Craig's comparison of 44.1 vs 48k

[Dan Worley]

Craig,

 

About two years ago I watched a Sweetwater Sound webcast of you doing a promo workshop on the Panasonic DA7--or something like that.

 

My memory is so bad it's a wonder I find my way home, so forgive me if I have misconstrued this, but, if I remember right, in that webcast you mentioned something about the small - and what you thought - unimportant difference there was in recording at 48kHz over 44.1 (besides just 3900 samples). You expounded on this (a bit of a diatribe, actually) in a way that really made sense to me at the time, but I can't remember exactly how you explained it. Do you? If you do remember, I would like to hear your take on it again. Or you could just make something up right now.

 

Thanks a lot,

 

Dan Worley

[Ethan Winer]

Dan,

 

> you mentioned something about the small - and what you thought - unimportant difference there was in recording at 48kHz over 44.1 <

 

I'm not Craig, but until he finds a moment to answer...

 

The highest frequency you can capture at 44.1 KHz. is about 20 KHz. and the highest at 48 is about 22 KHz. The highest most people can hear is 20 KHz., so anything higher than 44.1 is a waste of resources. Further, if you plan to put your work on a CD eventually - and who doesn't? - recording at 48 means you'll have to resample the files. Which at best is an extra step and at worst may degrade the audio quality.

 

--Ethan

[musicman1ovation.net]

Originally posted by Ethan Winer:

The highest frequency you can capture at 44.1 KHz. is about 20 KHz. and the highest at 48 is about 22 KHz. The highest most people can hear is 20 KHz., so anything higher than 44.1 is a waste of resources. Further, if you plan to put your work on a CD eventually - and who doesn't? - recording at 48 means you'll have to resample the files. Which at best is an extra step and at worst may degrade the audio quality.

 

--Ethan

 

If one is recording in a computer DAW, I would agree that tracking at 48 Khz, only to have to down-sample to 44.1 Khz for a CD mix might be that "at best is an extra step and at worst may degrade the audio quality," as you said. But if one is using a standalone digital recorder, whether it is some ADATs or a DASH or something like either the Tascam, Mackie or Alesis 24-bit 24 tracks, and they are using an analog mixer, then 48 Khz, or for that matter, 88.2/96 Khz, if they have a very high end console and support equipment, would be an improvement on the sound. The overall multi-track playback would have a slightly more transparent high end, since the top wouldn't arbitrarily come to a screeching halt within the hearing range of a very few people but a large number of devices. I think that when these devices mix their "upper atmospheres" in the analog realm, the higher the frequency range, the better. After all, in analog, 20 - 20 Khz is actually 20 - 28 Khz (just a grabbed number) in the case of some equipment, because the 20 Khz spec is generally something like +/- 1db in moderate equipment these days. It might well only be down 3 db at 28 Khz. No, I can't hear that, but I think my brain and the equipment's "brains" perceive how it effects lower frequencies (under 20 Khz), and I believe there is a subtle but real psycho-acoustical difference. It isn't hearing or not hearing, but a very fine difference that probably shows up merely as the subtlest transparency shift.

 

Someone else is going to know way, way, way more abut this than me, but I do believe this is so. Meanwhile, I do mix through an analog board, and I would always track at 48 Khz for that reason. I think it is actually needed.

[Lee Flier]

Well...

 

For years when CD's came out, I couldn't stand the sound of them. The high end was very definitely painful to my ears. I always presumed at the time that it was because the sampling rate wasn't high enough. But really it was just that the converters of the time were crappy, and that engineers really hadn't figured out how to master stuff digitally yet. I didn't buy a CD player until about 1993. By then, things had much improved.

 

I too was willing to buy into the "perceived psychoacoustic effect" of sub-20Hz or +20KHz with analog. However, in reality you never heard those frequencies on analog recordings anyway, because most of the other components in the chain, from the mic to the outboard gear to the recorder to the mixer to the listener's stereo system, was not designed to record or reproduce those frequencies in the first place. Even today there are not many components that will pick them up, and even if there were, as soon as the signal hits the first link in the chain that CAN'T handle them, they're gone.

