Current Status of GPU Mining Profitability

If you’re in an area where your power costs are still pretty low or, better yet, you’ve made upgrades to incorporate renewable energy and have access to free power, you’re in luck. I did some experimenting this week, and GPU mining may still be feasible for you. For some hobby miners who can cover the cost of electricity out of pocket, it’s enough just to be bringing in coins and hodling them in anticipation of coin prices going up, hopefully in the near future. Those miners have a little more flexibility in their choices of algorithms and coins, based on what they’re looking to hodl. However, I know many miners prefer to make sure they’re bringing in enough to cover the cost of electricity and still come out in the green so that they’re left with less risk in the end. While I urge you to always do your own research, I’m going to offer a run-down of what I found when testing out my RTX 3070. I tested out the 7 algorithms supported by PROHASHING that were originally popular with GPU miners before the Ethereum merge. Some of the GPU algorithms now, also, have started moving toward ASIC mining, which generally has an impact on the profitability, but that is outside the scope of this post. I also tested KHeavyhash, both alone and dual mined with other algorithms. I’ll give you the rundown of which of these 8 algorithms  would be worth my effort if I were not living in New Hampshire where, paying $.25/kW/hr, I’m doing way better than many residents who didn’t switch their providers quickly enough, and still may still have to soon consider selling my first born to pay my electricity bills. 

KAWPOW:

I started out with KAWPOW, as I’ve seen it listed a few places as one of the more profitable GPU algorithms at the moment. I used Gminer, and my card was able to get 23 MH/s, using 255 W. Using the PROHASHING worker configurator to estimate daily profits, I entered $.03/kW/hr as the electricity cost. This is a pretty low number pretty much everywhere, but as I said before, I’m assuming you’re either in an area with low costs like this, have access to renewable energy either as your primary source or to supplement and lower your overall costs, or you’re hobby mining to hodl coins. At these numbers, I could expect an estimated overall profit of about $.02 a day mining KAWPOW. I’m definitely not going to be able to quit my job off of that, but any time you’re in the green, it’s worth running the worker(s). 

Autolykos2:

Autolykos is the other algo I’d seen floating around the internet at one point as still pulling its weight. Here, I had to use $.02/kW/hr as an estimated electricity cost to come out in the green. My 3070 was getting 83 MH/s on this algorithm, again using Gminer, and it was using 142W. 6 months ago, Autolykos may have been holding its own, but this summer it’s declined a bit.

At $.02/kW/hr, I can expect to make a whopping $.01 per day. 

Etchash:

Around the time of the ETH merge, I had seen several people writing about having high hopes for ETC mining, but it doesn’t look like it was quite able to hold up. While people who have access to renewable energy can still get away with mining this algorithm, currently it looks like with a card like mine, you could only expect to make a profit if you’re paying $.01/kW/hr or less. Using Gminer, my card was getting 46 MH/s at 176W, which  would have me earning about $.03 a day in estimated profits.

Ethash:

I believe many GPU miners started out mining ETHW on Ethash right after the merge because of the ease of continuing on with the same algorithm until they could get the lay of the land. However, it doesn’t appear right now, with a similar card to the one I have, to be profitable to do so unless you have access to free/very cheap power. I was getting 48 MH/s with my card at 173W, using Gminer, and I would have needed to be able to get power for $.01/kW/hr or less to to expect an estimated daily profit of about $.03/day at that price. Still doable if you have free power or are hobby mining and can pay the electricity cost out of pocket, but not profitable for anyone with higher costs and a need to pay utility bills with the mining earnings.

Octopus:

Mining CFX also falls in that low end of our GPU algorithms, profit-wise. My card was pulling 34 MH/s at 202W, using Gminer. At $.01/kW/hr, I could expect about $0.02  per day, which, again, is really only worth it if I can swing free/renewable electricity.

Equihash 144 5:

My results for Equihash were better than those of Octopus. On Equihash 144 5, my worker mined BTG. I averaged a hashrate of 111 S/s at 269 W. The worker configurator estimates a daily profit of about $.04 / day with these numbers if you can swing the low electricity cost of $.03/kW/hr, which makes this algorithm a little more worthwhile if you can get your hands on the cheap electricity.

