TranceAddict Forums (www.tranceaddict.com/forums)
- Production Studio
-- 16 bit vs. 24 bit AUDIBLE DISCUSSION (NO TECH SPECS)
Pages (5): [1] 2 3 4 5 »

16 bit vs. 24 bit AUDIBLE DISCUSSION (NO TECH SPECS)

As emphasized in the title, I want to talk about the audible differences here. Not the fact that mp3's will convert it all to 16 bit, and 99.9% of music is spread via mp3, so 24 bit is worthless.

To me, from an audible perspective, I hear quite a difference when working with 16 bit and 24 bit samples. The biggest difference is crispness and clarity. I've got a few loops that literally trip me out when I play them through my AKG because they're so "real sounding" - It's as if my headphones are not even on and the sound is producing itself in perfect harmony inside my brain. Hard to explain!

Even when filtered down to 16bit and mp3'd, I still hear differences. I even hear differences with 32bit, although they are very subtle.

I've actually got a hell of an example of a 16 bit vs. 32 bit track, broken down into mp3, that produces a very noticable difference. The 32bit version has a much richer and fuller bass and is can clearly be distuingished if you have a good monitoring source.

But, let's focus on 24bit here since that seems to be the way things are slowly but surely going. I personally LOVE 24bit sample packs. I think it's why the deadmau5 sample pack is the highest quality pack I have ever heard (and I despise deadmau5) - There is no arguing the audible differences in WAV form, but there is a big argument when arguing the audible differences when the mix is filtered down to mp3. Some will say its impossible for there to be improved quality, because mp3 is 16bit, but I swear that is not true. I can tell with my own ears. How about you and your experiences?

Take a blind test - if you can spot 24 bit being better than 16 bit then work with 24 bit samples.
Dont notice any difference myself really.

well, that's not quite fair because you're likely using different samples and what are you monitoring with?

Maybe I do need to set up a blind test and filter a 24bit loop down to 16bit and see if people can differentiate.

Re: 16 bit vs. 24 bit AUDIBLE DISCUSSION (NO TECH SPECS)

quote:

Originally posted by DjStephenWiley Even when filtered down to 16bit and mp3'd, I still hear differences. I even hear differences with 32bit, although they are very subtle.

Well, this makes sense. When you convert down to 16 bit, dithering is applied. Apperently its artifacts have a pleasing effect on you. A 32 bit sample converted to 16 bits, however, has no more than 16 bits of dynamic depth.

Which raises the question, is it possible to apply "converting artifacts" to 16 bit recordings, to make them sound more pleasing to you, without ever using a 32 bit source....



Note: something similar was going on with wav/mp3 comparisons. In blind tests, a fair amount of people preferred the mp3 sound over the wav. It is not a matter of objective quality, but of subjective enjoyment of certain processing techniques applied to a sound as part of the converting process. Can you apply those techniques on a wav without actually converting to mp3....

I want to say that fuller bass will not be a product of using a higher bitrate, and would really like to hear the samples that you claim show a big difference in the bass between bitrates. Higher bitrates are useful for two main reasons:

1. You are working with very dynamic (wide range of loud parts to soft parts, i.e. not dance music) material, and want the absolute largest range of volumes possible in order to avoid the noise floor becoming an issue.
2. You are working with lots of individual tracks, and want to avoid the noise floor becoming an issue.

All the addition of bits does in audio processing is allow you to represent a larger range of volumes without the noise floor becoming an issue...

[Edit: As Pjotr G says, it's possible the conversion artifacts are actually enjoyable to you as well.]

32bit FTW lol

quote:

Originally posted by MrJiveBoJingles I want to say that fuller bass will not be a product of using a higher bitrate, and would really like to hear the samples that you claim show a big difference in the bass between bitrates. Higher bitrates are useful for two main reasons: 1. You are working with very dynamic (wide range of loud parts to soft parts, i.e. not dance music) material, and want the absolute largest range of volumes possible in order to avoid the noise floor becoming an issue. 2. You are working with lots of individual tracks, and want to avoid the noise floor becoming an issue. All the addition of bits does in audio processing is allow you to represent a larger range of volumes without the noise floor becoming an issue... [Edit: As Pjotr G says, it's possible the conversion artifacts are actually enjoyable to you as well.]

Remind me tomorow and I'll get them up. I gotta crash for now

quote:

Originally posted by DjStephenWiley well, that's not quite fair because you're likely using different samples and what are you monitoring with? Maybe I do need to set up a blind test and filter a 24bit loop down to 16bit and see if people can differentiate.

No i think someone needs to upload a few loops, each in 16 bit and 24 bit, and then see if you can tell the difference

not again. yes it might sound a little better (headroom? dynamics?) but all the shit u get with it? just dont do it imo. stick with 44,1kHz 16bit beacuse its the standard.

In Bob Katz's book he says that when bouncing down to 44.1/16, working at a higher samplerate makes no audible difference to the final file, but working at a higher bitrate does.

16 bit is nice, but 24 bit will give you smoother reverb and delay decays, among other things. It will sound less "stepped" as it drops from one volume level to the next.

24 bit provides a greater advantage to material with a wide dynamic range - classical music would benefit much more than a hard-hitting trance song.

Re: 16 bit vs. 24 bit AUDIBLE DISCUSSION (NO TECH SPECS)

quote:

Originally posted by DjStephenWiley I've actually got a hell of an example of a 16 bit vs. 32 bit track, broken down into mp3, that produces a very noticable difference. The 32bit version has a much richer and fuller bass and is can clearly be distuingished if you have a good monitoring source.

What dithering algorithm was used?

