24bits..ok, but 44.1 or 48? ..yes again..

Kryogh

New member
With all we can read in mags, and see almost everywhere, people are talking about samplerate now, what do or will you do?

______

I'll get back to 44.1 instead of 48. For the simplest reason, I'll record the same way it will end on the Cd. I've been doing some 48 for about ...1 1/2 year, did I saw the diff? not really. In fack, I TOUGHT I was hearing a little diff on VERY clean stuff, but, as soon has I started to read "new" info about samplerate, I had doubt about myself! So I paid attention to this, used both, compared...and, the only diff I could approve (didnt heard any!), is that the 48 take more HD space.

The simplest way to see the point is, if we end up on CD, use 44 or 88. If it will end on DVD, you can use 48 or 96. The reason is simple, the way it divide 48 to 44 is way more complex than 88 to 44. So some "approximation" are made. Obviously same thing about 48/96.

So, we can read many reasons, explanations, etc about this point, but this one is the simple I read, and it convince me to reanalyse what I'm doing.

..and ya know what?, too me, the last Korn CD sound like shit. They still proud of them cuz they recorded it in 96k and good stuuf here and better thant better stuff here etc. and I even didnt listen all of it. It's clean, but it sounds dead to me.

So, back to the point, what do you think about all the samplerate thing?

I know it's not the first thread about this, but, I you read mags, you certainely notice that this subject came back to the surface!
 
It's EXACTLY the same as when dealing with graphic images. If you start with low-res images, do some image processing (colors, brightness, enhancing, etc...) you find that the image quality degrades VERY quickly.

But if you start with a hi-res image and do ther same processing, image quality is retained and no loss becomes apparent.

You apply the same principle to digital audio recording. Unless you plan to do zero digital processing after you record (and keep in mind that even a volume change via a fader is a "digital process"), you will get the best results by recording at the highest rate and largest word-size possible.

The fact that the audio ends up on a relatively lo-res format compared to the hi-res recording is pretty much irrelevant, your source is hi-res and the last dither-down phase is simply a consequence of current audio formats (which may change in the future).
 
hmm..

Sounds like an other good point to me...DAMN!

but not entirely cuz, I'm also an infographist and, your not wrong, but, it's just a normal thing. You simply cant get a better image resolution, you have to deal with it. You cant get a better rez, but it's sometime possible to ameliore the image.

I also dont think that, pushing the fader up, affect the sound quality. If you can hear something like that, dude, you must be...dunno..must be...uh. damn english, cant find something appropriate :) hehe ..lets say god!

All this end on a CD that people will listen in their cars, gettoblasters, nightclubs, computers, etc. I dont think they will be able to say "man this tech at this studio have been way too far with that "X" thing! There is missing that 10011010 11010110 in the guitar!

I'still have to think and analyse all this cuz...honestly, I dont think I can hear any difference.

Maybe thats because I use some -medium- gear...

Thx for the awnser, makes me think about all this again..I dont think it will ever finish anyway :) I'll get back to my old Fostex R8, and everythin will be fine.. :)
 
Last edited:
Kryogh said:
but not entirely cuz, I'm also an infographist and, your not wrong, but, it's just a normal thing. You simply cant get a better image resolution, you have to deal with it. You cant get a better rez, but it's sometime possible to ameliore the image.
Using Behringer gear is "good enough" for some people too.... that doesn't make it "right"! If the sound quality matters, then you don't "deal with it", you do it the right way!


Kryogh said:
I also dont think that, pushing the fader up, affect the sound quality. If you can hear something like that, dude, you must be...dunno..must be...uh.
Well - you're wrong about that - in the digital realm, ANY change results in re-calculation, which means round-off error, which means degradation. Whether it's audible or not depends on many things, but it DOES degrade.


Kryogh said:
I'still have to think and analyse all this cuz...honestly, I dont think I can hear any difference.
That isn't really the point -- you don't record to the lowest common denominator of listening formats.... if this were the case we could all record to cassette decks and leave it at that!


If sound quality matters, again - use the best resolution you have at your disposal -- I guess the real question is how much does maintaining sound quality matter to you in the end?!
 
Exact.

And will I hear a difference between recording a band at 48k or 44k? nope, I dont think so. So why should I do this?

Just to write 48k instead of 44.1 in my publicity, in a weekly newspaper?...maybe it's the best reason to continue my work in 48!

