[Cast XMR] high speed XMR/CryptoNight miner for RX Vega GPUs (2 KHash/s)

[P00P135]

I've noticed that when Windows blocks CastXMR from accessing a GPU because of aggressive overclocking, the software doesn't handle it very gracefully. The affected GPU stops reporting its hashrate every five seconds, and after a few minutes the software hangs entirely and multiple ^C are needed to terminate the run, assuming that the system doesn't freeze entirely and need a warm reset.

A useful feature would be for CastXMR to recognize that if it is accessing a list of GPUs, each of them should report results periodically, and if one of them goes missing for, say, twice the running average time between results, then it should be considered dead, and a clean restart of CastXMR should be initiated.

Or better yet, if CastXMR could perform a disable/enable of the problem card and then reinitialize it individually, so that none of the other cards had to be bothered.

With a Mining Expert 19-slot motherboard running a pile of aggressively overclocked Vegas, you're more likely to wedge one of the cards than usual, and being able to recover gracefully from that situation would substantially improve the average hashrate.

It's not really a problem with CAST-XMR now is it. That is you choosing to overdrive the card and then demand that that the tool fix it for you. Maybe you should spend more time accepting a stable hash rate per card and live with it. Its really a driver issue anyway and you are taking risks when overclocking that you will put the system into an unrecoverable state.

I personally don't need a feature in CAST that disables the cards and I don't think that would be helpful for people who have configured their systems correctly. I reboot once a week for windows updates, other than that I don't really look at the rigs.

If you need these bandaid scripts to monitor your hash rate cause it drops or your cards hang you have other problems that are not associated with the miner. I would rather the dev spend time improving the miner efficiency rather than adding some lame monitoring function that is completely unnecessary.

Oh and by the way I use Wattman. YIKES!

Really? Unbelieveble!
I'm using cast_xmr_vega with 6x VEGA56. Everyday and very often I need to restart miner or PC. VEGA's are not modded.
Everything I do before running the miner is run devcon.exe for cli enable/disable GPU's and then Overdriven tool with overclock settings.
Very often is output paused... until S is pressed and after is all messages displayed with OUTDATED result.
I never had a ban from poolserver but last 2 days 3 times! Sometimes freezing after Difficulty changed message.
When I press Q displayed 'quitting..' and nothing happend.

 Right click top of command window > properties > options tab > uncheck quick edit mode

[Mrsmoker]

this is so good miner, ty sir! got the most high hash than other miner. and cooler GPU, i like it

[commonorx]

I've noticed that when Windows blocks CastXMR from accessing a GPU because of aggressive overclocking, the software doesn't handle it very gracefully. The affected GPU stops reporting its hashrate every five seconds, and after a few minutes the software hangs entirely and multiple ^C are needed to terminate the run, assuming that the system doesn't freeze entirely and need a warm reset.

A useful feature would be for CastXMR to recognize that if it is accessing a list of GPUs, each of them should report results periodically, and if one of them goes missing for, say, twice the running average time between results, then it should be considered dead, and a clean restart of CastXMR should be initiated.

Or better yet, if CastXMR could perform a disable/enable of the problem card and then reinitialize it individually, so that none of the other cards had to be bothered.

With a Mining Expert 19-slot motherboard running a pile of aggressively overclocked Vegas, you're more likely to wedge one of the cards than usual, and being able to recover gracefully from that situation would substantially improve the average hashrate.

It's not really a problem with CAST-XMR now is it. That is you choosing to overdrive the card and then demand that that the tool fix it for you. Maybe you should spend more time accepting a stable hash rate per card and live with it. Its really a driver issue anyway and you are taking risks when overclocking that you will put the system into an unrecoverable state.

I personally don't need a feature in CAST that disables the cards and I don't think that would be helpful for people who have configured their systems correctly. I reboot once a week for windows updates, other than that I don't really look at the rigs.

If you need these bandaid scripts to monitor your hash rate cause it drops or your cards hang you have other problems that are not associated with the miner. I would rather the dev spend time improving the miner efficiency rather than adding some lame monitoring function that is completely unnecessary.

Oh and by the way I use Wattman. YIKES!

Really? Unbelieveble!
I'm using cast_xmr_vega with 6x VEGA56. Everyday and very often I need to restart miner or PC. VEGA's are not modded.
Everything I do before running the miner is run devcon.exe for cli enable/disable GPU's and then Overdriven tool with overclock settings.
Very often is output paused... until S is pressed and after is all messages displayed with OUTDATED result.
I never had a ban from poolserver but last 2 days 3 times! Sometimes freezing after Difficulty changed message. https://ibb.co/n9Ra4w
When I press Q displayed 'quitting..' and nothing happend.

I would suggest not using useless tools like Overdriven. I know everyone raves about how awesome it is but I have never seen the need for it frankly, even though everyone says Wattman is the worst piece of software in the world.

I would make sure you started with a completely clean install of Windows. I would make sure you set your virtual memory correctly. I would make sure you configured windows power settings correctly. Then I would make sure to cleanly install the blockchain drivers. I would start would not start OC'ing anything until you can stay up for 24 hours at the same hash rate with no issues. Also make sure to add an exclusion for Windows Defender. AKA EVERYTHING THAT HAS ALREADY BEEN POSTED IN THIS WHOLE THREAD!

