Will GPU mining end after the Merge (formerly called ETH 2.0)?
Here we are in 2022, and the merge is still 3-6 months away. Just like it has been for the past 4 years. It’s practically a punchline at this point. But despite the countless delays – we really are approaching the finish line. The beacon chain has been running for years. Millions of ETH have been staked. Testnets have been successfully merged. I know we’ve all been hearing this for years. But our team has been following Ethereum development closely, and they are legitimately very close.
What’s the big deal anyway? Miners will just mine something else. It’s not the first time a coin has been lost to GPU miners, and everything turned out just fine before, right? There are lots of other choices, and the mining calculators show that profitability for ETC and RVN is still pretty good.
Sure, you will technically be able to mine something else. But it won’t matter. When Ethereum turns off the lights, there are going to be no good options left. This isn’t mere speculation – this is math. And the math isn’t pretty.
So, without further ado, let’s do the math.
Right now, in mid-2022, things are a little tight. But most GPU miners are in the green regardless of what they mine. And there are lots of good choices. A 6X RTX 3070 rig will pull about $8.60/day on ETH, and $6.25/day on ETC. If all anyone was facing was a 28% cut in profitability in profits post-merge, it would be rough, but it would be survivable. But that’s not what’s going to happen. For anyone. To understand why you must look at the big picture. All mining income is based on a simple formula:
Price per coin x Block Reward x Daily Blocks = Total Daily Income.
The block reward is the number of coins rewarded for finding a block. The block time is the amount of time between each block, and from this, we can determine the number of blocks produced per day.
Taking ETH as an example:
$1,975 per ETH x 2.01 block reward x 5760 daily blocks = $22,865,760.
That’s the total amount of income all ETH miners share, every single day.
Now let’s do the same for ETC. It’s the second biggest coin, and if it puts out a comparable total income as ETH, there will still be plenty to go around post-merge. Cue drumroll…
$24.14 x 3.1 x 6000 = $449,004.
Ugh. Not even close. We’re not even in the same galaxy as ETH. We’re two whole orders of magnitude off here. But… there are lots of other coins, so if the total aggregate income between them adds up to a nice healthy number, things will be fine! Let’s survey the landscape of GPU mineable altcoins:
Things are NOT fine. Every other altcoin combined doesn’t even add up to the income of ETC, which was itself a tiny sliver of the income of ETH. Yes, I did the math correctly. This is all publicly available data. Check it yourself. Even if I was off by 50%, it’s still game over. It’s like the Titanic is sinking, and there’s only one rickety old lifeboat to share. There is no universe where almost everyone doesn’t drown.
This is why I hate mining calculators with the burning passion of a thousand suns. If they’re your primary way to predict the future, things don’t seem so bad. But mining calculators are not showing you the relative hashpower and income of the various coins when they show you all these alternatives to ETH. They lull you into a false sense of security. Without ETH, GPU miners are screwed. The total amount of income available to ALL GPU miners is going to drop by 97%.
History is not a useful guide for what comes next, because nothing like this has ever happened before. In the past, when GPU miners lost LTC, ZEC or XMR to ASICs, they were just one small part of a very big pie. This situation is completely different. This is like the whole pie vanishing overnight, and miners will have to fight for the leftover crumbs. It is essential to understand this – this is an unprecedented and catastrophic reduction in income for GPU miners.
To put this into perspective, let’s walk through a few potential scenarios of how this actually plays out. We still must make some assumptions and approximations first. Rough calculations based on total hashpower suggest roughly 10-15 million GPUs are mining. Using the 10 million number, that puts the average income per GPU at $2.36/day – this suggests the average GPU is a RTX 3080, so this is being VERY conservative. Certainly, some amount of that hashpower is ETH ASICs, but given that they can mine ETC as well, and all the other coins add up to practically nothing – we can ignore that factor. Unfortunately, in this case, ASICs get to share the lifeboat.
A warning – the numbers are going to be so shocking you will probably think to yourself that they must be wrong in some way. They are not.
#1 – The Base Case
In this oversimplified base-case scenario, nothing changes between now and the merge. All crypto prices, total hashpower and block rewards stay the same. On merge day, all GPUs divert to other coins. 10 million GPUs are now left to split approximately $775,000. Average income per GPU? $0.0775.
