Claymore's Dual Ethereum AMD+NVIDIA GPU Miner v15.0 (Windows/Linux)

[LexLima]

Thanks everybody for help,

I will sure try this tool but as i am in business of selling complete rigs customers i sell to are used to simple solutions (few clicks and mining starts).

I am afraid if i start to use this tool which i am sure i will learn how to use, i wont be able to give them support as i do that as a part of my service.Everybody i sell to are used to MSI because mostly those are the people that dont know so much about computers but wants to mine.

About MSI AB i use latest version with latest AMD drivers.

About that windows update i disabled it (first thing i do), i also disabled everything in Task Scheduler which is connected to Windows Update.

What i notices about Trixx is that before he shuts down it seem like something is happening in windows that is causing this shutdowns. Like some programs are opening by themselfs.

Any other soultion guys?

overdrivenNTool is easy to use, it might not have a fancy UI but it gets the job done.

You just have to run it once before the miner starts to set the clock and voltages, like I mentioned in my previous reply.

Trixx and Afterburner run in the background all the time, and that might be the reason of your rig crashing.

Eliminating those programs will increase stability.

Try it out, if the problem persists, let us know

I am guessing thay cause this problems but on Windows 10 old version eveything was fine. The strangest thing is that i have one rig that was updated by force on Windows 10 1709 and still have Windows Update assistant, I literally didnt change anything and it is working fine for 2 months. Problem began when i started to install fresh version of Windows on my rigs. I read somewhere that if I use miningtweaks. bat for removing Windows Apps (Cortana, Xbox,Mail, People...), Windows UI starts messing up (Windows Defender pop up, Control  Panel Pop Up) but as i watch and monitor rigs i cant seem to catch that moment. Only when i sleep it happens. What an irony i cant sleep at night ))

[1713567319]

1713567319

[1713567319]

1713567319

[1713567319]

1713567319

[mu_enrico]

Hi people,

Can anyobody help me i am going mad.

I have 5 rig with 6 x RX 580 Special Edition. On all five rigs i have a fresh install of Windows 10 Home 1709. Fisrt my problem started when i could make MSI AB work with all 6 cards (it only recognize 3 cards) so i had to switch to Trixx.

I installed Claymore Dual Miner V 11.3, 11.4 and 11.5 and on all machines same problem. I watch them for few hours and everything is working just fine.No memory errors no nothing everything is perfect. Then suddenly overnight or in the morning some of them shutdown randomly(Blue screen). First i thought it was because of overclocking then when i realize it isnt (beacuse i set them to default), i switched to 11.2 because i read here it is the last stable version.

Now i have new problem. Trixx sudennly bugs out randomly on different rigs at a different time, stop working and all my cards stops working too.

I am desperate in need of help. I tried everything dont know what is the problem.

What i did before this:

Install Windows 10 64 bit Home V1709
Install new latest AMD drivers (compute mode)
tried MSI Ab but cant read 6 cards
uninstall all programs i dont need via CC cleaner and Miningtweaks. bat like Cortana, People, Xbox etc.
Set virtaul memory, power plan all other stuff that i was doing on the previous version of Windows 10. The reason i cant go back to old version of Windows is because it keeps updating.

So my question is what should i do to stop shutdowns?
How can i use MSI with new AMD drivers and Windows 1709?
What Claymore should i use?
Why is Trixx suddenly stop to work?

Hope someone gives me a solution.

Thanks

Me also experience similar issue after windows update,

What i do:
Uninstall AMD software via control panel
Uninstall (and delete gpu driver) via device manager
Restart
Reinstall newest AMD driver -> switch to compute mode
Install Trixx (since Afterburner seems not compatible after this update)
Restart
All good

[Tr1n1tY]

Hi, @Claymore! Since 11.2 Windows miner began to crash with no reason. First I thought it's becuase of new kernels, but 11.5 with -oldkerneks 1 does it also. It's very rare, maybe once a week - once a month on different rigs, but it appeared in 11.2. Screen shows standard error window about "application will be closed", Application event log has a record Application Error, code 1000. I'm catching this using scheduler and starting bat again, but I guess some bug came into 11.2...

Mine is also crashing on Linux after about 70 hours, did not crash before 11.4

[MinerosYV]

driver 9.14.10.01293, Crimson Beta for the GPUs
4.4.0 Beta for AF
Claymore 11.5
Remember the patch for the drivers
 Get some 31.4 for eth and 1700 for verge

Got a few Nitros 580 8 gb

Thanks for the answer but i alreadu tried that. For some strange reasong those drivers wont work for me because i can only get 22Mhs. Once i move to new drivers and set Compute mode they jump to 28 MHS. It is like its not working with none of my rigs. Am i doing something wrong? Bios moded, patched, everything by the book

You must be doing something wrong

the patch needs to be aplied before retarting for new drivers

mines have Hynix memory, they have been working for some time now

Usualy when it goes down to 22 means Windows updated and in the process updated the GPU driver

As far as Defender goes an exclusión for the foldres work

[Parade]

I'm using -cclock and -mclock for my AMD Miner. It was working well for v11.5 until I decided to restart my Windows.

