Calling Digital Experts! What Is 96khz Sampling Frequency, Really?

  • Thread starter Thread starter UnclePonto
  • Start date Start date
U

UnclePonto

New member
I feel like Charlie Brown in "Merry Christmas Charlie Brown" when he yells out, "Can't someone tell me what Christmas is really all about?". I need a Linus to answer my frustrated and despairing cry, "What is 96kHz Sampling Frequency and why is it important to me? And what's more, what's the significance of 24 Bits recording."

Could someone help me fill in the blanks?:

96kHz is better than 48.1 kHz because____________
24 Bit recording is better than 16 Bit recording because_______

Are these really audible differences or something else? Should I really concern myself with making sure that I have 24 Bit/96kHz recording or will I never know the difference? Can someone explain in layman's terms what these things mean?

I can honestly say that I have NO IDEA what any of this means. Linus... I'm looking in your direction.... help!
 
Quick inexpert answer. Correct me if I'm wrong!

96kHz is better than 48.1 kHz because you have twice as many samples per second, which should mean more accuracy (there's a recent thread on Prorec.com about this, check it out).


24 Bit recording is better than 16 Bit recording because each sample has exponentially (or is it logarithmically?) more resolution (volume levels it can represent--something like 65,000 levels in 16, and many millions in 24).

Are these really audible differences or something else? It seems to be the consensus that 24 bit is superior to 16 bit, though some will argue the point. There is less consensus about the merits of 96k or 88.2k over 48k or 44.1k, but many claim that 88.2k and up sounds better. There are pros who record in 24bit/44.1k, however.

This is my understanding at least. I can't wait until I can upgrade my slow-as-poop computer (funny, it was a FAST machine a few years ago ;) ) so I can try these out for myself! :)
 
My 2 cents

First off, welcome to the nasty world of digital audio! And secondly, welcome to the even nastier world of marketing-speak! Okay, that's unfair. They both offer (usually) audible increases in quality (which the manufacturers are quick to point out), but they also bring with them difficulties (which the manufactureres hope you won't hear about)!

I guess the quick Linus answer is, "It's what you make it, Charlie Brown." I mean, let's face it, in the end, we're stuck with 16bit/44.1kHz, and since CD's sound pretty darned good, it can't be THAT bad, can it? Let's try it this way: if you took two 2-track DAT machines, one at 16bit/44.1k, the other at 24bit/96k, then dithered the 24bit jobbie back down to 16bit/44k, it should sound ABOUT the same. The 'about' has to do with things like dither algorithms and such, but they should be pretty close, overall. So in this particular application, it would make no difference at all.

The problem is, we don't have theoretical converters, we have real ones, with noise, nonlinearity, and gain/offset errors. Let's break the two parts down, sample rate (44.1kHz vs. 96kHz), and resolution (16bit vs. 24bit).

Sample rate is pretty easy. The faster you sample (assuming your error rate doesn't go up), the higher a frequency you can capture. A natural consequence of this is that you can also capture faster transients (snare drum attack, pick strike, vocal noises) better. This helps you twice. First, you get more information to work with when you record, so you have more information to work with when mixing. Second, the slopes of all your digital filters can be more gentle (natural) and thus subsequent processing won't make the high end sound as forced. The two downsides here are bandwidth and sample-rate-conversion. Bandwidth simply means you need more than twice the hard-disk space, transfer speed, and processing power to handle the same audio stream for 96kHz vs. 44.1kHz.

Sample rate conversion is something a lot of people don't like to talk about. An exception here has been Sonusman, who in the past has spoken out vehemently against all sample rate conversion. The problem (especially when going from 96kHz to 44.1kHz) is that there are 2.176 samples in the 96kHz track for every one sample in the 44.1kHz track. Even if you used an 88.2kHz sample rate, just throwing out every other sample will not necessarily yield the most accurate 44.1kHz output. Throw in the non-integer multiple, and you need to start using math to "guess" what the sample was that was in between (the .176 th of a sample). That means that you are usually off by some amount, and someone with golden ears somewhere is going to say they hear it. Will you? twist the knobs, Charlie Brown! You'll need to see for yourself, as only you can tell if you can hear the difference. I can't afford the bandwidth, so it's a non-issue for me. Either way, if you can afford the higher sample rate, you're usually better off to use it.

