[NemosMiner] multi algo profit switching NVIDIA/CPU miner

[strowheim]

All of a sudden my 2.4.1 version which was working perfectly started failing and throwing errors. Running on ahashpool...

Anyone else see this issue? Looks like maybe the API call was getting an unexpected response. Switched back to 2.3 and its fine.

Transcript started, output file is .\Logs\miner.log
Loading BTC rate from 'api.coinbase.com'..
Loading pool stats..
Cannot convert null to type "System.DateTime".
At C:\Users\user\Desktop\NemosMiner-v2.4.1-master\Include.ps1:34 char:5
+     $Stat = [PSCustomObject]@{
+     ~~~~~~~~~~~~~~~~~~~~~~~~~~
    + CategoryInfo          : InvalidArgument: ( [], RuntimeException
    + FullyQualifiedErrorId : nullToObjectInvalidCast
Cannot convert null to type "System.DateTime".
At C:\Users\user\Desktop\NemosMiner-v2.4.1-master\Include.ps1:34 char:5
+     $Stat = [PSCustomObject]@{
+     ~~~~~~~~~~~~~~~~~~~~~~~~~~
    + CategoryInfo          : InvalidArgument: ( [], RuntimeException
    + FullyQualifiedErrorId : nullToObjectInvalidCast

Cannot find an overload for "op_Subtraction" and the argument count: "2".
At C:\Users\user\Desktop\NemosMiner-v2.4.1-master\Include.ps1:51 char:5
+     $Span_Minute = [Math]::Min(($Date - $Stat.Updated).TotalMinutes,  ...
+     ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
    + CategoryInfo          : NotSpecified: ( [], MethodException
    + FullyQualifiedErrorId : MethodCountCouldNotFindBest
Cannot find an overload for "op_Subtraction" and the argument count: "2".
At C:\Users\user\Desktop\NemosMiner-v2.4.1-master\Include.ps1:51 char:5
+     $Span_Minute = [Math]::Min(($Date - $Stat.Updated).TotalMinutes,  ...
+     ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
    + CategoryInfo          : NotSpecified: ( [], MethodException
    + FullyQualifiedErrorId : MethodCountCouldNotFindBest

[millsys]

All of a sudden my 2.4.1 version which was working perfectly started failing and throwing errors. Running on ahashpool...

Anyone else see this issue? Looks like maybe the API call was getting an unexpected response. Switched back to 2.3 and its fine.

Transcript started, output file is .\Logs\miner.log
Loading BTC rate from 'api.coinbase.com'..
Loading pool stats..
Cannot convert null to type "System.DateTime".
At C:\Users\user\Desktop\NemosMiner-v2.4.1-master\Include.ps1:34 char:5
+     $Stat = [PSCustomObject]@{
+     ~~~~~~~~~~~~~~~~~~~~~~~~~~
    + CategoryInfo          : InvalidArgument: ( [], RuntimeException
    + FullyQualifiedErrorId : nullToObjectInvalidCast
Cannot convert null to type "System.DateTime".
At C:\Users\user\Desktop\NemosMiner-v2.4.1-master\Include.ps1:34 char:5
+     $Stat = [PSCustomObject]@{
+     ~~~~~~~~~~~~~~~~~~~~~~~~~~
    + CategoryInfo          : InvalidArgument: ( [], RuntimeException
    + FullyQualifiedErrorId : nullToObjectInvalidCast

Cannot find an overload for "op_Subtraction" and the argument count: "2".
At C:\Users\user\Desktop\NemosMiner-v2.4.1-master\Include.ps1:51 char:5
+     $Span_Minute = [Math]::Min(($Date - $Stat.Updated).TotalMinutes,  ...
+     ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
    + CategoryInfo          : NotSpecified: ( [], MethodException
    + FullyQualifiedErrorId : MethodCountCouldNotFindBest
Cannot find an overload for "op_Subtraction" and the argument count: "2".
At C:\Users\user\Desktop\NemosMiner-v2.4.1-master\Include.ps1:51 char:5
+     $Span_Minute = [Math]::Min(($Date - $Stat.Updated).TotalMinutes,  ...
+     ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
    + CategoryInfo          : NotSpecified: ( [], MethodException
    + FullyQualifiedErrorId : MethodCountCouldNotFindBest

Delete all the pool stats (keep the ccminerxxxxxxx files) in STATS folder will fix the problem.

