Why I almost bit on bitcoin and nearly bought into the Butterfly Labs scam

Bitcoin and Butterfly Labs: What smelled like a rat apparently was. David Gewirtz tells the story of why he almost bought, but didn't and how the math worked in favor of the scammers.

I almost got hooked on the bitcoin gold rush, but my Spidey-sense eventually told me to walk away. Other than a few lost hours exploring the idea, and a few lost hours talking things over with my wife, I didn't lose a penny. Others were apparently not nearly so lucky.

Bitcoin is a form of digital currency (yes, you could call it fake currency, but it's traded for money, so there's that). Although it has no physical existence (like gold does), it does have some value. Like any other currency, that value is derived based on a set of standards that everyone who trades in it agrees upon.

In bitcoin's case, its algorithm is interesting. Each bitcoin comes to life once enough challenging CPU-intensive calculations have been made. Each bitcoin requires computational work to be expended in order to exist. This computational work is called "mining" because you have to grind and grind and grind processors to generate the numbers that become a bitcoin.

Of more interest, especially in our little scam story, is that as time goes on, each bitcoin is harder to make. What might take, say, five million CPU cycles to make the 900th bitcoin might take 50 million CPU cycles to make the 980th bitcoin.

In other words, every minute, the game gets harder to win.

When I first got interested in bitcoin, somewhere in the middle of 2013, it was almost impossible to mine bitcoin with conventional computers. Even so, that's what interested me. Like every techie, I have a garage filled with old computers, old cases, a bin of spare motherboards, and so on.

I figured it would be fun to rebuild these old machines into bitcoin miners. Filling a garage with money-making machines, what could be cooler?

But there was a gotcha to that plan. Bitcoin had gotten to the point where it would take far too long for a conventional PC to mine bitcoin. Just the power consumption vs. value ratio blew that away.

Bitcoin had outgrown the PC.

The new trend was building custom systems designed around the bitcoin architecture, with custom ASIC chips designed to calculate the bitcoin algorithm. They were purpose-built machines designed to do nothing but mine bitcoin.

Bitcoin mining operations were now measured in GHS (gigahash per second). The core bitcoin operation is the hash calculation, so 1 GHS was one billion hash calculations a second. Even at that rate, it could take days or weeks to mine a single bitcoin.

By the end of 2013, the bitcoin mining return-on-investment calculation had a number of variables: the performance of dedicated bitcoin mining computers in GHS, the degree of difficulty of bitcoin at that time, the cost of power and cooling for the bitcoin mining machines, available floorspace, and -- if you were going to scale it -- the cost of manpower to manage it all.

I had no intention of scaling bitcoin mining. My thought was it might be cool to fire up one machine, shove it in my air conditioned garage, and see if it made any money. If it did, I might add more.

Since using my old hardware was now out of the question, I started looking at where I could find bitcoin mining machines. Butterfly Labs, the now-notorious maker of custom-ASIC bitcoin machines, kept showing up on my radar. It appeared that if you were going to buy a bitcoin miner, that's where you had to go. There were other vendors, but the more I looked into doing it at a profitable level, the more the tea leaves (okay, the forums) pointed to Butterfly.

The seedier side of bitcoin

Want to make money mining bitcoins? Criminals have you beat

Bitcoins are becoming the "national currency" of criminals the world over and are becoming an increasingly poor investment for legitimate miners.

Read More

This is where my Spidey sense started to tingle. First, there was a waiting list for the machines. I wanted to do this project right then, not wait six months to do it. I'm like that. When I want to do a project, I want to do it right away.

But the waiting list wasn't like your normal waiting list. You had to pay upfront, and then, when Butterfly Labs got around to shipping your machine, they'd send it to you.

With machines ranging from $1,200 to $25,000, this was not something I'm a fan of. I firmly believe in the philosophy of don't charge until you ship. The idea that Butterfly Labs was willing to take money upfront, and then -- at some unspecified time in the future -- ship the computer, well, that idea did not appeal to me at all.

I did some additional research online and while some buyers indicated they liked the products from Butterfly Labs, others (quite a measurable number of others) were enraged that they hadn't gotten their machines after paying the full cost.

More to the point, there was some time-value-of-bitcoin math that was making me more uncomfortable with the prospect. Some of the machines Butterfly Labs was producing, if their gigahash per second claims were valid, would pay for themselves within a few months. But, if it became necessary to wait three or more months, the complexity of the bitcoin calculation would increase, and those machines would no longer generate nearly the value they would at the time of order.

In effect, by the time Butterfly Labs shipped the machines, they might not be fast enough to mine enough bitcoin to recover their cost, never mind generate any profit.

I fired up Excel and ran a bunch of spreadsheet simulations, using the GHS ratings provided by Butterfly Labs and factoring that against the constantly increasing difficulty of the bitcoin algorithm.

As it turned out, the numbers were good. Real good. But only if I could pay for and get the machine within about two weeks of placing the order. At that rate, it looked like I could double my money within about two months. After that, the value of the machine would plummet precipitously, almost to the point of having no value.

But Butterfly Labs wasn't delivering in two weeks, they were delivering in two to six months. I tried tracking them down and calling them, to see if there was any flexibility in their delivery schedule, but I never got a call or email returned.

By this point, my Spidey sense wasn't tingling, it was burning and yelling. It was yelling, "Run away. Run away as fast as your feets will carry you."

And so, I ran away. I did not get into the bitcoin game. Frankly, with all the hijacking of wallets and criminal activity surrounding bitcoin , I'm quite glad I stayed out.

But now, after the disclosure by the U.S. Federal Trade Commission that Butterfly Labs was pretty much running a scam operation, I'm very glad I have good Spidey sense.

Not only does the FTC have a bunch of complaints about Butterfly Labs, but their specific complaint that they used customer orders to mine bitcoins hits close to my spreadsheet calculations.

This is actually pretty big, because if the FTC allegations are true, Butterfly Labs effectively sucked all the value out of their hardware (on their customers' dime), before shipping the hardware to the customers.

Not only is Butterfly Labs alleged to have used the machines for months of mining, and thereby ship used machines, the real crime is Butterfly Labs used the machines for the bulk of their usable value-generating life to mine their own bitcoins.

Based on my spreadsheet calculations from that time, Butterfly Labs would have made a LOT of money off of those machines, because that two-weeks to two-months operating time was hugely profitable. If I saw double-your-money numbers paying Butterfly Labs retail price, the company likely would have quadrupled their money (if it had been their money to begin with).

But because they appear to have been using customer money, the company had apparently figured out how to use customer payments to fund machines that were hugely profitable for that specific short period of time where they could make money, at which they would then jettison the now-nearly-worthless machines to the customers that had paid for it all.

I'm glad I didn't get caught in this scam. This deal smelled like a rat, and I didn't bite.

Now it's time for the FTC to put everything Butterfly Labs has done through the magnifying glass and then burn the wings off those bastards.

By the way, I'm doing more updates on Twitter and Facebook than ever before. Be sure to follow me on Twitter at @DavidGewirtz and on Facebook at Facebook.com/DavidGewirtz.