Lastly, since you did not state what you were mining or what pool you were using you could do that favor as well. Perhaps the pool or your internet connection or your drivers or your ethernet or whatever is not working correctly or you have a riser problem? Really man can you troubleshoot or problem solve? There are endless things that can contribute to your problems and if you are having problems you need to stop mining and start testing!

There are so many things that are connected via hardware and controlled by software and you need to understand all that!

I have the same Vega rig as you do and guess what... they all run at about 1975h/s average. I never restart the miner except if I decide to change a pool or mine something else, and the last time I looked at that Vega 56 rig was last Sunday. I can see the output on the site I am mining to and the hash rate looks good! I am still getting paid!

So you can package up your rig and send it to me! I will take this problem child off your hands :>)

[QuintLeo]

Wattman isn't the worst junk softtware in the world - it's just massively inferior to Afterburner and quite a bit less capable when it comes to things like setting fan profiles, and the INTERFACE is very badly done and anti-intuitive.

 It's nowhere near as bad as the Unusable Interface on Windows 8, for example....

[Morgon]

I would suggest not using useless tools like Overdriven.

It certainly isn't the only way to go about tweaking, but it is quite versatile and thus hardly 'useless'. Would be interested in knowing what you use to manage a new GPU's settings.

[koinli]

Both are junk software Wattman and MSI afterburner, they have many problems when i try to set a clock on my gpu.

[Cyan_Antares]

hi

same problem with 0.81 :gpu3 hangs and crash the rig.

0.70 vers start at 4200 h for my 5 rx570 and all the card drop and get a 3400 hahsrate  :/

have you tried changing the riser of the offending card?

On a different note, I am now using WinMiner and I would like Cast XMR to be compatible with it, both are very good software, but if you run Cast XMR and even if WinMiner remains idle, it gets to a point where Cast XMR hangs, for some reason.

[commonorx]

Wattman isn't the worst junk softtware in the world - it's just massively inferior to Afterburner and quite a bit less capable when it comes to things like setting fan profiles, and the INTERFACE is very badly done and anti-intuitive.

 It's nowhere near as bad as the Unusable Interface on Windows 8, for example....

To be certain there is nothing extra needed in Afterbuner or Overdriven or any other tools that make them better than Wattman. A long while back I set some max temp and target settings, and some fan settings and that is all. There is absolutely nothing else that can be modified in any tool or wattman. You can only additioanly change the mem speed and GPU, but once you are done with that there is nothing to look at.  I have not needed to make any changes to that for a few months. I don't even look or open wattman.

So who cares and I would prefer to not install more software from some other source that does the same thing anyway. So yes these tools really are useless after your settings are made and they all control the same stupid functions. Use what you like. I don't like a ton of 3rd party redundant software.

[commonorx]

Both are junk software Wattman and MSI afterburner, they have many problems when i try to set a clock on my gpu.

I would say why do you need to do that? Are you using a Vega? Do you know about the reg settings? If you are having problems it mostly down to you pushing your hardware beyond its limits.

[commonorx]

I would suggest not using useless tools like Overdriven.

It certainly isn't the only way to go about tweaking, but it is quite versatile and thus hardly 'useless'. Would be interested in knowing what you use to manage a new GPU's settings.

I use the Vega's for mining. I don't care what the GPU is set to as it is not important for me. I used the regmods and I also am able to edit the reg file and update it with new values if needed. Frankly I don't need to keep tweaking forever. I was done with tweaking months back and now with these Vega cards and when or if I get any more Vega's I will immediately apply the regmod with the higher SOC and lower voltage, set the cards temp and fan speeds in wattman and call it good. Some cards can take a faster memspeed like 1105 or 1120 but really I don't care that much right now.

This applies only to the reference cards as the newer 3rd party designed cards may behave differently than the reference designs.

If AMD has a breakthrough in a driver for the Vega that speeds up hash rates for cryptonight based currencies then I would have a look at that and probably play around some again. But nothing to do now.

[R0mi]

I've noticed that when Windows blocks CastXMR from accessing a GPU because of aggressive overclocking, the software doesn't handle it very gracefully. The affected GPU stops reporting its hashrate every five seconds, and after a few minutes the software hangs entirely and multiple ^C are needed to terminate the run, assuming that the system doesn't freeze entirely and need a warm reset.

A useful feature would be for CastXMR to recognize that if it is accessing a list of GPUs, each of them should report results periodically, and if one of them goes missing for, say, twice the running average time between results, then it should be considered dead, and a clean restart of CastXMR should be initiated.

Or better yet, if CastXMR could perform a disable/enable of the problem card and then reinitialize it individually, so that none of the other cards had to be bothered.

With a Mining Expert 19-slot motherboard running a pile of aggressively overclocked Vegas, you're more likely to wedge one of the cards than usual, and being able to recover gracefully from that situation would substantially improve the average hashrate.

It's not really a problem with CAST-XMR now is it. That is you choosing to overdrive the card and then demand that that the tool fix it for you. Maybe you should spend more time accepting a stable hash rate per card and live with it. Its really a driver issue anyway and you are taking risks when overclocking that you will put the system into an unrecoverable state.

I personally don't need a feature in CAST that disables the cards and I don't think that would be helpful for people who have configured their systems correctly. I reboot once a week for windows updates, other than that I don't really look at the rigs.

