16bit vs 24bit 44khz vs 96khz

  • Thread starter Thread starter artCROSS
  • Start date Start date
A

artCROSS

New member
I know this topic was discussed in this forum before, but couldn't find it again by searching.

Does anyone know where I can find downloads of recordings, one recorded in 16bit 44khz and one 24bit 96khz? I want to be able to compare the two, as i have read about the differences and whatnot, but haven't really heard them. I want to eventually record an album that will hopefully sell commercially one day, but wondering if i could get away with it using an maudio quattro.
 
Art,

> Does anyone know where I can find downloads <

There's a comparison of different bit depths you can download and audition. It's 10th in the list on my Articles page:

www.ethanwiner.com/articles.html

I'll tell you in advance that you probably won't hear any degradation until you get to the 11 and 9 bit versions.

--Ethan
 
Thanks Ethan, that is exactly what i was looking for. To be honest, i couldn't really tell a huge difference between any of them...i guess for the untrained ear, it doesn't make too much of a difference. I might just stick with 16bit 44khz for a while under i can muster up enough cash to get a motu or the new edirol fa101. Thanks again.
 
I recently made the jump up from recording at 44/16 to 44/24. It makes a world of difference especially at the mixing stage. If you can swing 44/24 I would advise doing that because while you may not hear the difference in a 2 channel mix, you will definitely notice the difference in a multi-channel mixing environment.
 
This subject is constantly being discussed on forums all over the web......lmao.

I had a very reputable, experienced ME say to me recently that until very recent times, most clients projects were coming in for mastering on either 1/2" tape or at 16/44.1. Basically the message I got from a couple of engineers was not to worry about 16 or 24 bit, just get my tracks done to the best standard I can and if the songs are any good (BTW, they aren't mine), then that's more important than what bit rate they are recorded at.
 
Indeed. Anybody listening (buying) would much rather hear a less than stellar recording of an amazing tune than an astounding representation of a dogshit song. Concentrate hardest on the song and the performance. That's been a problem for me which I'm trying to correct. I've been spending so much time messing with hardware configurations and plugins and editing tools and other crap that I noticed some of my earliest recordings (on a 4-track casette) were actually a better listen than my newest stuff. The quality of recording is way better now, but the quality of song has suffered. I'm not saying that's the case for you, but it's something a lot of us should think about.
 
Indeed. Anybody listening (buying) would much rather hear a less than stellar recording of an amazing tune than an astounding representation of a dogshit song. Concentrate hardest on the song and the performance.

Isn't that a bit of a moot point? If you have the hardware to record at high resolution, press record. If you don't, then record how you can. The performance aspect is minimal.

Unless I am missing something, I "set it and forget it" on Adobe Audition. I record mine at 32bit and 96khz as the default. It's no harder to do at 44.1/16
 
randyfromde said:
Isn't that a bit of a moot point? If you have the hardware to record at high resolution, press record. If you don't, then record how you can. The performance aspect is minimal.

Unless I am missing something, I "set it and forget it" on Adobe Audition. I record mine at 32bit and 96khz as the default. It's no harder to do at 44.1/16
My point is that he DOESN'T as of now have the hardware to record at higher resolution. I'm saying that the difference between a solid performance of a good song vs. a poor performance of a bad song will be far more apparent to 99% of listeners than whether it was recorded at 16/44.1 or 24/96.
 
I see :) For some reason, I thought you were implying that it was somehow harder to record at high resolution...that it takes concentration of some sort!
 
Remember that [besides being more difficult] recording at 96K is going to produce much bigger sound files and take up a LOT more room on your hard drive. I record at 44.1K and 32 bit, then convert once after final mix to 16 bit for CD burning.
 
Remember, too that it is much easier to stay away from recording with compression going in if you use 24 bit... Might be worth for you to consider, especially if you are not that experienced. Alas, I can't as I always record with my vs880ex - a 16bitter...

Remember next, that a lot of EQ algorithms sound much better when being used in higher freqs... Simply too lazy to explain, why. May make a HUGE difference, too...

aXel
 
what is more important, sample rate or bit rate?
is it better to go say 24 bit 44.1 khz than go 16 bit 96khz?
 
[size=-1]-------------------------------------------------
DEFINITION - BIT RATE

The ratio of the number of bits that are transferred between devices in a specified amount of time, typically one second.

Bit rate is the same as data rate, data transfer rate and bit time.
-------------------------------------------------
[/size]

Note that "bit rate" has NOTHING to do with word size or bit depth of digital audio (ie - 16-bit or 24-bit)

Carry on..................
 
so...after you do your mix and convert to 16-bit (for cd) do you get any degradation in sound quality? Is that what dithering is for?
 