 

So if there are differences between the quality of different sampling rates, it has nothing to do with whether we "perceive" frequencies that aren't there. Doesn't mean there isn't some other reason though. Bottom line, you should compare your particular equipment recording at 44.1 vs. recording at 48 and then downsampling. Some people notice no difference while others seem to vastly prefer one over the other, depending on their equipment configuration. The quality of the converters and how DSP is applied probably has a lot to do with the differences.

 

--Lee

[John Worthington]

There is one case where I believe 48k is better than 44.1. Digital Video. If you're working with video, keep in mind that for DV at least, 44.1kHz isn't supported as a "locked" sample rate. The amount of drift isn't bad, but it's there.

 

For CD's, I tend to record at 44.1.

 

jw

[Ethan Winer]

musicman,

 

> ... 88.2/96 Khz, if they have a very high end console and support equipment, would be an improvement on the sound ... No, I can't hear that, but I think my brain and the equipment's "brains" perceive ... I do believe this is so. <

 

You are welcome to your beliefs, but my belief is that this cannot be verified scientifically. There really is no reason to think a frequency response beyond what you can hear can or will make a difference. In my experience, the things that most make a real improvement in a recording's quality are good mikes and good talent in front of the mikes.

 

--Ethan

[Anderton]

Well, everybody's pretty much said what I was going to say, but I wanted to add something to Lee's comments. BTW Lee - are your postings here first draft stuff, or do you do a lot of editing? Just curious.

 

>>For years when CD's came out, I couldn't stand the sound of them. The high end was very definitely painful to my ears. I always presumed at the time that it was because the sampling rate wasn't high enough. But really it was just that the converters of the time were crappy, and that engineers really hadn't figured out how to master stuff digitally yet. I didn't buy a CD player until about 1993. By then, things had much improved.<<

 

A BIG factor is that a lot of early CD players actually used 12 or 14 bit converters. Yuck. Also, the smoothing filters were awful, and oversampling hadn't been invented yet.

 

>>Even today there are not many components that will pick them up, and even if there were, as soon as the signal hits the first link in the chain that CAN'T handle them, they're gone.<<

 

Very true. If someone claims that the ultra-high frequencies more closely simulate what happens with a live performance, fine. But we never heard anything around 40 kHz come out of our home entertainment systems back in the all-analog days.

 

>>So if there are differences between the quality of different sampling rates, it has nothing to do with whether we "perceive" frequencies that aren't there. Doesn't mean there isn't some other reason though.<<

 

The higher the sampling rate (including what oversampling does), the gentler the filter slope you can design into a system, and the gentler the slope, the less violence done to your signal. This is why oversampled 44.1 kHz sound so much better than non-oversampled, even though the sampling rate is the same.

 

In theory, 48 kHz give a slight theoretical advantage, which some people can hear with the right types of program material. Also, the idea of recording to 48 kHz when mixing through analog makes perfect sense -- why not optimize the capture medium?

 

Okay, enough picky technical BS for now...think I'll go create some new guitar loops for an upcoming project...

[Chip McDonald]

Originally posted by Lee Flier:

Well...

 

For years when CD's came out, I couldn't stand the sound of them. The high end was very definitely painful to my ears. I always presumed at the time that it was because the sampling rate wasn't high enough

 

I had a big awakening when my Sony cd player broke down on me, and I was given a friend's Fisher (complete with "human assisted caddy" problems...). At the time - back in the Jurrassic era of CD players, the common thought was "they all sound the same", barring the ever elusive Magnavox that has some sort of immediately discernable differance.

 

Playing a CD through the Fisher sounded really different, really bad to me. Which was odd, because again - it was common knowledge that all CD players were pretty much generic in sound quality, which made one temporarily suspect a mental deficiency....

 

 

So I begged a few friends to let me borrown their CD players...

 

... and learned more about "digital sound" in an evening than I ever have since.

