Quick question about sample rates.

xgarmothx

Shredmaster Alpha
Hey there,

I've always used my home setup for jotting down ideas, etc. and recently bought some better gear - I am about to attempt some real recordings. I did a lot of research into bit rates/sample rates, and did a search on here that resulted in my doing about 2 hours of straight reading. So thanks for that!

I've gathered that the most common rates are 24 bit @ 44.1 kHz, as this provides the most quality without really taxing your setup. I've definitely gotten the impression that people stress using a 24 bit rate over a 16, and this provides more of a difference than the sample rate.

My question is, as someone who has encoded mp3s many times at many different sample rates, I can hear a distinct difference between 192 kHz and 128 kHz, and anything below 128 is barely listenable. How then is 44.1 kHz considered acceptable for recording? Is there a difference between what the term implies between in recording wavs vs. mp3s?
 
What you've done is mixed up bit depth and bit rate. the 16 and 24 are bit depth, the 192 and 128 are bit rate. Recording at 44.1Khz with 24 bits means you are recording 24 bits 44,100 times per second, or 1,058,400 bits per second.
 
My question is, as someone who has encoded mp3s many times at many different sample rates, I can hear a distinct difference between 192 kHz and 128 kHz, and anything below 128 is barely listenable. How then is 44.1 kHz considered acceptable for recording? Is there a difference between what the term implies between in recording wavs vs. mp3s?



first off it's bit depth when you're talking about the 16 or 24 numbers. The amount of data stored in each sample. Bit rate has to do with the amount of data that is stored per unit of time.
so 24 bits in one sample (when talking about uncompressed files) or 128 thousand bits for every second of audio (when talking about compressed)

Also, MP3s are encoded in kbit...or kilobits. This does NOT have to do with the sample rate. You confused 192kbp/s and 192kHz. MP3 are a compressed format so you will definitely hear a quality difference when compared to CD. In other words, techniques are used to make the file smaller when creating an MP3. And these techniques ruin some of the aural content.

128kbps and 192kbps are standard MP3 formats. Commonly used. 128kbps is sometimes mislabeled as "CD quality" when it's no where near CD quality.
For comparison purposes CD audio is stored at 1411 kbit/s. This is achieved with this formula:
44,100 samples * 16 bits * 2 (stereo channels) = 1,411,200 bits/second or roughly 1411 kbit/s

CD audio .... 1,411,200 b/s
MP3........... 128,000 b/s

Definitely something is missing when you reduce it to MP3

It's important to know the difference between all the suffixes and where to use them each properly
kHz -kiloHertz
MHz -MegaHertz
b -bit
kb -kilobit
Mb -Megabit
MB -MegaByte
 
Yeah, cd quality is considered 16bit/44.1kHz. If that is your intended medium, it may be a good place to start at. 24 bit can have advantages also, but 16 is still respectable for home recording.
 
That's still a different subject -

But in any case, the advantages to recording in 24-bit - Even if (almost especially if) your target is a 16-bit medium is pretty huge.
 
Ive done a test before recording at 24bit and 16bit and didnt notice any big difference. Now if your going to be listening to your music on nothing but audiophile quality music equipment then you may hear a difference. If your burning to a cd then its going to be rendered at 16bits anyway. Try doing a simple mix at both 24 and 16 and have someone else listen to the two. You will know which is which and that may sway your decision. Also remember that when you render a 24 bit recording down to 16 that dithering should be employed. I do all mine at 16bit and Im fine with it. I do agree with you on the mp3 thing (128 vs 192). I converted my LP's at 192 because I found it much better than 128. And the quality difference between 192 vs. higher wasnt much of a concern when playing a tune on an ipod or PC
 
Ive done a test before recording at 24bit and 16bit and didnt notice any big difference.
With only one or two tracks, you won't notice a big difference. When you are mixing 32 tracks together, having the noise floor another 40db down on each of the 32 tracks really helps. It also saves you from having to record really hot, which will give you more clarity and focus in your mixes.
 
With only one or two tracks, you won't notice a big difference. When you are mixing 32 tracks together, having the noise floor another 40db down on each of the 32 tracks really helps. It also saves you from having to record really hot, which will give you more clarity and focus in your mixes.


Nice to know. Thanks. I actually have at the most done maybe 16 tracks in a mix. When I did the test mentioned above I used 7 for drums and 2 for guitars and that was it. So that was 9 for the test. And the difference was negligable to me. I'll re-test with more tracks on my next mix and see how it does. I like doing these kind of experiments. :)
 
Back
Top