If you need these bandaid scripts to monitor your hash rate cause it drops or your cards hang you have other problems that are not associated with the miner. I would rather the dev spend time improving the miner efficiency rather than adding some lame monitoring function that is completely unnecessary.

Oh and by the way I use Wattman. YIKES!

Really? Unbelieveble!
I'm using cast_xmr_vega with 6x VEGA56. Everyday and very often I need to restart miner or PC. VEGA's are not modded.
Everything I do before running the miner is run devcon.exe for cli enable/disable GPU's and then Overdriven tool with overclock settings.
Very often is output paused... until S is pressed and after is all messages displayed with OUTDATED result.
I never had a ban from poolserver but last 2 days 3 times! Sometimes freezing after Difficulty changed message.
When I press Q displayed 'quitting..' and nothing happend.

I would suggest not using useless tools like Overdriven. I know everyone raves about how awesome it is but I have never seen the need for it frankly, even though everyone says Wattman is the worst piece of software in the world.

I would make sure you started with a completely clean install of Windows. I would make sure you set your virtual memory correctly. I would make sure you configured windows power settings correctly. Then I would make sure to cleanly install the blockchain drivers. I would start would not start OC'ing anything until you can stay up for 24 hours at the same hash rate with no issues. Also make sure to add an exclusion for Windows Defender. AKA EVERYTHING THAT HAS ALREADY BEEN POSTED IN THIS WHOLE THREAD!

Lastly, since you did not state what you were mining or what pool you were using you could do that favor as well. Perhaps the pool or your internet connection or your drivers or your ethernet or whatever is not working correctly or you have a riser problem? Really man can you troubleshoot or problem solve? There are endless things that can contribute to your problems and if you are having problems you need to stop mining and start testing!

There are so many things that are connected via hardware and controlled by software and you need to understand all that!

I have the same Vega rig as you do and guess what... they all run at about 1975h/s average. I never restart the miner except if I decide to change a pool or mine something else, and the last time I looked at that Vega 56 rig was last Sunday. I can see the output on the site I am mining to and the hash rate looks good! I am still getting paid!

So you can package up your rig and send it to me! I will take this problem child off your hands :>)

This is your third post as a jr. member that I've seen on this thread and every post of yours is insulting and offensive.  I think you need to give it a break.

[Aris_Eng]

Hi gents,

I might get some "pain" if this has already been answered, but have been looking for an answer for a few hours now and no joy...

Anyway, here is the question:

In cast-xmr (latest), after a few hours of running(mining successfully 2000+ h/s each card), suddenly my shares stop getting accepted with message "unauthenticated", but the miner keeps on running (in vain).

I am using blockchain drivers on a 3xVega 64 system, mining Alloy and IntenseCoin, on various diff pools, happens on each pool.

Any help appreciated.

Cheers

[ducho74]

I've noticed that when Windows blocks CastXMR from accessing a GPU because of aggressive overclocking, the software doesn't handle it very gracefully. The affected GPU stops reporting its hashrate every five seconds, and after a few minutes the software hangs entirely and multiple ^C are needed to terminate the run, assuming that the system doesn't freeze entirely and need a warm reset.

A useful feature would be for CastXMR to recognize that if it is accessing a list of GPUs, each of them should report results periodically, and if one of them goes missing for, say, twice the running average time between results, then it should be considered dead, and a clean restart of CastXMR should be initiated.

Or better yet, if CastXMR could perform a disable/enable of the problem card and then reinitialize it individually, so that none of the other cards had to be bothered.

With a Mining Expert 19-slot motherboard running a pile of aggressively overclocked Vegas, you're more likely to wedge one of the cards than usual, and being able to recover gracefully from that situation would substantially improve the average hashrate.

It's not really a problem with CAST-XMR now is it. That is you choosing to overdrive the card and then demand that that the tool fix it for you. Maybe you should spend more time accepting a stable hash rate per card and live with it. Its really a driver issue anyway and you are taking risks when overclocking that you will put the system into an unrecoverable state.

I personally don't need a feature in CAST that disables the cards and I don't think that would be helpful for people who have configured their systems correctly. I reboot once a week for windows updates, other than that I don't really look at the rigs.

If you need these bandaid scripts to monitor your hash rate cause it drops or your cards hang you have other problems that are not associated with the miner. I would rather the dev spend time improving the miner efficiency rather than adding some lame monitoring function that is completely unnecessary.

Oh and by the way I use Wattman. YIKES!

Really? Unbelieveble!
I'm using cast_xmr_vega with 6x VEGA56. Everyday and very often I need to restart miner or PC. VEGA's are not modded.
Everything I do before running the miner is run devcon.exe for cli enable/disable GPU's and then Overdriven tool with overclock settings.
Very often is output paused... until S is pressed and after is all messages displayed with OUTDATED result.
I never had a ban from poolserver but last 2 days 3 times! Sometimes freezing after Difficulty changed message. https://ibb.co/n9Ra4w
When I press Q displayed 'quitting..' and nothing happend.

I would suggest not using useless tools like Overdriven. I know everyone raves about how awesome it is but I have never seen the need for it frankly, even though everyone says Wattman is the worst piece of software in the world.