 

The same CD through 4 different players yields 4 different sounds. The "bad" sounding ones - strident upper mids - theoretically, price wise, were in the same category as the "better" sounding players. The weird thing - is that listening to all of them side by side made me aware that there is in fact a digital "sound": I don't care about trying to quantify it in the physical realm, it works practically enough for me to be able to hear it.

 

Which makes me a snob to a lot of friends, because cheap digital effects has that "sound" more than others. The newer higher bit rate stuff sounds better; perhaps not from a "resolution" standpoint as much as side effects of the electronics surrounding the translation process.

 

That's a quality that I'm not sure is measureable, any more than the apparently non-tangible aspects of upper tier preamps and such. Still, I'm curious as to the nature of the way the signal is converted, the *quality* of the parts involved in that. In which case, the magnitude of the increase of the sample rate makes me wonder if there are gains from the engineering required for making such a converter on those grounds alone, regardless of the math involved. I would not be surprised that as a whole 96k capable converters sound better if for no other reason than by products of better engineering specs surrounding the process.

 

At the time CD players first emerged, it was easy to find people to argue that an 16/44.1 converter would sound *exactly* like any other, because after all - mathematically it *has* to yield an identical output...

 

 

http://www.mp3.com/chipmcdonald

[alphajerk]

"as soon as the signal hits the first link in the chain that CAN'T handle them, they're gone."

 

which is in most cases the convertors and then the speakers [depending on quality] i think its all about the filters.

 

once its gone, its gone forever which is why i dont buy the whole make it 44.1 because thats what its going to be anyways... well it wont be forever. and i can hear a difference to 48 so im catching it the best i can.

[lisencocasema.net]

Originally posted by John Worthington:

There is one case where I believe 48k is better than 44.1. Digital Video. If you're working with video, keep in mind that for DV at least, 44.1kHz isn't supported as a "locked" sample rate. The amount of drift isn't bad, but it's there.

 

For CD's, I tend to record at 44.1.

 

jw

 

 

That's right!

 

Listening tests prove: 44.1 kHz is as good as 96 kHz.

 

 

------------------

[nhcomp45aol.com]

Hey guys,I have owned my sine wave generator since the mid 70s. I use it fo setting up PAs,ATRs, and power amp repairs and mods.Over the years countless people have been present while the unit was hooked to a power amp with quality speakers. Whenever I think of it just for fun I check how high they can hear. In over 25 years I have never met anyone who could hear over 17K.For me the limit has been about 12-13K as long as I can remember.Has anyone else tried this simple test.On a different note after having all my Pioneer,Sony, ETC CD players burn out I got a Rotel for fifteen dollars at the local pawn shop. To this day I have not heard a better sounding CD player. Any comments , Paul.

[Ethan Winer]

Paul,

 

> In over 25 years I have never met anyone who could hear over 17K. <

 

I think I used to be able to hear 18 KHz. http://www.musicplayer.com/ubb/smile.gif Of course, that test requires speakers that can reproduce the highest frequencies, and many cannot. Also, many speakers are very directional when playing sine waves that high, so sometimes pink noise and a sweepable filter is a better method. You play pink noise through the system, start with the filter set to 30 KHz. or whatever, and then slowly sweep it downward until you can hear a loss of highs. Then read the frequency dial.

 

--Ethan

[the stranger]

What frequency is the whistle that a TV makes?

[Lee Flier]

Originally posted by Anderton:

BTW Lee - are your postings here first draft stuff, or do you do a lot of editing? Just curious.

 

I'm pretty much "stream of consciousness"... I just type everything into the little box, I don't use the spell checker, and I've only done any editing if the BBS says I have. Before I hit the send button I check for typos and obvious stuff, but that's about it. There is no way I could manage the volume of stuff that I type on here if I nit picked over it! http://www.musicplayer.com/ubb/biggrin.gif

 

I try not to nit pick anything I do too much anyhow - it's not much fun. http://www.musicplayer.com/ubb/smile.gif Boy do I hate it when I'm mixing somebody on a DAW and they insist on nit picking every bar to death.

