Gemini's Thoughts on Solo Mining Miner Efficiency

Miner efficiency is crucial for solo miners, and it's not just about how much you're willing to spend on electricity. While a higher electricity budget allows you to run your hardware for longer, a more efficient miner will generate more hashes for the same amount of power, directly increasing your chances of finding a block and earning a reward.

Think of it this way: solo mining is like a lottery. The more tickets you have, the higher your probability of winning. In the world of cryptocurrency mining, your "tickets" are the number of hashes your hardware can produce, known as your hashrate.

The Role of Miner Efficiency

Miner efficiency is typically measured in Joules per terahash (J/TH or as I like to call it W/TH, watts per terahash). A lower J/TH value indicates a more efficient miner because it consumes less energy to produce one terahash of computing power. Here's why this is critical for a solo miner:

Maximizing Your Hashrate: For a fixed electricity budget, a more efficient miner will allow you to achieve a higher hashrate. This is because you are getting more computational power for every watt of electricity consumed. A higher hashrate translates directly to a greater share of the total network hashrate, increasing your odds of solving the next block and receiving the entire block reward.

Sustainability of Your Operation: Even with a significant budget, inefficient miners can quickly become unprofitable, especially as the network difficulty increases. A more efficient miner extends the economic viability of your operation, allowing you to continue mining for longer periods before your costs outweigh the potential rewards.

Beyond Just Electricity Costs

While electricity is a major operational expense, it's not the only factor. The initial investment in mining hardware is also a significant consideration. More efficient miners often come with a higher price tag. Therefore, a solo miner needs to find a balance between the upfront cost of the hardware and its long-term operational efficiency.

In essence, a willingness to spend a lot on electricity doesn't negate the importance of miner efficiency. An inefficient miner will simply waste a larger portion of that electricity budget, producing fewer hashes and diminishing your chances of success. For a solo miner, where the probability of finding a block is already low, maximizing your hashrate through efficient hardware is a critical strategy for improving your odds.

------------------------------------------------------

Use my referral code to sign up for The Gemini Credit Card and get an extra $50 in crypto! Check out offer and benefits. Referral code = xbqyduqdb

Antminer S3 Overclock / Underclock Testing

I know there's not a lot of need for Antminer S3 overclock / underclock testing since this Bitcoin miner was first introduced by Bitmain in June of 2014 and was superseded by other faster ASIC miners not long after that. But there are some good reasons to use the Antminer S3 (if you can find one) for home mining. I know there are a few Antminer S3's still out there in 2022 and they needn't be in a closet or gathering spiderwebs in the basement. 

1. It makes a great lottery miner. If you don't think 30 or 40 dollars per month is a lot to pay for lottery tickets, you could run the S3 for that much electricity (more or less depending where you live) and have the chance of winning 6.3 Bitcoin. The lowest underclock would cost around $15/mo. With this setup you don't have to drive to the convenience store all the time to buy lottery tickets. That said, solo mining for Bitcoin offers a very slim chance of solving a block, just as does a Powerball drawing offer a very slim chance of winning. You can also pool mine if you want to stack Sats. Additionally, you can mine for other SHA-256 coins which will have lower network difficulty levels and thus be "easier" to solve blocks for (i.e., Bitcoin Cash, Digibyte, XEC, and others).
2. The miner gives the house heat in the winter. So it serves the dual purpose of mining for SHA-256 coins and warming the house, thus reducing the work of the main heat system by whatever small amount. In summer you can underclock the S3 (if you're keeping it in the house) so that the heat it adds to a room is negligible.
3. The Antminer S3 is more powerful than USB stick miners and less hassle. 
4. It is easy to make it quiet. If you are going to use a miner in the house it has to be reasonably quiet. This miner doesn't get as hot as an Antminer S7 or Antminer S9 or any of the other multiple terahash miners that exist these days. Since it doesn't get as hot, it doesn't need as vigorous a fan as the S3 originally came with. Replace the original fans with quiet PC case fans of the same size and with 4 pin connectors (~3000 RPM). Even at the highest hashrate these are more than enough for cooling the S3. Make sure your PSU is quiet as well.

The testing data below is going to be frustrating for people with Antminer S3's that have all the chips available for hashing. My S3 doesn't, so I had to do the testing with 24 working chips when there are usually 32. I really did this testing for myself. Still, you may be able to glean some useful information from this testing, such as wattage differences between settings and hashrate differences from low clock speed to high. 

Testing was done using the firmware version "Fri Jan 9 20:41:34 CST 2015". This version of firmware has the ability to adjust clock speed (Frequency) and voltage. However, my testing and the testing of other miners has concluded that the voltage setting in the firmware doesn't do anything, or much of anything at all. If you want to adjust voltages, getting a power supply (PSU) that has adjustable output voltages is your best bet. In my testing the only thing that made a difference was changing the clock speed.

Different clock speeds often caused 1 or 2 chips to display an X in the miner's interface which means they aren't hashing. This causes a lower hashrate and lower wattages. Sometimes a reboot at that clock speed would help those chips, sometimes not. 

Fan speeds were affected by clock speed, hashrate, and by the HVAC heat coming on in the room, but you can see in the chart that lower clock speeds result in lower fan speeds and thus quieter miners.

Hardware errors were never much more than 0.01%, but when they were under that I noted it as low hardware errors in the chart (LHWE). Perhaps those are more efficient or stable speeds to run at.

In the 0725 voltage part of the chart I tested with one cable into each hashboard instead of two. There was no difference between the two options. 