I would make sure you started with a completely clean install of Windows. I would make sure you set your virtual memory correctly. I would make sure you configured windows power settings correctly. Then I would make sure to cleanly install the blockchain drivers. I would start would not start OC'ing anything until you can stay up for 24 hours at the same hash rate with no issues. Also make sure to add an exclusion for Windows Defender. AKA EVERYTHING THAT HAS ALREADY BEEN POSTED IN THIS WHOLE THREAD!

Lastly, since you did not state what you were mining or what pool you were using you could do that favor as well. Perhaps the pool or your internet connection or your drivers or your ethernet or whatever is not working correctly or you have a riser problem? Really man can you troubleshoot or problem solve? There are endless things that can contribute to your problems and if you are having problems you need to stop mining and start testing!

There are so many things that are connected via hardware and controlled by software and you need to understand all that!

I have the same Vega rig as you do and guess what... they all run at about 1975h/s average. I never restart the miner except if I decide to change a pool or mine something else, and the last time I looked at that Vega 56 rig was last Sunday. I can see the output on the site I am mining to and the hash rate looks good! I am still getting paid!

So you can package up your rig and send it to me! I will take this problem child off your hands :>)

First thank you for your comprehensive answer! Currently I mining XMR and ETH on supportxtm.com but same issues. I thing the internet connection is not problem.  This thread is very long but ok I read it all.
This is my current logs:

[Job: #41 | Difficulty: 439427 | Running: 16.0 sec | Avg Job Time: 47.4 sec]
[Hash Rate Avg: 11788.4 H/s]
  1977.6 H/s GPU0
  1960.8 H/s GPU1
  1965.5 H/s GPU2
  1970.5 H/s GPU3
  1966.0 H/s GPU4
  1947.1 H/s GPU5

[Shares Found: 66 | Avg Search Time: 29.5 sec]
  63 ( 95%) Accepted
   0 (  0%) Rejected by pool
   0 (  0%) Invalid result computation failed
   0 (  0%) Could not be submit because of network error
   3 (  5%) Outdated because of job change


[13:55:53] GPU5 | 42°C | Fan 4447 RPM | 1952.8 H/s
[13:55:53] GPU3 | 46°C | Fan 4552 RPM | 1972.1 H/s
[13:55:55] GPU1 | 44°C | Fan 4441 RPM | 1968.5 H/s
[13:55:56] GPU2 | 45°C | Fan 4538 RPM | 1968.9 H/s
[13:55:57] GPU0 | 45°C | Fan 4495 RPM | 1979.7 H/s
[13:55:58] GPU4 | 43°C | Fan 4494 RPM | 1968.1 H/s
[13:55:58] GPU5 | 42°C | Fan 4451 RPM | 1951.7 H/s
[13:55:59] GPU3 | 45°C | Fan 4560 RPM | 1972.5 H/s

My second rig:


[Pool: 'etn.fairhash.org:7777' | Connected: 2018-01-05 19:06:58]
  17:50:03 (100%) Online
  0:00:00 (  0%) Offline

[Job: #1217 | Difficulty: 250406 | Running: 19.0 sec | Avg Job Time: 52.8 sec]

[Hash Rate Avg: 5889.9 H/s]
  1965.7 H/s GPU0
  1955.6 H/s GPU1
  1968.5 H/s GPU2

[Shares Found: 1632 | Avg Search Time: 40.8 sec]
1575 ( 97%) Accepted
   0 (  0%) Rejected by pool
   0 (  0%) Invalid result computation failed
   0 (  0%) Could not be submit because of network error
  57 (  3%) Outdated because of job change

[12:57:02] GPU2 | 56°C | Fan 4541 RPM | 1970.7 H/s
[12:57:04] GPU0 | 53°C | Fan 4482 RPM | 1963.8 H/s
[12:57:12] GPU1 | 57°C | Fan 4487 RPM | 1954.9 H/s
[12:57:12] GPU2 | 56°C | Fan 4535 RPM | 1970.7 H/s
[12:57:12] GPU0 | 53°C | Fan 4477 RPM | 1962.8 H/s
[12:57:12] GPU1 | 57°C | Fan 4486 RPM | 1954.6 H/s
[12:57:13] GPU2 | 56°C | Fan 4533 RPM | 1970.7 H/s
[12:57:15] GPU0 | 53°C | Fan 4473 RPM | 1962.4 H/s
[12:57:16] GPU1 | 57°C | Fan 4479 RPM | 1955.3 H/s
[12:57:18] GPU2 | 55°C | Fan 4527 RPM | 1967.4 H/s

[ducho74]

I've noticed that when Windows blocks CastXMR from accessing a GPU because of aggressive overclocking, the software doesn't handle it very gracefully. The affected GPU stops reporting its hashrate every five seconds, and after a few minutes the software hangs entirely and multiple ^C are needed to terminate the run, assuming that the system doesn't freeze entirely and need a warm reset.

A useful feature would be for CastXMR to recognize that if it is accessing a list of GPUs, each of them should report results periodically, and if one of them goes missing for, say, twice the running average time between results, then it should be considered dead, and a clean restart of CastXMR should be initiated.

Or better yet, if CastXMR could perform a disable/enable of the problem card and then reinitialize it individually, so that none of the other cards had to be bothered.

With a Mining Expert 19-slot motherboard running a pile of aggressively overclocked Vegas, you're more likely to wedge one of the cards than usual, and being able to recover gracefully from that situation would substantially improve the average hashrate.

