Topic: Claymore's Dual Ethereum AMD+NVIDIA GPU Miner v15.0 (Windows/Linux) - page 502. (Read 6590757 times)

Hows your device manager settings? are they in normal without any errors? I would suggest to do the test one by one, exclude nvidia first and let the AMD to be installed and do the mining. this is to isolate what may be the culprits. riser may look fine but in a while they seem to be faulty try to replace them if you have any extras.

I did test the riser it works fine, the card tested with no OC same problem.

Hi all!

I recently had an issue with my miner and fixed it so I wanted to share the steps I took. We had a slight snow storm and power was going on and off so I shut down my computer. Later when I turned it on Claymore's miner would just crash due to a Watchdog error. Then my screen started looking funny (some pixels and different task bar colors).

Here;s my setup:
4x RX 480 8GB
32GB DDR3
2TB HDD
AMD ATHLON IIx4 630 CPU
MSI 970A-G43 MOBO

I unplugged everything and did a hard power reset. I unplugged EVERYTHING! Drives, power, wall motherboard, psu. I plugged everything back in and turned the computer on. One card was no longer showing up. So I took all but one card out and started testing them all on one pcie riser. Every card worked.
Same test, next riser. They all worked.
Same test, next riser. They all worked.
Same test, last riser. None of them worked.

Ok so I thought maybe a riser went bad. Luckily I have a few spares so I plugged a new one in and same result. Nothing worked.
It must be a power issue then, I thought.

So I started swapping the lines to power each graphics card while they were on risers I knew worked.
All the GPU power lines were fine.
So then I started swapping the riser power lines and low and behold: one power line, from the psu to the riser was bad.
I connected them all back up and it works fine now.

I think during the power outages there was a surge through a line causing it to no longer work.

Hope this helps someone out there.

Try to lower your clocks, and adjust your -tt to at least 75, are you using risers? the burn smell could be your riser as the AMD rx580 is a power hungry unlike the GTX, check your cables from riser to PSU, probably there burnt connectors out there.

Hello,
I was mining with claymore 9.8 & 1 card NVIDIA 1060 and I did some OC with msi afterburner, I was getting BSOD & freezing on windows 2-3 times a day, recently I had bought 2 XFX 580 4G & start mining with Claymore 10.2 using the command -tt 69,69,0(0 for NVIDIA), I lowered the OC of the gtx 1060 but i was still getting freezing in windows but no more BSOD, until yesterday the rig mined all the night without any problem then in the day the rig crashed on windows and the fan of the cards stopped and i smelled burning smell coming out from one of the XFX 580, i restarted the pc & I found that the card can be detected with gpu-z but when i start Claymore i got BSOD so the card is dead, now i am afraid that the same think will happen to the other cards, so I am looking for help to avoid that problem will happen again.
i was using "-dcri 30,30,38 -ftime 10 -cclock 1100,1100,0 -mclock 2045,2035,0 -cvddc 900,900,0 -mvddc 900,900,0 -tt 69,69,0" i got eth +decri 29,29,21 & 875,875,274.
Regards

Is the performance of mining with claymore on ATI graphic cards comparable to sgminer?

Managed to reset everything and start again, now pulling 840w-845w.  alot better than the 960w it was pulling. And managed to squeeze out another 1MH/s and lower temps a degree or 2

I see huge influx of people here and some are way more clueless then the others...
so here some tips for you there, mining with Polarises on Win10:

1. make sure you have sufficient power on your PSU(10-20% spare from the peaks)
Note: properly configured polaris should be just below 150W on dual ETH
2. molex power to raisers is probably the most common point of failure (bad/unstable hash, card disconnects, crashes, and such)
Note: it seems there's no such thing as bad raiser, only bad contacts in power connectors
3. disable all you can via Group Policy, especially windows update and drivers update
4. use latest DDU and install 17.11.1 driver, pointing manually from Device Manager to the dir with unzipped driver
Note: Do not use blockchain beta driver as it has some issues with latest API and also doesn't support more than 8 AMD GPUs
5. set compute mode for all GPUs. here's the example for dev0 (special thanks to @ober for that):
6. Flash bios mods with tighter straps. Look for PolarisEditor v164 with "one click modify" bios timings and use atiflash without gui
7. Use overdriven tool (RX5* Polarises normally do well on 1180 & 1950-2150 @ 850mv, RX4* may need lower core freq)
Note: do not use AB.
Note: do not use claymore for core&mem&fans control
8. Use sysinternals autologon and built-in MS task scheduler
9. make reboot.bat and set "-r 1", as restarting miner most likely will NOT work

Bonus:
* I'm using RealVNC for remote access - flawless
* you may find useful: display dummy dongles and usb watchdogs
* and as always openVPN is your friend

If I just made your life easier and increased your profits PM me and I'll give you my ETH address:)
...or you can just find it in my profile yourself

Hey Claymore.

