GeForce GTX 1080 8GB Mining Performance

Looking for something of a middle ground between the GTX 1070 and the GTX 1080 Ti? That ground belongs to the GeForce GTX 1080, formerly the fastest mining GPU that was readily available. With the 1080 Ti now playing king of the hill, prices on the 1080 have dropped by about $120. The result is a GPU that falls right between the 1070 and 1080 Ti, both in price as well as performance.

The GTX 1080 is a great card, and you can still potentially do six GPUs in a single rig -- but you'll be getting pretty close to maxing out most PSUs if you go that route. I generally prefer the GTX 1070, but the 1080 isn't terrible. As usual, I'd recommend getting a custom cooler design rather than the blower-style Founders Edition, since they have larger heatsinks and don't run as hot -- or as loud! I've got Gigabyte's GTX 1080 WindForce OC, and at $500 even it's one of the least expensive models -- only the blower-style Asus 1080 Turbo is typically cheaper, but it's also less desirable and not worth the $10 saved.

Here's how the WindForce OC performs (which is similar to many other factory overclocked 1080 models):

The GTX 1070 averages between $3.50-$4.10 per day, with a power draw of about 180W. That means at $0.10 per kWh in electricity, it will cost about $0.43 per day, giving a profit of $3.10-$3.70 per day. As you might expect, ROI isn't quite as good as the 1070, requiring 135-160 days for a single card, added to an existing system. But that's not too bad either.

As usual, here's the math:

Basic PC setup: $870 (no case -- use a wire shelf)
Ryzen 5 1600: $230 (I like the 6-core/12-thread models, for when you're doing other stuff)
Asus Prime-X370 Pro motherboard: $150
16GB DDR4-2667: $97
240GB SSD: $63 (boot faster, install updates faster)
1350W PSU: $240 (Platinum isn't really much worse than Titanium in practice)
PCIe riser adapters: $50 (don't use the SATA to Molex power adapter, though!)
Six GTX 1080 cards: $500 each = $3,000
Total = $3,870, income of $21.60 per day ~= 180 days to ROI

Cutting costs on the CPU could shave off $170 from the total, bringing you down to around 171 days to ROI with a Pentium G4560 and a Z270 motherboard. Nine days isn't worth the savings IMO, but do what you feel is best. Basically, the ROI time is equal to the 1080 Ti, so I'd lean toward the faster GPUs personally. Nvidia GPUs currently do well with Lyra2REv2, Equihash, and occasionally X11Gost and Lbry when mining via NiceHash. Note that some drivers/hardware combinations will do better than what I show in the above image, but this is how one of my 1080 cards is currently running.

GeForce GTX 1070 8GB Mining Performance

One of the best graphics cards around right now, in terms of gaming performance as well as mining performance, is Nvidia's GTX 1070 8GB. There are many reasons why this is a great card, the most important factors being:

1. Availability -- you can find GTX 1070 cards all over the place
2. Price -- the latest price cuts bring the 1070 down to less than $350
3. Power -- the 1070 needs a single 8-pin power connector at most
4. Temperatures -- the 1070 runs relatively cool and quiet

If you're in the market for a bunch of GPUs to use for mining, right now I'd give serious consideration to the 1070, specifically EVGA's GTX 170 SC, which is one of the least expensive models right now and still comes with a decent factory overclock. I've got several of these cards, and while they're not the fastest option, they're some of the best overall rate of return. Here are the stats for one of the factory overclocked cards:

Currently, the GTX 1070 is averaging $3.00-$3.50 per day, with a power draw of 150W -- yes, about the same power requirements as an RX 470/480 or RX 570/580. It costs more, but it also earns more, with a power cost of around $0.35 per day.

With a current price of $344 on the EVGA SC model, that works out to 109-130 days to hit ROI on each card. Even better, due to power requirements you could actually hook up six of these to a single system, provided the motherboard supports such a setup. Which of course you can certainly find. As I did with the GTX 1080 Ti recently, let's look at the complete PC ROI numbers.

