TIP: recording at 24bit 96k then mix down to 44.1 16bit

  • Thread starter Thread starter Alchemist3k
  • Start date Start date
TJ;

You may not believe this (and I'm sure there will be someone here who will have an arguement that's quite the opposite ;) )
but I've read about tests that have been done that show some of these 1/8" jacks are actually of higher quality than some of the 1/4" ones...go figure eh? It probably depends a lot on the card manufacturer I'm sure. A lot of us only have SB Lives and the little mini jacks and they sound great to me. As to your patching scheme...maybe that's it but I'd rather doubt it. Heck I used a patchbay (RCA jacks on it) for years with no problems in sound loss with my old Teac real to real machine.

If I were you though, I'd certainly spring a few bucks for a correct stereo Y-cable. It sure can't hurt and for all you know there might be a problem in one of the cables or the splitter. It wouldn't cost much. (You may actually have a hard time getting a 1/8" to stereo Y-RCAs at a music store though. You'll likely have better luck just grabbing one at Best Buy where they have plenty of those. They're not so bad....)
 
ok, i will play with the idea a little. I was thinking of just using my md8 as a mixer and doing all my track recording in my computer. any input? i got an athlon 800 , 256megs pc133 ram, geforce,soundblaster live,and an asus k7v mobo. i can get any programs necessary, whats the best? i got cakewalk,, (which i didnt try yet), cool edit pro, and cool edit 96, and cool edit 2000. I like the 96 better than the 2000. and i really like the cool edit pro.
 
It sounds like you've already got some great recording software there. The only suggestion I can make is to make sure you record everything at 16/48. That's the default for SB Live and if you try to record at any other resolution or bit rate your tracks will fall out of sync with each other. After you've got a finished stereo 2-track mix that you'd like to put on CD just use Cool Edit to dither it down to 16/44.1 for you. It's as simple as that and will sound just fine.
 
This is starting to bother me

If 24 bit recording does not produce any better sound than 16 bit during the recording/mixing stages then why does every major studio on earth do it ????

Capitol, Paramount .... They all do it for a reason.

If you have a bigger word length you get a bigger signal to noise ratio. Do you want more signal or more noise ????

A cd has a signal to noise ratio of 150 dB. But unless you go into a a CD player other stuff generates noise. Effects, pre amps, compressors all add noise. You want more signal.

Why is this ?? read a book on encoding. In addition you want to have a large word size and sampling to make all of those silly plugins you homewreckers insist on using work well. Digital signal processing is a data transformation and as such works better with more data.

To look for a way around this is basically fighting physics. Sound plays a pretty tight ballgame with physics so, good luck.

Dithering produces better sound. The pro guys here already know that and would like to move on to discussing newer and more controversial topics. So, they get tired of the same holes in the same arguments.

I am no pro. I am at school studying up to be one though.

Higher bit rates, word sizes are better. Effects on those signals will sound more real. It does take better hardware to record at 24 bit. It does take a better computer. But two inch tape also costs more than 1/4 inch tape right ?

If it does not sound better on your machine.
1. You are doing it wrong.
2. Your gear is maxed or crummy.
3. Your dsp has a shitty clock.
4. You recorded it badly.

Listen to a DVD on a nice stereo and then to a CD. Can you not hear the difference of 24 bit rec ? If not then turn the shit down deaf boy.

THE END
 
Re: This is starting to bother me

If 24 bit recording does not produce any better sound than 16 bit during the recording/mixing stages then why does every major studio on earth do it????

Capitol, Paramount .... They all do it for a reason.


Because you can make yourself believe just about anything if you want to I would guess. Sci-fi author, Phillip K. Dick, had a wonderful story along these lines:

He said he had a friend that was a real classical music connoisseur who put a lot of stock in having only certain recordings of his favorite music. He would often rave about a particular recording he had of this piano concerto or that symphonic piece and would go on and on about how this was the defining recording of that particular tune. He often would bring albums by Dick's house and they would listen together.

One day Dick had the idea to pull a prank on his friend. He had brought over a recording of some piece of music that Dick also had a recording of. Dick's recording was much older, by a completely different orchestra, and it even sounded scratchy because he had played it so much. While his friend was in the restroom he switched recordings. When the man came back into the room the other recording was now playing. He remarked to Dick, "See how wonderful this is! I tell you, there's no better recording in existence of this particular tune. Just listen to how the violinist plays this part coming up next. There, see! Isn't it marvelous?"

Dick never told him he had switched recordings but he and his friends always had great fun joking about it behind his back.

