Topic: Claymore's Dual Ethereum AMD+NVIDIA GPU Miner v15.0 (Windows/Linux) - page 468. (Read 6590565 times)

Claymore said above -ethi 8 increased shares is minimal, but stale shares increases a lot, so it's not profitable to use higher than -ethi 8

same problem with 12GPUs
over internet (same with local network) http requests takes more than 30 secs to respond

Can't say for high numbers, but the resulting difference between -ethi 8 and -ethi 16 is the no of submitted shares and hence - effective pool hashrate. I recall 16-vs-8 intensity gives 5-7% more shares per hour.

trojan detected in latest 10.5 downloaded from mega / kaspersky in cuda 9.1 folder

Have you turned off auto update in Ubuntu? If not, then you probably got an update for the Spectre/Meltdown. Fix -> https://bitcointalksearch.org/topic/m.28208217

I have also try to mine DCR while mining ETH

But the hashrate of ETH has falling arround 20MH/s at my 550MH/s rig,
But the main problem was the rising to the Power of my 2 RIGs what is needed,
So it overloads my PSU massive, and I also get much more heat out and need additional fans too
So I Stop mining DCR and has stayed on ETH only

When I have the settings to mine max amount of DCR with no effect on the ETH hashrate, the extra DCR covers the whole electricity bill. So it definitely makes sense to dual mine.

I'm having a big problem that I cant solve!
1 rig with 6 GPUs and simplemining cant set clocks. stay in 300/300 and mines 4mhs.
No matter if I put something in overclocking table, the clocks stay in 300/300
There is this error before it starts mining:
AMD ADL library not found.

Can this be the problem? How can I solve it?

Thanks!

Hardware:
Asus B250 Mining Expert
4Gb ram
1x rx480 8gb G1
5x rx580 8gb asus ocdual

I run Ubermix straps for samsung mem, 2000 seems to be about stable limit, some I lower slighly.
I use custom Elpida strap, 1750MHz cards easily run about 2020MHz XMR miner all day long.
I using a custom Hynix strap, on 1750MHz cards, seems stable at least to 1970MHz XMR, testing in steps.

In my experience, Ethash algorithm requires a bit lower (20-40MHz) lower clock than CryptoNight. I don't know how people supposedly run stable above 2050 though with good straps.

Install HwInfo64. Check for memory errors and PCI link errors. Can also show as high system interrupts for corrected errors in Task Manager, and Windows Event Viewer.   I have issues with with some boards, recently Biostar TB250 having PCIE issues.  One I had to disable miner mode, only runs 4 cards anyway. Another I had to just replace board, would get PCIE link errors and restart dozens of times per day. I gave up on your motherboard a while ago because they seemed to not work with the Blockchain Drivers. I was using Aug 11 drivers so maybe the Aug 23 ones work?  Had issues on 2 of about 15 Biostar TB250s. My favorite board is Asus Prime Z-270A, never had an issue, does 8 cards, also 270P version does 7. Good mining mobos high price now though.

Until figured out, I would leave memory @ 2000MHz

If you think you will have a stable rig with modified timing and over 2000 mclock that will never happen. Start at like 1900 mem clock and move up until you find whats stable.

is the  "remote management" set to read-only on all versions  or version x.xx and later? I just want to check to see if i need to upgrade to latest ASAP

Which Brand did you get? Were you able to mine at all with the faulty ones?

How I can find out, witch riser are fine ?

miningguy

This happened to me same build.

I had to build 4 rigs at the same time.

1st one, 30mins build to start mining - EASY (using some risers I got 6months ago)

Next 3 rigs took me 2 weeks!

I bought 30 new risers and 28/30 were duds!

So got some branded ones from China, plugged them in, all working no issues!

Quality of risers has deteriorated massively.

Hi Guys.

I am really desperate. Ive been trying to set up my mining rig for 7days + now and i just can't get it stable, no matter what i try.

