Comment by KerrickStaley
7 years ago
Question: Is 192 kHz better when you want to slow down (or speed up) a track significantly while keeping pitches the same? Does it produce less noticeable artifacts?
When DJing, I often speed up or slow down a track I'm cueing in order to match the tempo of the playing song. So having 192 kHz tracks might be better (although usually you try not to change a song's tempo too far from the original anyway).
No, it's not. You answered your own question here:
> while keeping pitches the same
All 192khz does is preserve higher frequencies. If you're keeping pitches the same, there's no advantage to using an extremely high sampling rate for your source material. The advantage comes if you're going to lower pitches.
(Note that some algorithms need higher sampling rates to avoid aliasing. That shouldn't be the case anymore, but if you're hearing a substantial increase in quality just going up to 192 khz, most likely one of your algorithms is faulty.)
(Note 2: I say "substantial increase" because some people can detect up to 27khz.)
I personally read the article to be addressing 192 kHz as a consumer of the music, I have a feeling for those producing (or mixing, etc.) it's a bit different.
It's kinda like how there's advantages of recording at 8k, better cropping, supersampling, etc. But for the average consumer there's no perceivable difference between the pixel density of 8k footage and 1080p footage on their 7" screen anyway.
Yeah, the author is not arguing against using 24 bits when recording, just when distributing to end users.
If the producer is planning to slow down the audio (and wants the ultrasonic components to become audible), then recording at higher sample rates makes sense, and the author doesn't address this; probably this is pretty rare in practice. You'd also need ultrasonic-capable microphones.
The much more common operation is to filter or amplify the signal, and for that, more bits per sample is better to avoid amplifying your quantization error. The author covers this in the "When does 24 bit matter?" section.
> and the author doesn't address this
No, it's literally excluded from consideration in the article's title. This is about music downloads, not music production.
I record and mix metal bands and have stuck with 48 kHz for years like many of the engineers I know. 96 kHz sounded better to my ears last time I checked in the studio (it's been years, maybe I wouldn't notice now that I'm older) but it's not worth the heavier storage and processing impact when nobody is actually going to use my stuff that way. I certainly don't feel limited by working at 48 kHz, either, but the hit to my workflow would be significant. Additionally, a lot of converters start imposing track limits when you go beyond 48 kHz, so that's one more reason to stay put.
More important than sample rate is AD/DA quality. I'll trust a new high-end converter at 48 kHz than an old prosumer device at 192 kHz.
Plenty of the albums we love as listeners were recorded at 44.1 or 48. Plenty were recorded with absolutely horrendous equipment but played and mixed by professionals who created magic. MANY modern vinyl releases where people brag about superior sound quality are just the CD master in all its 16/44.1 glory remastered for vinyl. Little of it matters when the end result is special.
When slowing down the track, you're changing the effective sampling rate (e.g. 192 kHz turns into 96 kHz at half the speed). This article is about regular playback, so in your case it might make sense to have a higher rate.
Not likely. The DAC will almost invariably oversample anyway, and even though stretching may not happen in a high-sampling-frequency domain, the result eventually does.