Topic: [NemosMiner] multi algo profit switching NVIDIA/CPU miner - page 61. (Read 289464 times)

Thanks for the feedback on using ahashpoolplus.
Interested to know if you see improvements in switching rate (less than current_estimate and more than Actual24hr) and earnings (more).

There should be more of the same coming for other pools. Just need to find the time to work on it...

Seeing good results on ahashpoolplus batch as well with 6 x 1070 rig, actually seems to make more then my 6 x 1080 ti Rig on blazepool at the moment. Will switch it to ahashpoolplus once payout is reached.

Using the Group Policy Editor
Start gpedit.msc.

Navigate to Computer Configuration > Administrative Templates > Windows Components > Windows Error Reporting.
Display Error Notifications : DISABLE
Disable Logging : ENABLE
Disable Windows Error Reports : ENABLE

AHashPoolPlus vs BlazePool Test using 72 GPU FARM (36 GPUs per Pool)

Six Rigs (Miner1-6)  : 6 x 1070ti (85%, +100 Core, +500 Mem)
Six Rigs (Miner7-12): 6 x 1070ti (85%, +100 Core, +500 Mem)

Time Period 2/12 7:30am to 2/14 7:30am
Allowed Pools to Settle after stopping miners
Algorithm: phi,skein,x17,Nist5,neoscrypt,blake2s
ActiveMinerGainPct 5
Interval 60


AHashPoolPlus generated : 0.021524928 ($198.78 USD)
BlazePool generated        : 0.016864128 ($155.74 USD)

Sorry Blazepool you lost this round.  

I am having an intermittent issue that I've never had before -- on the latest nemo / latest miner files occasionally "lyra2z" will hang up with a ccminer error. If I just go click "okay" NEMO will resume as normal. But that system will simply idle until I arrive and click OKAY.

Any fix for that ?

Having issue with hsrminer when mining neoscrypt, when it wants to save speed it gives error:

Cannot convert value "10320000000000,00" to type "System.Int32". Error: "De waarde is te groot of te klein voor een Int
32."
At C:\Nemos\Wrapper.ps1:57 char:4
+             $HashRate = [int]$HashRate
+             ~~~~~~~~~~~~~~~~~~~~~~~~~~
    + CategoryInfo          : InvalidArgument: ( [], RuntimeException
    + FullyQualifiedErrorId : InvalidCastIConvertible

Any idea how to solve this?

Your overclocking settings may cause your rigs to go down. I'm also not sure if overclocking your memory that much actually helps because timings of memory change as well. Having a Rig with 6 x GTX 1070 running at 75% with 85+ clock speed (100 would crash miner) and 200+ mem.

Update at 96 hours into the test, the btc/24hr is approximately as follows:
blazepool - 0.001757
ahashpool - 0.001452

Blazepool still ahead by approximately 20%, but is definitely trending downwards, I think they had a lucky streak in the first day or two of the test. I will try to keep going for another couple of days and see if it evens out more or if one starts pulling away further. Again, these results may be inaccurate due to unknown rig downtime; I've tried to account for it whenever I've been aware of it happening.

never ever, use only original source!!

Thank you for taking the time to respond.

For lyra2v2 in nemosminer you can use Excavator for me its the fastest and most stable across all my rigs ( 1050ti, 1060, 1070ti, 1080ti)

I also uploaded my copy here, download the zip and extract to your miner folder

NVIDIAExcavator-1.zip

Most of your questions are answered in this post: https://bitcointalksearch.org/topic/m.30205761. The comparisons I am doing are rather informal and should be treated as such. Any difference of 5% or less should be considered a tie on account of the rather low hashrate of a single GTX 1080 and I tend to runs these tests for 24, 48 or 72 hours when 7 days or more would likely be necessary. But since you all aren't paying me to do these tests I get to pick the contestants, run time etc... 

I am not using NemosMiner with NiceHash's pool - that doesn't make a lot of sense to me, frankly; miner scripts like NemosMiner along with algo-switching pools like Zpool, MiningPoolHub, HashRefinery, etc., are /alternatives/ to NiceHash, after all.

The running total of BTC does not take into consideration the 2% fee that NH charges to move "theoretical BTC" from your local miner to their wallet (no BTC is actually transferred, hence "theoretical"). I have a Coinbase account so I can transfer from the NH wallet to there for free.

I have no good explanation for why Zpool/NemosMiner are so far behind NH, but prior to this comparison I tried out Zpool in "single algo, convert to a coin besides BTC" with two different coins (same algo - Lyra2v2) and got excellent results. I therefore feel like the pool is trustworthy - it's not intentionally shortchanging me, in other words - so the difference in earnings may very well point to a flaw in the algo-switching.

For example, it takes vertminer - a specialized fork of ccminer strictly for lyra2v2 - a good 20 minutes before it gets fully up to speed on my 6x GTX 1060 rig: it starts out at around 120 MH/s and climbs to 133 MH/s - that right there is 11%.

I've noticed something similar with Neoscrypt, both with HSRminer and ccminer KlausT.

But as for why this is occurring, I have no clue.

Few questions about NH Vs Zpool:

Are you mining the same algos in both pools ? If not what algos are you mining in both ?
Are you using the same miners and settings ?
Are you using Nemos w/nicehash pool ?
Does you estimated 11% difference accounts for Nicehash fees to pay external wallets, as this would be the most similar of using zpool ?

Looking forward to your response, thanks for sharing your findings.

 

I just passed the 24 hour mark comparing NemosMiner on Zpool vs. NiceHash, since NH is the only reasonable metric I could come up with to evaluate the performance of these multi-algo mining pools.

So far NH is beating NM/Zpool by about 11%, but I am going to run the test for at least 2 more days - 72 hours total - because I am just using a single GTX 1080 in each rig.

Okay got the results of the full 24 hour testing Ahashpool and Blazepool ( not 24h pools ) .

So my results :

Rig1 with 5 x 1060 6gb Palit Jetstream at Ahashpool - 0.00088961 BTC - Estimated by nemos miner are ~0.00115 = so minus 30% of estimated ( subject of change)
Rig2 with 5 x 1060 6gb KFA  at Blazepool                - 0.00083065 BTC - Estimated by nemos miner are ~0.00105 = so minus 23% of estimated ( subject of change)

Give then fact that my Rig2 cards are doing like 10% less hashrates then Rig1 blazepool is pulling a little more then the ahashpool for the time of the test. ( like 0.00003 ).

Profits are down in everything right now and like it seems ppl need to pick a pools that are steady and dont have alot downtime. This is the only thing that matter as it seems that the differences are too small.

same here

hi guys, just got updated from GitHub and saw this error below
any idea?

thanks


PS>TerminatingError(Invoke-WebRequest): "Unable to connect to the remote server"
Add-Member : Cannot add a member with the name "CryptoNight" because a member with that name already exists. To
overwrite the member anyway, add the Force parameter to your command.
At C:\NemosMiner-v2.4.2-master\NemosMiner-v2.4.2.ps1:111 char:62
+ ... h {$Pools | Add-Member $_ ($AllPools | Where Algorithm -EQ $_ | Sort  ...
+                 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
    + CategoryInfo          : InvalidOperation: (@{Bitcore=; Bla...Hodl=; Decred=}:PSObject) [Add-Member],
InvalidOperationException
    + FullyQualifiedErrorId : MemberAlreadyExists,Microsoft.PowerShell.Commands.AddMemberCommand
Add-Member : Cannot add a member with the name "CryptoNight" because a member with that name already exists. To overwrite the member anyway, add the Force parameter to your
command.
At C:\NemosMiner-v2.4.2-master\NemosMiner-v2.4.2.ps1:111 char:62
+ ... h {$Pools | Add-Member $_ ($AllPools | Where Algorithm -EQ $_ | Sort  ...
+                 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
    + CategoryInfo          : InvalidOperation: (@{Bitcore=; Bla...Hodl=; Decred=}:PSObject) [Add-Member], InvalidOperationException
    + FullyQualifiedErrorId : MemberAlreadyExists,Microsoft.PowerShell.Commands.AddMemberCommand

To all using ahaspoolplus be it with NemosMinerPlus or NemosMiner 2.4.2.

There will be a reboot of the server performing calculations some time tomorrow.
Impacts on your rig will be:
•   During reboot (couple minutes) = Your rig will stick to mine what they are mining. [=> no impact]
•   After reboot (20 to 30 minutes) =  Your rig will act as if using ahashpool (Not 24rh). After that, back to normal ahashpoolplus estimations. Code needs around this time to learn. [=> low impact]

So… Should not have a big impact on earnings but wanted to make sure you’re aware of what happens.

I mine on zergpool do you recommend me x17 blake2s and phi on that pool?

Can anyone please provide some guidance on the popular over/underclock settings for each algo?

I got 3x 1070 TI and currently running all 3 on 75% power limit, +200 core, +700 memory in AfterBurner and it's stable for everything. Just wanted to optimize per-algo now since it's a possibility with the latest Nemos update. Thanks!

top 3 profit algos are x17, blake2s, and phi

So if you mine neoscrypt for 24 hours you will have low profits