Topic: Claymore's Dual Ethereum AMD+NVIDIA GPU Miner v15.0 (Windows/Linux) - page 612. (Read 6590757 times)

add -"dcoin dcr" option and add "miner1" as an worker in dcr.suprnova.cc site.

Try to change "miner1" to your "worker name" and write "-dpsw x" command too.

Appreciations here (ETH) 0x57137Aecf8d11b60522d53eD2b0Baf0215FD0b61

Hi all,

I have done quite a lot of research but haven't found a resolution to this problem - hopefully someone can help me on this.

Mining rig:
3 x MSI RX 580 8 GB cards (not Armor/Sapphire etc) - has single 8 pin power connector, single fan, no BIOS switch
Windows 10
SSD with 4 GB RAM
Installed Claymore version 10 Dual miner - Ethereum mining running well and stable but issue is that I cannot get Decred to authorize

Getting:
Socket was closed remotely (by pool)
 DCR: Job timeout, disconnect, retry in 20 sec...

Start.bat command:
EthDcrMiner64.exe -epool us1.ethermine.org:4444 -ewal 0x31be6e3~~~~~~~~~~~~~~~~~98c6.miner2 -dpool dcr.suprnova.cc:3252 -dwal myloginname.miner1 -dpsw 2222 -allpools 1

Troubleshooting:
- I have changed the password for the worker to match Claymore (x, 2222, other) - unable to leave the password blank for the worker
- verified Suprnova login (though there is an "x" in the active column?)
- attempted both weblogin name as well as email address used for profile
- added port 3252 in Windows firewall for both incoming and outgoing connections
- added miner to router's DMZ
- tried with/without stratum+tcp
- attempted -dpool dcr.suprnova.cc:3252 -dwal suprnova.1 -dpsw x -allpools 1 as detailed on suprnova's website
- though I can't recall which other decred pools I used but all got the same error

Any assistance or suggestions would be greatly appreciated!!

https://www.monitortests.com/atikmdag-patcher-1.4.5.zip for bypass pixel clock limit on AMD drivers.

what drivers are u using, if you are using something other then block chain drivers then that is the answer as ETC is on a higher DAG then ETH

Tried to mine cryptonight with rx Vega , got error:AMD HSA Code Object loading failed

Then claymore Said unable to allocate memory.
Got 32gb ram and alot of virtual.

My 1080 ti works fine (not claymore miner)

I can't even begin to process the logic involved in your first statement. So I decided to create an account.. to call you a fool. Touche sir, you grabbed my attention.

3 rx470   gpu clock: 1150 volt: 900, mem clock 2000 volt: 900

Claymore 9.8 works fine (0 mem errors)
same settings via .bat file
Claymore 10 gives me a lot of memory errors in hwinfo
why?

Thanks for your answer.

Hi!

I'm using Clay 10.0 and block chain 23 aug drivers now and experiencing stability problems. No memory errors 27-28 mh per card (I have rx570/580 with Samsung and Hynix), but still having messages about hanging opencl.

I use -cvddc 900, -mvddc 900, -cclock 1150 (around it), -mclock less than 2000, ram overclocking gets my millions of errors in HWInfo.

Sometimes it freezes rig, so - r 1 options useless.

Memory straps are the same, I have used them with clay 9.8 and Crimson July 12. Rigs were working for hundreds of hours without rebooting.

What should I do to get more stability and mem freq?

Better hashrate (~130MH/s with 4xRX580 in dual mining), but undervoltage doesn't work anymore ... only way is to lower clock voltage with AfterBurner, but it's limited (-100mW max ... and only for core voltage). Temp and fan speed are also read to "0" by Claymore, ans HW read right temp and speed, but read a insane core voltage "98628262,12V" si maybe a link with Claymore. They have to be updated for this driver ?
Any problem on my side or I have to wait a Claymore update ? Thanks
Claymore V10.0 and Blockchain driver.

thanks for the info but what is the "+ patch" in your post ?

TYIA

Well it says pretty clearly in the changelog for v10 that its not complete.

"- fixed issues with voltage/clocks management for latest AMD blockchain drivers (not completely)"

when i mine ETH with 5X RX470 i am getting 130 mh/s

but when i mine ETC with same 5X RX470 i am getting only 115 mh/s


why 15 mh/s  diffrence?

Hit the number keys to disable the GPU's in claymore before you close if you are having crashing issues.

It is disabled. Please do not think I am so new in rigs.
Moreover, I never seen when Radeon drivers enable crossfire for mixed cards automatically. Yes, if you install a pair of identical cards, driver will enable crossfire by default but never in my case.
And when crossfire enabled, it is easily reconizable in Claymore's monitor - paired cards will show equal temperature and fan speed.
As well as I am talking about mixed rig only, 470-480 rigs do not have the problem described above and the problem occured only after moving to "blockchain driver" and Claymore 9.8.

Hello,

I was wondering if anyone else has the problem of when you close the miner, it lock up the screen and I have to reboot. Windows 10 AMD processor

I dont have "AMD Crossfire" option

Any other ideas?

Big nvidia boost was 9.6 so anything after that and you'll see it.

You need to disable "AMD Crossfire" from Radeon Settings: go to Gaming -> Global Settings -> Global Graphics and you find here the option