For that matter, I once put up a recording of a song I made, in the Microphone forum here. I never told anyone the details of the recording and everyone just thought it was the most marvelous thing. Several wanted to know what board I used or what preamp etc. Including one who had regularly blasted people for using SB Live cards saying they were noisy and would turn recordings into mud. Well, it was done without a mixing board, just a cheap $100 Marshall mic through a little Art Tube preamp and into a Boss AD5 Acoustic Guitar preamp (yes, a preamp into a preamp) and then into an SB Live and recorded at 16/44.1. (I only needed one stereo track so there was no need to record at 16/48 to keep them in sync.) It was dead quiet and the high end sparkled. I can't imagine it sounding better.

If you have a bigger word length you get a bigger signal to noise ratio. Do you want more signal or more noise ????

I've got dead silence now and a pristine sound.

A cd has a signal to noise ratio of 150 dB.

There is no such thing. 100 db is as high as it gets. You're talking about a percentage of noise against the recorded sound.

But unless you go into a CD player other stuff generates noise. Effects, pre amps, compressors all add noise. You want more signal.

I don't have any noise. And if you're getting noise from your preamps etc, then "if" it were true that recording at a higher bit rate made a more intelligible sound, you'd only be making your noise more prominent.

Why is this ?? read a book on encoding.

Huh?

In addition you want to have a large word size and sampling to make all of those silly plugins you homewreckers insist on using work well. Digital signal processing is a data transformation and as such works better with more data.

Home wreckers? (That's two words by the way.) That makes no sense at all. My silly plug-ins work great. Signal processing "appears" to work better with more data. The fact remains that there's a point at which more data becomes redundant and a waste of time. I think we've reached that point with 16/44.1 and any further adding to the data stream seems fruitless and a waste of time and system resources.

Dithering produces better sound.

Prove it.

The pro guys here already know that and would like to move on to discussing newer and more controversial topics.

Name one who's told you this; and I don't know any pro guys who come here.

So, they get tired of the same holes in the same arguments.

You mean like I'm getting now….

I am no pro. I am at school studying up to be one though.

You certainly are not; you're a kid in dire need of a spanking and an attitude adjustment.

Higher bit rates, word sizes are better. Effects on those signals will sound more real.

Let's hear it. Those DVD's don't sound any better to me at all.

It does take better hardware to record at 24 bit. It does take a better computer. But two inch tape also costs more than 1/4 inch tape right ?

Yes but you don't see anyone rushing out to build a tape machine and tape that's a foot wide do you? It becomes redundant and a waste of resources. We're not dogs; our ears are only so good and no more.

If it does not sound better on your machine.
1. You are doing it wrong.
2. Your gear is maxed or crummy.
3. Your dsp has a shitty clock.
4. You recorded it badly.


If it "does" sound better on your machine maybe you're only fooling yourself. People are very good at that. People spent thousands of years building pyramids totally convinced that those laid to rest within would resurrect. Not one of them ever did; their bodies are lying in state at museums around the world….

Listen to a DVD on a nice stereo and then to a CD. Can you not hear the difference of 24 bit rec ? If not then turn the shit down deaf boy.

Oh to be 14 again….
 
it would still take forever to do anything.
should make no difference...

i think this depends mainly on your audio hardware (ie soundcard or DAW)

i can hear the differences between recording with 24/96k and 16/44k but 96kHz doesn't sound much better than 24/48k. I recorded with an AKG 414uls thru StudioMaster 20/8/2 console pre into my creamware Pulsar II card. The latency is fixed to 7ms in all cases. (Pulsar ASIO 2 24bit drivers to Logic Platinium). My guess is, that the digital headroom of the 24 bit recordings makes the difference if your A/D converter is fine (Pulsar uses low noise AnalogDevices converters) (so all soundblaster users can stick at 16bit resolution).
 
Well, when we're talking about it taking a long time to do things I mean that the files you create at 24 bit are a lot bigger than the ones at 16 so it takes a lot longer to apply effects and that sort of thing. A resolution of 96 compared to 48 or 44.1 would create a bigger file still. Of course if you've got the money and incentive to buy a really big hard drive and ram along with a fast clock speed etc, then it might be worth trying. Like I said though, I haven't really heard anything at 24 bit so I'll suspend judgment till I do.

The exception is that I've heard some of the recordings guys here have done and put up, but since they've been converted to MP3 format something may have been lost in the process, but they didn't sound any better to me.

Here's the thing though: I've yet to find the human that could destinguish between a 320k MP3 and a CD with three or four times more information. So why would I think I'm going to hear a difference by going to a wave file that's twice as big (or thereabouts) as a 16/44.1 wave? Heck, I don't think anybody could hear the dif between a really well recorded metal cassette with Dolby C or SR and a CD wave file.

