Bit Depth?

[Michael O'Regan]

Can someone explain this in laymans language?

How does 16bit vs 24bit affect sound?

-Mike

 

[Steenamaroo]

forgive me if i'm off BUT

as i understand it

the bit depth represents the number of degrees of amplitude that can be portrayed.

i know audio at 4 bit will represent 16 levels,2 two the power of 4

and audio at 16 bit will represent 65,536 levels, 2 to the power of 16

24 bit can represent sh1tloads,lol(thinks its like 20 million or something)


also,,as far as i know,,,,regardless of bit depth,,you will always cover the same range of amplitude,,,

the difference lies in how accurately it's portrayed.
(its kinda like 16 bit and 32 bit colour. you sill see em all,,,,there just arent as many degrees in the spectrum)


personally i use 24 bit at 48khz


the only other big difference i can think of is audio file size,,but its hardly an issue with the size and price of hdds today.

if a 16 bit audio clip is 10mb,,a 24 bit version will be around 15,ish.



hope that was accurate,and hope it helps

 

[hazerlazer]

16 bits was once the industry standard for recording audio, 24 bit has now become the standard for everything short of CDs and online music. 24 bit audio allows a lower noise floor and it is less likely to make artifacts from digital processing audible.

Or at least that's how I understand it. Basically, 24 bit audio gives you more levels of "loudness", so it has better resolution and dynamic range.

Same idea as 96k vs 44.1k, only along the vertical part of the waveform, rather than the horizontal. 96 has better resolution and a larger frequency range.

 

[Massive Master]

16-bit = 65,000 points of resolution. 24-bit = 16.7 million points. Ridiculous accuracy by comparison, a 48dB lower floor (the dynamic range of 24-bit is essentially greater than human hearing and almost any piece of gear).

It allows for much more accurate reproduction of a signal at lower ("more normal") levels, with no dither noise (which can build up quickly using 16-bit) at only 50% more drive space.

One of the few "no brainers" in audio.

The sample rate is another story -- I'd recommend 44.1kHz for audio projects (48kHz is a video rate). The damage from resampling is usually much worse than if you just record it at the target rate (which around 70-80% of full time industry professionals do) anyway. YMMV. Higher sample rates have a wider frequency response - But very (VERY, VERY) few recordings can actually benefit from it, few pieces of gear can reproduce it, fewer speakers can support it, no human can hear it.

 

[Michael O'Regan]

OK, FELLAS..........I ask for laymans terms and I get noise floors and amplitudes and resolutions!!!!!!!!

From what I can see, a lot of the Multitrack recorders - 8 track and 16 track Fostex, Tascam, Boss, etc. - are 16 bit.

How will this affect my sound? Should I look for a 24 bit machine or is it not that important at my (beginner) level?

-Mike

 

[Dogbreath]

You can make good recordings in 16 bit.

There's more room to achieve clarity and headroom in 24 bit.

Optimum (imho) is 24 bit at 44.1...

 

[dementedchord]

OK, FELLAS..........I ask for laymans terms and I get noise floors and amplitudes and resolutions!!!!!!!!-Mike

dont over think it...

consider a graph... if the verticle is the amplitude(think amp voltage maybe) and the horizontal is the freq response... obviously the number of slices gives you better detail... so 96k slices horizontaly is finer detail than 44k slices...
and 65k slices of voltage is not as fine detail wise than 16million...

as a practicle concern there's no big deal... either way for a newbie... but most of us go for 44k sample rate and 24 bit depth...

 

[.guitar]

OK, FELLAS..........I ask for laymans terms and I get noise floors and amplitudes and resolutions!!!!!!!!

From what I can see, a lot of the Multitrack recorders - 8 track and 16 track Fostex, Tascam, Boss, etc. - are 16 bit.

How will this affect my sound? Should I look for a 24 bit machine or is it not that important at my (beginner) level?

-Mike

You asked a question about a pretty complex topic and got a pretty simple answer...

To try and make it more simple: Just record in 24-bit.

Some non-recording sound cards don't actually record in 24-bit. But I would personally still use 24-bit in your software as it can be beneficial when working with effects. It will just be more 0's but it's better to perform the math using 24-bit precision than 16-bit.

Programs like Reaper and probably most others perform all math internally in either 32-bit float or 64 bit, and maybe even higher now. But a lot of FX plugins might use whatever precision the host (DAW) is telling it to use (project bit-depth).

But it's easier to just say: record in 24-bit.

 

[Michael O'Regan]

Ok. Thanks folks.

Next question.

Which Multitrack Recorders have 24 bit?


-Mike

 

[mixsit]

Their all going to include that in their specs. It's also often an option to select 16 vs 24 to increase record time on the compact or 'one-box recorders.

 

[thebigcheese]

Edited for stupidity (I was talking about the entirely wrong concept).

 

[Neeps]

But will the cd standard stay at 44.1k and is it not worth recording at 96k for when your remastered cd is re-released on DVD-Audio or SACD format?

 

[Massive Master]

Up-sampling (usually) isn't nearly as damaging as down-sampling.