Resolution is a very different beast. First off, esactun was right about both topics. So with higher resolution, the finer a detail you can resolve between different input amplitudes. That's how it applies to recording, but it's even more vital to the processing that happens during mixdown. We all know that if you multiply 4.2 * 3.2, you get 13.44 (at least my calculator knows that, anyway), but if I limit myself to that one decimal place of 'resolution,' I can only say I have 13.4 . Now what if I really want to know 4.2 * 3.2 * 2.2 * 1.2 ? My calculator says the answer is : 35.4816. But lets say I do it as three separate multiplications, and only keep 1 decimal place of resolution? We've already got 4.2*3.2=13.4, so now it's 13.4*2.2*1.2.
13.4*2.2 = 29.48, so I call it 29.5, now I've got 29.5*1.2 = 35.4. But if I round my original answer, 35.4816, to 1 decimal place, I get 35.6! So I'm off by (in effect) a whole bit! And that's only for 3 operations! What if I need 8 operations on a single sample to calculate the EQ in my Waves Q10? I could be off by even more! So they usually keep much higher precision inside the plugin (so you'd actually GET 35.4816, rounded 35.6), and only do the roundoff at the end.

But what if you use 10 different plugins on a single track? Bingo! Same problem, only this time you don't have the extra internal precision to save you. Here's where that high resolution REALLY helps you out! Since we're only keeping 16bits of info anyway, we can afford some error down in that 24 th, 23 rd, or even 22 nd bit without hurting anything very much.

So what's the cost? Well, about 50% extra bandwidth, just like the increased resolution. But there's another problem. All converters are NOT created equal! I believe Bob Katz has said that he has yet to hear a 24bit converter that yields as much useful information as his own (super-top-of-the-line) 20 bit converters. And others have said that some poorly designed 24bit converters actually show residual noise (not from the input signal) up around 16bits!!!

I've spent very little time researching 24bit converters, since I'm happy with the 20-bit units in my Yamaha 01V. Are they the best? NO WAY! Do they work for my crappy home recording? YEPers, baby! Can I afford even 2 channels of Apogee or Lucid conversion right now? Nope, but it's not that big a deal to me right now. At some point you've just gotta stop buying gear and start recording/making music. That's where I am right now.

Okay, last thing here is DITHER, that nastiest of nasties!!! It's just like sample-rate conversion, only now you're trying to figure out what series of 16-bit samples will sound the most like the series of 24-bit samples you started with. There are a lot of good discussions of this here, and my own opinions are just that (I can't really hear the difference between the dither algorithms I've heard so far), so I'd just suggest hitting the SEARCH button for more info on dither.

I know this has turned into a MAJOR sermon at this point, but I hope I've cleared up some of the issues surrounding digital audio and its 'no-free-lunch' behavior. If you have specific questions about anything I've said, please ask away. And if this got WAY too technical (I never do that :) ) I apologize, and I think I can put any of this in to as layman's a term as you'd like (I hope :))
 
96kHz is better than 48.1 kHz because it's 48 more
24 Bit recording is better than 16 Bit recording because it's 8 more

I agree with the gist of esactum's post above. There's a lot of disagreement on both points.

To be wildly oversimplistic, picture an audio signal as a wiggly line on a graph ... time is passing as you move to the right, signal (voltage, the position of the cone in a speaker, pressure waves in the air) is swinging up and down, above and below zero. A digital recorder takes a series of snapshots of where the line is.

The sample rate is how often it takes snapshots. At 96kHz, it's taking 96,000 snapshots per second. At 48kHz (48.1 is not, so far as I know, a sample rate used by anything), it's taking 48,000 snapshots per second. At 44.1kHz (CD standard), it's taking 44,100 shapshots per second.

You can think of each snapshot is a value between -1 and +1, where 1 is the highest positive voltage you can produce and -1 is the highest negative. As everyone knows from the real world, when you measure something, you can describe it with more or less precision: "This board is 3 inches wide," "this board is 3.1 inches wide," "this board is 3.14159 inches wide" etc. To write a more precise number, you need to take up more space on the paper ... or, if you're writing it on a hard disk, you need to take up more space on the disk. A 16-bit system uses 16 bits (where 1 is represented as +000000000000001, 2 is represented as +000000000000010, -32,768 is represented as -111111111111111, etc.) So a 16 bit system has 32,768 separate steps between 0 and 1 (and between 0 and -1). A 24-bit system uses 24 bits, so it more precisely represents the signal (there are 8,388,608 steps between 0 and 1, and between 0 and -1).

What this means in actual practice is more complicated.

