11.5 version restarts rigs randomly, every 1-5 hour. But before restart everything is working. It 100% only this versions issue. 11.4. 11.5.
Try disabling QuickEdit Mode in the properties of your batch file
Topic: Claymore's Dual Ethereum AMD+NVIDIA GPU Miner v15.0 (Windows/Linux) - page 312. (Read 6590718 times)
March 17, 2018, 01:30:03 PM
Try disabling QuickEdit Mode in the properties of your batch file
March 17, 2018, 01:16:48 PM
AMD cards that are BIOS modded definitely use oldkernels
AMD cards that are overclocked right on the edge might need oldkernels
Go ahead and try new kernels. With everything the same as before. If theres any instability or increase in rejected/incorrect do not bother underclocking just go straight to oldkernels.
The main difference between the two I've found is you don't get the fluctuating hash rate with the newer ones. So if you can get away using them do it.
That's my story and I am sticking to it.
AMD cards that are overclocked right on the edge might need oldkernels
Go ahead and try new kernels. With everything the same as before. If theres any instability or increase in rejected/incorrect do not bother underclocking just go straight to oldkernels.
The main difference between the two I've found is you don't get the fluctuating hash rate with the newer ones. So if you can get away using them do it.
That's my story and I am sticking to it.
Weird, I just upgraded to 11.5 my two rigs (rx480 and rx570) and both increase hashrate, and both have mod bios and both are stable without using oldkernels under linux.
Maybe your mod is not radical enough.
March 17, 2018, 12:17:32 PM
Same here, it worked good at the begging and then it started to fluctuate (3 rigs on 11.5 oldknrl rest on 11.4)
https://i.gyazo.com/00514fdb5f9cf3aac91eba8c1121ea24.png
https://i.gyazo.com/00514fdb5f9cf3aac91eba8c1121ea24.png
Click in "effective current hashrate" to remove it and you will get better resolution for the other two. Try closing gpuz or hwmonitor.
You're right, closing gpuz solved it... Thanks big al!
Do you mine running HwInfo 24/7 can affect hashrate/valid shares and overall work of the miner?
March 17, 2018, 12:16:01 PM
Quote
How do you measure stale shares?
CJ
Ethermine shows stale shares as %
March 17, 2018, 12:15:38 PM
Also, does -mvddc do anything for Vega cards? As far as I know HBM voltages are fixed, and defined by which bios you go for (56 or 64) aren't they?
March 17, 2018, 12:10:59 PM
Same here, it worked good at the begging and then it started to fluctuate (3 rigs on 11.5 oldknrl rest on 11.4)
https://i.gyazo.com/00514fdb5f9cf3aac91eba8c1121ea24.png
https://i.gyazo.com/00514fdb5f9cf3aac91eba8c1121ea24.png
Click in "effective current hashrate" to remove it and you will get better resolution for the other two. Try closing gpuz or hwmonitor.
You're right, closing gpuz solved it... Thanks big al!
March 17, 2018, 12:07:47 PM
Not sure it's been reported yet but -fanmin doesn't work on RX Vega's on the latest Adrenalin driver (18.3.2 March12).
March 17, 2018, 12:07:33 PM
AMD cards that are BIOS modded definitely use oldkernels
AMD cards that are overclocked right on the edge might need oldkernels
Go ahead and try new kernels. With everything the same as before. If theres any instability or increase in rejected/incorrect do not bother underclocking just go straight to oldkernels.
The main difference between the two I've found is you don't get the fluctuating hash rate with the newer ones. So if you can get away using them do it.
That's my story and I am sticking to it.
AMD cards that are overclocked right on the edge might need oldkernels
Go ahead and try new kernels. With everything the same as before. If theres any instability or increase in rejected/incorrect do not bother underclocking just go straight to oldkernels.
The main difference between the two I've found is you don't get the fluctuating hash rate with the newer ones. So if you can get away using them do it.
That's my story and I am sticking to it.
Weird, I just upgraded to 11.5 my two rigs (rx480 and rx570) and both increase hashrate, and both have mod bios and both are stable without using oldkernels under linux.
What ram straps are you running? I think that has everything to do with it. Some work with new some don't.