After Windows restart, the Miner is hashing at a slower rate. When I opened Trixx, it's showing the default core clock and mem clock, not the ones in my .bat files.

I did not perform any updates. Does anyone has any idea how to resolve this?

[big al]

After installing the latest AMD drivers you need to run pixel clock patcher for any cards that are BIOS modded

[Vann]

I'm using -cclock and -mclock for my AMD Miner. It was working well for v11.5 until I decided to restart my Windows.

After Windows restart, the Miner is hashing at a slower rate. When I opened Trixx, it's showing the default core clock and mem clock, not the ones in my .bat files.

I did not perform any updates. Does anyone has any idea how to resolve this?

Windows Update could have modified the drivers. Do a DDU in safe mode to remove the current driver, reinstall the latest Adrenaline drivers. If you have RX 470/480/570's with a modded Bios, run the pixel patcher to bypass the driver signature check. You also need to toggle to Compute mode from Graphics mode in AMD settings.

https://www.monitortests.com/forum/Thread-AMD-ATI-Pixel-Clock-Patcher

https://support.amd.com/en-us/kb-articles/Pages/DH-024.aspx

To keep Windows Update from modifying your drivers, disable drivers updates in the Group Policy Editor.

Windows key+R ==> Type gpedit and hit Enter ==> Navigate to: Computer Configuration ==> Administrative Templates ==> Windows Components ==> Windows Update and set 'Do not include drivers with Windows Update' to enabled.

[Parade]

Windows Update could have modified the drivers. Do a DDU in safe mode, reinstall the latest Adrenaline drivers. If you have RX 470/480/570's with a modded Bios, run the pixel patcher to bypass the driver signature check. You also need to toggle to Compute mode from Graphics mode in AMD settings.

https://www.monitortests.com/forum/Thread-AMD-ATI-Pixel-Clock-Patcher

https://support.amd.com/en-us/kb-articles/Pages/DH-024.aspx

To keep Windows Update from modifying your drivers, disable drivers updates in the Group Policy Editor.

Windows key+R ==> Type gpedit and hit Enter ==> Navigate to: Computer Configuration ==> Administrative Templates ==> Windows Components ==> Windows Update and set 'Do not include drivers with Windows Update' to enabled.

I've disabled Windows Update and driver updates. No update should have been applied during Windows restart.

Anyone else facing this issue?

[big al]

So my question is what should i do to stop shutdowns?

Try:

- Disabling quickedit mode for console
- using -oldkernels 1
- using v11.0
- Do NOT run GPUZ or HWinfo while mining

How can i use MSI with new AMD drivers and Windows 1709?

- The newest drivers work fine you need to run pixel clock patcher
- Make sure the drivers are minimal install and you do custom install, unclick everything except the drivers

What Claymore should i use?

Single mining 10.6
Dual mining 11.0 (might fluctuate), 11.5, or 11.5 with -oldkernels 1

All are stable on my BIOS modded rx580s but each has different results

Why is Trixx suddenly stop to work?

Use OverdrivenTool instead.  Far better solution

Hope someone gives me a solution.

Thanks

We hope so too!  Report back on testing.  Many are having problems with rx580s.

[big al]

Update on my test:

Everything is good except for one small issue.  I am getting 1.5% incorrect shares.  But then I realized I had HWinfo open all night!  This is a tripling of incorrect over v11.0 or v11.5 with oldkernels.  I closed HWinfo and bumped up voltage 5 mv and will retest.  But it seems perfectly stable.  I set it to restart Claymore at below 62 mhs which is very aggressive and it ran all night long.

What you mean by "Disabled all power states in the video card BIOS except the top"? Can you share some picture?

http://www.overclock.net/forum/26417747-post3402.html

See point #2

If you click on P0 to P6 it will turn grey and be disabled.  Now your card will only run in the highest power state.   But beware this increases idle power consumption ten fold!  Do the same for memory.

Does it actually do anything? I don't know.

[Vann]

Hi people,

Can anyobody help me i am going mad.

I have 5 rig with 6 x RX 580 Special Edition. On all five rigs i have a fresh install of Windows 10 Home 1709. Fisrt my problem started when i could make MSI AB work with all 6 cards (it only recognize 3 cards) so i had to switch to Trixx.

I installed Claymore Dual Miner V 11.3, 11.4 and 11.5 and on all machines same problem. I watch them for few hours and everything is working just fine.No memory errors no nothing everything is perfect. Then suddenly overnight or in the morning some of them shutdown randomly(Blue screen). First i thought it was because of overclocking then when i realize it isnt (beacuse i set them to default), i switched to 11.2 because i read here it is the last stable version.

Now i have new problem. Trixx sudennly bugs out randomly on different rigs at a different time, stop working and all my cards stops working too.

I am desperate in need of help. I tried everything dont know what is the problem.

