That's the benefit of using NiceHash over running individual mining algorithms. While in some cases it might be a little more profitable to run the mining algorithms directly, you would need to keep a constant check on profitability. NiceHash automatically switches to the most profitable algorithm/coin to mine. For that reason, it's important to run the benchmarks so that it can calculate and keep a check on profitability.
One critical note about NiceHash vs Pool Mining. NiceHash, you're not mining for pool reward share; You are mining for other people. NiceHash states that they will not compete and buy their own hashing power, ONLY IF it does not dip far (they have never defined) below actual value of each coin (They want to make sure miners feel it worthwhile to stay with the platform). This has some very odd side effects.
First off, there is NiceHash Solo, a separate pool mining system, that they promote heavily inside of their hash buying platform. Solo is essentially a pool, but instead of shared profits based on inputs, it is essentially a lottery.
Secondly, a lot of the pricing in NiceHash ends up being based on speculation. If something is surging or showing signs of surging, you can end up making a fair chunk of money (automatically) based on surges. For example, Zcash the other week. That went clear ham, and ended up with mining profits up as high as I saw as about 75% over average. If you were in a pool, you might not see that, especially if the prices are swinging wildly or sell at the right time.
Now, on top of this, I did do some preliminary numbers, and I find that unless you use one of the automated balancing software packages, you will likely see more cash out at the end of the day with NiceHash because something's always surging and increasing the price. The biggest issue with NiceHash, is the slow change between mining different coins, and the changing in different jobs, which ends up eating much of those gains. I was hoping that NiceHash 2 would end up have some sort of pre-queuing system, where it would spool up the next job when it comes up, so there would be little to no downtime in the changeover. Unfortunately they don't see the benefit in the reduced job change time.
Third, while you're mining specific coins, you are paid out in bitcoin. Some people don't like this; To others, it makes it super easy. This is up to you. However, this in itself causes its own world of challenges. Not only are you dealing with altcoin fluctuations, you also are running on bitcoin fluctuations. As well, the transaction fees are climbing so high on bitcoin, that you will lose a lot more than if you mined directly and spent directly.
However, there are a few more quirks to mining in bitcoin. As much as the bitcoin INCREASING in value is nice to our wallets, it also means people don't pay as much in bitcoin for mining, due to the fact that in an increase scenario, you find that your dollars in will be lower than the gain in market on the upturn, than when bitcoin is flat and has been flat for a bit. ALSO, during a downturn, people will throw cash into mining fast, and pay high. I don't see much benefit when it is running downwards; Sure, you're making more BTC, but at a dropping price.
I looked up the specs on my graphics card (an NVIDIA GTX 1060 6GB) and it looks like idle power consumption is about 9 watts. Seems to be running about 70 watts while mining. So I'm using 59 more watts than sitting idle. I know my electricity cost in kWh. Any thoughts on how I calculate my costs of running this?
Unless you're running it in your daily workstation, you need to factor in your entire system W cost as cost for power. That's why many people run many cards in 1 system, to reduce the overhead. And if you want to be technical, in the workstation, you should be factoring the entire card, not just the difference.
We're going to dip our toe in this water as well - looks like fun (ok, clearly geek-fun, but whatever). We've got a nice Optiplex i5/8GB castoff machine, a GTX1060 6G card, and a kill-a-watt unit to measure the electricity cost. One upfront question: Is there any inherent advantage to running Linux vs. Windows as the host OS? I think the answer is "no", but I've only read a few writeups and don't have any anecdotal experience to draw from.
Really depends. End of the day, the OS does not put enough (any?) overhead on that is going to offset it. If you're using nicehash, Windows is the way to go; Sure, you can set up miners to work with NiceHash, but you'd need another balancer, unless you only intend at mining one coin in exchange for bitcoin.
The biggest perk apparently to Linux, is the handling of more than 5-6 GPUs in the system, and a lot more automation abilities as well as a free licence.
I was thinking about OS and for just dipping our toes in the water I don't think it matters. But if you're running a bunch of rigs and scaling then I'd probably shy away from Windows. I don't want a forced update restarting my system.
Personally, I don't recommend using anything but Windows 7 for a full-time mining system. Windows 10 becomes a nightmare on a few levels including the one you mentioned. For mass deployment, Linux could be easier, but again, if you're doing mass deployment, you could just clone drives too.
)