Blue Bear Sound said:
[size=-1]-------------------------------------------------
DEFINITION - BIT RATE

The ratio of the number of bits that are transferred between devices in a specified amount of time, typically one second.

Bit rate is the same as data rate, data transfer rate and bit time.
-------------------------------------------------
[/size]

Note that "bit rate" has NOTHING to do with word size or bit depth of digital audio (ie - 16-bit or 24-bit)

Carry on..................


Man, I was wondering if anyone would notice that...the 16,24,32 is bit DEPTH. It is how many "points" of reference for each sample taken. Also, 24 bit is not 1.5X the depth of 16, but 256X the depth. The 44.1, 48, 96 khz is teh rate samples are taken 44.1 thousand times a second, and so on.

Someone had mentioned file size, both depth and rate increases will increase your file size - due to the bit rate increasing.....

Thats my 2 cents.
 
dafunkewun said:
so...after you do your mix and convert to 16-bit (for cd) do you get any degradation in sound quality? Is that what dithering is for?

That's exactly what dithering is for. If you just truncate the 8 bits from a 24 bit file, the last bit is not rounded to a value that is more "musical". It's as if someone put a gate on the last bit with it chattering on/off. Dithering adds a subtle noise which "excites" this bit in order to make the result more random and musical.

More technically, without dithering you will get quantization errors. These errors can potentially make your audio sound "grainier", add alias frequencies that make things sound harsh, and in general create what I like to call digititis.

This debate on this thread of 24 vs 16 bit (and IMHO the 24 vs 16 bit experiments) hasn't really taken this into account. In order to be able to hear the real difference between 24 and 16 bit you will need to heavily process both 24 bit and 16 bit material in their respective wordlengths before reduction to 16 bit. Otherwise all you are doing is comparing dithering a 24 bit file to 16 bit vs recording at 16 bits.

Either the above or compare a 24 bit file on a 24 bit converter and a 16 bit file on the same converter set to 16 bit.
 
NL5 said:
Man, I was wondering if anyone would notice that...the 16,24,32 is bit DEPTH. It is how many "points" of reference for each sample taken. Also, 24 bit is not 1.5X the depth of 16, but 256X the depth. The 44.1, 48, 96 khz is teh rate samples are taken 44.1 thousand times a second, and so on.

Someone had mentioned file size, both depth and rate increases will increase your file size - due to the bit rate increasing.....

Thats my 2 cents.

Good point on the 256X that some folks may miss. However a 32 bit file is not 256X the depth of a 24 bit file since 32 bit uses a floating point notation to represent the bits, whereas 16 and 24 uses fixed notation.
 
Think of digital recording and playback as slicing a loaf of bread and then reassembling it. No matter what you do, you'll never have the perfectly continuous upper surface of the loaf after it's been sliced. Sample rate is how thick the slices are and bit depth is how close the slice is in heighth to the original loaf. Bit depth may be thought of as forcing the slice to conform to a predetermined level which may or may not be as close as you need it to the original. Higher bit depths allow more choices, so that a level that falls "in between" bit levels in a lower bit depth will end up closer to the original. The good news is that, at some point, your ear starts to hear the samples at a given bit depth as continuous. Obviously the higher bit depth will give smoother sounding results, all other things being equal, than a higher sampling rate.

Edited to conform to good usage at the behest of Blue Bear.
 
Last edited:
lpdeluxe said:
Think of digital recording and playback as slicing a loaf of bread and then reassembling it. No matter what you do, you'll never have the perfectly continuous upper surface of the loaf after it's been sliced. Sample rate is how thick the slices are and bit depth is how close the slice is in heighth to the original loaf. Bit depth may be thought of as forcing the slice to conform to a predetermined level which may or may not be as close as you need it to the original. Higher bit rates allow more choices, so that a level that falls "in between" bit levels in a lower bit rate will end up closer to the original. The good news is that, at some point, your ear starts to hear the samples at a given bit rate as continuous. Obviously the higher bit rate will give smoother sounding results, all other things being equal, than a higher sampling rate.

Nice analogy! Continuing on that thought there are things like smoothing filters on D/A converters that help reconstruct the orginal waveform to make it less "steplike". This is like putting butter on top of the sliced bread. Adding dither is like having someone adding the butter on a vibrating table so the butter goes on "fuzzy" and also helps to trick us into thinking that the top of the loaf is more like the original. Adding EQ is like using jelly on the bread to help "sweeten" it, compression is like when you sit on the bread and it gets flat, hypercompressing audio is like taking the bread and making toast.

Hey I'm getting hungry, time for lunch ...
 
Back
Top