It's not really a problem with CAST-XMR now is it. That is you choosing to overdrive the card and then demand that that the tool fix it for you. Maybe you should spend more time accepting a stable hash rate per card and live with it. Its really a driver issue anyway and you are taking risks when overclocking that you will put the system into an unrecoverable state.

I personally don't need a feature in CAST that disables the cards and I don't think that would be helpful for people who have configured their systems correctly. I reboot once a week for windows updates, other than that I don't really look at the rigs.

If you need these bandaid scripts to monitor your hash rate cause it drops or your cards hang you have other problems that are not associated with the miner. I would rather the dev spend time improving the miner efficiency rather than adding some lame monitoring function that is completely unnecessary.

Oh and by the way I use Wattman. YIKES!

Really? Unbelieveble!
I'm using cast_xmr_vega with 6x VEGA56. Everyday and very often I need to restart miner or PC. VEGA's are not modded.
Everything I do before running the miner is run devcon.exe for cli enable/disable GPU's and then Overdriven tool with overclock settings.
Very often is output paused... until S is pressed and after is all messages displayed with OUTDATED result.
I never had a ban from poolserver but last 2 days 3 times! Sometimes freezing after Difficulty changed message. https://ibb.co/n9Ra4w
When I press Q displayed 'quitting..' and nothing happend.

 Right click top of command window > properties > options tab > uncheck quick edit mode

What u mean ? Where ? Why ?

[Cyan_Antares]

Too much errors with cast_xmr_vega.exe on 6x VEGA56.
Very often freezing, or stoped/paused and after pressing 'S' is ouput report showed with outdated result....

make sure that your Vega 56 is running at 935MHz or below, that's where they start to be totally stable --if you modded the BIOS, the values differ but I have no experience with them.

I am having a very similar issue to ducho74. After a while (usually under an hour) CAST hangs on difficulty changed. When Q is pressed, it prints quitting... and doesnt actually quit.

The machine currently running the vega is hybrid AMD/NVIDIA rig running Window 10, with 2 1070s and 1 1060 as well as the VEGA 64. the 1070s and the VEGA are powered my a seperate HSNTS PL-11 server power supply, and the 1060 is powered by the ATX PSU in the PC.

I have tried re-installing the blockchain driver, as well as different risers, and the problem still persists.

Any input is greatly appreciated

make sure that you use the correct settings for the Vega 64. Try the soft powerplay tables.

Plus, check that you arent running into the known CMD issue, mentioned some pages before.

[Dell1503]

Have Somebody made here with the cast XMR miner a 8 or 12 VEGA GPU Rig?
If yes would be nice for showing components, like Mainbard.

Thy

[Permonik]

Have Somebody made here with the cast XMR miner a 8 or 12 VEGA GPU Rig?
If yes would be nice for showing components, like Mainbard.

Thy

I am using 8x rx vegas 64 sapphire limited edition with blockchain driver aug23. (Inside the rig are 10x vegas gpu, but due to blockchain driver limitation, only 8 are plugged in. There are drivers for 12gpus for vegas, but automatic switching for gpu workload option is not working for cryptonight.)
PSU: 2x Supernova 1600 P2
MOBO: Asus mining expert b250
CPU: core i3-7100T
RAM: G.SKILL 32GB KIT DDR4 3000MHz CL16 RipjawsV
HDD: SSD ADATA 256GB
Risers: gen006c

[trip9600]

Is there a way to set CAST XMR to restart on error? I’ve got great hash rate but if there is a network error they go down the drain and don’t recover.

[commonorx]

I've noticed that when Windows blocks CastXMR from accessing a GPU because of aggressive overclocking, the software doesn't handle it very gracefully. The affected GPU stops reporting its hashrate every five seconds, and after a few minutes the software hangs entirely and multiple ^C are needed to terminate the run, assuming that the system doesn't freeze entirely and need a warm reset.

A useful feature would be for CastXMR to recognize that if it is accessing a list of GPUs, each of them should report results periodically, and if one of them goes missing for, say, twice the running average time between results, then it should be considered dead, and a clean restart of CastXMR should be initiated.

Or better yet, if CastXMR could perform a disable/enable of the problem card and then reinitialize it individually, so that none of the other cards had to be bothered.

With a Mining Expert 19-slot motherboard running a pile of aggressively overclocked Vegas, you're more likely to wedge one of the cards than usual, and being able to recover gracefully from that situation would substantially improve the average hashrate.

It's not really a problem with CAST-XMR now is it. That is you choosing to overdrive the card and then demand that that the tool fix it for you. Maybe you should spend more time accepting a stable hash rate per card and live with it. Its really a driver issue anyway and you are taking risks when overclocking that you will put the system into an unrecoverable state.

I personally don't need a feature in CAST that disables the cards and I don't think that would be helpful for people who have configured their systems correctly. I reboot once a week for windows updates, other than that I don't really look at the rigs.

If you need these bandaid scripts to monitor your hash rate cause it drops or your cards hang you have other problems that are not associated with the miner. I would rather the dev spend time improving the miner efficiency rather than adding some lame monitoring function that is completely unnecessary.

Oh and by the way I use Wattman. YIKES!