[BitBustah]

Hello BitBustah,

Thank you for testing on Blazepool.  How many cards are you running and did you include the nemos benchmarking times in that 9 hrs?  Judging an autoswitching miner along with an autoswitching pool with 9 hours of data without comparison to any identical setup with other pool during the same timeframe is kind of unfair.  We welcome everyone to do a little more systematic testing.  Afterall, it's really for your benefit and the community's benefit too to investigate deeper.

Happy and good luck mining.

Not dissing your pool. Love the design. I can see a lot of efford went into it.

Took me hours for Nemos to end benching. For me the earnings are too low, but that's not because of your pool; It's because of the times we live in right now.

[MrPlus]

We should not use 24 hour actuals in our profit switching script, that makes sense.
But current estimate has a high fluctuation.

Does anyone see a reason why we should not just use the average of both values? (as long as current estimate is not zero)
Or we could just check if current estimate is not zero and still use 24 hour actuals?

I use a customized version of Nemos which does some of this.
I'ts less switching than CurrentEstimate / more than Actual24hr. Works great for me since about 2 weeks.
Moreover with the current down trend. Following Actual 24hr is not great.
Made it available here: https://github.com/MrPlusGH/NemosMinerPlus-v2.4.2/releases. Have a go if you like.

Added functionality:

prerun feature

  Ability to run a batch prior switching to a specific algo.
  For example, can be used to set OC via nvidiaInspector
  Simply create a file named <AlgoName>.bat in prerun folder
  If <AlgoName>.bat does not exist, will try to launch prerun/default.bat

ahashpoolplus

  Added ahashpoolplus as a pool.
  Simply use -PoolName ahashpoolplus in start.bat
  Uses calculations based on 24hractual and currentestimate ahashpool prices to get more realistic estimate.
  Includes some trust index based on past 1hr currentestimate variation from 24hr.
  This shows less switching than following Current Estimate and more switching that following the 24hr Actual.
  AND is NOT sensible to spikes.
  Better profitability on our rigs.
  Test and share the results on yours.
  Only for ahashpool. Working on expanding the feature to other pools.

Algo switching log

  Added simple algo switching log in csv. switching.log file found in Logs folder.
  Can be easily imported in excel as csv file.

[Max3k]

2x1080 and 24hr: 0.00034... this is almost 5 times less than on a regular pool, or 2.5 times less than what NiceHash pays 

[millsys]

We should not use 24 hour actuals in our profit switching script, that makes sense.
But current estimate has a high fluctuation.

Does anyone see a reason why we should not just use the average of both values? (as long as current estimate is not zero)
Or we could just check if current estimate is not zero and still use 24 hour actuals?

I use a customized version of Nemos which does some of this.
I'ts less switching than CurrentEstimate / more than Actual24hr. Works great for me since about 2 weeks.
Moreover with the current down trend. Following Actual 24hr is not great.
Made it available here: https://github.com/MrPlusGH/NemosMinerPlus-v2.4.2/releases. Have a go if you like.

Added functionality:

prerun feature

  Ability to run a batch prior switching to a specific algo.
  For example, can be used to set OC via nvidiaInspector
  Simply create a file named <AlgoName>.bat in prerun folder
  If <AlgoName>.bat does not exist, will try to launch prerun/default.bat

ahashpoolplus

  Added ahashpoolplus as a pool.
  Simply use -PoolName ahashpoolplus in start.bat
  Uses calculations based on 24hractual and currentestimate ahashpool prices to get more realistic estimate.
  Includes some trust index based on past 1hr currentestimate variation from 24hr.
  This shows less switching than following Current Estimate and more switching that following the 24hr Actual.
  AND is NOT sensible to spikes.
  Better profitability on our rigs.
  Test and share the results on yours.
  Only for ahashpool. Working on expanding the feature to other pools.

