ethminer-0.9.41-genoil-1.1

[clintar2]

Code:

ethminer.exe -C -S eth-eu.dwarfpool.com:80/0Myxa2541adressss/worker

doesn't work

no it doesn't. my fork doesn't support dwarf stratum. a. their stratum is not stratum and b. their faq says my stratum is buggy while their implementation is simply out of spec.

i will support it if i get paid to do it  

I hear you, Genoil. NO pools have standard "stratum" support, and the vardiff is also flakey.

Coinotron, ethpool/ethermine, miningpoolhub are pretty good. Coinmine.pl as well although there's a little hack in my code to support it.

I've added support at my pool for direct stratum support of this protocol if anyone wants to try it. http://eth.cncoin.farm I've needed someone to do a good test on it besides my tiny 60MH. I've tried with both amd OpenCL and cuda versions and they seem to work, but some more hammering would be appreciated.

[1715358501]

1715358501

[malekbaba]

Please answer me if any one knows, which driver works best for AMD cards, should I install newest version of the driver ?

[Slydrule]

Please answer me if any one knows, which driver works best for AMD cards, should I install newest version of the driver ?

Do not install the 16.x versions of the drivers.  They will reduce your hash rate.  I have used 15.12 and 15.7 install packages for AMD and both provide the same top hash rate. 

[smaxz]

why does pool mining generate additional DAG's after the initial one is created?

i have noticed nicehash does not use this additional dag for their purposes.

would there be any reason for another dag to be created while solo mining with geth?

[dodther]

Miner freze after DAG change.
http://s017.radikal.ru/i419/1605/02/8580152c40de.png

Win7x64, genoil-1.0.8

[adaseb]

Miner freze after DAG change.


Win7x64, genoil-1.0.8

yeah i got that problem also. Looking for a fix

[Ayers]

affecting only amd i guess, maybe only low memory card, 2-3gb? because my 970 work flawless after the dag change

[Genoil]

affecting only amd i guess, maybe only low memory card, 2-3gb? because my 970 work flawless after the dag change

Did this happen while mining and then after a restart it ran again? In this case I will have to build me some fake DAG switching code to finally put and end to these reports 

Or, does the miner not work any longer alltogether after restarting it? In that case could you specify the GPU type(s)?

[adaseb]

affecting only amd i guess, maybe only low memory card, 2-3gb? because my 970 work flawless after the dag change

Did this happen while mining and then after a restart it ran again? In this case I will have to build me some fake DAG switching code to finally put and end to these reports 

Or, does the miner not work any longer alltogether after restarting it? In that case could you specify the GPU type(s)?

Restarting the miner makes it work again, or closing the miner before the DAG change block.  But if you leave it running, the AMD cards crash.

[Genoil]

affecting only amd i guess, maybe only low memory card, 2-3gb? because my 970 work flawless after the dag change

Did this happen while mining and then after a restart it ran again? In this case I will have to build me some fake DAG switching code to finally put and end to these reports 

Or, does the miner not work any longer alltogether after restarting it? In that case could you specify the GPU type(s)?

Restarting the miner makes it work again, or closing the miner before the DAG change block.  But if you leave it running, the AMD cards crash.

thanks. on top of the list for 1.0.9!

[dodther]

affecting only amd i guess, maybe only low memory card, 2-3gb? because my 970 work flawless after the dag change

Did this happen while mining and then after a restart it ran again? In this case I will have to build me some fake DAG switching code to finally put and end to these reports 

Or, does the miner not work any longer alltogether after restarting it? In that case could you specify the GPU type(s)?

Restarting the miner makes it work again.

[restless]

I have no idea how is memory reserved, but what if on start miner attempt to reserve bigger buffer than needed for current DAG, say, big enough for (current epoch + 2)?

[Genoil]

I have no idea how is memory reserved, but what if on start miner attempt to reserve bigger buffer than needed for current DAG, say, big enough for (current epoch + 2)?

That's how Claymore's works. I have to assess the problem first to see if the problem even lies there. Judging by the fact that CUDA cards don't seem to crash (mine alos doesn't), it probably is an OpenCL related bug. 

[pehoko]

How to run ethminer for each single miner gpu as one instance - multi miners?
Tried but not working - ethminer crash.

[gami82]

I have no idea how is memory reserved, but what if on start miner attempt to reserve bigger buffer than needed for current DAG, say, big enough for (current epoch + 2)?

That's how Claymore's works. I have to assess the problem first to see if the problem even lies there. Judging by the fact that CUDA cards don't seem to crash (mine alos doesn't), it probably is an OpenCL related bug. 

Perhaps it helps...my Rigs are not crashing while DAG change. All my Cards are Sapphire R9 390 - Win 10 64BIT Crimson 15.12 8GB RAM Virtual 16GB Miner Version: ethminer-0.9.41-genoil-1.0.7

[io8621]

Hi,
some questions, i always used dwarfpool and their ethproxy,
but, after some problems with the pool in the last days, i started to move a couple of rig to ethermine.org,
and use genoil with builtin stratum.
My question is: with stratum is better get a more stable hashrate by increasing the value of --farm-recheck or not?
My rig are all with r7 370 or r9 380, and i use this bat file:

setx GPU_FORCE_64BIT_PTR 0
setx GPU_MAX_HEAP_SIZE 100
setx GPU_USE_SYNC_OBJECTS 1
SET GPU_SINGLE_ALLOC_PERCENT=100

ethminer -G -S eu1.ethermine.org:4444 -O xxxxxxxxxxxxxxxxxxxxx.home --farm-recheck 200 --cl-local-work 256 --cl-global-work 8192

Thanks for your opinions

[Genoil]

In stratum mode a higher farm-recheck only gives a more stable hashrate display. 2000 works well. It doesn't affect actual miner efficiency.

[pehoko]

Do we need farm-recheck if stratum is present?

[Genoil]

Do we need farm-recheck if stratum is present?

No. Also, 1.0.8. will default it to 2000 when stratum is enabled.

BTW Linux users of 1.0.8 are recommended to disable work-timeout (using --work-timeout 999) until we fix this bug: https://github.com/Genoil/cpp-ethereum/issues/38

If you still get it after this, please let me know.

[Vaccinus]

how to enable stratum mode with 1.0.8?