> every miner would have to win every block to stay in business
To my understanding that'd be true if the energy cost was $2M per miner, but I think jpcfl was suggesting $2M total. i.e. that an individual miner would spend $100 on energy for a $100 expected return, which may be a 10% chance at $1000 or a 100% chance at $100.
Could adjust $2M up slightly for whatever portion was done by crypto-mining malware or dabblers with higher expected energy costs than return, and down for the portion done by professional miners who will be expecting a slightly positive return even after factoring in other costs.
If I spend $2M to mine $2M of gold, and the value of that gold increases by 2x over the next 10 years, then I think I could stay in business. That would actually be a pretty solid business.
The value of BTC has increased nearly ~200x in the last 10 years. I don't anticipate it will do this again, but I'm sure some of these miners think it will continue to grow at some rate (plus, they have already capitalized on the last 10yrs of growth).
That wasn't my point. My point was that they could stay in business.
My original point was that it was in the ballpark of $2M. Could be more, could be less, all depending on a number of variables--I believe one of those variable just doubled.
I'm not sure what the cost is now, but back in 2012-ish when I briefly looked into mining, it cost about $1 in energy to mine $1 worth of BTC. I used your logic and decided it didn't make sense to invest in mining, so I didn't. I wish I had, or at least purchased some BTC, but I was a broke college student just looking to capitalize on the hardware I already owned, and I didn't really know a thing about investing (other than investing in a 6 pack of beer to meet girls at parties--my ROI was not great, BTW, so I don't recommend this strategy).
I sympathize! I considered running a miner in 2012, with the positive side effect of heating my office. Decided I didn't want to listen to the fan noise for negligible returns.
> it cost about $1 in energy to mine $1 worth of BTC
This is the crux of it (we may be saying the same thing!). There is no 1:1 relationship between mining cost and reward on a single block. If there was, no one would do it, because they do not win every block. And all blocks you compete on cost the same, win or lose, but of course the blocks you don't win, pay zero.
There's a (very) roughly 1:1 relationship between a miner's overall cost and reward, averaged over many blocks. If reward increases (BTC price spikes), more competition comes online and your win frequency drops.
So if the question is "how much energy (cost) did it take to mine this specific $2MM block", the answer is closer to "The average block reward divided by the winning miner's win frequency", which is e.g. 5% for one of the bigger miners (I did not check this block or this miner). This was a high reward block, so the real miner cost might have been more like $50K. Less for energy alone.
But it's like the guy who buys a $10 lottery ticket every day. He needs to win a few hundred dollars per month to maintain the habit (gambling addiction notwithstanding!). Today he got lucky and won $700 on the $10 ticket.
> There is no 1:1 relationship between mining cost and reward on a single block
Ah, I see what you mean.
I think I've been mixing up my costs. When I wrote the original comment, I was thinking about the total energy cost for mining this block, i.e., the energy cost of all the nodes that worked on it. But the truth is, I have no idea about the economics of BTC mining. I was just being facetious :)