Nope, but that isn't an issue. You can leave the GPUs supplied by power all the time via their 6/8 pin connectors without issue, and no, they don't draw very much power in that state (maybe 1W each?).

RX 550 isn't cheaper in the US. Check out this sorted/filtered list on newegg: https://www.newegg.com/Product/ProductList.aspx?Submit=ENE&N=100007709%20601296396%20601296397%208000&IsNodeId=1&bop=And&order=PRICE&page=1

If you can get an RX 560 for the same price, or even less, than an RX 550 why wouldn't you? Unless, of course, you could get an RX 570 for MSRP... but that hasn't been possible for a couple of months now.

Maybe, but considering that an errant powershell script can literally destroy a Windows installation, not really.

Look at the substance of what I wrote: there is not even a basic description of what NemosMiner does in the first post. It is *assumed* by the OP that everyone knows what a multi-pool miner does. Well, maybe that is true at the so-called "30,000 foot level", but I still want to know the details like, where is the data from API polling stored, where/how is that data manipulated, how are errors handled like, say, a benchmark failing (one of my test rigs was basically idled for 8+ hours because "timetravel" failed... that's dumb no matter how you slice it), how are miners configured, hell, what does every switch in the pool batch files do!?, etc... None of that is explained either in the first post or on the "github" - more on that below.


It's not that I am unsympathetic to your position - indeed, I always read the first 2-3 and last 2-3 pages of a thread before asking a (potentially stupid) question - but after *decades* of using forums (starting with BBSes, USENET, etc...) I have a fairly keen sense of how things ought to proceed in matters like this, and the first post is by far the best place to put updates. Github is a relatively new creation, and besides, few ever bother to write a decent readme.md file for their git repository, anyway. See, for example, the SGMiner-GM github with a readme.md that is 4+ years old. Seriously?

Well, I'm still waiting for an explanation of what, precisely, this collection of half-assed scripts does, so it's no surprise the first post wasn't updated with a changelog.

And btw, pointing someone to post #3421 and asking rhetorically, "not sure where you've been looking" is kind of a dick move. While we can all fantasize about every one reading 3000+ post threads before asking a question, that's just flatly unrealistic, so let's not pretend otherwise, mmm?

Believe me, after all the trash talking I've done about NiceHash (aka NiceHack), no one was more surprised as each and every multi-algo pool I've tested over the last month has come in 2nd place, and often by a startling amount. HashRefinery and Zergpool both brought in 30% less over their 1 week test period than NH. Granted, both pools happened to get on the wrong side of some hard-forks, but since that isn't something you have to worry about with NH I can't really toss out the comparisons as invalid. Both MiningPoolHub and Zpool came in around 10% less than NH over their test periods (1 week for MPH, only 3 days for Zpool) and given the limitations of my test setup I am inclined to call these a tie, but that's rather missing the point: the whole idea behind multi-algo, auto-switching pools is to beat single-coin mining AND NiceHash over time (if you are the type that either cashes out in fiat on a regular basis or just wants to accumulate BTC without buying a shitty ASIC).

And especially since NH is rather restricted in the algos it offers I am even more surprised at the poor showing of these multi-algo pools, but maybe it's not the pool's fault, maybe it's NemosMiner's, so I am contemplating trying another multi-algo miner manager next.

They do make breakout boards that are hardwired to turn on when AC is applied, and usually that version is cheaper. If you already have this board and it requires pressing a button to restart it then you'll need to get creative if you don't know how to solder.

If I knew what the top 6 algos were at any given moment I wouldn't need NemosMiner?!? I could just make the usual rolling list of ccminer, etc., command line entries in a batch file and loop through them every 60s (or whatever)!

I may well give ahashpool a shot (using the plus version of the batch file) but as of now I don't really have enough hashpower to make me comfortable trying out a pool with a 0.01 BTC minimum payout. I did take note of your test results, but I'd like to see how ahashpool compares with NiceHash, because so far every single one of these multi-algo pools has lost out to the 800lb gorilla.

I wish I could help you guys having trouble, but aside from not getting the correct memory the first time (Amazon's fault, not mine), and wasting time trying to use one of the GPUs to drive a monitor because the Aliexpress listing specifically said onboard video was disabled (it wasn't, and had to be used, actually) I haven't really fiddled with mine and only once checked the BIOS settings but didn't change any.

I can tell you that a code 99 during boot usually means some critical piece of the mobo hardware isn't working. Perhaps the PCIe bus expander or the like. This would likely be a fatal error unless you can find something set wrong in the BIOS (but, to reiterate, mine was set perfectly for its intended task as a miner).

@Fuzzy Kittens - your issue could be from the new BIOS changing the default video adapter from onboard/integrated to external/PCIe.

Yep, I have been adding cards to my 2nd Onda D1800 6-GPU mobo as I get them in and I really like these boards, BUT....

The 5cm PCIe slot spacing on my D1800's is just too close to allow the GPU fans to pull in enough air, and I am not running terribly powerful cards in either rig. One D1800 is kitted out with RX 560s and they aren't such a problem - only 75W max TDP and a bit thinner, too - but the other one is loaded up with GTX 1060s and it has been a real struggle to keep the first 5 cards cool (the last one is fine since its fan is unobstructed).