Neoscrypt:

This was my second experience mining Neoscrypt. I honestly don’t know a ton about it, but Neoscrypt is another one that it looks like you would basically need to be hobby mining or have access to free power to get anything out of right now. I used ccminer with CUDA because Neoscrypt is not supported by Gminer. During that time, I was mining NYX. My RTX 3070 got about 1.338 MH/s, and it was drawing 266W. At those numbers, I would be taking a loss at anything more $.01/kW/hr, but with free power, could expect about $.05 a day. So still doable if you’re running your worker(s) on solar power or something similar but not really feasible if you have to pay for your electricity and want the overhead covered by your mining.

KHeavyHash:

Kaspa mining is the newest addition to the PROHASHING offerings. Yesterday it was performing similarly to Etchash in terms of profitability, but it has dropped slightly between yesterday and today. I was getting 688 MH/s at 210 W, and at $.02/kW/hr, I could expect just barely over breaking even, as of yesterday. Today, I would need electricity costs of $.02/kW/hr to break even and could expect to make a little less than $.05 a day at that cost if mining solely Kaspa.

KHeavyHash + Autolykos2:

One of the draws to Kaspa mining for GPU miners is the ability to dual or triple mine KHeavyHash with another algorithm. The first dual mining combination I tested out for this post was KAS + ERG. I was able to get 305.8 M/s for KAS and 31.8 MH/s for ERG on this card, drawing a total of 174 W. At $.01/kW/hr I could expect to make about $.06 / day with this combo, and at $.02/kW/hr, I could expect to make a little less than $.02/day.

KHeavyHash + Octopus:

Mining KAS and CFX together, I got a lower KHeavyHash hash rate, and the mining was drawing more watts, at a total of 229. However, the addition of the Octopus mining at a hashrate of 11 MH/s allowed me to still anticipate an estimated daily earnings of about $0.04 a day at an electricity cost of $.01/kW/hr. Again, I would basically need access to free power to make it worth dual mining KAS + CFX.

KHeavyHash + Etchash:

KAS + ETC mining performed a bit better. I was able to get a hashrate of 497 MH/s on Kheavyhash and 38 MH/s on Etchash at a total of 269 W. This means that I could anticipate being able to make about $.06 per day at a cost of $.02/kW/hr. 

KHeavyHash + Ethash:

Last, I tested KAS + ETHW mining. I was able to get a higher KHeavyHash hashrate doing this, at 529 MH/s, and I got 36 MH/s on Ethash. It was drawing a total of 269 W. This combo could be estimated to make me a little less than $.05 a day at $.02/kW/hr

Conclusion:

If you’re in an area with high electricity costs at the moment, you’ll have some decisions to make with regard to whether you can cover the cost of power out of pocket in order to mine with your GPU(s) and hodl the coin(s) of your choice. Again, I strongly encourage you to do your own research based on your equipment, expenses, market conditions, etc., and I’m not in any way offering financial advice. However, if you’re able to get electricity for $.03/kW/hr or less, which is not currently possible anywhere in the US unless you’ve managed to negotiate a stellar deal, or you’ve invested in renewable energy like solar panels and can get your electricity essentially free from here on in, then hopefully this post will have saved you some effort in at least giving you a good starting point. At current conditions, with my RTX 3070, Equihash 144-5 and KAWPOW seem to be the two algorithms PROHASHING supports that are most capable of giving me some profitability if I were  looking to mine one algorithm at a time in a low electricity cost area, or could probably eek out some profits mining a combo of KAS and ETC, especially if I were willing to play around with overclocking/power limiting to see if I could improve the results a bit. 

It seems that GPU mining is starting to go the way of the CPU. CPU mining started out profitable, but with the advent of GPU mining, FPGA mining, and ASIC mining, CPU mining has been nudged aside, and what used to be profitable has become little more than a hobby for most. In this ASIC dominated industry, it’s more important than ever to support companies designing new ASIC tech that allows smaller miners to remain profitable or regain some profitability, such as Lokotech coming out with their GPU style ASIC hashblade. The little guys are down but certainly not out! Happy mining, all!