And what about the thing I mentioned at the beginning of the thread that, simply convert a 48 project in 44.1 at the end, will mess it up more than if it were completely made in 44?
 
Not so much between 44.1/48 - but between 88.2 and 44.1 - there's a significant difference.

BUT........ I've found that the bigger difference is using 24-bit vs. 16-bit....

Personally, I track at 24-bit/48KHz, and I mixdown (in analog) onto 24-bit/88.2KHz........
 
ok

I agree about the 16/24bits.

And, if I could, I certainely would record in 88.2. But it's to much for the ADAT outputs on my 01v. So I dont. I could record in 88 but it will give me only 4 track instead of 8, so...I dont.

Well, I'm a bit suprised only you awnsered my thread! :)

your in Ottawa uh? ...just curious

The drummer who I play with was from Embrun. You know where it is?
 
Yeah, I know where Embrun is -- fairly close to Ottawa, just slightly southeast of it actually.... what's the drummers' name?

I just noticed you're from Montreal (from looking at your studio) -- nice looking, BTW...

I was born in Montreal....!
 
thx!

When Serge LaVictoire (drummer) was in Embrun - he's in montreal for about 3 years now - his band was call "Latreille". Was a rock-pop band...

Now we do music more like..Queen of the stone age, some people said Soundgarden (??) even Alice in chains...anyway.. we have fun thats the important point :)

In fact, I'm not in Montreal, I'm close to it, about 20 min. I'm in a small town called Vercheres.


Why did u moved to Ottawa? The last time I went to Ottawa was to see TOOL live in the Corel Center :)

Do you speak french?
 
I moved here while I was still in my late teens/early 20's... (followed my parents...)

I went to CEGEP in Montreal for a while - Electronics at Vanier College - St. Laurent Campus.

J'ai aucune chance de parler en francais depuis des annees! Tu voix que j'ecrit tellement mal!
 
Last edited:
The french sentence is not that bad...I guess my english sentences are worst!

Anyway, it's nice to "web meet" you hehe
 
If you play PC games you're probably aware of the term AA. This is a similar process at work. The image is drawn 4 times as big, then the extra information is used to come up with the best low res pucture that your monitor can handle. It can have some artifacts depending on the method, but generally looks far better than simply running at the lower resolution.
The analogy isn't really directly applicable, but its similar.

After going through all that,
"I say", use 44.1 and as high a word size as you can and don't mess around with samplerate conversion. 48 strikes me as too little of an improvement to throw to the mercy of a samplerate converter.

Listen to Blue bear tho, he actually knows what he's talking about.
 
The reason I use 24/48, is that I'm using an analog mixer, so I use the highest res I can on my recorder (HD24)....

I mixdown onto 24/88.2 though....

If I were using an all digital chain, my resolution decision may be different!
 
I am allways wondering about this shit.
I use protools le and record at 24/48. But when i bounce to disk, i got to go 16/44.1. So i am degrading (???) my mix when i convert to a wave file to burn to cd. For some reason this bothers me.(no wonder im allways not totally diggin my mix) but i was told to record as hi as the system allows. Plus there is less latency when recording at 48.
 
detuned6 said:
But when i bounce to disk, i got to go 16/44.1. So i am degrading (???) my mix when i convert to a wave file to burn to cd. For some reason this bothers me.
Well - there really is no choice since Redbook audio specifies 16/44.1.... that is not likely to be a determining factor in the quality of your mixes though.... the difference is not going to be night and day relative to the importance of proper mix balance/structure/sonic characteristics anyways. (ie, dithering to 16/44.1 is not going to make/break the sound of your mix!)
 
48kHz is only 7% more samples than 44.1kHz.....
And you KNOW you're going to have to downsample to 44.1 if you're making CDs.

That downsampling isn't going to be perfect. Why introduce even more errors for such a slight increase?

24bit is an easy choice.....
Would you rather crawl through a 24" or 16" high air duct with the Alien behind you? That's a big percentage increase in headroom.


I'm sticking with 24bit/44.1kHz.
If I hit the lotto, I'll go up to 24/88.2......
 
Ditto with Bruce and Tim. I do everything at 24 bit, 44.1. This is only because my hardware only supports 44.1 or 48, and my ears can't really tell the slight difference between 44.1 and 48. Whatever I may gain doesn't seem to justify the issue of then having to resample everything back to 44.1 anyways.

If my hardware supported 96, or 88, I would do it.
 
Back
Top