Reluctant to chip in on this heated debate, but I have compared a UnitiServe FLAC rip to a D100 FLAC (dBpoweramp) rip and they are audibly different and have different sizes. A spectral analysis shows they are also visually different, so I would question not the AoB, but the laser and error correction on the two ripping systems.
The odds of two CD rippers producing the exact same count of bytes is incredibly improbable, but not impossible. As the MD5 is showing the same values I would question if the correct files had been sent in the first place.
Yes, there is snake oil out there, but don’t discredit someone hearing an audible difference time and time again as there may be valid reasons for this.