48.000? 41.8890? whatever the hell it is

P

PsychoBandito

New member
hey I was just wondering

lets say that I had a recording studio and all that and I was recording in 24/48hz or whatever and lets just say that the song was mixed but the best in the world so that its really good

bassically what I'm asking is this

would it make a difference if a good song is recorded in 24/96 or 24/48 or 24/41
 
i have no clue what you said in the first sentence.

probably not for the question in the second sentence.

danny
 
He's asking if he were the worlds greatest studio engineer and engineered an otherwise flawless work, would having a higher sampling rate effect the soundquality of the project. Yes it does. Higher resolution audio has greater detail.
 
ahahahahaahhahaahahahhahahaha

yeah sorry about that.... I don't know what I was smoking

I just was asking if I use a 24/96 sampling rate would it be better or basically the same
 
The same project done by the same worldclass engineer would sound better in 96kHz sampling. 95% of people probably couldn't hear the difference, but for the one's that can (and have the necessary monitoring systems) the 96kHz will sound better.
If you're asking if an amatuer working in 96kHz would sound better than a pro in 44.1kHz then the answer is no. A worldclass engineer would make better sounding records with a few SM57's and a portastudio than an amature with all the toys in the world.
 
Put it this way.... if you can't get your shit to sound good at 24/44.1, or even 16/44.1, then using 24/96 won't make the slightest bit of difference.
 
TFunkadelic said:
He's asking if he were the worlds greatest studio engineer and engineered an otherwise flawless work, would having a higher sampling rate effect the soundquality of the project. Yes it does. Higher resolution audio has greater detail.
Click to expand...

These kind of questions and answers, although mathematically correct, are by no means that simple to distill. Especially in light of the apparent inexperience of the original poster.

Nika Aldrich is one of the industries best at debunking and explaining the reasons why there appears to be no significant and tangible gains by recording above 48k. I've steadfastly stayed out of the equation vs physic's arguments but he has spent a great deal of time researching and ultimately writing a book on the subject. Of course all of his summations factor in the complexity and the limitations of the human ear...that being said his conclusions are not so boldly definitive.

Of course it goes without saying that some converters are more musical at 44.1 than others will ever be at 96 but that was not part of the original question. It is also wise to understand that an industry so glutted with competition tends to grab whatever ploy that may make them temporarily seem superior. A case in point might be 192k sampling rate.

I'd be more inclined to respond to the original post as to weather there would be any significant difference between 44.1, 48 or 96 as simply.......no.
 
yeah thanks guys

I just wanted to know if it would sound a lot better at 96 or higher cause I would rather not use that much cpu and ram on my comp.
 
PsychoBandito said:
I just was asking if I use a 24/96 sampling rate would it be better or basically the same
Click to expand...
24/96 is two figures. ONE of them refers to sampling rate (the second figure), the other refers to Bit depth. The consensus seems to be that recording at a bit depth of 24 is preferrable to 16 (it's certainly easier to work with in terms of headroom and noise floor).

Dan lavry wrote a paper on sampling rates and concluded that recording at higher sample rates is actually detrimental to sound quality:

"Sampling audio signals at 192KHz is about 3 times faster than the optimal rate.
It compromises the accuracy which ends up as audio distortions."

Go here and look at the white paper entitled "sampling theory" under the "support" heading.

Hence I record at 24/44.1 (I toyed with 48 for a while but really couldn't determine a difference).
 
Last edited:
PsychoBandito said:
would it make a difference if a good song is recorded in 24/96 or 24/48 or 24/41
Click to expand...
Why is it people wait until one brave soul brings up a relatively fresh topic in some thread before a hundred people come out of the woodwork and ask the exact same question over again in a new thread?

Pshycho, please read this thread right next door for a great discussion on this topic that is already going on as we speak.

G.
 
SouthSIDE Glen said:
Why is it people wait until one brave soul brings up a relatively fresh topic in some thread before a hundred people come out of the woodwork and ask the exact same question over again in a new thread?G.
Click to expand...


I can only speak for myself of course but it's mostly because I'm dim-witted.
 
Excellent link, KD. The most important point in that paper is that D-A converters do not connect the dots.

Here's another helpful article, Part I and Part II.
 
PsychoBandito said:
yeah thanks guys

I just wanted to know if it would sound a lot better at 96 or higher cause I would rather not use that much cpu and ram on my comp.
Click to expand...
Another consideration here is the math required of your system to down-sample the higher rates for use in 16 bit/44.1Khz systems (i.e your CD-Player). To down sample from Higher rates your computer needs to divide that sample rate by 44.1Khz here is the math required to do that.

From:
48Khz=1.0884353741496598639455782312925
96Khz=2.176870748299319727891156462585
192Khz=4.3537414965986394557823129251701

Bare in mind this figure has to be calculated for 44,100 times for every second of audio, and our CPU's obviously calculate this MUCH faster. This process alone is enough to risk causing quantization distortion from all but the best CPU's (Quantization distortion is one of the reasons we sample at higher rates to begin with). In summary, if you feel the need to sample at higher rates, sample in even multiples of your final products intended rate (such as 88.2Khz for material that will eventually be on a CD). Your computer's CPU will give you far more dependable results.
 
Blue Bear Sound said:
Put it this way.... if you can't get your shit to sound good at 24/44.1, or even 16/44.1, then using 24/96 won't make the slightest bit of difference.
Click to expand...

Ok, what if you have 16/44.1 for the source. Can you gain anything by loading them, say, into Sound Forge and saving them as 24/44.1?
 
Atterion said:
In summary, if you feel the need to sample at higher rates, sample in even multiples of your final products intended rate (such as 88.2Khz for material that will eventually be on a CD). Your computer's CPU will give you far more dependable results.
Click to expand...
Are you saying it is better to record with 44.1 than 48 if eventually you'll be burning a cd?
 
7string said:
Ok, what if you have 16/44.1 for the source. Can you gain anything by loading them, say, into Sound Forge and saving them as 24/44.1?
Click to expand...
No, nothing at all will be gained simply by going from from 16-bit to 24-bit. The data is still 16 bits, but the low-order byte will be padded with zeros. Any further processing you do to the files at that point, will benefit from the increased bit depth. But simply converting from 16 to 24-bit on its own has zero benefit.
 
Blue Bear Sound said:
Any further processing you do to the files at that point, will benefit from the increased bit depth.
Click to expand...

Now that I read my own question I see how obvious that was... I'm not sure if it was the 12th or 15th beer... ;)

Can you define 'further processing' for me and explain why it will benefit?
 
this article does a good job of explaining the computational impact on processing stages where the audio may under go several word length changes and ultimately the result will benefit from having a 24 bit length.

http://www.digido.com/portal/pmodule_id=11/pmdmode=fullscreen/pageadder_page_id=27/

basically if you take 16 bit audio and put it into 24 bit, it pads the lower 8 bits with zero. if you take 24 bit audio and place it into 32 bits (most DAW), again it pads it with zeros. Now if you perform a calculation on it (addition and/or multiply) - say EQ it, then the number of bits can easily become 48 bit or double depending on the programmer. So now, the application must begin to convert the 48 or 56 or 64 bit math results back into 32 bit for the DAW, and then from 32 bit back to 24 bit for the file. Because the lower order bits now are likely to contain data that has an impact on the audio if simply truncated (drop the bottom 8 bits which are no longer zero), then the algorithm has to find a way to gracefully drop those bits and still appear to have the lower bits influence the sound. rather than round off or truncate, most use a "dithering" computation to ensure a smooth sounding word length change...
 
You must log in or register to reply here.
Back
Top