Ether’s recent sky-high transaction fees are not all bad news for Ether developers in fields like DeFi — if you happen to have an NVIDIA or AMD graphics card on hand, you may want to use those idle resources to participate in ETH mining and thus gain some additional passive income from the high gas fees. In this tutorial, we will objectively review 7 mainstream eth mining software such as GMiner, T-Rex, EthMiner, lolMiner, Phoenix Miner, NBMiner, Team Red Miner, etc. and give definite test results, which hopefully will help you choose the right ETH mining software.
1.ETH mining software evaluation environment
We chose two configurations of platforms for the experiment: one with 9 Nvidia 1660Ti graphics cards and the other with 4 AMD 5700 graphics cards. We did not change the default settings of the ETH mining software. Each ETH mining software participating in the review ran for 3 hours each, and the experiment started when the mining software was launched.
For this experiment, we created a separate ETH mining pool with a low shared difficulty.
The current shared difficulty in 2miners’ ETH pool is 8.72G. We set the ETH pool for testing to a difficulty of 64M, 136 times lower than the original one, so we could get a lot of shares without having to run the experiment for several months. This is the best value to maximize shares and ensure that the ETH mining software can handle the load and the processor does not slow down the test.
The 3 hours in the ETH pool used for this test is equal to 408.75 hours (17 days) in the actual ETH pool.
2. Hash rate calculation for ETH mining software and ETH pools
We believe that the number of shares in the ETH mining pool at the end of the experiment is the most important parameter for evaluating the ETH mining software. Moreover, the final table of the obtained evaluation results contains two types of hash rates.
True hash rate from the ETH mining pool test: this is the average hash rate obtained during the 3-hour experiment.The hash rate reported by the ETH mining software itself: this is what is seen when the ETH mining software is running.
At the end of the experiment, we get this value from the ETH mining log, which is the last hash rate value before closing the ETH mining software.
The hash rate calculation of ETH mining software is very simple, it tries millions of hash calculations per second, not considering whether these results are satisfied or not, and the number of hashes is increased by +1 for each new hash obtained.
Suppose you see 200 MH / s in the ETH mining software interface. This means that your ETH mining software can perform 200 million hash calculations per second using your GPU card.
On the other hand, the hash rate in the ETH mining pool needs to be calculated to arrive at it. The difficulty of each share sent to the ETH mining pool is not less than 64M (the share difficulty is set by the pool). Suppose the ETH mining software sends 1200 shares in 10 minutes, then the average number of shares it sends per second is 1200/10/60 = 2. We multiply 2 shares by the 64M share difficulty to get a hash rate of 128 MH / s.
3.Nvidia graphics card ETH mining software and hardware configuration
ETH mining equipment used for testing.
- Motherboard: MSI Z270 A-pro
- Processor: Celeron G3930
- Memory: 12 GB
- Solid State Drive: 120 GB
- 2 PSUs: DeepCool KCAS-1000M 1000W
- 9 GPU: Nvidia MSI Gaming X GTX 1660 Ti 6 GB
- Overclock: core-650, mem 2290, PL 70
- Driver: Nvidia 460.39
- OS: Linux
Parameters and reviewed ETH mining software.
- GMiner 2.43
- T-Rex 0.19.9]()
- Ethminer 0.19.0
- lolMiner 1.20
- Phoenix Miner 5.5c
- NBMiner 36.1
Parameters and reviews of ETH mining software.
- GMiner 2.43
- T-Rex 0.19.9]()
- Ethminer 0.19.0
- lolMiner 1.20
- Phoenix Miner 5.5c
- NBMiner 36.1
4.Nvidia graphics card best ETH mining software review results
Gminer ranked first, followed by T-Rex. It is worth mentioning that these two ETH mining software were also winners in our test three months ago, the only difference being that Gminer was in second place at that time.
How accurate is the hash rate reported by the ETH mining software itself? We think it’s pretty accurate. 285 MH / s is shown in the ETH mining software interface window, 282 MH / s is calculated in our ETH mining pool – the difference is caused by developer fee pumping.
Developer fees for ETH mining software are charged in this way: the ETH mining software allocates a small portion of its running time to the developer, and the mining SHARE for this time goes into the developer’s wallet, usually less than one minute per hour is allocated to the ETH mining software developer in return.