You read that correctly. Income has dropped from $2.36 per day to 7 cents per day. And that “average” GPU is burning at least 25c in electricity per day at an average 10c/kwh. In other words, you need 3c/kwh electricity just to break even. And even if you have free electricity, your total takeaway per day is 7 measly cents.
But of course, it’s more complicated than that. Most miners know the merge is coming and many will turn off before then. And crypto prices fluctuate wildly every day. So, let’s play with the numbers and look at some other scenarios.
#2 – The Plausible Best-Case Scenario
Let’s assume that prior to the merge, half the GPUs shut off, and suddenly, the price of all crypto doubles. Pre-merge, the average GPU is making $9.44/day. Awesome! Enjoy that while it lasts, because post-merge, 5 million GPUs are splitting $1.549 million/day. The “average” GPU is now making 31 cents per day. After electricity, the average GPU is still effectively losing money. And that’s the plausible best-case scenario? WTF even has to happen to get us back to the status quo?
#3 – The Drunk on Hopium Best-Case Scenario
Things are already tight at current returns, so what needs to happen to maintain the $2.36 average? We’d need the price of crypto to 6X and at the same time, for 80% of GPUs shut off. 2 million GPUs are then splitting $4.65 million/day. Average per GPU is $2.32. Can you imagine a world where crypto goes 6X in the next few months and 80% of miners still shut off? I can’t. This isn’t happening.
#4 – The Plausible Worst-Case Scenario
Let’s take as a plausible worst-case scenario that pre-merge, the price of crypto drops another 25%, and despite that, only 25% of miners shut off. Average income per GPU? 7c, just like our base case. The takeaway here is that even though the base case was a contrived and oversimplified example, there are very realistic scenarios where that number still plays out.
#5 – The Minerdämmerung
Miner Twilight aka worst-case scenario
Just for fun, let’s imagine how bad things could really get. The price of crypto drops another 50%, and despite that, hashpower never declined. The average per GPU post-merge is 3.5c. That’s 50% less than the base case but if we’re being honest, it hardly even matters – half of effectively zero is still zero.
The takeaway here should be clear – realistically, there’s no good outcome here for miners on merge day. A miracle needs to happen just to keep things the way they were. Winter is coming.
So where do we go from here?
The long term picture for GPU mining
These scenarios are only what happens immediately after merge day. Mining has always been a self-correcting system. So what needs to happen to get us back to a point where GPU mining isn’t a complete waste of time – let’s say, $1.50/day revenue for the average GPU? We’d need crypto prices to stay stable and 95% of GPUs to shut off. If we’re heading into a crypto bear market and prices drop by 50%, we need 97.5% of GPUs to shut off. If you actually need to pay for electricity, realistically 99% of other GPUs need to shut off. No exaggeration. That’s really what needs to happen. The average miner won’t survive this. Even the most dedicated 1% of miners will just barely scrape by, losing money every day until we get to this point.
Miners won’t be able to just mine something else, near term
By now it should be very, very clear that the average miner won’t be able to “just mine something else.” And make no mistake – this must happen. This WILL happen. The math insists on it. The choice every miner will need to make is how long they can stand to lose money before they throw in the towel. Even if you turn your rigs off, GPUs are going to lose tremendous amounts of resale value every single day – so even if you’re not mining, you’re STILL losing money indirectly.
How long will it take for 95-99% of GPUs to peel off? Hard to say, but I think there will probably be a very steep decline upfront and then a slow bleed of miners giving up one by one when they’re fed up with losing money for weeks or months on end. This isn’t like proof of stake where you can just hold on to your coins and pray that the price recovers. Miners need to pay to play, and all of the altcoins combined simply can’t afford to pay.
This is going to have significant real-world consequences. I’ve been really taken aback by the prevalence of the sentiment that miners can just “mine something else.” That’s not going to happen. Businesses that were profitable the day before will be wrecked overnight. Lives will be ruined. Personally, I saw this coming a mile away and sold all my GPUs last year. But I’m still very sympathetic to the pain this will cause.