Hard ware:
- 6x rx580 8gb red devil hynix memory
  tried anoraknophobias bios & frosty s bios
- 1x rx580 8gb red devil samsung memory
   anoraknophobia bios
- msi z170a gaming pro carbon motherboard:
  newest bios
  above 4g encoding enabled
  pice max link speed gen2
  pice latency auto
  display output by integrated graphics by default
  windows whql enabled
- 60 gb ssd
- Intel Celeron Processor G3900 (2,80 GHz, 2 MB Intel Smart-Cache)
- 8gb RAM
- 2x 800w be quiet PSU (one for mobo + 3 cards, other for 4 cards)
- windows 10 home version 1703, build 15066.786
- aug. 23rd blockchain drivers
- claymore miner version 10.0
- virtual memory 30400mb

i have tried claymore settings:
-1100 cvddc -1100mvddc -1100 cclock -2100 mclock
to
-850 cvddc -850 mvddc -1250 cclock -2250 mclock

and pretty much anything in between in claymore v10.0 and v10.5, within 10 minutes max i get the infamous "GPU X hangs in OpenCL call, exit.." Error on one card after the other.

whenever i start the miner without any clock and voltage settings, the cards also ramp up to 30mhs, but fail within 10 minutes, same OpenCL error.

I have tried to troubleshoot:
- temperature is not an issue, all temps stay within 50-70 °C
- power doesn't seem to be an issue, i have added another 600w PSU to power one card of each of the other PSUs, didn't help
- i have fiddled with the pcie max link speed settings(gen1, gen2)
- i have swapped all risers
- CPU is not getting to too hot, although under heavy load during DAG creation & weirdly sometimes when one card fails or when i exit the miner
- I have tried the method of adding one GPU, testing it with the miner, adding another and so forth:
  inexpliccably, the rig seems to run fine with up to 3 GPUS, when i add a 4th GPU, the 3rd GPU starts failing. (not the newly added one).
- i have reinstalled windows several times
- tried to add claymore options -gser, -eres

The only thing i can think of is that maybe the CPU could be a bottleneck, since its under heavy load sometimes and can make the rig a bit slow for about 10 seconds. Still, the errors arise when the CPU is just sitting at 15ish %.


Another weird thing is that, whenever i let each card run individually, they all run fine with no issues for more than one hour. So i guess it cant be a riser issue.

But here comes the worst part:
I have set up another rig with 7x rx580 sapphire nitro+ 8gb hynix (anorak bios), same motherboard, same CPU, same amount of RAM same driver, same virtual memory, same version of Windows, claymore v10.0 and it happily runs on 850cvddc 850mvddc 1250cclock 2250mclock churning out a little over 200mhs for the past 3 weeks without any problems.

Its driving me nuts, seriously  

Please guys, if you have any idea what i could look for to get this baby stable it would be much appreciated.












  

i take it the latest amd drivers have yet again stopped claymore / afterburner from looking at or adjusting the fans ect?
i love the way amd forces their crap on us grrrrrrr

Thank for  the reply....
Well done

1. CVE-2017-16929 works only if you specified -mport > 0 and don't use password (-mpsw). Also it can crash miner:

From History.txt:

...
v10.2
- fixed critical issues in remote management feature (attacker could crash miner even in read-only mode).
...

That was the last known security issue and it was fixed in v10.2 and above.

Here is my opinon:
Yes my miners have "remote management" feature that allows administrators to upload files, restart miners etc. It's not a vulnerability itself. Also I know that this feature can be used by attackers, therefore by default miner uses "read-only" mode which ignores all commands except "miner_getstat1" (I mean recent versions of my miners, very old versions have no "read-only mode"). User must specify "-mport" option and positive port manually to enable potentially dangerous commands. Also I added "-mpsw" option that allows users to set a password for remote management. Also miner shows warning during startup if "read-only" mode is disabled and -mpsw is missed. Also I added necessary information to Readme file. And, of course, it's bad idea to make these ports public, they must be used in local network only, "-mport" option can expose the port to specified network interface only.
To make this feature dangerous, user must do the following:
1. Disable "read-mode", ignore miner warning.
2. Ignore "-mpsw", i.e. don't set a password, again ignore miner warning.
3. Expose the port(s) to internet.
So, in my opinion, I did everything to make this feature safe as much as possible, but some users have no idea/knowledge what they do when they change default settings.

Hello,
is it possible to get 27,5 mhs (Rx470) with linux? i only get <20 mhs (after DAG 160)

lg