Over at CloudPools, we were having a little discussion on our Discord with several members about the best miner.
Along with the DSTM vs EWBF general debate, there has also been the question as to whether EWBF has a true dev fee penalty if you turn the dev fee off by adding
--fee 0 to your configuration.
So, since I have two nearly identical NVIDIA GTX 1080 rigs laying running, I decided to point both of them at the CloudPools HUSH pool, tweak them to hash as closely together as possible, and then switch one to EWBF. Later, I then ran both using EWBF, one with the fee on, and one without.
Now, this wasn’t an extensive test, and I would love to see other numbers/tests/feedback (we have a few members running similar tests now to confirm), but it pretty much settles it in my book!
Testing DSTM vs EWBF
Setup / Baseline
First, I set both rigs up using DSTM pointing at our HUSH pool. I tweaked timings a little bit, and was able to get both hashing within .2% of each other (~3270Sol/s AVG over a few 10 minutes or so runtime).
Screenshot below was taken a few hours later, when the hashes were a little down from testing time.
Then, I took one of the miners and switched it to using EWBF with the fee still enabled.
For a little bit over an hour, I monitored them and pulled averages from the miners and from the CloudPools HUSH pool.
DSTM vs EWBF Results
Results indicated that the hash rates DSTM reported were closer to the pool’s reported hash rates for that machine, and that it out-hashed EWBF in general.
- DSTM AVG: 3276.4 Sol/s
- DSTM 30 minute miner average: 3276.7 Sol/s
- DSTM 30 minute pool average: 3170 Sol/s
- DSTM 60 minute miner average: 3274.5 Sol/s
- DSTM 60 minute pool average: 3240 Sol/s
- DSTM AVG: 3280 Sol/s
- EWBF 30 minute miner average: 3164.3 Sol/s
- EWBF 30 minute pool average: 3080 Sol/s
- EWBF 60 minute miner average: 3161 Sol/s
- EWBF 60 minute pool average: 3020 Sol/s
So, the clear winner in this, admittedly short test appears to be DSTM!
Testing EWBF with and without Dev Fee
Setup / Baseline
We already have a decent baseline for EWBF on rig 2, so what I did here was leave the existing EWBF miner alone, except I changed the miner name to get a clean average on the pool. Then I switched the other miner over to using EWBF, gave it a new name as well, and cranked them up.
EWBF Fee vs No Fee Results
The results here were basically as the EWBF dev describes they should be. With the fee on, you “donate” ~2% to the dev. If you turn off the fee, you lose some production. In this case, it was about -6% at the pool when turning the dev fee off (even though miner reported rates were actually up when using no fee).
EWBF Fee Rig
- Baseline: 3175.5 Sol/s
- 1 hr miner average: 3156 Sol/s
- 1 hr pool average: 3230 Sol/s
EWBF No Fee Rig
- Baseline: 3202 Sol/s
- 1 hr miner average: 3191 Sol/s
- 1 hr pool average: 3030 Sol/s
So, if you are going to use EWBF, which is slighty more stable depending on your card optimization, I would say leave the fee on, let the dev get his, and hash away.
However, if you want to squeeze a little more hash out (7.2% in this case), move over to DSTM, tweak your rig to run stable on it, and enjoy (even with his 2% dev fee).
What miner are you using? What coin are you hashing?
We want to know what you’re mining, where, and what you’re using to mine it! Let us know in the comments, on our Discord, or stop by the CloudPools Subreddit!