What sampling rate do you use?

What sampling rate do you record at?

  • 44.1

    Votes: 197 55.8%

  • 48

    Votes: 79 22.4%

  • 88.2

    Votes: 6 1.7%

  • 96

    Votes: 52 14.7%

  • 192

    Votes: 10 2.8%

  • 384

    Votes: 1 0.3%

  • I'm waiting for 768

    Votes: 8 2.3%
  • Total voters
    353
tkingen said:
When I went from 16 to 24bit there was definitely an improvement in the fidelity of raw tracks. I haven't dithered any finished mixes yet so cannot offer anything to that argument.
Click to expand...

Recording in 24bit from the start is better than 16bit, and the dithering down can be noticable if you have good converters in your software or hardware.

I use Samplitude 7 Pro, to make CD's and I have to say the CD's do somewhat still sound 24bit (sound like it is still being played in the computer).
 
I agree with everyone that 24 bits is the most audible parameter.
As for sample rate, I used to do 96 but this board convinced me that it was a waste of disk space, and started 44.1 then decided to go back to 88.2, again because of this board (well if this board told me to jump off a cliff, wooouuuld I ????? )
I think the higher sample rates might make a difference when plug-ins are appied, but this is theory - I can't hear any difference between 44 and 88.
I'm using 88.2 rather than 96 to make it easy for the resampling algos.
If it's ok with the board, maybe I can go back to 44.1 :)
 
I lied...

I voted 96 just because I'm excited that my new soundcard can, but currently I'm recording at 44.1/24bit.
 
HD storage is cheap

I've been using 96K for some time now. My theory is that it does make a difference, though I haven't gone through the hassle of actually recording and mixing an identical song in 44.1 and 96 to compare. My sound quality has greatly improved since the days when I used 48, but I've changed allot of things, not just sample rate.

Storage is becoming so cheap and abundant these days, that disk space argument against 96 is loosing validity. What is true in my experience is that 96K is much more taxing on the system in other ways: plugins use more CPU bandwidth and hard drive I/O can get bottlenecked.

The worst problem I've had is with hard drive I/O. At 96K, the number of tracks you can work with simultaneously is limited, and some systems won't be able to handle recording even a small number of tracks without skips. Since I starting using an AMD opteron platform, however, that whole problem went away. I also very much recommend western digital SATA hard drives, for their better performance and longevity (it's awful to have a drive die on you)
 
bdemenil said:
Since I starting using an AMD opteron platform, however, that whole problem went away. I also very much recommend western digital SATA hard drives, for their better performance and longevity (it's awful to have a drive die on you)
Click to expand...

I also use a Western Digital SATA HD and an AMD, mine's the new Athlon 64! I love it.
 
24/44.1 Recording at 24 bit allows you to get your audio signal as far above the digital noise floor as possible. Listen to a 8 bit recording, and you can appreciate why you would want to get above this ugliness as far as possible.

It still maintains that resolution, or being above that noise when I burn my Red Book CD in 16/44.1. I'm not saying it's still 24bit, but it was initially recorded that way so it still is a quieter recording that has kept that noise away from your precious music. Excellent recordings can be made at 16 bit, but I think it's trickier, because you need to be more vigilant with watching your levels making sure your as close to zero as possible without overs.
 
24 44.1

I'd move up to 88.2KHz, but it eats so much processing power. I'd love to have everything recorded at 88.2 for future use, but I need my plugins!
 
I'm at 24/96 thru a Motu 828mkII. At first I was going all 24/44.1 just because i thought "so it samples 96khz? Bfd." and wanted to save drive space (my first couple weeks recording with something newer than a SoundBlaster Live - cut me some slack), but i read a piece that explained a bit about sampling technologies, how it works, and why higher numbers are better.

When you're recording, you want to get as close to an analogue waveform as possible (to the digital equivalent of tape, or even vinyl) to capture the whole waveform in its entirety, which is something the higher sampling rates help achieve, even if it's only a step up to 88.2, or from 96 to 192, whatever. They're not perfect, and theres still very very subtle overtones, harmonics, frequencies, etc that may get lost, but i think even the sharpest human ear would have trouble picking it up at 192... From there, yes, it gets dithered down to 16/44.1 on a CD press, but someone posted an example about working with hi-rez graphics which really hits the nail on the head.

I sat down and did some comparisons with my trace elliot preamp pedal (love that thing), and my 12 string acoustic guitar, recording at 44.1 and at 96. I'll be damned if at 96k, with no other changes made, the sound was fuller, picking up the lower ends with more vibrancy.

Taking a photochopped image down from 2400dpi to 600dpi for a print job in my mind is always more preferable than working only in 600dpi. It translates the same to audio work.

And for a true sample comparison, don't post a sample in MP3s. Put something up that got Flac'd, or some other lossless standard. Artifacts *will* be present in mp3s, one way or the other.

All of this, is of course, only my humble findings on the long road to the middle, and is written with no implication of accurateness, or service contracts. Bumper to bumper warranty not included. Mileage will vary. Be kind, rewind.

Please drink responsibly.
 
Jumpin' In

OK- I just saw the title and hit post reply so I haven't read any of the previous posts but I bet it's a pretty heated debate right now.

Just wanted to state some facts for the record. Last night I was recording some mic tests and then running the tracks through my usual processing tools. I just couldn't get it to match up with the quality of my older recordings.

Then I looked at the sample rate and sure enough it was 44.1 instead of the usual 48kHz I've been used to. The difference in detail was obvious, particularly in the detail of reverb trails.

When I first got things rolling a few years ago, I arrived at the conclusion that the best resolution for the DAW I built was 48kHz/32-bit float.

Although my card could handle up to 96kHz, the computer could not. So I started experimenting. A lot. My keyboard and my amp simulator were both 48kHz. Matter-o-fact, when using the amp sim (SPDIF), it HAD to be set at 48kHz and the amp sim had to be in charge of sync or there would be many artifacts. My MiniDisc, which I use to capture audio in the field, also records @ 48kHz.

So that's my story. At the end of the day, they are my ears and they are the only pair I've got. So I've got to put some trust in them. The real trick is to get to the point where you've done the best you can to research it all and now it's time to actually record something.

Peace, Rez
 
24/41.1

I've A/B'd different possibilities before, but after reading the thread I think I may do some more tests to make sure I'm getting the most from my gear.

Tim
 
44.1 all the way. Keep things simple, keep hard drive space, and still get a quality-sounding product. The whole 96khz, 192khz sampling ability is a big marketing hype in my opinion so companies can make their product look better stat-wise next to others. A higher sample rate does not make up for poor quality converters.
 
Supercreep said:
96k, but only because I have a beefy PC and lots of disk space.
Click to expand...

I think that'll be my attitude/decision too
I have an 1820m coming,192/24, and a fast 250gb drive... so I'll just record @ that and say HEY! F*CK THE DISCUSSION, LET'S MAKE SOME MUSIC!
 
Recently read an article by one of Apogees designers. Being an expert in digital audio and conversion, he should know what he's talking about. He pointed out that the human ear can't hear much above 18-20khz and that since 48K gives you ability to record up to 24khz, there was little benefit to recording at sample rates above that. It seems to me that the higher rate you use, the more accurate the sound waves would be recorded, but he's the expert so I'm sure he is right. I've also read that 88.2 would be easier to convert to 44.1 due to simpler math processing but with newer dithering algorithms, that may not be necessary. I use 48k since it seems to provide adequate frequency response with minimal disk space and processing.
 
You must log in or register to reply here.
Back
Top