Here's the quick math:
Basic PC setup: $870 (no case -- use a wire shelf)
Ryzen 5 1600: $230 (I like the 6-core/12-thread models, for when you're doing other stuff)
Asus Prime-X370 Pro motherboard: $150
16GB DDR4-2667: $97
240GB SSD: $63 (boot faster, install updates faster)
1200W Titanium PSU: $280 (efficiency is really important!)
PCIe riser adapters: $50 (don't use the SATA to Molex power adapter, though!)
Six GTX 1070 cards: starting at $344 each = $2,064
Total = $2,934, income of $18 per day = 163 days to ROI

If you want to cut costs, you could probably go with 8GB RAM and a cheaper CPU, but I wouldn't go too low -- at some point, you might want to do something else with the PC, and having a better processor is really nice. But if you want to go with a Pentium G4560 and a Z270 motherboard, that will save about $170. If you do that, you could hit ROI potentially in 155 days.

Either way, that's a bit faster than building a GTX 1080 Ti mining rig. Another thing I like about going the Nvidia route instead of AMD is that there tends to be more flexibility on algorithms when you're mining via NiceHash. Most of my AMD rigs don't seem to like certain algorithms, and outside of DaggerHashimoto, they're outperformed by Nvidia equivalents. If Ethereum weren't so big, AMD mining would be far less useful.

Radeon R9 Fury X, R9 Fury, and R9 Nano Mining Performance

 A long time ago, in a galaxy far, far away, AMD launched their ginormous Fiji architecture, the final gasp from their power hungry GCN v3 days. And if you got one of the cards for free, I suppose they were worth using for mining purposes, but for everyone else the cards are a bit of a joke. Except, nowadays you can sometimes find a Fury or Fury X floating around for $300-$350. Could the Fury possibly be worth the price of admission at that point?

I happen to have a trio of Fury cards, and by that I mean I have the usual trifecta of Fiji products: R9 Fury X with a CLC, R9 Fury in gargantuan Asus Strix format, and the diminutive R9 Nano. They're all happily mining away, though as you'll see in a moment, the earnings are only mediocre given the price. And don't go thinking that it's only now that the Fury cards are a poor choice -- when they were new, they cost almost twice as much as the R9 390, which actually performs pretty similarly for mining purposes.

You see, the problem with the Fiji chips is that they were never popular for mining, so the mining software was never fully optimized for the chips, which means they underperform. Also, in many hashing algorithms having tons of memory bandwidth isn't all that important (it varies, naturally), so HBM doesn't always show much benefit.

Anyway, for better or worse, here are the numbers -- for historical context if nothing else.

Both the Fury and Fury X end up showing the same R9 Fury label, but the faster card (listed first) is the Fury X. Really, there wasn't a huge difference in specs -- Fury X had a few more cores and slightly higher clocks, but memory bandwidth was the same. So here we are, a year and a half after the Fury launched, and the cards are only slightly faster than an RX 470 that costs half as much.

As for the R9 Nano, it's running in a different rig, and as you can see the profitability is pretty pathetic -- worse than an R9 390 by a decent margin. And that's the benefit of sticking with the 'popular' mining cards, because in practice the R9 Nano should actually be faster than the R9 390.

The bottom line is that the AMD Fiji-based GPUs were never great picks for mining from a price and power perspective, which in turn makes them worse from a software optimization perspective. You can mine with them, yes, but the Fury X at least will use 275W, or about $0.66 per day in power, and while the liquid cooling works great for temperatures, it's cumbersome if you want to put a bunch of cards into a mining rig. The R9 Fury is a bit better in some ways, but still not a great pick, and the prices on R9 Nano still put it out of consideration.

Even today, if you could find one of these cards for $300-$350, you'd be looking at ROI in around 135-160 days at best, and potentially as long as 200 days. Sadly, there are simply too many better alternatives -- including the old Radeon R9 390, which to this day typically matches the Fury X on income potential. With used 390 cards selling for $240 or less, at least there the ROI is closer to the RX 470: about 100-120 days. It's still more power hungry, though, which means you might end up limited by your circuit breaker if you want to build multiple rigs.

How to get the fastest GPU in the world and have it pay for itself in less than six months

I wrote last month about the GTX 1080 Ti Founders Edition, and showed initial mining performance. Things are constantly in flux in the world of cryptocurrency, so what was once a great idea may suddenly become a terrible plan. Or, as is the case here, a decent idea can suddenly turn into an awesome idea.