Algo switching log

  Added simple algo switching log in csv. switching.log file found in Logs folder.
  Can be easily imported in excel as csv file.

Looks great, I am testing it on all my rigs now.  Thank you!

[primo1983]

The big problem of mining multiple currencies is the temperature of the GPUs, this had to be automatic as it is being complicated to maintain a good temperature. I accept suggestions of what to do I own GTX 1080 TI

[nickvasel]

I am wanting to try NEMOSMINER, What VER should I be using 2.4.1 or 2.4.2? Scratch that. I downloaded both edited the.bat file for the pool I want to use and this is what I get.
Cannot process argument transformation on parameter 'API_ID'.<----------- ANY HELP PLEASE...

[millsys]

The big problem of mining multiple currencies is the temperature of the GPUs, this had to be automatic as it is being complicated to maintain a good temperature. I accept suggestions of what to do I own GTX 1080 TI

MSI Afterburner does this for all my rigs.  Set 85% power limit, +100 GPU, +600 MEM, AUTO FAN, now click gear icon next to auto fan and add a 60 degrees / 60 % dot to the graph and a 69 degrees / 70% dot to the graph.
Now apply the settings to all cards.  Must set these values for different models, MSI will group similar cards together.

Now as the miner changes ALGO, if one particular algo runs the card hotter, MSI AfterBurner will adjust the fan speeds accordingly.  The extra 60 and 69 degree points help ramp up the fans a little sooner, default settings like to keep the temps in the 70s and I prefer keeping them in the 60s if possible.

[primo1983]

The big problem of mining multiple currencies is the temperature of the GPUs, this had to be automatic as it is being complicated to maintain a good temperature. I accept suggestions of what to do I own GTX 1080 TI

MSI Afterburner does this for all my rigs.  Set 85% power limit, +100 GPU, +600 MEM, AUTO FAN, now click gear icon next to auto fan and add a 60 degrees / 60 % dot to the graph and a 69 degrees / 70% dot to the graph.
Now apply the settings to all cards.  Must set these values for different models, MSI will group similar cards together.

Now as the miner changes ALGO, if one particular algo runs the card hotter, MSI AfterBurner will adjust the fan speeds accordingly.  The extra 60 and 69 degree points help ramp up the fans a little sooner, default settings like to keep the temps in the 70s and I prefer keeping them in the 60s if possible.

send the preent here would be legal for a base

[primo1983]

The big problem of mining multiple currencies is the temperature of the GPUs, this had to be automatic as it is being complicated to maintain a good temperature. I accept suggestions of what to do I own GTX 1080 TI

MSI Afterburner does this for all my rigs.  Set 85% power limit, +100 GPU, +600 MEM, AUTO FAN, now click gear icon next to auto fan and add a 60 degrees / 60 % dot to the graph and a 69 degrees / 70% dot to the graph.
Now apply the settings to all cards.  Must set these values for different models, MSI will group similar cards together.

Now as the miner changes ALGO, if one particular algo runs the card hotter, MSI AfterBurner will adjust the fan speeds accordingly.  The extra 60 and 69 degree points help ramp up the fans a little sooner, default settings like to keep the temps in the 70s and I prefer keeping them in the 60s if possible.

send the preent here would be legal for a base

see my preent
https://image.prntscr.com/image/9JkVfiFjT2OvRcOOl2SyDg.jpg

[millsys]

The big problem of mining multiple currencies is the temperature of the GPUs, this had to be automatic as it is being complicated to maintain a good temperature. I accept suggestions of what to do I own GTX 1080 TI

MSI Afterburner does this for all my rigs.  Set 85% power limit, +100 GPU, +600 MEM, AUTO FAN, now click gear icon next to auto fan and add a 60 degrees / 60 % dot to the graph and a 69 degrees / 70% dot to the graph.
Now apply the settings to all cards.  Must set these values for different models, MSI will group similar cards together.

