Topic: Claymore's Dual Ethereum AMD+NVIDIA GPU Miner v15.0 (Windows/Linux) - page 752. (Read 6590718 times)

Hi everyone
I am new to mining. Setup my rig with 5 x GTX1060 (the 6th GPU is on back order) and I am using Claymore V9.7
3 x Asus Strix OC 6GB
2 x Gigabyte Windforce OC 6GB
They are all using Micron memory chip. I am getting default Eth hashrate of 19.xMH/s. Power usage = 560W.
I tried to OC using MSI AfterBurner and have attempted various settings but with success. It keeps crashing.

1) PL 100% Core 0 Mem +650
[Artifacts on screen]

2) PL 100% Core 0 Mem +600
[Getting 23.2 MH/s - Power usage 610W. No ideal as I am only running 5 cards. If I add the 6th, power usage will be approx. 720W - this is bad]

3) PL 65% Core 0 Mem +500
 [Getting 22.5MH/s - Power usage 430W. Stable for 2 hours but crashed when I open Internet browser. Attempted 2nd time and it crashed after 15mins. Crash as in I am getting 1.44MH/s on GPU1 and GPU3 after some time and the PC is VERY SLOW, UNRESPONSIVE and had to reboot. Checked log and found error - 1bcc   NVML: cannot get fan speed, error 999 (an internal driver error occurred]

Not sure how to optimise to get a stable mining performance. Checked forum and saw some ppl did it with PL 65% Core +200 Mem +900-1000 but this didn't work for me. Immediate crash.

How do you guys optimise? I suppose we start from increasing the Mem, then find the sweet spot and adjust the PL leaving Core untouched? I can get good results but they wont last, probably crash after 5 to 20mins.

Do I have shitty cards?

its off, just checked it again.
What could it possible be? Can this scenario cause any problems:
I have asrock h81 pro btc, , i have plugged in the monitor to the 16x slot, that cant be it right?
When i added the cards, i just added them all and then installed drivers. All cards are shown in device manager.
Anyone recommend doing a Clean with DDU, then starting with 1 card on the 16x or the closest pcie from the cpu, installting nvidia drivers and then adding 1 card and rebooting?
Cant really Think this is the problem but i am desperate...

Check your power management policy (Start -> Run -> powercfg.cpl) and make sure that PCIE Link State Power Management is disabled (Off)

I noticed just Before the "cannot get current temperature, error 999" ( and 1 cards shuts down ), i see a small windows loading circle.

Anone knows anything to disable in windows 10 that can be the cause of these crashes? I have already disabled windows update.

already and nothing , same error.

I think I'm at -ethi 1 or 2

Thank you for this new version and special thanks for the nVidia boost :-).

I am interested in tithing my profits but would like to automate.  I am thinking that dual mode would be a good way to accomplish this.  But I do not see setting to allocate 10% resources to a specific pool.  I was thinking that -dcri option would be my best option, but am not sure.  I guess I could specify one GPU to a specific pool.  But I have 6 GPU's and that is more than I want to tithe (1/6 is about 16%). Are there any ideas or suggestions for me to accomplish this? TIA!

realized i was using v9.2    updated to 9.7 we will see if i have the same issues.

o ok, i didn't know if you had any other parameters in there. thanks

Did you use "-ethi 0" or something else ?

EthDcrMiner64.exe -mode 1 -epool eu2.ethermine.org:4444 -ewal 0xb3f095c58ccefe4a596350fbf1397720857f3971.YourRig -epsw x -allpools 1 -allcoins 1

GPU & other settings were posted just a few lines above, also you can read my blog for all the details: jukesays.wordpress.com

I've found that with my GTX 970, running about 90% load with the miner means I can do most other tasks without issue.  Can even play some old games like the first Far Cry and Half-Life 2.

increase your virtual memory

Claymore 9.7

MSI 1060 3G Hynix Memory
+200 Core
+550 Memory

Giving me about 19.5 Mh/s on ETH SOLO.
So this version is not for Hynix Memory?
Only who have 20+ Mh/s is Samsung

can you share your settings? i have 5 1060 and it locks up after 20 min of use and i need to reboot the system. i can only pin it on the miner.

Is there s way to limit GPU usage when using Claymore ? I own 1080Ti, but for mining purposes I would like to allocate only 50% of its resources. I need the other 50% of power to be able to work on my PC.

Found "-ethi" command, but putting "-ethi 0" in .bat file still doesn't seems to fix the issue. I get lower hash rates, but only by a bit (~30Mh/s) and GPU usage is at about 95%.

for Nvidia, do we still need to use the old 372.54 drivers or the latest 384.76?

Thanks

You are correct, it is ignored for Nvidia cards currently. I will fix it in next update.


It's bad idea because remote manager won't know that new port is used. Just set "-mport 0" if you don't use remote manager.

Yes, such option would be great.