I've got a pair of GTX 1080 Ti cards, chugging happily away. I conservatively estimated income for each card at about $3 per day (minimum), but that's proving far too pessimistic. Right now, I'm seeing income of anywhere between $4.50 at a minimum to as much as $7.50 maximum. Now, the maximum isn't likely to last for days at a time, but on average you can probably count on $4.50-$5.00 per day of net profit.

I'm partial to the non-reference designs for a variety of reasons, and my second card is the MSI GTX 1080 Ti Gaming X 11G. This is a large, heavily factory overclocked 1080 Ti. Check this out:

These are two examples taken during the past day of how the 1080 Ti is doing. $5.25 to $5.90 per day would mean that you can buy for a $740 and have it pay for itself in 140 days at a moderate return, or it might do even better and earn that back in just 125 days. That's four months to earn back the cost, after which the card is purely profit. Power costs are a factor, though, which means subtract about $0.60 per day from the earnings. You're still looking at breaking even in less than six months, easy!

But why not go the other way, and buy four cards for a single system, a 1350W PSU, and you're still looking at ROI in six months or less. The difference is that with four cards cranking away, once you hit ROI you're looking at around $21 extra in income each day, every day. Could you use $600 a month in spending money? I know I could! How about $2,400 extra a month? Build four such mining rigs!

And it really is that easy. I'm not saying the 1080 Ti is the best mining card on the planet, but it's super fast and you can do four cards per PC instead of trying for six cards. That's a bit less management headache.

Just a final caveat: don't plan on running 20 or more of these systems in your home. Each one would use over 1000W of power, and on a typical 200A circuit you're limited to 24,000W on the entire circuit -- total! You still need to account for lights, appliances, etc. And if you're on a 150A or 100A circuit breaker, you'll have to get by with half of that. Also: fire hazard. Still, doing 5-10 systems is certainly possible.

Here's the quick math:
Basic PC setup: $870 (no case -- use a wire shelf)
Ryzen 5 1600: $230
Asus Prime-X370 Pro motherboard: $150
16GB DDR4-2667: $97
240GB SSD: $63 (boot faster, install updates faster)
1200W Titanium PSU: $280 (efficiency is really important!)
PCIe riser adapters: $50 (don't use the SATA to Molex power adapter, though!)

Four GTX 1080 Ti cards: starting at $680 each = $2,720
Total = $3,590, income of $20 per day = 180 days to ROI
Five PCs = $17,950, income of $100 per day. Still 180 days to ROI
Ten PCs = $35,900, income of $200 per day. Still 180 days to payoff!

More importantly, once the initial hardware has paid for itself, you're not earning $6,300 per month. Okay, maybe more, maybe less. It doesn't matter too much, as bottom line is you have a very nice salary that will continue to pay you with little to no additional effort.

If you're the enterprising type, maybe rent a warehouse with a couple of 200A circuits and plunk down $150K on forty crazy powerful PCs. Then earn $800 per day, 365 days per year, and you'd have the equivalent of a $292K salary. Ha. But I'd suggest starting smaller and working your way up to that level, rather than jumping in whole hog. And don't be surprised if the market shifts and you see profits drop by half!

Radeon RX 570 4GB Mining Performance

Going along with my testing of RX 580 8GB mining performance, I also have a shiny new Asus RX 570 4GB card. It's running in the same Core i5-7500 system as the 580, so as you might expect the numbers are similar overall -- just lower. Here's how the card performs, at the stock OC, on the various current algorithms:

Again, DaggerHashimoto numbers are pretty low. I could try VBIOS flashing or various other tweaks, but this is running NiceHashMiner, and of course the RX 570 requires AMD's latest drivers -- which tend to not be as optimal for mining as some of the older drivers.

Based on the current numbers, looking at WhatToMine, you'll gross about $2.00-$2.25 per day, with a power draw of around 125W per card. With a power cost of $0.10 per kWh that puts net earnings at about $1.80-$2.00 per day. The Asus card sells for $190, so you'll hit ROI in about 100 days. This is a really great result, and it's why the 470 and now 570 are some of the best GPUs for cryptocurrency mining. Power draw is a bit less than the 580, prices are a bit lower, and performance is a bit lower.