Now as the miner changes ALGO, if one particular algo runs the card hotter, MSI AfterBurner will adjust the fan speeds accordingly.  The extra 60 and 69 degree points help ramp up the fans a little sooner, default settings like to keep the temps in the 70s and I prefer keeping them in the 60s if possible.

send the preent here would be legal for a base

see my preent

Click AUTO (its not colored! must be colored)
Click the squiggly lines on the mini GPU menu to 'sync', right now its set to NOT SYNC between similar cards
Now Click Checkmark to commit to similar cards

[primo1983]

The big problem of mining multiple currencies is the temperature of the GPUs, this had to be automatic as it is being complicated to maintain a good temperature. I accept suggestions of what to do I own GTX 1080 TI

MSI Afterburner does this for all my rigs.  Set 85% power limit, +100 GPU, +600 MEM, AUTO FAN, now click gear icon next to auto fan and add a 60 degrees / 60 % dot to the graph and a 69 degrees / 70% dot to the graph.
Now apply the settings to all cards.  Must set these values for different models, MSI will group similar cards together.

Now as the miner changes ALGO, if one particular algo runs the card hotter, MSI AfterBurner will adjust the fan speeds accordingly.  The extra 60 and 69 degree points help ramp up the fans a little sooner, default settings like to keep the temps in the 70s and I prefer keeping them in the 60s if possible.

send the preent here would be legal for a base

see my preent
https://image.prntscr.com/image/9JkVfiFjT2OvRcOOl2SyDg.jpg

Click AUTO (its not colored! must be colored)
Click the squiggly lines on the mini GPU menu to 'sync', right now its set to NOT SYNC between similar cards
Now Click Checkmark to commit to similar cards

see, but do not get colored in that skin is so even it is self it moves
https://image.prntscr.com/image/4WH7C9PiS2iCq8SwoiPJFw.jpg
see the fan
https://image.prntscr.com/image/kZp3LBT6QOmN2ey4awks_g.jpg

[GitPushForce]

We should not use 24 hour actuals in our profit switching script, that makes sense.
But current estimate has a high fluctuation.

Does anyone see a reason why we should not just use the average of both values? (as long as current estimate is not zero)
Or we could just check if current estimate is not zero and still use 24 hour actuals?

I use a customized version of Nemos which does some of this.
I'ts less switching than CurrentEstimate / more than Actual24hr. Works great for me since about 2 weeks.
Moreover with the current down trend. Following Actual 24hr is not great.
Made it available here: https://github.com/MrPlusGH/NemosMinerPlus-v2.4.2/releases. Have a go if you like.

Added functionality:

prerun feature

  Ability to run a batch prior switching to a specific algo.
  For example, can be used to set OC via nvidiaInspector
  Simply create a file named <AlgoName>.bat in prerun folder
  If <AlgoName>.bat does not exist, will try to launch prerun/default.bat

ahashpoolplus

  Added ahashpoolplus as a pool.
  Simply use -PoolName ahashpoolplus in start.bat
  Uses calculations based on 24hractual and currentestimate ahashpool prices to get more realistic estimate.
  Includes some trust index based on past 1hr currentestimate variation from 24hr.
  This shows less switching than following Current Estimate and more switching that following the 24hr Actual.
  AND is NOT sensible to spikes.
  Better profitability on our rigs.
  Test and share the results on yours.
  Only for ahashpool. Working on expanding the feature to other pools.

Algo switching log

  Added simple algo switching log in csv. switching.log file found in Logs folder.
  Can be easily imported in excel as csv file.

Dude, with all due respect, but editing nemo's tool and *removing* his devfee address to put yours instead is definitely not nice. I see the point in wanting to get benefit from your work, but I'd suggest:
1. Be transparent about it (IE: Communicate)
2. Do not fully remove the original author's devfee. Consider a mixed solution.