Ethminer, the only open source ETH mining software, comes in third. It does not charge developers, so the hash rate in the program and the pool are almost identical, with a difference of 0.01%. As you can see, even though ETH mining software is free (no developer draw), it does not mean that you can get more profit. If you choose Gminer or T-Rex, you can in fact get more share of mining.
LolMiner ranks fourth. LolMiner released its Nvidia mining software a few months ago, but in terms of efficiency, it has overtaken Phoenix, once the most popular ETH mining software.
Phoenix and NBMiner rank in the bottom two. Both report much higher hash rates in the ETH mining software interface window than the actual results. Regarding Phoenix, we have determined last time that we should not focus on its Eth speed, but rather on its effective Eth speed. This time the effective speed is 276.86 MH / s, which seems accurate and corresponds to the number of shares in the ETH mining pool. Phoenix shows the hash rate in 3 ways: speed, average speed (5 minutes), and effective speed. We think you should only refer to the effective velocity.
Example from Phoenix mining logs.
- Eth speed: 284.577 MH / s
- Eth average speed (5 minutes): 284.557 MH / s
- Eth effective speed: 276.86 MH / s; mining pool speed: 276.83 MH / s
The bad point is that the hash rate in the NBMiner miner window proved to be higher than the hash rate in the pool. nBMiner reported a higher hash rate than other ETH mining software. We checked the logs and did not see any rejected or outdated share records.
This means that Phoenix and NBMiner exaggerate their hash rates, which does not do you any good.
5.AMD graphics card ETH mining software and hardware configuration
ETH mining equipment used for testing.
- Motherboard: MSI Z270 A-pro
- Processor: Celeron G3930
- Memory: 4 GB
- Solid State Drive: 120 GB
- Power supply (PSU): Cougar CMX 1200W
- 4 GPU: AMD Power Color RX 5700
- Overclocking: Core 1300, Memory 930
- Drivers: AMD driver 20.40
- OS: Linux
ETH mining software participating in the review.
- Team Red Miner 0.8.0
- Phoenix Miner 5.5c
- lolMiner 1.20
- NBMiner 36.1
6, AMD graphics card best ETH mining software review results
TeamReadMiner is ranked first just like last time. This time, it widened the gap with its competitors, and its hash rate was 5% higher than the second place. The TRM hash rates in miners and mining pools are almost identical (TeamRedMiner may not take into account development costs when calculating hash rates).
Phoenix and lolMiner are in second and third place, with very little difference between them. phoenix hash rate is as high as ever, while the effective speed is very accurate: 214.58 MH / s. So, if using Phoenix Miner, be careful to focus only on its effective speed.
NBMiner ranks fourth, 10% behind TeamRedMiner.
We also tried to test GMiner with an AMD graphics card, but probably the memory was overclocked too much and it wouldn’t start. We have sent the bug report to GMiner developers and hope to see GMiner’s test results in a future AMD test.
7. Confidence level of ETH mining software review and future tests
This time, our experiments were 5 times more accurate than the last time (we tested 5 times the number of mining shares). TeamReadMiner shared their thoughts on comparing ETH mining software. They said it is important to consider mining as a Poisson process with all subsequent measurement errors.
In our 50,000share experiment, the confidence level was slightly less than 99%. If you look at the results, it is clear that 1 percent can make a big difference.
If we want to perform a truly accurate experiment, the ETH mining software should send 10 million shares each. this means that the testing of each ETH mining software needs to last a month! That’s too much! Because by the time we finish such an experiment, the developers will release an upgrade and the experiment will become useless. Having said that, we do hope to conduct new tests in the future using upgraded ETH mining software and higher accuracy.
It is worth mentioning that if the developers of ETH mining software are honest in reporting their hash rates and there are no errors in the code, you can trust the hash rates reported by the program when choosing ETH mining software.
8. ETH mining software review summary
Our review should help you, but please don’t take it as absolute truth, it is wrong to say that Gminer and TeamRedMiner are always the best and NBMiner is always the worst. Depending on your GPU, overclocking settings and gear configuration, you may get different results.