You'll want to shop around a bit, and potentially other systems could run this same GPU 5-10 percent faster. It's just sort of 'luck of the draw' combined with CPU and platform factors. As with the 580, you could run up to six RX 570 cards in a single rig without too much difficulty. Getting a good 1200W or higher Platinum or Titanium rated PSU is recommended -- Rosewill's 1200W Tokamak is another decent alternative, with a few quirks (like no power switch).

Radeon RX 580 8GB Mining Performance

I like to test out new graphics cards with Bitcoin mining, and on the hotseat today is AMD's new Radeon RX 580 8GB. I'm not Bitcoin directly, obviously, but using the cards for mining altcoins via NiceHash. The recent upward trend in Bitcoin prices -- 25 percent during the month of April 2017! -- has definitely helped my bottom line, but over the long term, what will really matter is the performance and power use.

I've also found it's difficult to get a clear view of mining performance from the various GPUs. People will show results from heavily overclocked cards, results that are not long-term stable in my experience. This is stupid. If you modify your VBIOS to use tighter latencies on the GDDR5, it can improve performance by 5-10 percent ... and a few weeks later your cards will all start failing. I've been there with the RX 480, and all that downtime just isn't worth the hassle!

My advice is simple: buy a factory overclocked card with a massive cooler, use PCIe risers to separate the cards from each other, and then just run at the default clocks. The cards will run stable, at reasonable temperatures, and you're on your way to profitability. In fact, rather than overclocking, sometimes it's better to underclock and undervolt your card, to reduce power and temperatures. You're far less likely to burn out a fan, or overheat, or any number of other potential problems.

Okay, let's get on to the RX 580 8GB. I'm using the Asus ROS RX 480 Gaming OC, and absolutely massive card that's bigger than many GTX 1080 models. The benefit is it has three large fans, the cooler keeps things frosty, and it comes with a healthy factory overclock. It uses a single 8-pin power connector, and in practice if you install it in a case, you'll use three slots -- you really can't put anything into the two adjacent slots due to the size of the cooler.

And here's how the card performs, at the stock OC, on the various current algorithms:

You'll notice perhaps that the DaggerHashimoto numbers are pretty low, which is basically variance between GDDR5 types, memory subtimings, and the test system. In this case, I'm running the RX 580 on a Core i5-7500 with a B250 motherboard. That might be a factor, or it might just be a combination of relatively early hardware and the lack of specific tuning for the RX 580. You might think the 580 and 480 are the same hardware and should perform the same, but in practice there are all sorts of contributing factors. As an example, I have an RX 470 card that does 29 MH/s -- at stock factory settings! -- for DaggerHashimoto. (The card is no longer in stock anywhere and tended to sell for $240+ if you're wondering.)

Anyway, based on the current numbers, you'll earn about $2.25-$2.75 per day, with a power draw of around 150W per card. With a power cost of $0.10 per kWh that puts net earnings at about $2.00-$2.40 per day. In the case of the Asus card, which sells for $280, you'll hit ROI in about 140 days, which is a really great result.

This is with one of the most expensive RX 580 cards, of course, and you could drop to a 4GB model without really losing any performance. As an alternative to the Asus, the MSI RX 580 Gaming X 4G costs just $215, and it has similar clockspeeds in the default 'Gaming' mode. ROI on a card like this is only 100-110 days, conservatively -- and if Bitcoin prices improve, you might hit ROI in as few as 60 days. Or things could go the other way and it might take 200 days.

Overall, the RX 580 is one of the best options for cryptocurrency mining right now. With a single 8-pin connector, it makes it relatively easy to put four or even six GPUs into a single mining rig. I recommend getting a great power supply for such a configuration, with the Enermax Platimax 1350W sitting at the top of my list.

GeForce GTX 1080 Ti Founders Edition Mining Performance

Last week, Nvidia launched their hot and sexy new graphics card, the GTX 1080 Ti. This is a beast of a GPU, with 3,584 CUDA cores clocked at 1480-1582MHz stock -- and really, that's a conservative estimate on the clockspeed, as the card will routinely boost above that level. But how does the card work as a mining GPU? I decided to do a quick check.