GPU mining here to stay? Still hopeful…long term
Years from now, it’s possible that GPU mining has a renaissance, and we do this all over again. It’s not out of the question that within a few years, the aggregate income for GPU miners increases by 6X from the post-merge baseline without ETH, after 80% of miners have already shut down. I’m not sure I’d bet on that happening anytime soon, but I can at least entertain the possibility. Or it’s possible that this catastrophe salts the earth and GPU mining is effectively dead from a profitability perspective and remains a niche hobby for the rest of our lives.
Where’s the GPU miner resistance?
Time will tell. Before I conclude, allow me to step onto my soapbox for one last moment here. A year ago, I was certain that miners would come together, in the end, to fight the merge using every dirty trick in the book. There are literally billions of dollars at stake here! Billion-dollar industries typically don’t just vanish overnight. I figured a coalition of large ETH miners, and maybe even Nvidia or AMD themselves would step up to the plate and defend their livelihood. And small miners all around the world would rally behind this effort, and it would make the bitcoin block size war look like a walk in the park. Because this wouldn’t just be a fight for the path forward, it would literally be a fight for survival. And here we are, on the eve of the apocalypse. And there’s just…crickets. All that most miners seem to care about is surviving another day mining ETH and then dreaming of a future where they’ll just “mine something else” at just a bit less income, completely ignorant of how dire the post-merge situation will be.
I think the reality of the situation is that despite what PoS advocates want everyone to believe, GPU miners are so legitimately decentralized that getting them to do anything together is like herding cats. Even the largest ETH miner, accounting for, let’s say conservatively 5% of the hashrate, is a small fish in a huge sea. GPU miners never stood a chance against the concentrated wealth of the highly coordinated whales that desperately want PoS to happen so they can pump their bags with virtually zero effort. I hope that magnificently blows up in their face because they created something truly beautiful here before they threw it under the bus. I think 5 years from now we’ll look back at a much lower-priced ETH and be able to clearly point to the merge as where things went off the rails in retrospect. The irony that the coin that created the most successfully decentralized PoW network is the same one that kicked it to the curb is almost too much to bear. It’s straight out of the twilight zone. They practically achieved the impossible, and it’s so, so disappointing to see that almost no one with the means, motive, and opportunity cared enough to fight for it.
Delay the merge years, avoid the worst-case scenarios
Realistically, the only path forward where GPU miners don’t get obliterated by the merge is if it doesn’t happen for another few years, and the price of ETH gradually falls to the point where miners can gracefully lay down their GPUs and walk away without too much carnage. I think that’s extremely unlikely at this point, but admittedly, the chances are non-zero. It’s certainly not too late for miners to come together and fight for their right to mine and prove to the world that Ethereum isn’t just another centralized shitcoin. Or at the very least, to come together and create something new worth fighting for before it’s too late. I just thought we’d be well down that path by now. In aggregate, GPU miners are going down in flames, there’s no two ways about it. But from an individual miner’s perspective, it’s not too late to cut your losses. If you’re daring, you can even work this to your advantage. Even if you don’t realize it, all GPU miners have been losing money since the start of 2022. Every single one of you. Here’s why.
At this point, we are 100% focused on assisting miners in their exit from GPU mining, temporarily or permanently. No matter the size or location of your farm, we can make a buyout offer and, in most cases, pay and pick up in person for an easy, no-risk transaction. Our offers will be realistic and market-based – for better or worse. Even if you’re still on the fence, we can assist with up-to-date market valuations so you can time your exit appropriately. I’m sincerely upset with the way things are turning out. I didn’t think it would end this way, and I still hold that tiny sliver of hope that it still doesn’t. This article isn’t some wild prediction designed to scare you into selling your GPUs to us so we can mine with them ourselves. We’re going to sell them back to gamers as fast as we can because no one can escape the brutality of the math. We’d prefer that GPU mining remains healthy, and people sell their hardware to us when they want to, not when they have to. But make no mistake – this is still an extremely challenging
period for our business as well. For those of you with larger farms, we’re still trying to figure out how to make this work at scale when the price of GPUs is falling so fast that not even the biggest GPU reseller in the world can move them into the hands of gamers before they’re worth less than we paid for them. As long as you’re realistic about what we’re all up against here, we’ll figure out a way to make it work.
However you choose to move forward, all of us at Bitpro wish every miner the best of luck navigating what lies ahead.