Can we hear the bottom bits of 24?

[watchnerd]

Not watched that one, cheers

Monty is the OG

 

[JRS]

Very coherent and cogent arguments re the marketing vs the reality of what is being called high rez audio, and why even Redbook when done right is all one needs. Hell he argues that few none can consistently discern the difference between well done 320kB and 44.1. I have tried and it is much more difficult than the disdain heaped upon say Spotify implies. And that was many years ago. Would I rather have FLAC? Of course, it only makes sense to maximize SQ throughout, especially when there is no associated costs. If I had to pay a quarter per tune, I'd likely be far more selective.

 

[MRC01]

... he argues that few none can consistently discern the difference between well done 320kB and 44.1. I have tried and it is much more difficult than the disdain heaped upon say Spotify implies. ...

I respect Monty but whoever says that is incorrectly overstating the case. Some people can consistently discern 320 kbps MP3 from lossless 44.1 in a DBT, when using the right material. I have done it myself, and I don't have golden ears. I know others here (our host @amirm among others) who also can do this. It's not that unusual for well trained listeners. I didn't say it was easy, the differences are subtle but can be reliably detected. That said, I don't believe people who say 320k MP3 is garbage or that the differences are obvious.
In short, 320k MP3, or even MP3 in VBR at high quality, is not transparent but it is very good.

 

[JRS]

I respect Monty but whoever says that is incorrectly overstating the case. Some people can consistently discern 320 kbps MP3 from lossless 44.1 in a DBT, when using the right material. I have done it myself, and I don't have golden ears. I know others here (our host @amirm among others) who also can do this. It's not that unusual for well trained listeners. I didn't say it was easy, the differences are subtle but can be reliably detected. That said, I don't believe people who say 320k MP3 is garbage or that the differences are obvious.
In short, 320k MP3, or even MP3 in VBR at high quality, is not transparent but it is very good.

When using the "right" material--in other words the stuff that is complex enough to beat the algorithm as opposed to average music. I'm not criticizing you, it is these claims should also be made with the admission that it is only discernible in a tiny minority of cases when carefully scrutinized. I'm sure most of play music to relax and not have to worry about slivers of difference that have no real world impact. So maybe consistently needs to be reemphasized. Noone questions that this can be done, but I'd bet my last dollar that not an audiophile on the planet can distinguish between 24/196 and 16/44 under normal conditions of playback and made from the same master done at say 24/48. Not one unless the recording is extraordinary.

 

[danadam]

Some people can consistently discern 320 kbps MP3 from lossless 44.1 in a DBT, when using the right material.

What percentage of all material is "the right material"?

(@JRS beat me to it)

 

[earlevel]

I respect Monty but whoever says that is incorrectly overstating the case. Some people can consistently discern 320 kbps MP3 from lossless 44.1 in a DBT, when using the right material. I have done it myself, and I don't have golden ears. I know others here (our host @amirm among others) who also can do this. It's not that unusual for well trained listeners. I didn't say it was easy, the differences are subtle but can be reliably detected. That said, I don't believe people who say 320k MP3 is garbage or that the differences are obvious.
In short, 320k MP3, or even MP3 in VBR at high quality, is not transparent but it is very good.

I've never done comparisons, but agree that 320k mp3 is totally listenable, I'll use it when needed to send a work in progress to someone for a listen, to fit in email.

FWIW, I think Monty sets the bar on audio quality to "for 95% of people, 95% of music, 95% of the time" or something. In particular, he said "maybe...maybe" dither would more important if CD had ended up 14-bit as originally planned. That's putting it too leniently. A digital fade at 14 bits will "tear", so you only need to be on a quality system, in a quiet room, monitoring reasonably loudly, to get routinely annoyed by non-dithered 14-bit. Though I'd agree it drops off quickly as you go to 15 and 16 bits. I consider that a minor quibble, since he's not advocating a change to 14-bit So, I agree with what he says 100%, even if a fraction of a percent of it is "in spirit".

 

[kchap]

I've never done comparisons, but agree that 320k mp3 is totally listenable, I'll use it when needed to send a work in progress to someone for a listen, to fit in email.

FWIW, I think Monty sets the bar on audio quality to "for 95% of people, 95% of music, 95% of the time" or something. In particular, he said "maybe...maybe" dither would more important if CD had ended up 14-bit as originally planned. That's putting it too leniently. A digital fade at 14 bits will "tear", so you only need to be on a quality system, in a quiet room, monitoring reasonably loudly, to get routinely annoyed by non-dithered 14-bit. Though I'd agree it drops off quickly as you go to 15 and 16 bits. I consider that a minor quibble, since he's not advocating a change to 14-bit So, I agree with what he says 100%, even if a fraction of a percent of it is "in spirit".

It's interesting to speculate what would happen to the weighted noise and distortion if pre-emphasis/de-emhasis was mandatory and noise shaped dither was also applied.