There you have it, the numbers from the latest version (1.7.5.3) of the NiceHashMiner. This is running fully stock, but I've found you can get at least 10 percent more performance by increasing the power limit on the card. At higher power draw, obviously, so it might not be worth doing for a lot of people.

Given this is a 250W TDP card, it's not realistic to try and slap six of them into a mining rig -- even the highest wattage PSUs would be a questionable choice. You could do two PSUs I suppose, but it's probably a better idea to limit your rigs to four of these if you're serious about mining.

In terms of performance, the GTX 1080 Ti is almost twice as fast as a single GTX 1070, or easily more than twice the performance of a GTX 1060 6GB card. It also costs twice as much as a GTX 1070, and nearly three times as much as a GTX 1060 6GB. Twice the power, twice the performance, twice the cost... but you could potentially double the hashrate per PCIe slot.

ROI at earnings of around $3 per day is around 233 days, so not awesome for a single card but not really that bad either. You can also underclock/undervolt a bit to improve ROI.

UPDATE: Currently, income from mining with the 1080 TI is way better than $3 per day, so RIO is now closer to about 150 days. Fluctuations will change that over time, but I expect a realistic ROI to be under 180 days.

Six GPU mining rig for Ethereum, ZCash, ROI, and more

Introduction

I've been dabbling in cryptocurrency for nearly six years now. Recently, I had a discussion with someone regarding an old article where I declared GPU mining dead. It's definitely not dead, though you have to know what you're getting into before you start. But here's an updated look at the subject of making money with cryptocurrencies.

The biggest issue is time -- things don't stand still, prices fluctuate, and you're going to need to stick with mining for a good long time before you're in the black. I'm not talking decades or anything like that, but volatility and your approach to selling/hoarding will play a role, along with some luck.

On the luck side, compare purchasing $10,000 worth of mining hardware with buying $10,000 worth of Bitcoin or Ethereum. If you did this a bit more than a year ago, Ethereum was sitting at around $1-$2, and today it's at about $20. $10,000 of Ethereum in 14 months would have become $50K-$100K! Bitcoin didn't do quite so spectacularly. It went from $375-$400 to about $1,200, so it would still be a massive increase in net worth.

Mining on the other hand won't pay out hugely, particularly if you're not mining and hoarding (and eating the power costs). I know from personal experience that roughly $10,000 worth of optimal mining hardware purchased around the same time would have generated somewhere between $12,000 and $16,000 during the past 14 months... but it would also cost about $4,000 in electricity, never mind the noise and potential upkeep.

Here's the thing. That $10,000 investment in mining hardware could be sold today for about $7,000. So where the risk with buying straight into cryptocurrency is 100 percent -- you could potentially lose everything -- with mining you've always got the hardware. But the risk if you buy BTC is arguably a lot lower, because BTC isn't going away. Worst-case, it might drop in value by half. But mining can be a lot more fun that just buying Bitcoin, and some people already have a fair amount of extra hardware available.

With that preamble out of the way, here's the current pricing and equipment needed to build a mining rig with support for up to six GPUs.

Parts List

6x Radeon RX 480 4GB (choose your favorite brand) OR
6x Radeon RX 470 4GB (choose your favorite brand) OR
6x GeForce GTX 1060 6GB (choose your favorite brand) OR
6x GeForce GTX 1060 3GB (choose your own adventure)

1x motherboard with six PCIe slots (any brand could work, but finding six PCIe slots requires a bit of searching)

6x Powered 1x PCIe to x16 risers (USB cable version)

1x Pentium G4500 CPU (dual-core, Hyper-Threading, and cheap!)

2x4GB DDR4 memory (no need to get higher performance RAM)

1200W 80 Plus Platinum PSU (get a good one with a long warranty!)

120GB SATA SSD (because I hate waiting for HDD to reboot)

Use a wire shelf for your case (much cheaper and easier, and cools well)

Don't forget: Zip ties, power meter, and a surge protector. These should be self-explanatory, but you'll want to know your real power draw (from the wall), and surge protection is a great idea.