[millsys]

The big problem of mining multiple currencies is the temperature of the GPUs, this had to be automatic as it is being complicated to maintain a good temperature. I accept suggestions of what to do I own GTX 1080 TI

MSI Afterburner does this for all my rigs.  Set 85% power limit, +100 GPU, +600 MEM, AUTO FAN, now click gear icon next to auto fan and add a 60 degrees / 60 % dot to the graph and a 69 degrees / 70% dot to the graph.
Now apply the settings to all cards.  Must set these values for different models, MSI will group similar cards together.

Now as the miner changes ALGO, if one particular algo runs the card hotter, MSI AfterBurner will adjust the fan speeds accordingly.  The extra 60 and 69 degree points help ramp up the fans a little sooner, default settings like to keep the temps in the 70s and I prefer keeping them in the 60s if possible.

send the preent here would be legal for a base

see my preent

Click AUTO (its not colored! must be colored)
Click the squiggly lines on the mini GPU menu to 'sync', right now its set to NOT SYNC between similar cards
Now Click Checkmark to commit to similar cards

see, but do not get colored in that skin is so even it is self it moves

see the fan

Try MSI Afterburner 4.4.2
https://imgur.com/a/Ni8zW

[nickvasel]

I am wanting to try NEMOSMINER, What VER should I be using 2.4.1 or 2.4.2? Scratch that. I downloaded both edited the.bat file for the pool I want to use and this is what I get.
Cannot process argument transformation on parameter 'API_ID'.<----------- ANY HELP PLEASE...

Is this where you get help for NEMOSMINER?
EDIT: Don't all answer at once now...

[MagicSmoker]

We should not use 24 hour actuals in our profit switching script, that makes sense.
But current estimate has a high fluctuation.

Does anyone see a reason why we should not just use the average of both values? (as long as current estimate is not zero)
Or we could just check if current estimate is not zero and still use 24 hour actuals?

I use a customized version of Nemos which does some of this.
I'ts less switching than CurrentEstimate / more than Actual24hr. Works great for me since about 2 weeks.
Moreover with the current down trend. Following Actual 24hr is not great.
Made it available here: https://github.com/MrPlusGH/NemosMinerPlus-v2.4.2/releases. Have a go if you like.

Holy Cannolis, I was trying to teach myself enough about powershell scripting so I could modify NemosMiner to do something almost like this! My idea was to frequently query the pool for the highest profit algo, but do a moving average of the results to reduce chasing some shit-coin just because it shot up in profitability for all of 5 minutes. Also, some auto-switching pools inexplicably use PPLNS (though, granted, I don't know what "N" is) so frequent coin - much less algo - switching could be even more painful.

[SpliceOfLife]

Howdy folks
I get this wee error on palgin neoscypt:
Cannot convert value "12280000000000.00" to type "System.Int32". Error: "Value was either too large or too small for
an Int32."
At C:\Users\The Beast\Documents\Cryp\miners\NemosMiner-v2.4.2-master\Wrapper.ps1:57 char:4
+             $HashRate = [int]$HashRate
+             ~~~~~~~~~~~~~~~~~~~~~~~~~~
    + CategoryInfo          : InvalidArgument: ( [], RuntimeException
    + FullyQualifiedErrorId : InvalidCastIConvertible

When I'm benchmarking and the hash rate is through the roof unrealistic when finished.


I  also got an error running the xevan benchmark. FAILED 1 is displayed under the profitability table in nemosminer. I commented out xevan in the miner that was displaying the problem and that seems to have fixed it.

Found the problem on github. Do you think you have a fix for this nemo ? Thanks for all the updates.
Hope everybody is hashing healthy!

~SOL

[primo1983]

how to lower the temperature because only the temperature goes beyond the limits
https://image.prntscr.com/image/cVdwDyVvQqSwLs9lwuCbQg.jpg

[MagicSmoker]

how to lower the temperature because only the temperature goes beyond the limits
...

Increase the fan %!