Hacker News Reader

Comment by ubermonkey | Hacker News Reader

I really wonder if you can do this in a double blind, ABX test.

Note that parent is a DJ and while DJing the files are bot just reproduced but usually there is some digital processing involved. This makes it easier to hear the difference.

Maybe but I don't do critical listening on club systems unless I'm checking my own mixdowns. Unless the DJ is redlining (which they shouldn't be if they're competent) the audio should come out of the mixer relatively unmolested. My critical listening rig is a broadcast amp with two passive studio monitors.

Should also note I can't hear the different between well-encoded 320kbit and FLAC. But lower bitrates (256kbit and lower) on mp3 are audible.

I showed a friend once how bitrate and sampling frequency start to really matter when doing stuff like slowing down a lot. But I guess it's not typical in DJing.

I generally record everything in 44khz, but I will record at 192khz if I know I will be doing any time stretching. The high sampling frequency matters for that.

That type of processing is super common in DJing, but the understanding of how it interacts with lossy files is not.

I can spot and recognize the audio on ABX if I focus with basically a 100% rate on the infamous ABX test website with a HD600 headphone, so basically a crap one. Now if you ask me which one would be the best sounding or lossless (for the "highest" quality one, obviously for other it's relatively easy to spot in a clean environment), I believe I would have a harder time.

Reply on news.ycombinator.com