File Size and quality

[Mbroady]

Curious to know if you experienced engineers record at 96k, 48K 44k.........and why? This assumes you have a quality recording chain as well as an excellent listening environment

Also From your experience , is there any advantage in taking a 96k track and converting it to 44k for final mixing as compared to just recording at 44k .

Is the standard response to a question like this ....the human ear can't tell the difference?

Thanks

[rick-slo]

Usually record at 24 bit 88,200 hertz for audio tracks. Sometimes at 44,100 hertz and can't say I really hear a difference, but I have very good gear.

96,000 is overkill IMO and can be a negative artifact wise (e.g. see link below). When all done I end up converting things to wav files at 16 bit 44,100.

https://productionadvice.co.uk/high-...c-sound-worse/

[Brent Hahn]

For sound design, some kinds of sounds are cool slowed way down, so 96 and 192 are beneficial. For music I don't bother.

[KevWind]

I usually record at 24 bit 48k, but sometimes 24-88k (honestly I don't hear much difference BUT then @ 70, my hearing is definitely not what it used to be, particularly above 14 khz

[keith.rogers]

I have my defaults set to 48kHz/24-bit.

I don't slow down things or record bats (et al, which some folks actually do) with microphones capable of capturing frequencies well above human hearing. (I don't believe there's an audible benefit to recording at higher sample rates, but some folks do.)

48k (vs. 44.1) is because I read once upon a time that is more common when using the audio with video, which happens more often than it used to. I very rarely resample/bounce to 44.1/16-bit because I don't make CD images.

I do bounce to 16-bit (with dither) before using in a video or converting to a lossy format, but whether that's sensible or just to give myself some false sense of control, I don't know - just what I do these days.

[lkingston]

My 58 year old ear (only the right one works) can only hear up to 12k. I record at 44.1k or 48k 24 bits. 24 bits is important so that you can normalize without losing resolution. The final product is always 16 bits.

[RRuskin]

When I know the final product will be an audio CD, I use 44.1kHz/24 bit. I do all recording on a RADAR 24 and only use a computer for editing and playback for mixing through a console. In addition, all signal processing is done with hardware. Therefore, conversion in either direction is only done once.

[TBman]

Why is the final mix down often 16 bits? Just curious.

[KevWind]

Originally Posted by TBman

Why is the final mix down often 16 bits? Just curious.

That was standard practice when CD became the prominent playback medium, because 24 bit had not really been developed and 16 bit 44.kHz was the standard Compact Disk format
In todays world that would only be for cutting actual CD's . Other mediums can use 24 bit which has more headroom before clipping than 16.

[TBman]

Originally Posted by KevWind

That was standard practice when CD became the prominent playback medium, because 24 bit had not really been developed and 16 bit 44.kHz was the standard Compact Disk format
In todays world that would only be for cutting actual CD's . Other mediums can use 24 bit which has more headroom before clipping than 16.

Thanks!

[rick-slo]

Originally Posted by TBman

Why is the final mix down often 16 bits? Just curious.

It's about dynamic range. Sixteen bits gets you 96 dB. That's plenty in the final released product. More bits during recording and processing is a
digital over buffer and moves possible repeated software rounding errors further from what could otherwise become audible.

[keith.rogers]

Originally Posted by TBman

Why is the final mix down often 16 bits? Just curious.

I do it because I've already normalized/pseudo-mastered to a target LUFS and I will choose my own dither down to 16-bit. This way I feel that whatever lossy conversion gets applied down the line that is out of my control (SoundCloud, YouTube, et al), the resulting [D/A'd analog] stream stays more consistent. As mentioned, there's no need for the 24-bit dynamic range at that point. Nobody is selling HD audio of the stuff I do ...