Sample Rates

[OME]

I'm looking for an outboard digital recorder in attempt to not be so

reliable on my desktop pc and relives some strain from it.
I've decided maybe on a Boss BR1600, 1200 or the Roland 2400CD because
of the tracking power and built in drums/vocal stuff.

My main question is...what is the difference between 44.1 Khz and 96Khz
How much different is the sound quality between the two?

All the machines I'm looking at are 24bit, but only the BR1600 and 2400CD
have 96Khz sampl rates. The 1200 does all the same stuff but only uses a
44.1 sample rate.

Does anyone own a Roland here, anything above a 2000CD?

[JESTER700]

You will get different answers, as with anything audio. My take: If you are going to end up on CD, don't worry about the 96k. You will eventually resample to 44.1k ANYWAY, so any gains you get (which are questionable in the first place) will be gone. If you'll end up on DVD-A or some other potentially 96k format, it's up to you. I've never seen a good test showing that 96k gives any benefit as a playback format, but it might make sense to have the option "just in case".

[Ibanut]

A sample rate is how many "pictures or samples" the recorder takes of the incoming sound waveform per second. So 44.1 takes 44100 samples per second of the waveform .96 takes 96000 per second. 96 will give you a better tracking of the original signal but like the other guy said the AES standard for a cd is 16 bit 44.1 sample rate. you can have a copy thats 24/96 but the store quality standard is 16/44.1
The bit rate is a measurement of the top to bottom sample rate of a waveform. or dynamic range . tracking in 24/96 gives great tracks and thats what most professional engineers do. some go even higher,but eventually they all end up 16/44.1 or an mp4 which sucks.digital downloads SUCK!!!

Buy the best you can afford you cant go wrong with 24/96

[JESTER700]

True, but keep in mind that there is no advantage of higher sample rate BEYOND INCREASED BANDWIDTH. IOW, 44.1k will give you response out to 20,000 Hz. That's all most people (ESPECIALLY adult males!) can hear. There is a common belief that within that passband higher sample rate will somehow add to the accuracy of the signal. Not so; it only increases bandwidth (well, there was some old gear that had dodgy filtering, but that's another story and no longer germaine). Of course, if you have great Shoeps mics, great tweeters, and great ears, knock yourself out!

Bit depth helps, especially when you're recording and mutilating...er, effecting the signal. In the real world with real systems and real environmental noise, I agree with 16 bit as a good choice of delivery format - you *generally* won't hear a difference going to 24 here. But there are rare exceptions.

Digital downloads usually DO suck, but that's all about bitrate. When you start hitting 256k/s, many of the codecs (wma, mp3) sound pretty durn good. WHY so many of them are at 128k is beyond me.

[OME]

That helps me plenty, I've converted some songs from a 44.1Khz to 96K
and I didn't notice too much difference, I thought it may have been the
SBLive soundcard, maybe not.

I was looking at the 96K stuff to make my final mixes fatter, I guess the
fatness of a mix is all in the EQ at the master

[frankfalbo]

Converting from 44.1 to 96 is an impossible premise for improvement because it's a 44.1k signal to begin with. The only benefit I can see of 96k is this: Every digital conversion will produce some artifacts. If you get enough tracks together, there can be some awkward manufactured overtones as a result of the combination of all the converted signals. You don't have that with analog. So I think 96k would reduce that possibility in a very large multitrack session, where a lot of things are doubled. Like if you had 10 guitar tracks, a drum loop + real drums, multiple vocals, etc. all with seperate effects tracks, you might hear a difference in your final master.

Even then, worrying about 96k and 24 bit is a waste of time IMO. Especially on a Boss or Roland box. That's so laughable, because you are inputing through their consumer grade preamps, EQ, and digital effects processing. (including the essential compression) If you're using COSM on anything, guitars, mics, whatever, then you're a slave to that level of sound quality, regardless of the sample rate or bit depth. Not that the Roland boxes sound bad, because they don't. They sound good. But the high conversion specs are simply marketing hype on those boxes, unless you're bypassing all the fancy features and using pro grade outboard gear. In that case you wouldn't be buying the Roland box. You'd get a stand alone multitracker, or you'd stay on the computer.

[7 Dying Trees]

CD's have a sample rate of 44.1kHz as the human ear hears up to roughly 20kHz and due to sampling laws the sample rate must be twice the maximum frequency in range. As such 48kHz may give a noticable improvement, but not much.

Any case, 96 kHz is just high end, and is NOT needed in anyway, all there is is a massive waste of bandwidth encompassing frequencies no human (and if anyone claims otherwise they are wrong, as it can't physically be done, you simply cannot hear into the upper ranges). So 96kHz just fits in with all the "a bigger number means it's better" brigade.

ALSO when you dump the music to CD format (44.1kHz) then note that you have to convert between the two sample rates introducing (however small) rounding into the conversion.

24bit, well, basically all you really need is a few more bits than 16 to catch rounding artifacts. Go for 24bit if you can, but as far as 96kHz is concerned I wouldn't be too bothered about it.

[JESTER700]

Originally Posted by frankfalbo

Converting from 44.1 to 96 is an impossible premise for improvement because it's a 44.1k signal to begin with. The only benefit I can see of 96k is this: Every digital conversion will produce some artifacts. If you get enough tracks together, there can be some awkward manufactured overtones as a result of the combination of all the converted signals. You don't have that with analog. So I think 96k would reduce that possibility in a very large multitrack session, where a lot of things are doubled. Like if you had 10 guitar tracks, a drum loop + real drums, multiple vocals, etc. all with seperate effects tracks, you might hear a difference in your final master.

I think this depends on the digital mixing engine. I've heard a few unfriendly things about the ProTools engine, but that was from a competitor's forum - though from a generally unbiased and experienced engineer. I wonder if anyone's compared these things? Vegas hasn't given me grief yet, but I haven't stacked tracks to the sky yet, either. I do mix to 24 bit (from 16 bit tracks), but that shouldn't affect sample rate anomalies.

But if the mixing is done in the analog domain (even if the tracks are recorded & stored digitally), this shouldn't happen in any case.

[OME]

Thanks for all the input, it's great, I definitely knew I needed some "outboard" opinions besides my own. I've decided to try a Boss BR1600. I will take advantage of the COSM and drum features, but I will also be inputing some of my own amps and mics into the machine. Most of the time I like to get a good gritty guitar onto tape. Somethin that sounds true, not too processed, I like hearing your handsweeps on the strings and fingers hiting strings and that type of stuff.

It'll be a few days before it gets here and I'll do a crash jam on it to see how it handles my playing and I'll let you guyz know what happened. I hope I get the best of both world from it cause I do seom techno type stuff everynowandthen. Thats when I need the most clarity.

[LGMT]

Maybe its too late but, go for 44.1/24

Great benefits at a low HD coast.