Everyone knows the pros only use 64bit.

You can't tell the difference between a 24 and 16 bit recording if they're both in mp3 at the time you listened to them. You must be hearing something else like dithering as others have mentioned.

I use 16 because I'm all software, if I was doing more than 3-4 audio I'd use 24.

quote:

Originally posted by derail 16 bit is nice, but 24 bit will give you smoother reverb and delay decays, among other things. It will sound less "stepped" as it drops from one volume level to the next.

These smoother decays would pretty much only make a difference when solo'd (i.e. from reverb to total silence). In "our" music, sounds are usually stacked on top of each other pretty much all the time. Meaning you don't need the bit-depth to "either have sound going on or not" (sound falling above or below recordable thresholds).

16 bit depth gives us 65536 dynamic "steps" (2^16).
Now, imagine a speaker going from step 55254 to step 55253 in 1/44100th of a second. It does not move in a "stepped" way, the speaker has to physically move from one position to another, and in that short time, in practice it will go in a smooth motion. Compare to going from 55254 to 55253,5 (possible with 17 bits, double the steps, 2^17). A speaker is a physical thing, and theory is one thing, but actually producing the actual digital waveform as stored is another. The speaker cannot jump from point a to point b without going through all points between it in time.

[edit] there IS a difference of course, and 24 bits naturally allows for more dynamic precision. I'm just argueing its practical use in dance music.

quote:

Originally posted by kitphillips You can't tell the difference between a 24 and 16 bit recording if they're both in mp3 at the time you listened to them. You must be hearing something else like dithering as others have mentioned.

This makes absolutely no sense. You're not the only person who has said this either. How can you say it doesn't make an audible difference, and then in the very next sentence say that I must be *HEARING* .....(we'll stop there because you just said I am hearing a difference, regardless of what that difference may be)

Re: Re: 16 bit vs. 24 bit AUDIBLE DISCUSSION (NO TECH SPECS)

quote:

Originally posted by DigiNut What dithering algorithm was used?

I would have to ask the producer. His name is Mikey W. I think he floats around on these forums. The track was called Habituate. It's available online.

quote:

Originally posted by derail In Bob Katz's book he says that when bouncing down to 44.1/16, working at a higher samplerate makes no audible difference to the final file, but working at a higher bitrate does. 16 bit is nice, but 24 bit will give you smoother reverb and delay decays, among other things. It will sound less "stepped" as it drops from one volume level to the next. 24 bit provides a greater advantage to material with a wide dynamic range - classical music would benefit much more than a hard-hitting trance song.

This may explain why I am more prone to hearing "differences" - I work with a lot with classical elements and very little with mega compressed bass and kicks.

But beyond classical, I can still hear a difference in other things. That difference is clarity, and it's quite noticable with my AKG 701's which are known for having some of the best hi-end hifi monitoring available. Lately, 24 bit or 16 bit has been a deciding factor in buying sample packs.

quote:

Originally posted by derail 16 bit is nice, but 24 bit will give you smoother reverb and delay decays, among other things. It will sound less "stepped" as it drops from one volume level to the next.

dunno what the fuck you are talking about here really, care to explain?

quote:

Originally posted by sixofour.604 Everyone knows the pros only use 64bit.

everyone knows only idiots dont follow good quality, well established standards. if ur recording vocals or guitars to make some quiet acoustic music then go for it. if not forget about it. dance music sounds completely static anyway which is part of the charm imo. i hate music fucking with dynamics up and down (classical)

Super high bitrates (32, 64 bit) only really matter in the internals of a DAW imo. High bit rates allow you to do crazy stuff with the audio (maximizing, compression, etc.) and still achieve a high sound quality.

Even if you made your track half as loud and then maximized it, in 64-bit, then you would still have WAY more dynamic range than if you did the same thing in 16-bit.

When you are rendering off the final product then lower bitrates are fine.

Edit: screw just one half the volume. You would have to make a 64-bit signal INCREDIBLY quiet to get it even close to 16-bit dynamic range.

wouldnt that distort as hell when converting back to 16 bit?

quote:

Originally posted by palm wouldnt that distort as hell when converting back to 16 bit?

If you maximize the signal before converting to 16-bit then no. Maximizing the signal will give you the total dynamic range that 16-bit offers.

16-bit range is +- 32,767
64-bit range is +- 9,223,372,036,854,775,808

Luckily most DAW's these days are 64-bit internal

quote:

Originally posted by coroknight Even if you made your track half as loud and then maximized it, in 64-bit, then you would still have WAY more dynamic range than if you did the same thing in 16-bit.

Just to illustrate how working with bits works for those interested:

if you use half the dynamic range in a 17 bit file, you have the same dynamic range as a 16 bit file.

so if u produce on 24 bit and think your track sound good it will actualy be reduced by 50% 8 times? ud end up with 1/256 of the dynamic area u thought u had? lol. this is mumbo jumbo, pointless shit.

When you are mixing your tracks their are all sorts of things that change the dynamic range of each channel. Even setting the volume lower than max reduces the dynamic range. Also, effects change the dynamic range, some effects will cause the volume of a signal to drop and then compensate by boosting the signal before outputting it.

This also lowers the dynamic range because, lowering the signal rounds off certain values.

Luckily, DAWs generally operate in 64-bit so that when you lower the dynamic range you still have a MASSIVE amount of range left.

However, 16-bit is ok for the finished product since we won't be manipulating the signal after that point.

Also, by finished product I mean after mastering. It's still a good idea to have a really high quality render pre-master.

Powered by: vBulletin
Copyright © 2000-2021, Jelsoft Enterprises Ltd.