Basically, think of your line as a combination of sine waves with different frequencies, phases and amplitudes. Theory says that, with your series of snapshots, you can describe any sine wave perfectly, so long as its frequency is less than half of your sample rate. If the sine wave has a frequency that's half the sample rate or higher, you aren't taking snapshots often enough to know what it is. SO - run your audio signal through a low-pass filter that removes everything above, say, 20kHz. You can't hear anything above 20kHz anyway. Now your system, with it's 44.1 kHz sample rate can describe it accurately.

So you don't really gain anything by going to 96kHz. Well, maybe you do. There's no such thing as a perfect low-pass filter. It's easier to make a low-pass filter that cuts out everything above, say, 40kHz; and it produces fewer artifacts below the cutoff point. So there's that advantage, though people disagree about how much difference it makes in practice. There are others who say that, even though tests indicate that people can't hear anything above about 20kHz, it somehow affects what you do hear. There are others who believe a wide variety of other things, some of which involve magic and, possibly, angels.

As for bitrate (number of bits in each snapshot), things are also confusing. One way to think of it is to think about the size of the biggest signal (1 or -1) you can produce, realtive to the size of the biggest error in measurement that the system produces. Call the errors "noise" ... and you realize that what you're thinking about is the signal to noise ratio. In a purely theoretical world, a 16 bit system should be, at most 1 bit off in its measurements, which would produce a very high signal to noise ratio indeed. In the real world, things aren't so pure, so you don't get quite that huge a signal to noise ratio. Is the real S/N ratio achieved by 16-bit systems "high enough"? Well, that's a tough question. Is the ratio produced by a 24-bit system worth it? People disagree. Is there something else, other than S/N ratio going on: "smoothness," "soundstage," "I felt like the saxophone player was looking over my shoulder and reading my magazine"? Who knows.
 
A Note

One note, relevant to Johnboy's (Johnboy?!) post (which appeared between the time I started writing mine and when I finished it) and bitrate. I agree with him on the advantage of higher precision when you're manipulating the signal (mixing, processing, etc.). You lose precision with every arithmetic operation, so I do tend to think (partly on faith, maybe) that having 24-bit signals going into a digital mixer makes sense, even though at the end of the day you may only want to wind up with a 16-bit stereo CD.
 
Whooooeeee! Thas' a lot of typing! Thanks for the great job guys (also saves me from doing all that typing myself!).

One other note on rounding errors: you'll notice that many plug-ins and digital boxes of different types actually have a higher internal bit rate for doing their calculations (32, 48, 56, etc.) This is precisely so that the multiple calculations (like in the excellent example that JohnBoy gave) don't result in a noticeable amount of cumulative rounding errors. After the digital processing is finished internally, the final number is then rounded back down to 24 or 20 or 16 bits (depending on the device). This is an effective way to keep rounding errors from becoming so egregious.
 
roundoff error

littledog:

Yeah, I was basically just trying to point out some of the less obvious merits of the various increases in precision. As to roundoff error, it affects me more when I'm running a bunch of plugins. Every change you make to a track reduces the amount of useful information left, so when I process my 20-bit converted tracks at 24 bit, I've got to do quite a bit of damage before the roundoff errors begin to overtake my original precision.


sjjohnston :
Thanks for hitting the more basic stuff!!!! I re-read my post after I put it together, and realized I'd probably missed the foundation stuff that would have made it mean anything. Thanks for covering my gross oversight, I really appreciate it.

"Angels" - I LOVE it!! Yep, there's enough voodoo in this topic to make you go blind, and enough of it comes from people I trust implicitly to make me question our ability to quantify digital audio right now.

One thing having to do with high sample rates also has to do with high-frequency equalization. In the analog realm, we can literally calculate the attenuation of a low-pass-filter at, say 200kHz. It's gonna be huge, but there's still SOME signal getting through. At 96kHz, only information below 48kHz is considered "valid." In your algorithm, you have to decide what to do with information above that point. You can probably dump everything above 30kHz and still be okay, and not get anywhere NEAR your upper limit.

44.1kHz, on the other had, demands VERY sharp cutoffs to keep 20kHz and totally attenuate 22.05kHz. So I think that's where a lot of the perceived benefit of high sampling rates comes in. It keeps us from getting "iffy" around our upper frequency limit.




But before anyone goes and says that, "Well, Johnboy said if you don't have 24bit/96k, you're missing all this stuff," remember, people made some pretty impressive recordings on 16-bit ADATs. Heck, there's another thread going around right now saying people loved their old 4-track cassette machines for their "immediacy."

