recording at 24 bit 48k?

Knivez

New member
i know that recording at higher rates is the better way to go but i never fully understood why?

is it going to make a huge difference on the mix or the master?

its definitly a huge difference on memory usage

is it going to make your sound better? cleaner?

im just kinda lost on the whole recording bit rates and what not so if anyone can give me a good run down of it that would be awesome
 
24/48 or 24/44.1 are both fine. Stay at one setting for the whole project until mastering and go to 16/44.1 at that stage.
 
I'd say keep to 44.1 for music. If you can, and have the space to do so, record at 24, 32... or whatever you can put it to. Then bring it down to 16 after recording to put to CD. If you are doing audio for DVD/film, 48 is definitely better for you than 44.1.
 
Recording at 24bit I would say is the norm these days. Compared to 16bit, 24bit recording mean you have much much more bits to work with so you can keep your recording levels lower. I tend to record to around -6dB peak.

As for 48kHz sample rate, make sure you stay at 48kHz the whole project until the Mastering, then you want to have your master finished to CD redbook standards, 44.1kHz, 16bit.

Recording at 48kHz is known as oversampling, this spread the same amount of noise that you would get in 44.1kHz recordings over the wider 48kHz, so when you down sample back to 44.1kHz there is less noise.

G
 
The only good reasons to go 48k is if you're doing video (where the final file is 16bit/48k) or use hardware that is native 48k.

Otherwise just lock it down on 24bit/44.1k and worry about other more important things...
 
The only good reasons to go 48k is if you're doing video (where the final file is 16bit/48k) or use hardware that is native 48k.

Otherwise just lock it down on 24bit/44.1k and worry about other more important things...

Amen, brutha. Never understood why anyone would record at a video rate and then do the sloppiest possible resample for the lowest possible gain in fidelity (which is gone and then some by the time it's downsampled)... I see the rappers doing this a lot for some reason -
 
Knivez... Ditto what the others said.

Here is a good description that I read on another forum:
"Yes, you now have a faster sampling rate, with more layers to sample from. You've about doubled the size of the file.

With 16 bit audio, there are 65,536 possible levels to sample from at any frozen moment of time. Think if it as a picture, the more information the clearer the picture. With every bit increase of resolution, the number of levels double. By the time we get to 24 bit, we actually have 16,777,216 levels of information.

Then we add the sample rate. This is the number of times your audio is sampled per second. With a sample rate of 44.1 kHz there are 44,100 slices sampled every second. With 96khz there are 96,000 slices of audio sampled evert second.

Your original file is 16bit, 44.1 so that is 44,100 slices, each having 65,536 levels. Increasing to 96kHz, would give you 96,000 slices a second with nearly 17 million levels for every slice.

Recording at 24/96 yields greatly increased audio resolution-over 250 times that at 16/44.1. That is why for VO I always record at 24 bit (24/44.1) for radio. For video I record at 48 khz -most editors only use that sample rate- and anything for HD or Blue-ray I use 96kHz. (Sticking to 48kHz is not a bad idea across the board for even general VO work)

If you are hearing a difference make the change." Posted by Mike Sommer


Maybe more than you wanted.
Dale
 
i know that recording at higher rates is the better way to go but i never fully understood why?

Recording at high sample rates and large bit depths is not better than CD quality for most situations. If your end product is a CD, then it's better to just record at 44.1 KHz and avoid a potential degradation from the sample rate conversion that's ultimately needed. As for bit depth, the only thing that affects is the noise floor. When was the last time you heard hiss on a CD that was from the CD itself? :D

The only time I'd bother recording at 24 bits is for a live event where there's a potential for unexpected loud volumes. With 24 bits you can record at -20 and still have a silent noise floor. Then again, even with "only" 16 bits, recording at a peak level of -20 is still quieter than the finest analog tape recorder.

--Ethan
 
With 16 bit audio, there are 65,536 possible levels to sample from at any frozen moment of time. Think if it as a picture, the more information the clearer the picture. With every bit increase of resolution, the number of levels double. By the time we get to 24 bit, we actually have 16,777,216 levels of information.

The reason for recording at 24 bit is that it provides a lower noise floor so you can record at a lower level and not ever have to worry about clipping. Recordings done at 16 bit have exactly as much resolution as those done at 24 bit but with a higher noise floor.
 
Plugins will have to work harder at higher sample rates since they will have to process more samples per second and this will use more CPU or DSP power depending on what kind of plugs you use.

Also converters *May* be more comfortable working at some sample rates than at others


whether any of this makes any audible difference to you is really going to depend on how well the plugin is coded, how well the converters are designed and how much expectation bias you have picked up surfing recording sites

Some people swear they can hear a big enough difference in the recorded and plugin processed output to warrant recording at a higher sample rate and have to downsample to 44.1k or MP3 for distribution. Others say it is bunk

I'd say, as with so many audio questions, so long as you have the disk space and CPU/DSP power to handle it, you should do whatever floats your boat

YMMV
 
Thinking that CD is gonna be your release format is pretty short-sighted. You should look beyond CD quality for everything you do, and always consider that everything you do might be part of video project at some point.

Tracking at 24/48 is the minimal for the year 2012. Apple has recently stated that they'd like to receive mixes 24/96 files as much as possible going forward. Encoding from that 24/96 master to AAC can yield a better sounded file now... and the high-res file is there for future high-res consumer release.