Maybe when I hear the 24 bit stuff for myself I'll be greatly surprised but I'm doubting it very much.
 
teached ears can hear the difference - most people can not.
But what i ment primary is: recording with 24 bit is much easier than with 16 bit, because you' will lose information with each plugin. The analogue noise problem is equal to nowadays digital resolution problem. You cannot make good info out of bad info. (Sorry for my english but i hope you see what i meen)

greets
 
"recording with 24 bit is much easier than with 16 bit, because you' will lose information with each plugin."

I'm glad you mentioned that because I just got Cakewalk GT Pro and one of the things they mentioned in the literature was that if you record at 24 bit and you use plugins they would sound better. So that's been your experience too eh? Well, I guess that's something to consider. I don't use plugins much though so I'm not sure how much it would benefit a guy like me. My cheap little SB Live can't do 24 bit anyway. It's sounds good at what it does though so I doubt I'll be changing anytime soon. I mostly record solo fingerstyle guitar stuff and use an outboard reverb and compressor so I rarely need to use plugins and I certainly don't need midi. So there you are. I'm a guy with low tech needs in a world of high tech gadgets. I'm stuck in the wrong era.:rolleyes:
 
Windowman,

If you have not ever recorded at 24 bit then why are you knocking it? Especially when you new software tells you it will work better with plugins.

I have recorded on cassette tape to 2 inch otari and from 16 bit to 24 bit by 96khz.

I absolutely feel that 24 bit recordings sond better in general over 16 bit recordings. I think 192 khz sampling sounds amazing I think 96 khz sounds good and 48/44.1 sounds all right.

I have recorded operas, theater productions, rock bands and latin music as well as acoustic guitar. Still sounds better. I have never used a soundblaster live or whatever card you have so I don't know what difference having a creative labs card makes over a MOTU interface or a Akai hard disk recorder.

I know about encoding from my previous job designing telco networks. More data is better for a more accurate transformation.

I don't feel a need to prove anything to you.

I am a black belt in Akido. If you try spanking me I will break your wrist when I throw you on your head.

Stick with the science fiction. It suits you.
 
hoo fela - my jiu teacher (5th dan) tells anyone not to be aggressive, because otherwise you will loose your energy...

so keep it with janapese suavity: smile and relax
 
i guess someone got me wrong !?!
So there you are. I'm a guy with low tech needs in a world of high tech gadgets.
This is ok! For some kind of music it works fine.
Do you know 'all the time' from Nicole Willis? It was done on a cheap analog 8 track, and has been voted under the 10 best RnB tracks of the whole 90's! check it out.
 
Well, I have posted the following before, but since this thread is sort of embracing redundancy, what the heck...

About a year ago I bought a Gadget Labs Wave 824 card. It records 16 or 24 bit, 44.1 or 48 sampling rate. In the course of my initial playing with it I did a few tests. I recorded tracks - both a live vocal from a mic and playback from a CD - at different bit rates and sampling rates. I then played them back both through monitor speakers and headphones.

My purely un-scientific results were that I could hear a difference between 24 and 16 bit, but no clear difference between 44.1 and 48. So I have been recording at 24/44.1 (using Cakewalk 9) and then converting to 16 bit using either Cakewalk or SoundForge.

Now, if my hardware supported 96 I would certainly try that as well, but its a moot point for now. But I generally agree with the concept of diminishing returns from more expensive hardware. However it should be no suprise that large studios use all the very best tools available to them, because their high-paying clients expect the very best results they can buy. We "home recorders" have to make more pragmatic purchasing decisions.

I should also mention that simply recording at 16 vs 24 bit is only a small part of the picture. A LOT of it is in how good your analog-to-digital converts are. My Gadget Labs card does sound better than my soundblater card(s) when recording at 16/44.1. Not to mention the impact of cables, mixer, any outboard effects, etc. Your recorded sound is only as good as the weakest link in the whole chain!

Now I should also say that I have a Soundblaster Live in my PC at my day job, and keep a cassette player hooked up to it. I have a work-related reason for this (I sometimes am required to convert wave files of customer phone calls to cassette) but it also makes it convienent for me to work on transfering my vast collection of cassettes to CD. Since many of the cassettes I have are live recordings, the sound quality is not great. Hence the lower A/D quality really doesn't matter that much - a lower quality source means lower quality output regardless.

And one more thing - I do record on said work PC w/ SB Live, using Soundforge, at 44.1 in stereo, and it seems to work fine. But after reading this discussion I will go back again and do some test recordings at 48 vs 44.1 and see if I can hear the latency problem which was described.
 
Back
Top