Putting it together

Now, the practical experience of piecing it all together. First, put a piece of non-conductive material under the motherboard (wood will suffice, though it's flammable, so....) If you want to install some mounts for the motherboard, that's not a bad idea, but it's not required -- it just keeps the board from sliding around.

If you don't know how to install an OS and the rest of the hardware (PSU, SSD, RAM, etc.) you should probably stop right now. It's not particularly difficult, but if setting up a new PC scares you, go search out some videos on the subject and get comfortable with that before investing a bunch of money into mining hardware.

For the GPUs, hang them from the top shelf and secure them with Zip ties. Use the PCIe risers, but make sure to skip the included Molex to SATA power adapters -- just use a Molex connector direct from the PSU. You might also want to get a few extra risers, as sometimes one or two are bad/nonfunctional. Keep a moderate amount of space between the GPUs so they all get plenty of fresh air.

I'd get the system up and running with a single GPU first. Whether you use Windows or Ubuntu or some other flavor of Linux is up to you. I tend to use Windows because I'm most familiar with that OS and I know how to get all the tools for modding and flashing the VBIOS, mining, etc.

Once the system is running, install your mining software (Nicehash is easy and pretty good). If you want to be thorough, start with one GPU and test all of your riser cables and PCIe slots -- make sure they all function properly. This will require a bunch of shutting off and restarting, obviously, but it can save you some headaches in the long run. Once you've verified all six risers and slots are working, you can hook everything up so that it looks something like this:

Both of those are pictures from the web, with far nicer cabling than I usually do. If you want to do multiple miners, so two systems with six GPUs each, that's entirely possible. Note that you'll want to have sufficient power (and cooling for the room!) and that usually means not putting more than two rigs on a 15 Amp circuit.

To keep power down, and help avoid killing your GPUs, for RX 470 4GB cards I recommend doing a VBIOS flash (Google it if you don't know how) that sets the RAM clock to 1900-2000MHz, and limits the GPU core to 1100MHz. You might want to adjust the fan speeds as well. That should keep each 470 at around 125W. (Feel free to ask me for more details if you need help.)

Expected performance / ROI

With everything assembled, your actual performance and earnings will vary depending on the market, the hardware you selected, and what you plan on mining. I use Nicehash because it will switch algorithms automatically and pays out in BTC, which is 'safe'. Others prefer mining Ethereum or ZCash (or whatever the hot new coin is) directly. Your choice. Generally speaking, here's what I would expect for performance/power use, in order of increasing ROI time.

6-way RX 480 4GB: Earnings of around $5.50-$8.50 per day, depending on the market, with a power use of about 900W per mining rig. I expect the average over the next year to be close to $6.50 per day. Total cost (at present) is $1,650, and power use will be 21.5kWh. If you pay $0.10 per kWh, that's about $2.15 per day for power, netting you $4.35 per day, maybe more on good days. ROI would be about 380 days.

6-way RX 470 4GB: Earnings of around $5-$8 per day, depending on the market, with a power use of about 800W per mining rig. I expect the average over the next year to be close to $6 per day. Total cost (at present) is $1,575, and power use will be 19kWh. If you pay $0.10 per kWh, that's about $2 per day for power, netting you $4 per day, maybe more on good days. ROI would be about 400 days.

6-way GTX 1060 3GB: Earnings of around $4.50-$7 per day, depending on the market, with a power use of about 550W per mining rig. (Yes, Nvidia cards use quite a bit less power than AMD.) I expect the average over the next year to be close to $5.50 per day. Total cost (at present) is $1,750, and power use will be 13kWh. If you pay $0.10 per kWh, that's about $1.30 per day for power, netting you $4.20 per day, maybe more on good days. ROI would be about 415 days.

6-way GTX 1060 6GB: Earnings of around $5-$8 per day, depending on the market, with a power use of about 600W per mining rig. (Yes, Nvidia cards use quite a bit less power than AMD.) I expect the average over the next year to be close to $6 per day. Total cost (at present) is $2,000, and power use will be 14.5kWh. If you pay $0.10 per kWh, that's about $1.50 per day for power, netting you $4.50 per day, maybe more on good days. ROI would be about 450 days.

