Many Ethereum Miners Likely Lost Money During the Past Two Years (2024)

But we were curious about the potential profits miners may or may not have realized over the past two years, so we decided to do some research based on the price of Ethereum versus the cost of hardware and power. In short: if you didn't sell your currency before the crash, you have probably spent more than you earned.

Below, we've selected ten of the most popular graphics cards for cryptocurrency miners. These also happen to match up nicely with the

, assuming you could actually find any of the GPUs in stock for a reasonable price. To keep things simple, we're only looking at Nvidia's original non-LHR models to give us the standpoint of people who might have picked up a card right on the day of the GPU launch.

The numbers we'll present assume 24/7 mining, with no pool fees. We've used our own calculations for power draw and hash rates, erring perhaps on the higher end of the profitability spectrum. We've also used a price per kWh of $0.10, which may or may not be similar to what you'd pay, but it represents at least an easy starting point. Finally, we've calculated things with no selling of mined coins — the "HODL" mindset, in other words.

These are strict numerical estimates, which are almost certainly better than what anyone doing real mining experienced over the past 20 months or so. We're also not including the cost of the rest of the PC, and prices on GPUs were all over the place so we've used a somewhat conservative estimate, meaning most people paid more than what we're showing in the table. You should probably add at least $750 for the cost of a CPU, motherboard, PSU, RAM, storage and case. We'll stop there and let the numbers speak for themselves.

Depending on whether you're a miner or someone looking for some schadenfreude, that table ought to make you rethink the value of cryptocurrency mining. The fastest mining GPU on the planet for most of the past two years, the

, would have netted a sum total of 2.7 Ethereum, probably less. Two months ago, on April 1, that would have been worth about $7,425 — not a bad haul for a graphics card that might have cost $2,000–$2,500. Even after paying the nearly $500 in electricity costs for all that mining, you still would have had a nice little chunk of profit.

But today, if you're still holding? Chances are you barely broke even, unless you got a good deal on the card. We know almost no one was paying MSRPs on the GPUs when they launched, and eBay prices were often double or more the MSRP. Factor in the cost of the rest of the PC (roughly $750–$1000 for a full mining rig) and the time to get things set up and running, and things look even worse.

We'll get to our second estimate of potential profits (losses) in a moment, for those that didn't start mining as soon as the various GPUs were available, representing a more likely scenario. But first, let's go through some of the other popular GPUs of the past year or two.

The RTX 3080 was supposed to cost around $700, but for much of 2021, it sold for $1,500–$2,000, perhaps more. If you got a good variant right around the time the GPU launched and mined 24/7 since then, you might have netted over $800, despite the crash in crypto prices. Maybe. The RTX 3070 and RTX 3060 Ti were also potentially profitable, depending on when you started mining and how much you paid for the cards, while the RTX 3060 that launched in February 2021 fell short even if you got the LHR v1 that was "cracked" by Nvidia's own driver team within weeks of launch.

What about AMD? All of the Navi 21 GPUs have similar hash rates and power requirements, while

ranged from as little as $1,000 (briefly) for the RX 6800 to as much as $2,000 for the RX 6900 XT, depending on when in the year you bought them. The

launched several weeks before the RX 6900 XT, netting an extra $100 from mining, but all of the AMD GPUs were likely unprofitable if you're looking at how things stand right now.

The RX 6600 XT is in an even worse position since it hasn't been available for even a full year yet, with current potential earnings after power sitting at just $146. Hopefully you didn't purchase a bunch of those cards for mining at $650 each, which would put you about $500 behind on each GPU!

But what if you weren't quite able to begin mining right when the GPUs launched? Here's a different look at the same data, only this time we're going with mining since June 1, 2021 — we've removed the RX 6600 XT from the list since it wasn't available yet. We used slightly lower prices than we were tracking last year on eBay as the GPU cost estimate. As you'd expect, less time mining means fewer coins mined, and the raw profits in Ethereum have been on a steady decline during most of the past two years.

The RTX 3090 wouldn't have mined even a single ETH in the past year. Subtract the power cost and we estimate only $756 in the value of mined coins, for a card that very likely cost $2,500 or more. You would still have the GPU, but the balance sheet would be sitting about $1,750 in the red — without factoring in the rest of the PC cost, or for institutional miners the warehouse and infrastructure cost.

Other GPUs that didn't cost quite as much fared a bit better, at least in terms of how much money you would have lost, but none of these GPUs would be anywhere close to "profitable" if you didn't start mining until June of last year.

This is likely why many mining groups are now looking to sell off all of their mining equipment. If they can recover a decent chunk of the cost, maybe they broke even. But we didn't include air conditioning, warehouse space, or infrastructure and personnel costs, either. So it's not hard to imagine some mining farms being very far in debt right now.