 

--Lee

[Hippie]

Well, if some technician, way back when, determined that 48k is better (if only on paper),than 44.1k, and it is just a strike of a button away, why not use it? If you're mastering through an analog board and you are not going to be dithering the signal, I don't see a reason not to use 48k.

 

-Hippie

[Philip OKeefe]

Originally posted by dr destructo:

What frequency is the whistle that a TV makes?

 

I think it's a 15 KHz oscillation. Maybe it was 16 KHz. It still drives me batty.

 

Phil O'Keefe

Sound Sanctuary Recording

Riverside CA

http://members.aol.com/ssanctuary/index.html

email: pokeefe777@msn.com

[Guest]

The TV scream drives me up the wall, one reason I don't have a TV anymore. Growing up without a TV probably makes me more aware of it, though.

 

High-pitched noise and eye-strain are the worst thing about the DAW environment, I'm looking forward to a flat screen and high-quality computer for that aspect alone.

 

Generating sine waves in Cool Edit (-6 dB, Sennheiser HD 600 phones) I can hear the signal to 19khz in the right ear and 21khz in the left.

 

The 20hz lower limit seems to hold true for me- maybe it's the limits of the phones, but I don't think so. Below that it's silence, unless there's for example straightout physical rattling of a set of speakers, then I know it's there but I can't hear it. Obviously on a monster system you can feel subsonics.

 

From my experience, I'd have to say that the 20-20k rule of thumb is pretty good. I also believe that people percieve MUCH higher frequencies subconciously, and that most people hear much better than they think they do- it's a matter of health, weariness, attention, the way you listen, etc.

 

I believe that the kind of mindset you can easily achieve when turning slowly on the rotisserie deep in the valley of the Stoned Age is very conducive to hearing- can't do it much myself because it dogs my tonsils, but I do think it's a good thing.

 

A good friend who is abnormally physically strong once told me it was largely a matter of "thinking strong" and not selling yourself short of your potential and I think there is a similar phenomenon here.

 

The human being is one bad-ass creation, and that's you. Don't sell yourself short.

 

 

- Cameron Bobro

[nhcomp45aol.com]

Cameron, I want your ears please,I will pay any price. Back in the mid 60s when I was about 12 years old I had a cheap Radio Shack FM stereo tuner and once in a while I could hear the 19K Sent to trigger the stereo beacon.A quality tuner would filter this out.Wish I could hear that again.

[Guest]

Well the other kids always made fun of my big ears and nose. http://www.musicplayer.com/ubb/frown.gif I have to say, my wife (professional opera singer) can hear at least as high (she is younger it is true) and we both had childhood asthma. Hearing loss and ability to hear high frequencies are both paradoxically possible side-effects of childhood asthma, or so a doc told me, which doesn't make it necessarily true.

 

Any pediatricians reading, chime in.

 

I think almost everyone hears much more and better than they think they do. It's like perfect pitch- a cousin who usually sings VERY off-key when singing along with the radio can be psyched into singing perfectly just by saying "that song, you know, the words are...how does it go?" Boom! dead-on version, in the correct key.

 

And I think the contemporary focus on youth has screwed up many perceptions- the SKILL of hearing can easily outpace the shear physical ability, just as in reflexes, physical presicion, love-making, etc.

 

-CB

[rumpelstiltskin.]

Originally posted by nhcomp45@aol.com:

In over 25 years I have never met anyone who could hear over 17K.For me the limit has been about 12-13K as long as I can remember.Has anyone else tried this simple test?

 

i am an advocate of this test after i realized my hearing falls off around 16 or 17 kHz. i was using a sine wave generator at work through a pair of JBL control5 monitors. it wasn't scientifically done, but i was playing with an amp after using it to check the bandwidth of a circuit at high current. i cranked up the frequency in 100Hz intervals and stopped hearing somewhere after 16kHz.

 

i've been to a few concerts, but now, even while rehearsing, i insist on using ear plugs if it gets particularly loud.