As with any predictions, take the above as estimates only. If the cryptocurrency market does well, you might hit ROI in six months. If it does poorly, it might take two years. Your power bill and computer parts are also tax deductible, if that helps -- and you'll have a huge power bill if you build out six of these rigs, I can promise you that!

Bitcoin's Late 2016 Surge

Bitcoin is a strange beast, fluctuating at times wildly between highs and lows. It's like the stock market, only with fewer regulations. Is that a good thing? I'm not entirely convinced. What I do know is that here at the end of 2016, Bitcoin prices are climbing rapidly yet again.

In the last week, BTC prices have shot up about 15 percent, which is impressive--and a nice end-of-year bonus for anyone holding BTC right now (assuming you sell at a profit, of course). But that's a really narrow view. In the last month, the cryptocurrency is up more than 30 percent, and over the last year it's a 125 percent gain. So everyone should go all-in on BTC yet again, right?

Not so fast. BTC is nothing if not volatile, and while we're now at a 3-year high (the last time BTC was in the $900+ range was in January 2014), the all-time high was in November 2013, of around $1100. If you bought into BTC on that last massive wave (attributed to China hysteria over BTC by many), up until the last month or so you were likely sitting on a loss of 50 percent or more.

What's behind the new interest in BTC? Could it be that it's finally gaining mainstream acceptance? I seriously doubt that -- in fact, I offered a friend a $50 gift of BTC recently, and he turned it down flat. "I don't believe in that stuff and don't want to deal with the hassle" was his reply in a nutshell. Talk about looking a gift horse in the mouth, but that attitude is far more widespread than proponents would like to admit. News of the IRS going after Coinbase records doesn't help, though at least that would offer some clarity on the tax situation. We're all supposed to pay taxes on our BTC gains, right -- but if you're mining, calculating 'gains' is a rather complex proposition.

But there's good news as well, like the India and Venezuela monetary policies potentially pushing people into BTC (and gold and other commodities). And I love the "I told you so" posts by people who predicted BTC hitting $1000 earlier this year -- it's happening, so of course they're pointing at their old posts and tweets and showing us all how prescient they were. Never mind the fact that most of these pundits have predicted BTC surges into the $1000+ range every year going back to 2013.

What's funny about all of this to me is that everywhere people are making the same mistake. Correlation does not equal causation. India and Venezuela make some policy changes and BTC started to go up in price. Did the policy changes cause the BTC price increase, or did they merely happen at the same time? Or maybe Trump winning the US election caused the increase! This is extremely difficult to prove.

The good news is that the lack of regulations on Bitcoin -- and the Internet echo chamber -- means that people can spout hype till they're blue in the face, and if enough people start to believe the hype, it becomes a self-fulfilling prophecy.

I got into Bitcoin/cryptocurrency back mid-2011, at the first surge to $31. I watched BTC tumble to less than $2, and it took nearly two years (19-20 months) before it surpassed the previous high. But when it did, it shot up to $200+, which seemed incredible at the time, only to be eclipsed by the record-setting explosion six months later. In other words, I've seen this all before ... and I still can't say I have any real knowledge of what will happen next.

2017 could be the year BTC breaks the $2000 mark. Or it could be the year where BTC briefly breaks $1000 again only to fall back to $500. And that's why, rather than trying to guess at the investment game of Bitcoin, I stick with mining when it's profitable. And lately, BTC (cryptocurrency mining) has been doing quite well.

The current king for cryptocurrency mining profitability: GTX 1060.

What's the best bet for a new mining system? You might be surprised to know that it's NOT going to be something built on AMD GPUs -- not even the Radeon RX 470 4GB or Radeon RX 480 4GB. Best case right now, you're looking at breaking even in around 195 days for the RX 480 and 219 days for the RX 470. Which isn't terrible, but you can do better. Enter the GeForce GTX 1060 6GB, which at current rates will hit ROI in about 167 days. If you have an existing system, that works out to around $1.50 per day, give or take.

Here's the thing: if you don't have a system ready for the GPU, the cost of building such a system totally changes the equation. Put three of these into a modest build and you're looking at $1,100 minimum, and possibly $1,350 or more. With three GPUs earning $1.50 each, you're looking at $4.50 per day, which means it's now going to require the better part of a year to break even -- assuming prices stay relatively consistent.