I think someone put it best recently by saying, "It's not the tools, it's the carpenter." We all want the Midas consoles and the Manley pres and comps, but when it all stops being fun, it's time to stop and look around. I have improved my equipment over the last 2 years, but more than that I've learned that it's the tracking that makes the recording. When I started out, I couldn't get a good drum sound because I couldn't mic drums! I still can't get the drum sound I want, but I've got a better idea what I did wrong during tracking.

Okay, too much preaching. Anyway, I hope everybody's having fun recording and/or mixing tonight!!!
 
Hey Guys,

WOW!!! Thank you so much for a thorough explanation of what this is all about. I feel like I know so much more. I had to read your posts a couple of times to take it all in but I think I have a decent idea about what you are talking about. I looked around for information on this and could not find anything particularly useful - which is why I turned to the experts on this site. It's funny - I consider myself to be fairly technically "able" but I would have never guess that Sampling Frequency is what it is. Makes sense though.

I am only making audio CD's at the moment so it's probably fine for me to go 16bit/44.1 kHz. But I will "trun the knobs" and see if I can detect a difference, as suggested. But let me just throw out this metaphor for you and, if you have a minute give your opinions: do you think the capability of CD's will change in the coming couple of years to accept 24bit/96kHz? What I am getting at is: if I were setting up my home studio now, today, would it be wise to spend the bucks on the 24/96 capability so that my system doesn't go obsolete in a couple of years? Example - like my computer - when I got it in 1997 the 266 processor was the fastest you could get. In 8 months my parents bought a 333 for $600 less than I paid for a 266... and now you can't GIVE away a 266. Is that likely to happen with record devices where component makers will stop producing 16/44.1 compatible components and all of the 16/44.1 gang will end up having to rebuild?

Back to Charlie Brown - now I need to see Dr. Lucy (Psy.D) to diagnose my fear of technology!

Thanks again guys.
Uncle Ponto
 
I read something about a month ago about a firmware update to become available for several burners in the near future, that will quadruple the storage space on a CD-R. I think it mentioned a change in media too, but still: This would allow the same length album at 24/96 on a CD with room to spare, or 5-6 songs at 24/96 in 5.1 surround (subs don't need 96k, so might get their own sample rate, say 1/4... Plus, there is DVD-r, DVD-RW, DVD+R, etc, with 4.6 GB on a disk - sooner or later, enough people are going to be exposed to 5.1 audio and want it; not to mention if you get lucky and get to score a movie, they're going to want it in 5.1 at who-knows-what sampling rate/depth, things in this field are not static. I've seen specs for 5.1, 7.1, 6.1, 9.2, etc - My point is that any more you have to be ahead just to be a little bit behind. If you want to stay with any of these changes, you shouldn't limit your options you already know about, because tomorrow there will be even more. Hey, it's only money, right? On the other hand, Tascam 424's are nice and simple - aahhh, simple.... (sigh) Steve
 
Believe me when I say I was afraid that was going to be the next question. Before I jump in over my head, I record at 20bit/44.1kHz, because that's what my 01V supports. Now here's what I've heard and what I think of all of it.

Assuming equal quality converters, you're probably best off getting the highest sample rate/resolution you can afford. But as you go up in resolution, you also go WAY up in price. For example, the Apogee Rosetta is currently $1100 at Musician's Friend, and only does 2 channels of 24/48k. The 96k version is $1900!!! Now compare that with the M-Audio Delta 1010 PCI I/O card for the PC. It's only $600, and does 8 channels at 24/96. There are many people (both around here and friend of mine who, despite my constant urgings, won't come here) who have been very happy with the 1010. And my 01V isn't even as good as the 1010.

Is the Rosetta that much better? I don't know. My ears are still pretty green, I'm really subject to suggestion and tech-based prejudice, and I just haven't had time to sit down with a really nice converter for a few days and compare it to what I've got right now. But from what I know about digital processing, it's really easy to add bits to your converter and increase your sample rate. It's another thing entirely to make those changes yield increases in recorded information. Just because a converter outputs 24 bits of data doesn't mean that the lower 2 (or 4 or even 8) aren't just garbage and noise.

Go to Analog Devices' website (www.analog.com) and look up AtoD converters. Then find the graph that shows the various converters' error rates vs. input value and sample rate, temperature, etc. You may find that their best 16 bit converter is guaranteed to never be off by more than 1 bit (so it's really a 15-bit converter), but their cheapest 24 bit converter (which may cost less than that 16-bit unit) may have cumulative error right up to the 16 th bit! So when you buy a piece of equipment with that 24 bit converter (which costs more just 'cause it's 24 bit, right?), you're only getting 1 more bit of accuracy! Then we get into clock jitter (another topic well-covered on this site and worthy of a good ol' search) which takes all that other stuff and mucks it up even more. GRRRR.

