The pool shows me the same hashrate, so this guy is either lying about the hashrate or he is taking more dev fee than he says he does.
I want to give bminer a chance despite everything so I just did another 24 hour test and while bminer 5.2.0 claims it is 2.7% faster than dstm (1875 vs. 1825), the average hashrate at the pool came in slightly lower (1835 vs. 1850).
Since performance is nearly the same other metrics have to be considered such as the console output and stability. In this case, stability was the same for both, with less than 1% rejected shares and no crashes. I prefer some things about dstm's console output - reporting each GPU temperature, in particular - but I like bminer's running total of accepted vs. rejected shares. Both of these miners would benefit greatly from some Claymore-style color coding, however.
I won't go so far as to say bminer's author is lying or scammy, but I have said that I don't think he deserves to collect a 2% devfee while the program is still in alpha/beta stage. Yes, that is the same devfee as dstm, the best performing/most popular Equihash miner, but bminer is neither the best performing nor most popular.