I just have to say thanks for this GPU Miner.
I have been using the miner for over a year now without any troubles 

So I'm pretty new to mining and could use some help

I have ran 5 of my cards successfully but when I added the 6th I got this

09:39:52:657   19a4   º     Claymore's Dual ETH + DCR/SC/LBC/PASC GPU Miner v10.2      º
09:39:52:674   19a4   ÈÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍͼ
09:39:52:679   19a4   
09:39:52:895   19a4   ETH: 5 pools are specified
09:39:52:895   19a4   Main Ethereum pool is eth-us-west1.nanopool.org:9999
09:39:52:911   19a4   AMD OpenCL platform not found
09:39:52:911   19a4   cudaGetDeviceCount failed (35, CUDA driver version is insufficient for CUDA runtime version), probably no CUDA devices
09:39:52:911   19a4   No NVIDIA CUDA GPUs detected.
09:39:52:926   19a4   No AMD OPENCL or NVIDIA CUDA GPUs found, exit

6x XFX RX580 8GB
Win10-64Bit-Crimson-ReLive-Beta-Blockchain-Workloads-Aug23

Any help would be greatly appreciated

hi guys
my RX 470
core clock set in bios to 1169
i change it with -cclock option to 1150
but does not change
any reason?
any solution?

-mclock works fine

Yep, that's the way I prefer it - changing only the mem timings in bios. Volts and frequencies (which actually may vary between different algos)
However you still should be able to override gpu config from overdriven tool disregarding the bios values.

Do clean windows 10 install and  17.11.1 driver
Or:
Try OverdrivenTool

Do you have a separate install directory for each instance?

i have problem (as always)

I decided to mix gpus

6x amd + 8x nvidia

I plug 6x RX570  + 4x1050ti (10gpus)

all works fine without problems (z270-p asus + pci-e multiplier)

when I add 11th GPU after start claymore its freeze (PC works fine, just claymore freeze, if i try to close it then entire PC will freeze)


so
i tried to open 2x claymore and use command -platform 1 for amd and -platform 2 for nvidia, but only one can work (first claymore works ok, but when I run second claymore i am getting (cannot create DAG error/gpu failed etc

of course i have to use -mport 0 command bcoz without it i got sever10048 error

You could start here for tips on how to mod your BIOS https://bitcointalksearch.org/topic/guidetutorialxmrethhow-to-mod-bios-rx470570480580-vega-mining-1954245

To reboot miner read the readme file about the "-r" switch and if PC reboots then to auto start miner use this https://support.microsoft.com/en-gb/help/4026268/windows-change-startup-apps-in-windows-10

hi guys,

still experiencing the same issue. claymore refuses to set clock, core and voltage settings on my gigabyte aorus 580 xtr cards. i have to use afterburner which isnt reliable. every now and then my cards default to normal clock settings which affects the hashrate and power use.

Any ideas why this might be the case?

running win 10 pro, and claymore 10.2.

Thanks

The key should be to add an offset in VoltageObjectInfo (google it)
https://forum.ethereum.org/discussion/12970/undervolting-through-hex-editor-simplemining/p3
http://www.overclock.net/t/1621800/would-appreciate-help-in-finding-editing-the-voltage-offset-in-the-rx480-bios/0_100#post_26047272
https://bitcointalk.org/index.php?topic=1465128.20

Well, power consumption is important, its pulling lot of power on wall without undervolting.
Flashing bios for higher clock is what I wanted but not really understood how to do, specially undervolting.. Say I want to undervolt 900 mV , but numbers there are 65423 ...

Any link or indepth guide for bios modding ?

Hello.
Why does monitoring reflect information selectively? 
Miner Manager ver. 3.5 | 3.6
https://i.imgur.com/ADRsf0t.png