There's another alternative we should consider: the "sell as you go" mentality for mining. The idea here is to recover initial costs as quickly as possible, and only start holding cryptocoins once you're in the black.

The above assumes that you jumped on the mining bandwagon on March 1, 2021, when things were still looking pretty awesome from a mining perspective. The prices come from our

, for the month of March, except for cards that launched after March 1 (like the RX 6700 XT and RX 6600 XT), where we used the first month of pricing data. Then we assume selling off all the mined Ethereum on a daily basis — that's probably not realistic, but even if you did it once per week the results would be similar to what we're showing here.

If you did this to try and recover the cost of your hardware as quickly as possible — and again, we're only looking at the GPU cost; the rest of the PC could easily add another $1,000 if you were building six GPU mining rigs with dual high-end power supplies — a few of the GPUs might have come close to being profitable, but other factors mean pretty much everyone lost money. No one would have broken even, despite mining for 483 days. Unless you got a lower price on the GPUs, naturally, or if you held your coins and sold at the peak of Ethereum's pricing bubble.

How much money, or at least potential money, you could have gotten out of Ethereum mining over the past 18 months or more isn't a straightforward answer, just in case that wasn't already clear. There's a relationship between when you started mining, what GPU(s) you used for mining, how much you paid for the hardware, and when or if you decided to sell the coins for "normal" money. So far we've looked at three possibilities, but let's take the fourth: Absolute maximum theoretical profits from Ethereum mining.

This is, quite literally, the best-case scenario for how things stand right now. We assume you started mining the day each GPU was released, that you mined Ethereum 24/7 and had no extra fees, and that you sold whatever Ethereum you were currently holding at the peak price, relative to when you mined the coins. Oh, and you still have the GPU(s), which continue to be worth at least a decent chunk of change.

To give a specific example, Ethereum hit its highest price ever (at least so far) on November 8, 2021. That price was $4,811, at least for the daily average price according to Etherscan.io. Maximum profits would come if you held all your ETH until that day, which in the case of the RTX 3090 would have been 2.30 ETH mined. After that date, the next highest remaining price for Ethereum occurred on November 9, 2021, at $4,732. This time, a 3090 miner would have only had 0.00214 additional ETH available to sell, worth about $10. We continued holding or selling up until the end of the data we used, which was on June 26, 2021.

No one actually made this much money off of newly mined Ethereum (per GPU) since the various current generation GPUs launched, which is why it's a theoretical maximum. Realistically, quite a few people sold off some Ethereum at the previous peak of over $4,000 in mid-May, 2021. Actually, a lot more likely sold after the peak, when prices dropped by nearly 50% and stayed in the $1,800–$2,500 range for a couple of months. A more realistic scenario then is to have sold most of the mined Ethereum for $2,500 or so, but that would be yet another table of potential returns.

If you were savvy enough to mine and hold Ethereum from 2020 up until sometime between April 2021 and May 2022, you likely ended up with some decent returns. But the same people that held from 2020 all the way until late 2021 were just as likely to keep holding, which puts them back at the initial table. Some cashed out and made money, some cashed out at least enough to cover their costs, and some are holding for the next big wave.

This almost certainly won't be the last time we hear or write about cryptocurrency mining. However, it should be a sobering look at what you can expect long-term. If you get in right at the start of the next wave, pocketing the profits as you go, you could definitely make back your investment. Getting a good deal on the graphics card in the first place will go a long way toward putting your endeavor into the black as well.

The biggest factor of course is the gambling aspect of cryptocurrencies. Some people bought a few coins right before the bubble and managed to sell those coins for as much as 10X what they initially paid. And they did all of that without mining. You could also end up making money if you mined coins and sold at the right time. Either way, though, it's basically speculation and gambling.

For those that didn't get in early, or overpaid for hardware, or got nervous and sold at the wrong time, or any other number of possible routes, they lost money — some people and companies lost a lot of money. The real winners were mostly the graphics card manufacturers and scalpers. Both happily took in people's money, laughing all the way to the bank.

This past cryptocurrency bubble had some similarities to the previous bubbles we saw in 2017 and 2014, but it lasted a lot longer. Perhaps that was due to the pandemic and supply chain issues, and we certainly wouldn't try to bet on what might happen if or when we see the next cryptocurrency hype wave. However, some are still in it for the long haul, and if they can cover the costs for many years and Ethereum shoots back up to its previous high, and then continues on to set new records, they'll look incredibly prescient. On the other hand, those who can't hold out for perhaps five or more years of waiting will likely get caught holding the bag yet again.

Jarred Walton is a senior editor at Tom's Hardware focusing on everything GPU. He has been working as a tech journalist since 2004, writing for AnandTech, Maximum PC, and PC Gamer. From the first S3 Virge '3D decelerators' to today's GPUs, Jarred keeps up with all the latest graphics trends and is the one to ask about game performance.