And there's the rub. Prices are never consistent in the cryptocurrency world. New coins launch to much hype, and if you're just mining via Nicehash (which is what I've been doing), you can avoid all the pump and dump nonsense and often make a healthy profit. I've got about eight systems doing mining, and typically they earn a combined $30-$40 on most days. That rate has been pretty steady for at least six months, except for things like early November where it shot up to $100-$200 per day for a week.

My advice is that if you're interested in Bitcoin / mining, take a very long-term approach. Don't hope to pay off a computer mining rig in three months or even nine months, but if you keep it humming along for more than a year, you'll almost certainly hit ROI, and everything after that is profit. It may not be a lot of profit (though sometimes it is), but this is how I've managed to keep at it. Happy New Year, and happy mining!

Build an Ethereum Miner with Four RX 470

It's been a while since I last talked about cryptocurrency mining. The past two years have seen plenty of changes if you drill down, but pull back and look at the big picture and the cryptocurrency world hasn't radically altered our finances. Scamcoins and hacks have come and gone, Bitcoin continues plugging along, but Litecoin has dropped to under $4, Darkcoin has changed its name to Dash, and the second most popular cryptocurrency is now Ethereum.

I'm not going to get into the discussion of whether or not Ethereum is a better or worse solution than Bitcoin -- or any number of other coins -- but the fact that it's actually supported on Coinbase says a lot about it's potential. It's currently trading at over $13 as well, which is pretty impressive. So let's talk mining potential.

First, if you hopped on the Ethereum boat at launch and saved your coins, congratulations! You've struck it rich and should be quite happy. For us Johnny-come-lately miners, it's going to take a bit of work to break even. I've poked around at looking at Ethereum mining (including dual mining with Ethereum and Library Coin). You've got a few options, and here's what I've found in terms of hashrates, using the Claymore Dual Miner 0.7.1:

RX 470 (stock): ~20MH/s ETH, ~25MH/s LBRY, 150W
RX 470 (OC): ~25MH/s ETH, ~32MH/s LBRY, 180W
RX 480 (stock): ~24MH/s ETH, ~32MH/s LBRY, 150W
RX 480 (OC): ~28MH/s ETH, ~36MH/s LBRY, 180W
R9 390 (stock): ~26MH/s ETH, ~37MH/s LBRY, 275W
R9 390X (stock): ~28MH/s ETH, ~40MH/s LBRY, 275W
R9 Nano (stock): ~26MH/s ETH, ~37MH/s LBRY, 175W
R9 Fury (stock): ~28MH/s ETH, ~40MH/s LBRY, 275W
R9 Fury X (stock): ~30MH/s ETH, ~42MH/s LBRY, 275W

If you look at hash rates, power requirements, and perhaps most importantly, price, it should be pretty obvious why the RX 470 and RX 480 are your best bet. Overclocking will give the RX 470 performance relatively close to the RX 480 (overclocked as well), and power requirements and price are substantially lower. At $200 per card, it's possible to build a complete mining rig with four RX 470 cards for around $1350.

Now, a few notes about the above. First, most of the systems are running AMD's latest Crimson drivers, on Windows 10 -- not the ideal scenario apparently, as Crimson 15.12 is recommended. I don't know how much 15.12 drivers would improve things, though with RX 400 cards you need more recent drivers regardless. Anyway, running Windows 7 appears to improve performance by perhaps 10 percent, so keep that in mind.

Here's the thing you need to ask: How much money can you make per day with such a system? Or in other words, how quickly with the PC pay for itself? The cryptocurrency market is always a bit volatile and difficult to predict, but with services like Nicehash, you can at least eliminate some of the system management time and generally get a decent return. Best guess, you'll make a net income of around $7.50 per day after power costs with the above mining PC.

Bottom line, then: 180 days of mining to break even.

And if you're wondering, Nvidia mining with GTX 1060 cards is also possible, and the time to break even really isn't any worse than the AMD cards. You'll consume less power and likely have to mine different coins (I'd recommend just going with Nicehash and letting it manage switching), so you could drop down to an 850W Platinum power supply and still be safe.