ETH, 1.25 ETH.
Stale shares will be calculated by mathematically expected uncle block reward.
1 level stale share will be treated as (4.375 / 5) than normal value.
2 level stale share will be treated as (3.75 / 5) than normal value.
What can I do to reduce invalid shares?
->Stale shares that are too late (more than 6 level stale) can be treated as invalid shares, but mostly it's just wrong calculated share.
Seems like one of your GPU is generating invalid shares continually. Maybe it's hardware error.
It would be better if you separate each GPU and watch which is causing error.
Pool can't distinguish which GPU has problem, since they are all reported as '6eae0a' workername from you.
In another thread someone suggest core clock could be an issue?
-> Core clock seting could affect this issue. Or releasing excessive undervolting setting would solve it.
Or GPU may need to be fixed. You have to test each GPU first and see what is causing error.
Another suggestion was a problem with DAG. I am using ethOS so each time I startup the DAG is regenerated right?
-> AFAIK yes. I haven't tried ethOS yet, so don't know much about it.
Nowadays, latest release of genoil-ethminer, claymore works well, generating DAG in GPU.
Have a try.