 

all i can say is that a number on a spec sheet doesn't equate to all things being equal. there are a lot of ways that a $2500 20bit/48kHz standalone could sound better than $200 24bit/96kHz soundcard.

 

if you really want to get theoretical, check out george massenburg's forum, 96kHz discussion or

http://www.musicplayer.com/ubb/Forum3/HTML/000822.html

if you want to really know about the math and science behind the sampling rate debate.

[rumpelstiltskin.]

Originally posted by dr destructo:

What frequency is the whistle that a TV makes?

 

i think it's about 1kHz. not very high at all.

[Chip McDonald]

Originally posted by synaes:

i think it's about 1kHz. not very high at all.

 

 

It's from the flyback transformer and it's 15.7k I think. On computer monitors it's the horizontal sync rate.

 

I *wish* OSHA (if King George hasn't completely dismantled them yet) would make standards to quiet flyback whine. I can't get near some monitors because of that. Had to return a brand new television once because I couldn't be in the same room with it...

 

 

http://www.mp3.com/chipmcdonald

[rumpelstiltskin.]

well, i obviously don't know what i'm talking about.

 

i thought we were talking about the tone they played when the test patterns were up at the end of the broadcast day (does that happen anymore?).

 

i don't watch enough tv, i guess, to notice a 15.7kHz power supply hum. i can barely stand the hideously compressed advert audio. it makes me nuts to have the volume on during advertisements.

[Chip McDonald]

Originally posted by synaes:

i thought we were talking about the tone they played when the test patterns were up at the end of the broadcast day (does that happen anymore?).

 

Oh!

 

Ok.

 

I HATE that.

 

When I was a kid we didn't have cable. When they did that it affected all of the channels. You're sitting there, watching whatever, and suddenly the announcement comes on and

 

 

BERRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRR

 

 

It meant only one thing to a precocious kid with a very elaborate and vivid imagination:

 

 

 

 

IMMINENT NUCLEAR HOLOCAUST

 

It was just certain that a very painful and dramatic death was about to occur. Combined with that terse, scratchy and band-narrowed voiceover - the Competant Voice of Dispassionate Doom - it always scared me and got my heart racing.

 

To this day it gives me the willies, I HATE it.

 

http://www.mp3.com/chipmcdonald

[Philip OKeefe]

Man Chip, I can RELATE! That old "emergency Broadcast System stuff made mem feel exactly the same and think of all those similar things. I HATED sine waves for years and used to always turn off the volume on those broadcasts until they were over.

 

So what do I do? Go out and get involved in a profession where 1 KHz sine waves are used on a daily basis! http://www.musicplayer.com/ubb/smile.gif (back in the stone age when I was getting into audio engineering, everything was analog and you used 100 Hz / 1 KHz / 10 KHz test tones on every reel and if you were serious, you checked your azmuth / alignment / bias etc. regularly...)

 

Our experiences and thoughts were so similar, I could have signed your post.

 

 

Phil O'Keefe

Sound Sanctuary Recording

Riverside CA

http://members.aol.com/ssanctuary/index.html

email: pokeefe777@msn.com

 

 

PS The other thing that gets my heart going is those "we interrupt this broadcast for a special report" notices they flash on the screen before a newscaster comes on with whatever urgent news. It's usually something bad, but when I was a kid it seemed like it was always someone getting shot - JFK, MLK, RFK, et cetera.

 

 

Originally posted by Chip McDonald:

Oh!

 

Ok.

 

I HATE that.

 

When I was a kid we didn't have cable. When they did that it affected all of the channels. You're sitting there, watching whatever, and suddenly the announcement comes on and

 

 

BERRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRR

 

 

It meant only one thing to a precocious kid with a very elaborate and vivid imagination:

 

 

 

 

IMMINENT NUCLEAR HOLOCAUST

 

It was just certain that a very painful and dramatic death was about to occur. Combined with that terse, scratchy and band-narrowed voiceover - the Competant Voice of Dispassionate Doom - it always scared me and got my heart racing.

