PSA: No, you will not be able to profitably mine an alternate altcoin with your 4GB GPUs once the 4GB Ethereum DAG limit hits.

Tl;dr – Seriously…you will not be able to profitably mine something else with 4GB cards next year.

Update: As of April 2021, Zombie Mode keeps 4GB cards hashing profitably. The creativity within the mining community continues to be amazing. The next big change for miners will come with EIP-1559 scheduled for incorporation in July, which will reduce the reward to miners from fees. Make sure to look into it before investing in hardware!

A common response I keep seeing to “what should I do with my 4GB cards once the DAG limit hits?” is “just mine something else” – and while this is technically possible, in practice it’s not going to work out for nearly anyone.

Right now there’s a few GPU mineable coins worth mining. Often it’s even more profitable to mine stuff other than ETH, so it feels like you have lots of good options once ETH is no longer mineable. But this all changes the moment the DAG limit hits. This is not an opinion, this is factually inevitable based on the math. Let’s work through it.

Daily Revenue for a 4GB GPU right now, assuming a highly optimized card with 30MH is $0.80c. With 10c/kwh power (well below the national average) on a 120W card, there’s roughly $0.30c of power costs per day, for a profitability of $0.50c/day. Far from the glory days of 2017, but not bad considering where things have been for the past few years.

It’s estimated that anywhere from 25-50% of ETH’s hashrate comes from 4GB ASICs and GPUs. The majority of that is ASICs, so let’s use a really, really conservative estimate for the hashrate contribution from 4GB GPUs – 10%. The current ETH nethash is 258TH. That conservatively puts the estimate of 4GB GPUs at 26TH.

Total reward for ETH in the last 24h is roughly $6.5m. That means those 4GB GPUs are conservatively pulling in $650K/day.

Total reward for ETC? $115K. Nethash? 4TH

Total reward for RVN? $100K. Nethash? 2TH, but given a well-optimized card gets 10MH, let’s say that’s roughly equivalent to 6TH Ethash.

Total reward and nethash for everything else 4GB GPUs can mine? Not even worth calculating.

You probably already see the problem here. Let’s pretend ASICs don’t exist for now.

Once 4GB GPUs are ejected from ETH, let’s say they split evenly between ETC and RVN. That 26TH coming off ETH adds 13TH to ETC and RVN each. New nethash for ETC is now 17TH, for a roughly 4X increase in difficulty. That $0.80/day per GPU will rapidly degrade to $0.20/day. On a GPU that costs $0.30c/day to run. The math doesn’t work, and it’s roughly the same for RVN. Unless the vast majority of people are willing to mine at a loss, a lot of these 4GB GPUs need to turn off permanently. And remember, we are using *conservative* estimates for the amount of nethash contributed by 4GB GPUs – and even then it’s already an untenable situation. The reality will probably be worse.

Now let’s add ASICs to the mix. They can’t mine KAWPOW, but they will be able to mine ETC due to recent developments. Let’s say 20% of ETH’s nethash is currently 4GB ASICs. If all E3s moved over – and they literally have nowhere else to go – that’s going to add another 52TH to a blockchain with a nethash of 4TH. Revenue per unit is going to drop to ~10% of where it currently stands, and needless to say, that’s going to mean the vast majority of ASICs get turned off (boohoo for them). But the part that we care about here is that it’s going to push pretty much every 4GB GPU that even wanted to mine ETC off to RVN instead. So realistically 100% of that 26TH moves over to RVN, and now daily revenue is ~ $0.10/c day. Profit will be a net negative. Again, this is using conservative figures. Reality will be worse.

Doing the math a different way, you’ve got a set of GPUs that used to make $650K/day trying to mine a blockchain that only outputs $100K/day – with most GPUs still producing a meager profit at $650K. It is simply impossible for the vast majority of these 4GB GPUs to mine at a profit when the total rewards available to them are a small fraction of what they used to be. With free electricity, the GPU is netting the full $0.10c/day, but unless the GPUs and your mining infrastructure are free, this is still essentially a net loss.

So unless there is an EPIC price increase in RVN (5-10x) between now and late December, you are not going to be mining with 4GB GPUs. Unless you like mining at a loss. I know anything is possible in crypto, but this is a massive stretch from where we are right now. From that point forward, both ETC and RVN will be dominated by miners with ultra-cheap or free electricity, and even then they’re barely scraping by. Even after most other 4GB GPUs turn off.

The fact that you can mine RVN profitably today is irrelevant. This is the new reality. You will not be able to mine RVN profitably next year. Short of that extraordinary rise in RVN prices, profitably mining with 4GB GPUs is all but dead at ETH block height 11,520,000.

Normally this is the point where I’d mildly shill for our company (Bitpro, the #1 seller of used GPUs at retail by a very large margin). We’re always purchasing GPU mining equipment of all types, and months ago we’d suggest that you sell your farm of 1000+ 4GB GPUs to us ASAP because no other miner in the world will want them next year. A few of you even took us up on that offer. But that ship has sailed for now. Once the reality of the situation hits people, the price for 4GB cards will fall through the floor in an unprecedented way, and I hesitate to even guess where the bottom limit is. Because these GPUs are still capable of producing $.50c/day between now and Xmas eve, there is basically no price we could offer that would make sense for you today but also make sense for us for a qty of cards that would take months for us to resell into the market through our various channels. For smaller miners, we can talk, but in most cases, you’ll simply make more mining with them for a few more weeks than we’d be willing to offer for them. That being said, we *will* be purchasing a whole lot of these late this year and early next year, when the numbers can start to make sense for both sides. So if you’ve got a huge lot of 4GB cards to unload, it won’t hurt to reach out to us early, because while we’ll be able to absorb quite a few of these 4GB GPUs, we won’t be able to buy them all, and we’ll probably have the deals for as many as we can handle signed by EOY.

I know there’s a lot of new people here nowadays, so let me finish by unequivocally stating DO NOT BUY 4GB GPUs FOR MINING AT THIS POINT. They are not a bargain – they are on the precipice of being completely obsolete and there are going to be a lot of people trying to sell them to noobs in the next few weeks while leaving that inconvenient truth out. We’ve already had a bunch of new miners reach out to us to purchase huge quantities of 4GB GPUs and we simply refuse to sell them to miners, because we know how it’s going to work out for them and we’ve got no interest in ripping people off like that.

The silver lining here is of course that anyone with 6GB+ GPUs is going to get a nice Xmas gift from the difficulty of ETH dropping by 25-50%, revenues multiplying 2-4x overnight, and profits multiplying even more. We’re hoping next year is as fun a year for mining as 2017 was…but 4GB GPUs are not going to be invited to the party.

Originally posted to the gpumining subreddit