Lest my next rant sound like I'm arguing with Nightfly (who it appears knows a LOT more about this whole recording thing than he's preferring to let on just yet :) ), he's right: if you can get yourself set up for 24/96 right now, do it, it's the way things are going. Assuming they don't just skip right to 24/192, in which case I'll have to shoot over to Apogee and kick some butt!!!

Me going Postal at Apogee:
"What are you guys DOING! You know how hard it is to make a good home recording without you nutballs figuring out how to QUADRUPLE storage, processing, and transmission bandwidths!"

That aside, it's like he said, 424's are simple. What exactly are you trying to accomplish? If this is your first foray into home recording, might I suggest doing what I did:

1) Get a 2-track cassette deck and a radio-shack 4-channel mixer.
2) Record buddies, grimace at sounds, try to fix it, play with mics, guitars, etc.
3) Give up on 2-track, get Tascam 4-track cassette deck with mixer and
EQ with (oooo boy!) sweepable mids!
4) Record buddies again, grimace less this time, keep changing around mics, guitars etc.
5) Get a compressor (Alesis 3630, of course!), keep grimacing, only now at the compressor pumping
6) Keep playing music (almost forgot that one)
7) Get a Soundblaster Live! With cakewalk LT, program the drums and bass, lay down your killer-mad guitar-playing skills.
8) Grimace at cheesy canned drums, but groove that overall sound is better.
9) FINALLY move up to full-blown computer setup. Or something else.

Not knowing where you are in this progression (and not wanting you to make most of the mistakes I have), let me suggest jumping right to the Tascam and Soundblaster (if you're not there already). This will give you MORE than enough crap to fool with while you learn about the stuff that has nothing to do with reolution and sample rate. It will also give you a minimal investment if you decide it's just not for you. For the record, my latest recording rig (minus mics) has ended up costing over $5000, by the time you factor in monitors (which that 24/96 soundcard will make PAINFULLY necessary). And that's a lot less than many people around here have into it. Yes it's a lot of fun to play on, and I'm getting some decent sounds (watch the MP3 forum over the next few days for a new mix), but it's been a long, hard road, and like I said in an earlier post, my weakness is still my tracking ability. If I'd spent more time on the Tascam and Soundblaster combo (and if this forum had existed back them), I think I would be much further along than I am.

But like Nightfly said, if you're beyond this point, the higher up you start, the further you are from obsolescence.

Man, I've GOTTA stop my ranting. I love digital audio (and I HATE the myths that the marketing guys are spewing), but this is starting to get outta hand. :)
 
Thanks Johnboy. I appreciate all the information. As for me, I've been home recording since about 1991. And to date I've done it all on a Fostex X-28 four track and some outboard equipment (Alesis effects unit, dbx EQ, BBE Sonic Maximizer... pots, pans, shoes, chop sticks, shot glasses, CROSS pens, and so forth). I believe that everything can be made to sound musical in the right setting. I struggled for years but untimately figured out how to get decent tracks and maintain some semblance of quality even after bouncing tracks. So now I'm going digital but I'm not into recording on computers so I'll end up with a self contained 16 track unit, a la, KORG, AKAI, or YAMAHA AW4416. After going through the specs on all these machines and speaking with a number of people I realized that I had no idea what I was hearing as far as BIT Rate and Sampling rate are concerned. But now that I see why one is deemed better than the other I feel more well informed. I never shop for gear w/o massive reconaissance. I would not say that money is 'no object' but seeing as though the difference (for the entire package of things I am planning to purchase - including genuine nearfield monitors) is about $1,000 I am not ruling any machine out based on price alone. I mean, if a Fostex machine can give me 10 years and 4+ albums of material, I would expect a 16 track to do at least that and likely more. So a grand over 10 years is not that much money. Anywho, I think you've all done a great job navigating through this information and laying it out and I appreciate it. Additional thoughts will be happily digested. By the way, do you really like the Alesis 3630 Comp or was that a more sarcastic comment? I need to throw a compressor on my rig and am trying to keep the costs reasonable ($200 or so). Yeah, I've lived without a compressor fot ten years. I'm not proud of it... and I vowed after my last album that I would never ever record w/o a compressor again...

Thanks.
 