 

To this day it gives me the willies, I HATE it.

 

http://www.mp3.com/chipmcdonald

 

 

 

This message has been edited by pokeefe777@msn.com on 04-30-2001 at 04:43 PM

[Anderton]

DUCK AND COVER!!

[Chip McDonald]

Originally posted by pokeefe777@msn.com:

PS The other thing that gets my heart going is those "we interrupt this broadcast for a special report" notices they flash on the screen before a newscaster comes on with whatever urgent news.

 

Oh yeah, definitely.

 

Someone should chronicle the Cold War years from that perspective. It's easy to forget... Reagan is on tv posturing, always reports about which side has more ICBM's, Russia is doing that we're doing this....

 

Meanwhile a wonderful little film came on television called "The Day After". Remember that?

 

Man, really good timing. The odd thing was they made us watch it at school "to talk about it". Great, I don't want to talk about it, I've already read too many books about megadeath ratings, MIRV's, cobalt encased bombs, Nevil Shute's _On the Beach_, _Damnation Alley_.... hey, gimme a break....

 

 

Then there was the time around about the age of 12 I was asleep having a really wonderful dream about....... being in bed, half asleep, staring at the pulled window shade and suddenly seeing the darkness around the window shade get intensely bright (nothing like a brain with Rod Sterling built in)....

 

BOOOOOM!!!!!

 

Here's the really impossible odd part:

 

There was a power transformer on the corner 3 houses up the street from my parents house.

 

Someone, *JUST AT THE IMPOSSIBLY RIGHT MOMENT*, ran off the road and struck it - made a really nice loud explosive sound. For Extra Bonus Spooky Effect, it also meant the power went out: so now I'm laying in total darkness, no night light from the bathroom, no led display on the clock, nothing....

 

That *really* sucked for a good many moments....

 

The irony is we're probably more in danger now of an accidental or rogue attack than ever. Oh well.

 

http://www.mp3.com/chipmcdonald

[GT40sc]

Hey Chip,

 

I want to go back to your post about borrowing CD players and finding that they all sound different. So much depends on "the quality of the electronics surrounding the translation process." Absolutely correct. I have no friends left at the local stereo shops because I've returned five CD players in the last two years. And when I did find one with a good tone,(Parasound CDP-1000) it had trouble playing CDRs.

 

Damn.

 

SC

[hiraga]

It's also about SPL and the sound source.. My speakers are essentially flat to 40kHz (on axis - measured by me, I do loudspeaker designs for a living amongst other things..), and I can hear way over 20kHz, if the SPL is high enough. Ever been near a ultra sonic pool cleaner? Makes anyone go nuts after a few secounds.. most of them run at 40kHz.

 

As for the low frequencies.. well, it's difficult to say when you're hearing and when you're feeling. The SPL needs to be at least 90dB to be able to hear a 20Hz tone, and frankly very very few speaker systems can do that. You need something like the big 2x18" Genelecs AND (much more important!) a room that can support it..

 

I'm a firm believer in that we, humans, can percieve much more than '20-20kHz'. It doens't matter if we're hearing or feeling, imho. My own (monitor) system sounds a LOT better, when I don't limit it to 20-20kHz. My LF drivers has an Fs @ 8Hz and are flat, if the room is big enough, to well under 20Hz.

 

I'm really looking forward to do 96kHz permantly..

 

TVs fucking sucks.. never owned one.

[KenElevenShadows]

I'll just mention quickly here that when I was in college (undergrad, not when I was going for my M.A.), I took a hearing test and could hear up to 19 kHz. Only one or two of my classmates were able to hear this high, although many of them could hear up to 17 or 18 kHz. One lady, who was in her 60s or 70s, could not hear past 11 kHz.

 

------------------

Ken/Eleven Shadows/d i t h er/nectar

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

music*travel photos*tibet*lots of stuff

"Sangsara" "Irian Jaya" & d i t h er CDs available!

http://www.elevenshadows.com

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~