Really? Unbelieveble!
I'm using cast_xmr_vega with 6x VEGA56. Everyday and very often I need to restart miner or PC. VEGA's are not modded.
Everything I do before running the miner is run devcon.exe for cli enable/disable GPU's and then Overdriven tool with overclock settings.
Very often is output paused... until S is pressed and after is all messages displayed with OUTDATED result.
I never had a ban from poolserver but last 2 days 3 times! Sometimes freezing after Difficulty changed message. https://ibb.co/n9Ra4w
When I press Q displayed 'quitting..' and nothing happend.

I would suggest not using useless tools like Overdriven. I know everyone raves about how awesome it is but I have never seen the need for it frankly, even though everyone says Wattman is the worst piece of software in the world.

I would make sure you started with a completely clean install of Windows. I would make sure you set your virtual memory correctly. I would make sure you configured windows power settings correctly. Then I would make sure to cleanly install the blockchain drivers. I would start would not start OC'ing anything until you can stay up for 24 hours at the same hash rate with no issues. Also make sure to add an exclusion for Windows Defender. AKA EVERYTHING THAT HAS ALREADY BEEN POSTED IN THIS WHOLE THREAD!

Lastly, since you did not state what you were mining or what pool you were using you could do that favor as well. Perhaps the pool or your internet connection or your drivers or your ethernet or whatever is not working correctly or you have a riser problem? Really man can you troubleshoot or problem solve? There are endless things that can contribute to your problems and if you are having problems you need to stop mining and start testing!

There are so many things that are connected via hardware and controlled by software and you need to understand all that!

I have the same Vega rig as you do and guess what... they all run at about 1975h/s average. I never restart the miner except if I decide to change a pool or mine something else, and the last time I looked at that Vega 56 rig was last Sunday. I can see the output on the site I am mining to and the hash rate looks good! I am still getting paid!

So you can package up your rig and send it to me! I will take this problem child off your hands :>)

First thank you for your comprehensive answer! Currently I mining XMR and ETH on supportxtm.com but same issues. I thing the internet connection is not problem.  This thread is very long but ok I read it all.
This is my current logs:

[Job: #41 | Difficulty: 439427 | Running: 16.0 sec | Avg Job Time: 47.4 sec]
[Hash Rate Avg: 11788.4 H/s]
  1977.6 H/s GPU0
  1960.8 H/s GPU1
  1965.5 H/s GPU2
  1970.5 H/s GPU3
  1966.0 H/s GPU4
  1947.1 H/s GPU5

[Shares Found: 66 | Avg Search Time: 29.5 sec]
  63 ( 95%) Accepted
   0 (  0%) Rejected by pool
   0 (  0%) Invalid result computation failed
   0 (  0%) Could not be submit because of network error
   3 (  5%) Outdated because of job change


[13:55:53] GPU5 | 42°C | Fan 4447 RPM | 1952.8 H/s
[13:55:53] GPU3 | 46°C | Fan 4552 RPM | 1972.1 H/s
[13:55:55] GPU1 | 44°C | Fan 4441 RPM | 1968.5 H/s
[13:55:56] GPU2 | 45°C | Fan 4538 RPM | 1968.9 H/s
[13:55:57] GPU0 | 45°C | Fan 4495 RPM | 1979.7 H/s
[13:55:58] GPU4 | 43°C | Fan 4494 RPM | 1968.1 H/s
[13:55:58] GPU5 | 42°C | Fan 4451 RPM | 1951.7 H/s
[13:55:59] GPU3 | 45°C | Fan 4560 RPM | 1972.5 H/s

My second rig:


[Pool: 'etn.fairhash.org:7777' | Connected: 2018-01-05 19:06:58]
  17:50:03 (100%) Online
  0:00:00 (  0%) Offline

[Job: #1217 | Difficulty: 250406 | Running: 19.0 sec | Avg Job Time: 52.8 sec]

[Hash Rate Avg: 5889.9 H/s]
  1965.7 H/s GPU0
  1955.6 H/s GPU1
  1968.5 H/s GPU2

[Shares Found: 1632 | Avg Search Time: 40.8 sec]
1575 ( 97%) Accepted
   0 (  0%) Rejected by pool
   0 (  0%) Invalid result computation failed
   0 (  0%) Could not be submit because of network error
  57 (  3%) Outdated because of job change

[12:57:02] GPU2 | 56°C | Fan 4541 RPM | 1970.7 H/s
[12:57:04] GPU0 | 53°C | Fan 4482 RPM | 1963.8 H/s
[12:57:12] GPU1 | 57°C | Fan 4487 RPM | 1954.9 H/s
[12:57:12] GPU2 | 56°C | Fan 4535 RPM | 1970.7 H/s
[12:57:12] GPU0 | 53°C | Fan 4477 RPM | 1962.8 H/s
[12:57:12] GPU1 | 57°C | Fan 4486 RPM | 1954.6 H/s
[12:57:13] GPU2 | 56°C | Fan 4533 RPM | 1970.7 H/s
[12:57:15] GPU0 | 53°C | Fan 4473 RPM | 1962.4 H/s
[12:57:16] GPU1 | 57°C | Fan 4479 RPM | 1955.3 H/s
[12:57:18] GPU2 | 55°C | Fan 4527 RPM | 1967.4 H/s

What is your memory speed at? Your fans seem really high. Looks like you are able to get over 1500 shares at the same hash rate? Have you used the --fastjobswtich to help lower the number of Outdated shares?