The Celeron CPU is also woefully underpowered - you pretty much have to stop mining if you need to interact with it at all. I'm not saying it needs an i5 or a Ryzen or anything, but even something about 2x faster would be a tremendous improvement. That said, the wimpy CPU is a main reason why the mobo only needs about 30W...

But will that half of an RX 550 do at least half the hashrate of a full one? 

Yeah, I had another crash after about 36 hours of operation. That's pretty good, but my 6x GTX 1060 Equihash rig will run for a week without a reboot; maybe longer, but Windows Update rarely lets me go more than a week without forcibly rebooting. This despite manually disabling its service... One of the things I absolutely despise about W10.

GenX dinosaur checking in; I did have a spendable merit so I gave it to you. Really excellent work.

Actually, concise but helpful posts are even more likely to get merited because of the "tl;dr" types! 

All of the 2GB XFX RX 560's I've received so far have Elpida memory so I've gotten lucky in the silicon lottery. A 4th is on the way and I hope I didn't just jinx myself 

Once I get all 6 slots on my Onda D1800 filled out I am going to revisit xmr-stak and claymore because SGMiner is just so finicky to get working right and it has been abandoned for about a year now. The supposed replacement - gateless gate sharp - is still very much in beta testing (lots of crashes being reported).

Hmm, yes, you mentioned pruning the algo list before when timetravel got stuck during benchmarking, but after thinking about it some more I couldn't see the reasoning behind removing any algo from the list as long as the pool supports it, the algo hasn't been "ASIC'ed", the pool uses the Proportional payout scheme (so you don't get screwed if no blocks are found) and the pool auto-converts everything to BTC. Maybe skunk or tribus or timetravel suck right now, but maybe they won't in the future. And I know for sure that Lyra2RE2 has good days because I keep my eye on a couple of coins that use that algo.

I'm just about to install the newest version - 2.4.2 - so this would be as good a time as any to do some pruning, but I don't want to shoot myself in the foot here.

In order reach the minimum balance on Zpool in a reasonable time I copied my existing NemosMiner installation on to my other GTX 1080 rig. Everything is running smoothly and since both rigs are more or less the same my existing benchmarks and algos should work fine. I changed the refresh interval to 300 seconds on both setups and started them within a few minutes of each other so API checks aren't exactly in sync, but they aren't, say, 10 minutes apart.

What's really strange is that they have not been mining the same algo, nor even showing the same ordering of profitability. And I am not just talking about 1 or 2 switched places, I mean the profitability estimates are way, way different. For example, on rig 1 the first 5 algos listed are Skein, Tribus, HSR, Skunk & Phi, while on rig 2 the order is Lyra2RE2, Skunk, Phi, HSR, Tribus.

It's almost like the profit switching script is randomly guessing?!?

There is so much misunderstanding in those few few sentences I hesitated replying...

The rate at which you earn coins is proportional to the hashrate of your cards, but also inversely proportional to the hashrate of the entire network; that is what dictates "difficulty". So the more miners that decide to mine ETH the more difficult it becomes to mine ETH.

Furthermore, Ethash is not the 1070's strong suit: Lyra2v2, Equihash, Skein, Nist5 are several algos that perform much better on it; AMD cards do a lot better at Ethash (and Cryptonight).

So if you really want Ethereum I would suggest either selling your mining equipment and buying ETH directly or mining a coin your hardware excels at, such as ZEN, ZEC, VTC, XVG, etc., then trade it for ETH (NB - you may have to trade for BTC first).

In all seriousness, you should stick to NiceHash. No shame in that game - I've usually got one rig on NiceHash at all times now, because it is at least predictable, if not the most profitable.

Low market cap / sub-rosa coins that aren't listed on major exchanges are the deep end of the speculative alt pool and like with any speculative investment just a tiny portion of your assets (ie - money or hashpower) should be allocated to them.

I've been using it on Zpool the last few days with no issue. In fact, I was doing a comparison between NemosMiner/Zpool vs. NiceHash (spoiler - NiceHash won; see this post here: https://bitcointalk.org/index.php?topic=1751272.msg30341376#msg30341376).

In fact, every single comparison I've done between NiceHash and a multi-algo, auto-switching pool - always using NemosMiner - has ended with NiceHash winning. As a result, I've concluded that it's just not worth trying to beat NiceHash at its own game. YMMV, etc. and so on.

At the 72 hour mark Zpool fell behind NiceHash again, with a net earnings of 0.90309 mBTC for Zpool vs. 1.01949 mBTC for NH. While that is an 11.% difference, I can't really say it is significant statistically because just yesterday Zpool was nearly tied with NH while the day before it was behind NH by 12.1%, so daily earnings fluctuate quite a bit.

What I can say for sure, though, is that the estimated profit data which NemosMiner is obtaining/using to select which algo to run is persistently overstated by anywhere from 1.5x to 3x the actual earnings so because of this, as well as a few other glaring deficiencies in these profit switching scripts, I think I will stick to using Zpool in single-algo mode.

Meh, I'm not so worried about being an exit node for a VPN as long as there is full encryption of all internal communications, and, frankly, any VPN that allowed intermediaries in the network to observe what was going on inside it would no longer be /private/.

But I vote with my hashpower and DERO has the most promise of any Cryptonote coin I've come across; ITNS seems to be struggling a bit here.