What i did before this:

Install Windows 10 64 bit Home V1709
Install new latest AMD drivers (compute mode)
tried MSI Ab but cant read 6 cards
uninstall all programs i dont need via CC cleaner and Miningtweaks. bat like Cortana, People, Xbox etc.
Set virtaul memory, power plan all other stuff that i was doing on the previous version of Windows 10. The reason i cant go back to old version of Windows is because it keeps updating.

So my question is what should i do to stop shutdowns?
How can i use MSI with new AMD drivers and Windows 1709?
What Claymore should i use?
Why is Trixx suddenly stop to work?

Hope someone gives me a solution.

Thanks

MSI Afterburner won't work with the latest drivers and more than 3 cards. There is no Group Policy Editor in the Home version of Windows 10, so I don't think you can disable Windows Update driver updates. RX 580's don't need the pixel patcher with a modded Bios, only the RX 470/480/570's.

I'm using v11.5 with the latest Adrenaline drivers in Compute mode on a 13 card RX 480/580 rig with ZERO problems or invalid shares and it's very stable. You can also set the overclock/undervolt/fans directly in the Claymore config or .bat file.

https://bitcointalk.org/index.php?topic=1433925.msg32024794#msg32024794

[big al]

Are you running oldkernels? single mining?

[RYXES]

Anyone else seen a progressive amount of stale shares in the last hour or so?

One of my rigs has hit 16% stale shares in the last hours.

Nothing has changed in my rig so I can assume maybe it has something to do with Ethermine perhaps?

[Vann]

Are you running oldkernels? single mining?

No. These are the only parameters I use for the Claymore config. Dual mining was also very stable with only a small amount of invalid shares. 1-2 per 1000 shares max.

Code:

-platform 1
-mode 1
-gser 0
-etha 0
-ethi 8
-ftime 3
-r 0
-cclock 1200,1150,1200,1200,1200,1200,1200,1200,1200,1200,1200,1200,1200
-mclock 2050,2050,2050,2200,2100,2100,2100,2220,2220,2220,2100,2220,2050
-powlim -10,-5,-10,-15,-10,-15,-10,-10,-5,-10,-10,-10,-10
-cvddc 950,950,950,850,900,850,850,900,900,900,850,900,900
-mvddc 950,950,950,850,900,850,850,900,900,900,850,900,900
-tstop 75

[big al]

No. These are the only parameters I use for the Claymorte config.

Thanks.  So it's possible some of the issues are more related to dual mining?  I haven't tested single mining.

BTW new optional drivers came out yesterday (18.3.3) and they did seem to need pixel clock patcher even with RX580s.  OverdrivenTool wouldn't work without the patch. Or maybe user error? That's always possible.

[Vann]

No. These are the only parameters I use for the Claymorte config.

Thanks.  So it's possible some of the issues are more related to dual mining?  I haven't tested single mining.

BTW new optional drivers came out yesterday (18.3.3) and they did seem to need pixel clock patcher even with RX580s.  OverdrivenTool wouldn't work without the patch. Or maybe user error? That's always possible.

I just updated another RX 570/580 rig to the 18.3.3 drivers. Without the pixel patcher only the RX 570's showed up with the code 43 error in Device Manager. There was no change for me dual mining between v11.3 and v11.5, which was also very stable.

https://bitcointalk.org/index.php?topic=1433925.msg31997079#msg31997079

[big al]

I wasn't getting 43 without the patch just overdriventool wouldn't work.

[Vann]

The pixel patcher bypasses the driver Bios signature check. With a modded Bios the signature is invalid and causes Windows not to be able to load the driver for the card (code 43). AFAIK a RX 580 with a modded Bios has never needed the pixel patcher to bypass the Bios signature check. If OverDriveNtool wasn't working it may be because the drivers changed something in the API, which has nothing to do with the pixel patcher.

The same thing happened when AMD switched from the Crimson Relive to the Adrenaline drivers, which caused the Claymore overclocking parameters to stop working until he updated the miner in v10.3 so they were compatible with the Adrenaline drivers.

https://bitcointalk.org/index.php?topic=1433925.msg27476189#msg27476189

[zloyneo]

nvidia 1060 3gb stopped mining Ethereum
DAG files problem
Windows 10

[ascaloon]

The pixel patcher bypasses the driver Bios signature check. With a modded Bios the signature is invalid and causes Windows not to be able to load the driver for the card (code 43). AFAIK a RX 580 with a modded Bios has never needed the pixel patcher to bypass the Bios signature check. If OverDriveNtool wasn't working it may be because the drivers changed something in the API, which has nothing to do with the pixel patcher.

The same thing happened when AMD switched from the Crimson Relive to the Adrenaline drivers, which caused the Claymore overclocking parameters to stop working until he updated the miner in v10.3 so they were compatible with the Adrenaline drivers.

https://bitcointalk.org/index.php?topic=1433925.msg27476189#msg27476189

Try this if OverdriveNtool won`t work. It helps to me.

https://forums.guru3d.com/threads/overdriventool-tool-for-amd-gpus.416116/page-13