Here's some something to read on this.... be sure to read to the very end:
What 'Mastered For iTunes' Really Means : The Record : NPR
 
It all depends on what you're aiming for and what you can "afford" to record at,

Understanding what these numbers actually mean will help you get a head for it,

In the digital world the sample rate and bit rate are how exact and how quick the converter takes a "snapshot" of the incoming audio an audio waveform is first sampled by asking a digital question, is it above or below the halfway mark? if the wave is above 0 amplitude then it gets a first bit of 1 and if its below then it gets a bit of 0 then it asks again, and if its above 0.5 ± depending on which the first bit was it gets another 1 or 0, and as each question is asked it asks the next depending on wether or not the last question was 1 or 0 and this goes on till it gets more and more accurate, so basically a higher bitrate dictates a more accurate image (think of resolution) of where the wave existed in the analogue domain

Next the sample rate is how often this wave is tested, at 44.1 its is checked 44100 times a second, and as this goes up we get a more and more accurate idea of where the wave was at that point in time,

This brings us to the Nyquist point which is normally half the sample rate, the nyquist point is the point at which digital artifacts start getting introduced into the equation for instance if you're recording at 44.1 after 22050hz these artifacts get introduced. humans can hear up to around 20,000hz so 22,050 is above human hearing and thats why it was chosen for cd's but now a day higher sample rates are chosen to get these artifacts even further away from our monitors and more importantly our ears,

so if you are recording in 16bit 44.1hz you are using 16*44100 bits a second 705600 bits all in all every second, so as you increase these numbers you also increase your hdd usage, this is why WAV files are huge

but if you can afford to use higher rates you get a more accurate digital image of the audio (coloured only by your source mic outbound gear and interface) and this is becoming more and more common as hdd's become cheaper and converters can handle this, so people are expecting higher and higher sample rates before it hits cd just incase it may go on tv or dvd or blu ray or any other new technology but if you plan on recording a band that only ever wants to release songs on CD and have stated that openly then feel free to jump down to 16/44.1 and stay at that until the end because any artifacts that are there will be unnoticeable to the average john doe just rocking out to it on his car stereo.

The decision lies ultimately on you're own personal situation.
 
Tracking at 24/48 is the minimal for the year 2012.
I can't even imagine why.

I've said it a hundred times before and I'm sure I'll say it a hundred times again -- If you can't make one of the greatest recordings ever - EVER - at 44.1kHz, bumping the sample rate up to 48kHz isn't going to equate to a bowl of warm sinus fluid. It's not the fault of the sample rate.

Half the stuff I get in at 48kHz (or higher for that matter) has absolutely nothing north of 19 or 20kHz. The high-res (double-rate 88.2/96kHz) stuff frequently has some rather nasty schmutz up top.

And again - I'm not arguing "against" it necessarily -- But there are so many FAR more important things...
 
This brings us to the Nyquist point which is normally half the sample rate, the nyquist point is the point at which digital artifacts start getting introduced into the equation for instance if you're recording at 44.1 after 22050hz these artifacts get introduced. humans can hear up to around 20,000hz so 22,050 is above human hearing and thats why it was chosen for cd's but now a day higher sample rates are chosen to get these artifacts even further away from our monitors and more importantly our ears,

I believe that this is no longer an issue with over-sampling ADCs, which sample at a frequency way outside the audio range, and then digitally filter back down to the nominal sample rate. The analog input still requires an anti-aliasing filter, but the corner frequency is now well ouside the audio range (say, a few hundred kHz), but still below the oversampling Nyquist frequency. The PCM4222, for instance, over-samples at 128 x PCM sample rate for 44.1kHz and 48kHz - that is, 5,644,800Hz for PCM sample rates of 44.1kHz.

Paul
 
And again - I'm not arguing "against" it necessarily -- But there are so many FAR more important things...

I agree. That's why I think it's sensible for most in the year 2012 to just "lock" into 24/48.

Your work will be relatively "future" and "video" ready, and you will have kept down the real-time processing demands on your rig compared to 96k.

Set it and forget it... get on to more important things.
 
One thing I will say about stupid high sample rates is that they decrease latency of input monitoring that goes through the converters.
 
if you can afford to use higher rates you get a more accurate digital image of the audio

Accuracy is defined by frequency response, distortion, added noise, and time-based errors (jitter). Nothing else affects the quality of digital audio. It might seem like having more samples or more bits would get a more accurate capture, but it really doesn't.

I've said it a hundred times before and I'm sure I'll say it a hundred times again -- If you can't make one of the greatest recordings ever - EVER - at 44.1kHz, bumping the sample rate up to 48kHz isn't going to equate to a bowl of warm sinus fluid. It's not the fault of the sample rate.

Exactly. Not using a high sample rate is never the reason a newbie's mixes suck.

--Ethan
 
Last edited:
Setting my equipment up to use 96kHz for the first time was amazing, hearing the difference was just awsome.

Personally, I wouldn't use anything below 24bit and I'm trying to use 96kHz or 48kHz (I'm stuck with 48kHz on my old machine). I am having trouble with converting down to 16bit 44.1kHz because of dynamic range and compression though but I'm not really planning on using CDs just at the moment.


Daniel.
 
Setting my equipment up to use 96kHz for the first time was amazing, hearing the difference was just awsome.

Personally, I wouldn't use anything below 24bit and I'm trying to use 96kHz or 48kHz (I'm stuck with 48kHz on my old machine). I am having trouble with converting down to 16bit 44.1kHz because of dynamic range and compression though but I'm not really planning on using CDs just at the moment.


Daniel.
After reading your other thread, you are having a problem with getting your stuff a competitive volume. That has nothing to do with downsampleing or bit reduction.
 
As for 48kHz sample rate, make sure you stay at 48kHz the whole project until the Mastering, then you want to have your master finished to CD redbook standards, 44.1kHz, 16bit.
 
Back
Top