[ducho74]

I've noticed that when Windows blocks CastXMR from accessing a GPU because of aggressive overclocking, the software doesn't handle it very gracefully. The affected GPU stops reporting its hashrate every five seconds, and after a few minutes the software hangs entirely and multiple ^C are needed to terminate the run, assuming that the system doesn't freeze entirely and need a warm reset.

A useful feature would be for CastXMR to recognize that if it is accessing a list of GPUs, each of them should report results periodically, and if one of them goes missing for, say, twice the running average time between results, then it should be considered dead, and a clean restart of CastXMR should be initiated.

Or better yet, if CastXMR could perform a disable/enable of the problem card and then reinitialize it individually, so that none of the other cards had to be bothered.

With a Mining Expert 19-slot motherboard running a pile of aggressively overclocked Vegas, you're more likely to wedge one of the cards than usual, and being able to recover gracefully from that situation would substantially improve the average hashrate.

It's not really a problem with CAST-XMR now is it. That is you choosing to overdrive the card and then demand that that the tool fix it for you. Maybe you should spend more time accepting a stable hash rate per card and live with it. Its really a driver issue anyway and you are taking risks when overclocking that you will put the system into an unrecoverable state.

I personally don't need a feature in CAST that disables the cards and I don't think that would be helpful for people who have configured their systems correctly. I reboot once a week for windows updates, other than that I don't really look at the rigs.

If you need these bandaid scripts to monitor your hash rate cause it drops or your cards hang you have other problems that are not associated with the miner. I would rather the dev spend time improving the miner efficiency rather than adding some lame monitoring function that is completely unnecessary.

Oh and by the way I use Wattman. YIKES!

Really? Unbelieveble!
I'm using cast_xmr_vega with 6x VEGA56. Everyday and very often I need to restart miner or PC. VEGA's are not modded.
Everything I do before running the miner is run devcon.exe for cli enable/disable GPU's and then Overdriven tool with overclock settings.
Very often is output paused... until S is pressed and after is all messages displayed with OUTDATED result.
I never had a ban from poolserver but last 2 days 3 times! Sometimes freezing after Difficulty changed message. https://ibb.co/n9Ra4w
When I press Q displayed 'quitting..' and nothing happend.

I would suggest not using useless tools like Overdriven. I know everyone raves about how awesome it is but I have never seen the need for it frankly, even though everyone says Wattman is the worst piece of software in the world.

I would make sure you started with a completely clean install of Windows. I would make sure you set your virtual memory correctly. I would make sure you configured windows power settings correctly. Then I would make sure to cleanly install the blockchain drivers. I would start would not start OC'ing anything until you can stay up for 24 hours at the same hash rate with no issues. Also make sure to add an exclusion for Windows Defender. AKA EVERYTHING THAT HAS ALREADY BEEN POSTED IN THIS WHOLE THREAD!

Lastly, since you did not state what you were mining or what pool you were using you could do that favor as well. Perhaps the pool or your internet connection or your drivers or your ethernet or whatever is not working correctly or you have a riser problem? Really man can you troubleshoot or problem solve? There are endless things that can contribute to your problems and if you are having problems you need to stop mining and start testing!

There are so many things that are connected via hardware and controlled by software and you need to understand all that!

I have the same Vega rig as you do and guess what... they all run at about 1975h/s average. I never restart the miner except if I decide to change a pool or mine something else, and the last time I looked at that Vega 56 rig was last Sunday. I can see the output on the site I am mining to and the hash rate looks good! I am still getting paid!

So you can package up your rig and send it to me! I will take this problem child off your hands :>)

First thank you for your comprehensive answer! Currently I mining XMR and ETH on supportxtm.com but same issues. I thing the internet connection is not problem.  This thread is very long but ok I read it all.
This is my current logs:

[Job: #41 | Difficulty: 439427 | Running: 16.0 sec | Avg Job Time: 47.4 sec]
[Hash Rate Avg: 11788.4 H/s]
  1977.6 H/s GPU0
  1960.8 H/s GPU1
  1965.5 H/s GPU2
  1970.5 H/s GPU3
  1966.0 H/s GPU4
  1947.1 H/s GPU5

[Shares Found: 66 | Avg Search Time: 29.5 sec]
  63 ( 95%) Accepted
   0 (  0%) Rejected by pool
   0 (  0%) Invalid result computation failed
   0 (  0%) Could not be submit because of network error
   3 (  5%) Outdated because of job change


[13:55:53] GPU5 | 42°C | Fan 4447 RPM | 1952.8 H/s
[13:55:53] GPU3 | 46°C | Fan 4552 RPM | 1972.1 H/s
[13:55:55] GPU1 | 44°C | Fan 4441 RPM | 1968.5 H/s
[13:55:56] GPU2 | 45°C | Fan 4538 RPM | 1968.9 H/s
[13:55:57] GPU0 | 45°C | Fan 4495 RPM | 1979.7 H/s
[13:55:58] GPU4 | 43°C | Fan 4494 RPM | 1968.1 H/s
[13:55:58] GPU5 | 42°C | Fan 4451 RPM | 1951.7 H/s
[13:55:59] GPU3 | 45°C | Fan 4560 RPM | 1972.5 H/s

My second rig:


[Pool: 'etn.fairhash.org:7777' | Connected: 2018-01-05 19:06:58]
  17:50:03 (100%) Online
  0:00:00 (  0%) Offline

[Job: #1217 | Difficulty: 250406 | Running: 19.0 sec | Avg Job Time: 52.8 sec]

[Hash Rate Avg: 5889.9 H/s]
  1965.7 H/s GPU0
  1955.6 H/s GPU1
  1968.5 H/s GPU2

[Shares Found: 1632 | Avg Search Time: 40.8 sec]
1575 ( 97%) Accepted
   0 (  0%) Rejected by pool
   0 (  0%) Invalid result computation failed
   0 (  0%) Could not be submit because of network error
  57 (  3%) Outdated because of job change

[12:57:02] GPU2 | 56°C | Fan 4541 RPM | 1970.7 H/s
[12:57:04] GPU0 | 53°C | Fan 4482 RPM | 1963.8 H/s
[12:57:12] GPU1 | 57°C | Fan 4487 RPM | 1954.9 H/s
[12:57:12] GPU2 | 56°C | Fan 4535 RPM | 1970.7 H/s
[12:57:12] GPU0 | 53°C | Fan 4477 RPM | 1962.8 H/s
[12:57:12] GPU1 | 57°C | Fan 4486 RPM | 1954.6 H/s
[12:57:13] GPU2 | 56°C | Fan 4533 RPM | 1970.7 H/s
[12:57:15] GPU0 | 53°C | Fan 4473 RPM | 1962.4 H/s
[12:57:16] GPU1 | 57°C | Fan 4479 RPM | 1955.3 H/s
[12:57:18] GPU2 | 55°C | Fan 4527 RPM | 1967.4 H/s

What is your memory speed at? Your fans seem really high. Looks like you are able to get over 1500 shares at the same hash rate? Have you used the --fastjobswtich to help lower the number of Outdated shares?

Temperature in the room is currently about 25C because it is warm outside. I'm not using cooling in the room. When temperature in the room is about 14C then fans are lower and temp on GPU about 46-49C.
I'm using this configuration but currently testing Fan_Target=45 and Fan_Min=2800

There is two rigs.. The one of them (3x GPU) is more stable. I already had 10k shares. when I think about it, it's about that room temperature..
But it's interesting that every rig has very different values.. Sad
What do you think?

[Profile_0]
Name=VEGA56
GPU_P0=852;800
GPU_P1=991;900
GPU_P2=1138;950
GPU_P3=1269;1000
GPU_P4=1312;1050
GPU_P5=1474;1100
GPU_P6=1077;950
GPU_P7=1112;950
Mem_P0=167;800
Mem_P1=500;800
Mem_P2=700;900
Mem_P3=950;950
Fan_Min=2000
Fan_Max=4900
Fan_Target=50
Fan_Acoustic=2400
Power_Temp=85
Power_Target=0

I'm not using --fastjobswtich.

On second rig :

[Hash Rate Avg: 11840.6 H/s]
  1979.2 H/s GPU0
  1961.8 H/s GPU1
  1969.8 H/s GPU2
  1973.9 H/s GPU3
  1968.7 H/s GPU4
  1934.2 H/s GPU5

[Shares Found: 616 | Avg Search Time: 40.6 sec]
 591 ( 96%) Accepted
   0 (  0%) Rejected by pool
   1 (  0%) Invalid result computation failed
   2 (  0%) Could not be submit because of network error
  22 (  4%) Outdated because of job change

[22:12:07] GPU3 Found Nonce, submitting...
[22:12:07] Share Accepted -> +1
[22:12:07] Shares: 592 Accepted, 3 Errors | Hash Rate Avg: 11787.6 H/s | Avg Search Time: 40.6 sec
[22:12:07] GPU0 | 45°C | Fan 3311 RPM | 1991.5 H/s
[22:12:08] GPU3 | 41°C | Fan 3416 RPM | 1969.6 H/s
[22:12:08] GPU2 | 41°C | Fan 3286 RPM | 1966.0 H/s
[22:12:08] GPU4 | 40°C | Fan 2989 RPM | 1967.8 H/s
[22:12:09] GPU5 | 45°C | Fan 4199 RPM | 1927.9 H/s
[22:12:10] GPU1 | 44°C | Fan 3663 RPM | 1957.8 H/s
[22:12:12] GPU0 | 45°C | Fan 3502 RPM | 1983.0 H/s
[22:12:13] GPU3 | 41°C | Fan 3423 RPM | 1973.2 H/s
[22:12:14] GPU2 | 41°C | Fan 3295 RPM | 1970.7 H/s
[22:12:14] GPU4 | 40°C | Fan 2995 RPM | 1970.0 H/s
[22:12:15] GPU5 | 43°C | Fan 4212 RPM | 1931.0 H/s
[22:12:15] GPU1 | 44°C | Fan 3661 RPM | 1958.8 H/s
[22:12:17] GPU0 | 45°C | Fan 3618 RPM | 1979.0 H/s
[22:12:18] GPU3 | 43°C | Fan 3571 RPM | 1974.7 H/s
[22:12:19] GPU2 | 42°C | Fan 3340 RPM | 1971.4 H/s
[22:12:19] GPU4 | 42°C | Fan 3158 RPM | 1967.8 H/s