Another tweak i use - introducing MSIAfterBurner profile changes to switch the appropriate profile (I have all NVIDIA)
e.g. In Miners i invoke EthDcrMiner64.cmd which contains the following
title %*
"C:\Program Files (x86)\MSI Afterburner\MSIAfterburner.exe" -profile1
"C:\crypto\Megaminer.new\Megaminer-3.1\Bin\Ethash-Claymore\EthDcrMiner64.exe" %*
I have profile1 for memory intensive overclock (Ethereum), and profile5 for GPU clock specific (the rest)
So i have a cminer.cmd in each allowing me to add the relevant Afterburner profile swap
Overall this affords me 10%+ additional performance without any loss of stability
Very interesting!
"Miners"- it is directory MINERS in MM or a part some data code in some file?
"cminer.cmd in each" - that means you have cminer.cmd file for each miner?
Can you tell us step-to -step manual how we can add this feature to MM?
In the MegaMiner/Miners directory you have a json file for each miner supported
In ClaymoreETH.json i change
"Path" : ".\\Bin\\Ethash-Claymore\\EthDcrMiner64.exe",
to
"Path" : ".\\Bin\\Ethash-Claymore\\EthDcrMiner64.cmd",
And create a new file EthDcrMiner64.cmd in
Megaminer\Bin\Ethash-Claymore with the following contents:
title %*
"C:\Program Files (x86)\MSI Afterburner\MSIAfterburner.exe" -profile1
"C:\crypto\Megaminer.new\Megaminer-3.1\Bin\Ethash-Claymore\EthDcrMiner64.exe" %*
I chose to specify the FULL directory of my installation structure so i did not depend on what directory the miner was launched from
Also check your MSI AFterburner is installed in the directory and update y our .cmd accordingly
You do this for EACH miner you are using so change ccminerAlexis78.json
"Path" : ".\\Bin\\NVIDIA-Alexis78\\ccminer.exe",
to
"Path" : ".\\Bin\\NVIDIA-Alexis78\\ccminer.cmd",
and drop a new file ccminer.cmd with the following
title %*
"C:\Program Files (x86)\MSI Afterburner\MSIAfterburner.exe" -profile5
"C:\crypto\Megaminer.new\Megaminer-3.1\Bin\NVIDIA-Alexis78\ccminer.exe" %*
Sometimes you might find another miner name - say cminer-x64.exe - be careful these match correspondingly
So each Bin/*/cminer.cmd file is slightly different according to each .exe and any other setting you might like to differentiate
Note i used profile 1 for Claymore ETH and profile 5 for ALL other miners
Thats is because Ethereum mining is know to be memory speed intensive and most other algorithms are more GPU clock sensitive - including ZCASH
In AfterBurner you save your required overclock settings to profile 1 & 5 accordingly (or any other as you like
Choosing appropriate setting for a stable rig is an art in itself - there will always be the odd poor card in a rig that can cause instability if clocked too far
It s also VERY difficult to get a stable optimal rig unless all the cards are identical, the main problem is that the settings are generally specified as OFFSETS from the norm for each card and the base can vary so +100 for 1 card means something slightly different to +100 for another. This is ESPECIALLY important for the power limit setting
I have 2 rigs and a few other stray cards:
- 4 x 1060 GTX - 4 identical cards - i can set them the same but 1 card in particular tends to fail before the others - I have settled on
profile 1: +500 memory, +50 for the GPU /80% power limit (cap 96W) - if was was sole ethereum mining i might choose to lower the GPU overclock to zero and possibly -ve
profile 5: -500 memory, +50 for the GPU /80% power limit (cap 96W) - I am edging the GPU up until i lose stability and hoping to get to 150 (cap 96W)
- 6 x 1070 GTX - this is a mess - 6 cards, 1 Gigabyte, 4 MSI (3 models), 1 Palit - the PL are all different but i have targeted a CAP of around 120W each but otherwise using similar memory/gpu clock to the 1060 rig above
- i also have a 1080ti/1060 pair in a standalone PC and another 1070 on its own which i set accordingly, My 1080ti Caps at GPU 210 & memory is set just above +/-500
Clocking a single card is far easier than getting optimal settings for a rig - it is possible to have different settings for each card but its a pain to keep up with - i prefer mild overclock and more stability. I can get individual cards to GPU 150-200+ and memory to 650+ but then i have to pay more attention to resetting the rig/miners more often - and the power consumption starts to erode the extra revenue above a certain point - I don't have a lot of headroom on my PSU
I have 2 profiles:
5 for GPU clock focused (cminer, zecminer) , 1 for Ethereum memory focused (Claymore, ethminer) - you can use the relevant profile in each .cmd accordingly
Note that the .cmd files will not generally execute past the miner launch command as they are generally killed by Megaminer ahead of the next candidate so i choose to set the profile ahead of the miner launch and don't attempt to reset it back to "normal" after the miner completes (because it mostly doesn't)
), where exactly is it that I can set the destination pool for each coin? 
!