The chart did not convert well onto a blog page so it is presented here as an image which you can click on to enlarge. If you right-click on the image you can save it and thus enlarge it when viewing.

Check out: The Essential Guide to Bitcoin Mining: A Cryptocurrency Tutorial

Paperback, Kindle, Nook, Kobo, Google Play

Use my referral code to sign up for The Gemini Credit Card and get an extra $50 in crypto! Check out offer and benefits. Referral code = xbqyduqdb

On Bitcoin Mining 8

Here we go, just another chapter in a Bitcoin mining adventure.

So, I had a perfectly good Bitmain Antminer R4 earlier this year and when I saw how much I could make selling it compared to the heat, noise, electric bill and slower return on investment I decided to sell it. What I made was almost double what I bought it for (and for a lightly used miner) that's how hot the market for Bitcoin miners in May/June 2017 was and still is. 

The Antminer R4 was my favorite Bitmain product because even though it is a powerful miner it is pretty quiet compared to other miners. Why they stopped making it is beyond me but I have a feeling that they wanted to concentrate on making, noisy but modular, Antminer S7's and S9's which they could sell to mining farms by the thousands and hence make a better profit, sort of giving up on the home-oriented miners market.

At any rate, since there was a more immediate profit to be made for me by selling Bitcoin miners as opposed to mining Bitcoin with them I decided to buy a couple Antminer S9 Bitcoin miners from Bitmain and resell them. There was a wait to receive them in June but it was only about a month or so at this time. Buying from the Bitmain company in China was (and is) not a simple process, as most people will tell you who try to buy miners from them - one of the reasons that people are willing to pay more on eBay or Amazon. It is simpler there and less of a hassle. The communication is inefficient with Bitmain, which is understaffed in their sales department. In addition, they are always changing the accepted method of payment, sometimes it is with Bitcoin, sometimes with bank transfer, and now with Bitcoin Cash.

Anyway, eventually I received the two Antminer S9's and wanted to test them to make sure they were operable before I listed them and shipped them to a paying customer. Luckily I had a 1375 watt power supply, so I hooked one up and tried to find it on the network. It seems like this is always difficult when you first turn on a miner and try to find it on your network. When I did find it and assigned a mining pool to it, the machine started mining and I was treated to the trademark Antminer "jet engine" noise that I discovered with my Antminer S7. I don't know if the S9 is louder than the S7 or if they are the same but there is no way I would be keeping either one of these loud ASIC miners since I only have a garage and not a mining farm. Both S9's operated well, hashing away at around 14.5 TH/s. After my brief testing was complete I listed these on eBay and eventually sold them for a 70% profit. Not bad.

Since this fiscal adventure was a success I decided to do it again. However, the next time I wanted to buy some S9's to sell I would have had to wait two months to receive them, so I decided to look elsewhere, if possible, for Bitcoin miners that I would receive sooner. (Apparently other people were having this idea too.) The company I settled on was another Chinese company called Canaan who were making and selling a nice miner called the Avalon 741. Unfortunately, they were just as hard to communicate with as Bitmain, also being woefully understaffed to meet customer demand. But once the back and forth about price and shipping were taken care of and the bank transfer went through, they shipped and I received the machines within two weeks of my initial contact with them. This was probably the last good experience anyone would ever have with them as the company was beset with orders because people wanted and needed miners and Bitmain became even more backlogged. Now Canaan would be backlogged too. 

But for now this first buying experience from Canaan was pretty good. They accepted bank transfer, shipping cost from China was reasonable and I received the machines fairly quickly. This time I bought five machines and controllers and tested one or two of the Bitcoin miners. Within a couple of weeks they all sold and though there was a profit it was significantly less than on the Antminers. The reason the profit was less was that these miners didn't have the same reputation in the mining world as the Antminers do. Also the Avalon 741 hashes at a slower speed than the S9 and so for these reasons it was somewhat less sought after. One of the controllers and miners I had sold did not function properly and were sent back by the customer. After refunding him I determined that one of the hashing boards was inoperable. Canaan support was very helpful and sent me a new hash board which was pretty easy to replace. I was then able to sell the refurbished miner.

The Avalon 741 runs quieter than the Antminer S9. This is because it has only one variable speed fan and because it runs at half the speed of the S9 (7.3 TH/s vs. 14.5 TH/s). It is a good, sturdy and steady Bitcoin miner with a good user interface. It uses a controller (Raspberry Pi) which you have to buy, hook up and power separately and this is unappealing compared to Antminer's onboard controller. I did run the Avalon for about a month and made some pretty good Bitcoin with it, above and beyond electricity cost, so it was a positive experience overall.

As of December 6, 2017 it is almost impossible to get a Bitcoin miner unless you want to pay at least double the price. Bitmain has a 2 to 3 month waiting period and only accepts Bitcoin Cash. Canaan doesn't even bother to sell to individuals any more and their distributors either have a waiting period, no miners to sell or sell out as soon as they put them up for sale. Jhejiang Ebang, who makes the Ebit series of miners, which I will talk about in the next blog posting, is having the same problems. They don't respond to e-mails after you make an order and apparently don't want to deal with individual buyers.

With Bitcoin's precipitous rise in price I think I am back to feeling that it is better to mine for Bitcoin if you want to make money from Bitcoin rather than selling the machines -- that is if you can get your hands on the latest version (or any version) of a Bitcoin ASIC miner. Just buying Bitcoin and holding may indeed be the wisest strategy, even at this point when it is 12,783 dollars for a single Bitcoin.
 
The next blog post will cover my Ebit E9 plus Bitcoin miner experience.