If the comment about the Aleses wasn't sarcastic, it should have been - you need a RNC - stands for Really Nice Compressor, and it truly is. People who use them have said they would pay over $1000 for them compared to the rest of the bunch - yet the cost is only a couple hundred. Only have unbalanced I/O, but they kick the crap out of Alesis, and the company is NOT in Chapter ll, or wherever Alesis is now... Steve
 
Thanks Kinghtfly. I'm all over the RNC Compressor based on the swooning, gushing, and glorious reviews of this very same site! Found it at HumbuckerMusic... I think that's the name... at $175.00 with free shipping. That race is over. Thanks!!!!!

That's one decision made. Feels good.
 
Regarding sound quality.....

I've heard two "reasonable" arguments why higher sample rates sound better.

1. While the Nyquist Theorem shows that 44.1kHz sampling rate can reproduce signals up to 20kHz, this sampling rate cannot accurately maintain the phase relationships of signals in the region. I haven't investigated the mathematics of this argument, but it seems like a reasonable concern and is worth investigating.

2. The relatively low frequency filters needed to prevent aliasing distortion in 44.1kHz sampled signals causes significant degradation of the relative phase of high frequency information. This argument I have little doubt about.

The main consequence of these two factors, if they are indeed significant, would be to smear the high frequency stereo phase relationship and cause a degradation in imaging since high frequencies are where we derive the bulk of our spatial perception.

Another personal bias have is for using 88.2kHz rather than 96kHz. CD is still the primary medium and everything ultimately needs to be resampled to 44.1kHz. So it just seems like dividing by 2 is probably less prone to error than dividing by 2.176870748....

Like I said, I haven't investigated the validity of the maths behind any of this, but the arguments do at least seem to be within the bounds of reason.

barefoot
 
I have news for you all, which I'm sure that some of you know, DVD is 192k so 96 is old already!
 
I've been researching it for a while, and for about 7000 bucks, you can get a really nice 24/96 system.

>Mackie 1604VLZ Pro Mixer
>Mackie HR824's
>2 Delta 1010 Recording Cards from Midiman
>Shure KSM 32/sl Vocal Mic
>2 Alesis 3630 Compressors
>Alesis Midiverb 4 Reverb Unit
>ROTO Rack 12 spc rack
>Cakewalk Sonar 2.0XL
>Antares AutoTune3.0
>>COMPUTER SYSTEM
>>>>2.0ghz Pentium 4
>>>>Soyo RAID/Scsi Motherboard
>>>>420W Power Supply and Full Size Case
>>>>512mb of RD-RAM Rambus Ram
>>>>2 7200 RPM 120GB Hard Discs
>>>>40x12x40 Plextor CD-RW
>>>>16x Artec DVD-Rom
>>>>Keyboard And Mouse
>>>>EVGA nVidia GeForce ti4600 Dual Monitor Video Card
>>>>2 15" ViewSonic LCD Display Panels
>184 sq feet of wedges
>a 6x4x8 foot enclosure
>Desk
>Chair
>cords
 
It's certainly not how I'd spend my $7000. Any list with two (!) 3630's on it makes me wonder how "well researched" the whole list is!

You have one mid-grade microphone, only the Mackie for preamps, zero useable compressors, the Midiverb for your only fx, and Autotune. Exactly what kind of music are you recording that qualifies this as a "really nice" 24/96 system? Maybe a "bare bones" system would be a more accurate description, and then only if you were doing mostly sequenced recordings with perhaps a couple of overdubs.

Recording at 96k is not some sort of magic bullet that will automatically make your recordings sound great, no matter how mediocre the rest of your gear is! There are too many weak links in your chain to use the term "really nice". You obviously are more enamored with sexy editing capabilites than with good sound, based on all the bells and whistles you are getting for your computer at the expense of the audio part of the chain.

:eek:
 
Aw, c'mon Littledog, that is too a really nice system - compared to a Soundblaster, Cubasis and a Recoton mic... Steve
 
tyler657recpro said:
I have news for you all, which I'm sure that some of you know, DVD is 192k so 96 is old already!
Click to expand...
Actually DVD-Video, which was launched in 1997, has only CD quality sound.

DVD-Audio and SACD which can support 24bit 192kHz audio are effectively only about 1 year old and relatively few people have players yet (DVD-A was launched in 1999, but it took quite a while for titles to become available).

My guess is that it will still be a few years before you can stop worrying about converting down to 16bit 44.1kHz.

barefoot
 
You must log in or register to reply here.
Back
Top