March 17, 2018, 12:00:23 PM
11.5 with -oldkernels 1 gives me less and fluctating reported hashrate.... 11.4 works perfect for me but i am getting 3-6% stale shares and someone mentioned 0-2% stale shares with 11.5
How do you measure stale shares?
CJ
March 17, 2018, 11:44:27 AM
11.5 version restarts rigs randomly, every 1-5 hour. But before restart everything is working. It 100% only this versions issue. 11.4. 11.5.
March 17, 2018, 11:39:19 AM
AMD cards that are BIOS modded definitely use oldkernels
AMD cards that are overclocked right on the edge might need oldkernels
Go ahead and try new kernels. With everything the same as before. If theres any instability or increase in rejected/incorrect do not bother underclocking just go straight to oldkernels.
The main difference between the two I've found is you don't get the fluctuating hash rate with the newer ones. So if you can get away using them do it.
That's my story and I am sticking to it.
AMD cards that are overclocked right on the edge might need oldkernels
Go ahead and try new kernels. With everything the same as before. If theres any instability or increase in rejected/incorrect do not bother underclocking just go straight to oldkernels.
The main difference between the two I've found is you don't get the fluctuating hash rate with the newer ones. So if you can get away using them do it.
That's my story and I am sticking to it.
Weird, I just upgraded to 11.5 my two rigs (rx480 and rx570) and both increase hashrate, and both have mod bios and both are stable without using oldkernels under linux.
March 17, 2018, 10:58:57 AM
11.5 looks stable on 6 x 470 rig with modded BIOS - avg 180 Mh/s
Thank you Claymore!
Thank you Claymore!
March 17, 2018, 10:34:32 AM
Keccak or Blake2s has lower power consuption in dual mine? Please advise
no, all dual mining coins are the same on power consumption, 30 - 55 watts each card.
March 17, 2018, 09:55:52 AM
Issue with Linux/SMOS changing from 11.4 to 11.5
Random restarts all rigs every few minutes.
Rigs are nVidia's p106, 1060, 1070, 1070Ti
I revert back to 11.4 and all rigs are stable now.
Random restarts all rigs every few minutes.
Rigs are nVidia's p106, 1060, 1070, 1070Ti
I revert back to 11.4 and all rigs are stable now.
March 17, 2018, 06:01:49 AM
Keccak or Blake2s has lower power consuption in dual mine? Please advise
Blake2s
March 17, 2018, 04:04:48 AM
Keccak or Blake2s has lower power consuption in dual mine? Please advise
March 17, 2018, 03:34:41 AM
Click in "effective current hashrate" to remove it and you will get better resolution for the other two. Try closing gpuz or hwmonitor.
Rigs are running Win 10 lite edition, aside from Claymor nothing else runs.
Don't see how turning off a state on graph can help
March 17, 2018, 03:34:23 AM
To everyone saying that there are hard and fast rules to using old kernel vs new, that is simply not true. Each of my GPUs is within 2mhz of giving incorrect shares on old kernels and doesn't give any on new kernels either. I think you just have to try for yourself and see what works best for you.
That said when the mining craze nears the end, old kernels are more power efficient!
That said when the mining craze nears the end, old kernels are more power efficient!
March 16, 2018, 11:57:26 PM
Hi. With Claymore 11.5 i often recieve "socket was closed by pool". This is miner problem or my interner connection issue?
Probably Yes its connection issue not the miner, sometimes problem is on the pool itself.
whats your pool?
post your .bat file
setx GPU_FORCE_64BIT_PTR 0
setx GPU_MAX_HEAP_SIZE 100
setx GPU_USE_SYNC_OBJECTS 1
setx GPU_MAX_ALLOC_PERCENT 100
setx GPU_SINGLE_ALLOC_PERCENT 100
EthDcrMiner64.exe -epool eu1.ethermine.org:4444 -ewal WALLET.rig1 -epsw x
March 16, 2018, 11:47:12 PM
Hi. With Claymore 11.5 i often recieve "socket was closed by pool". This is miner problem or my interner connection issue?
Probably Yes its connection issue not the miner, sometimes problem is on the pool itself.
whats your pool?
post your .bat file
Jump to: