All posts with more than 3000 Hits, prior to 2008

CD Audio

Sat May 17, 2003 7:08 pm


I'll do my best to explain this without being condescending, please bear with me.

A CD isn't really an audio device. It's simply a storage medium for digital data, which in this case happens to be digitized music. In order for a CD player to understand the information stored on a CD, the data on the CD must conform to to a specific format or the CD player won't know what to do with it. In the days leading up to the introduction of the compact disc, a bunch of technical dudes from the music industry, and some geeky techies that know all about computer stuff, formed a committe to define the format for CD audio so that all CD players could understand the data on the CD. They came up with this thing called the Red Book standard that declared that in order for a CD player to understand what was on a compact disc, the CD had to be the following:

2 tracks, to allow for conventional stereo music
16 bit resolution (supposedly sufficient to detect all amplitude variations MAN can hear)
44.1KHz sampling rate (supposedly gathers all frequencies MAN can hear)
Table of Contents must be in order (tells the player how many tracks are on the CD and how long the tracks are, etc)
CDs must be able to hold 74 minutes minimum (the length of one of the committee member's favorite symphonies)

So the Red Book was accepted and they started making CD players.

Then commercial compact discs started showing up. Everyone thought the Analog Days were over, because CDs would always sound better, right? But, they forgot about the human part of the equation. So, many of the old analog sources sounded bad on CD. This wasn't because there is anything wrong with CDs, it was because the record companies, in a rush to get out products, put any old album master on CD, without regard to what generation it was and what condition it was in. At the same time, engineers found all kinds of new digital tools at their disposal and thought "I can take this old classic and add EQ, noise reduction, digital echo, this is going to be so fun!"

You probably think I have lost track and not answered your question, but it all has a point. While all this compact disc craziness was going on, no one bothered to think about what to do to old mono recordings. So, some guys did all digital editing in a one track format, then made an identical track so they could meet the red book standard, and the result is two identical tracks on the CD, and true mono coming out of your speakers. Bravo! Other guys when faced with this situation and having a "two track mono" tape source, simply digitized both output channels of the twin track tape machine, and the result is a historically accurate twin track mono presentation, not necessary but not bad either. Kudos to these guys.

Then there were the OTHER guys. They would apply digital stereo echo or reverb to a mono recording. Well, this produces different information in the left and right channels, meaning any tool you used to try to distinguish between stereo and mono would mistakenly call this stereo, when it's really tampered mono. BAD BAD BAD. Some of Elvis' 50's tracks had this done to them on official releases. Shame on you BMG!

I am not aware of a tool that is used specifically to determine if a given track is stereo or mono. The best way is to listen, spend some time on the Master & Session website, and if you are unsure about a particular track, ask around.

The techniques I mentioned in my last post are pretty effective. If you run this process and the result is just noise or little echo artifacts, then you have some sort of mono recording that may have stereo reverb on it or it might have been a twin track mono source originally. If you have nothing after running the operation, then it was completely mono copied on to two tracks for CD release. Anything else, it is a stereo recording*

*we've had disagreements on this board about the definition of stereo. In this context, when I say stereo, I mean that the two tracks have different blends of musical instruments on them. This definition works for what Ernst is advertising on the upcoming box set. These recordings weren't made specifically for stereo release, so the mix won't be conventional stereo and shouldn't be considered as such, but these will be authentic twin track recordings with different blends of instruments and voices in each track. These are real, not tampered or 'fake' stereo. But not stereo in the modern sense either. But a joy to listen to nonetheless.

So to finally answer your question:

"if all cd tracks are 2 channel 16bit 44.100khz then on the new box set Close Up coming out in July, these are touted as being stereo because they are 2 track...but wouldn't they be 2 track anyway when ON THE CD
following on with that info that all cd tracks are 2 channel 16 bit 44.100khz ?"

The answer is.....yes, anything you hear on a CD is two track, but that doesn't make it stereo. This is only because of the Redbook Standard. If the information on both tracks is identical, they are mono recordings. You could have asked the same question in the late 1950s, when it was common practice to send the same mono mix to the twin tracks of a tape machine to generate what is commonly called known as Twin Track Mono. Still mono. A stereo* recording must have different blends of the instruments and voices on each track.

Does that help?

And one last, unrelated thing. These days, everyone is talking about the new 24 bit/96KHz Japanese paper sleeve releases. These don't conform to the redbook standard, so why do they work? Well, they actually do conform to redbook. What this means is that digital editing was done at 24/96, but the final step of the mastering process is to downsample the music to 16/44.1 redbook. So the actual data on these CDs is still 16bit/44.1KHZ like any other CD, and if you use your little tester on a Japanese Paper Sleeve CD, it will tell you it is stereo, 16 bit, 44.1KHZ.


Sat May 17, 2003 8:13 pm


Techinically speaking isn't that why the Love Me Tender is a kind of stereo. The Voice reference tape(?) was added to the mix right? It seems to me that Heart And Soul DSD remaster seems to sound superior to the DVDAUDIO version of Love Me Tender.

Incidently your last paragraph ties in with one thing I have argued about extensively. I believe that the original transfer retains some of its benefit even when downsampled(DSD or 96/24). You being a little more knowledgable then I(or at least better at explaining lol), can you tell me in techinical terms how this is possible. I have my own ideas and have expressed them, but I really would like to know your take. Or do you think there is no benefit and all is lost?

I enjoyed this last explanation and agree one hundred percent. I also like the bit where you said "though pleasing to the ears" This is true. Sometimes I don't mind a little tinkering. I don't think this is always bad.

More Stereo Thoughts....UGH!

Sat May 17, 2003 10:02 pm


I actually have not studied up on exactly how the 'stereo' version of Love Me Tender was created. For me, I would call it true stereo, because it has two discrete tracks with different mixes of the band/voices on each track. There was no tampering involved at making this, it was recorded this way in '56.

We had a guy on the old board, I think it was Kiwi Alan (if I am wrong, I apoligize) that brought up an interesting point, that this shouldn't be called stereo because it wasn't mixed or intended for stereo release, and that it doesn't have the conventional left-right-center type of image of a conventional stereo release.

So, stereo means a lot of different things, so to avoid trouble, I'll just say that back in '56, the songs Elvis made for the Love Me Tender movie were recorded to at least two discrete tracks with different mixes on each track, and that means that today we can listen to an authentic twin track of these old songs, and it is darn cool.

Regarding your question about downsampling, I have not tried this myself. If you used the same source, same encoder, and did everything the same on two identical projects, except one was endoded to 24/96 and downsampled to 16/44.1, and the other was encoded at 16/44.1 in the first place, you should be able to view the spectral plot in cooledit or whatever your tool of choice is and see which one represents the high frequencies better.

I think the reason that the 24/96 recordings sound so good is because in addition to mastering at the superior bit rate, people are taking the time to find better masters, use more conservative noise reduction and EQ, and A/D converters have come a long way since the early CD releases.

So, I didn't answer your question because I just don't know. Some rainy day when I am bored, I will break out an old analog tape and encode it at both rates with all other things the same, then downsample the 24/96 one and perform a spectral analysis to see which is better. My gut instinct is that they will be much the same.

I'll bet someone has done this and will have a better answer than mine.



Sun May 18, 2003 1:17 pm

Greg, thank for all that info, I did know about the Red Book Standard....but I'd never heard it explained so well before/especially when it comes to stereo/mono, it's like you know it digital but your still thinking like vinyl....that' again for the great explanation, it's cleared my mind up completely. I just hope alot more people than just Genesim & I read this as I think it would have enlightened many. Thanks.

Mon May 19, 2003 5:31 am

OK greg answer me this. The cd is 16 bit 44.1 khz when downsampled. Now does that mean the samples are at the same point no matter which transfer it is done with?

In other words if one has a better transfer and on the first transfer one sample is at the up part of the wavelength and one point is at the peak of the wavelength and another is at the lower part of the wavelength.

Then lets say on the lower quality transfer it has the same EXCEPT the peak of the wavelength(I know these numbers don't match up, but my point is that the lower quality tranfer would have less points..actually up to what 60 times less with a DSD transfer?).

Now using this example the Better tranfer is a better representation because the lower quality missed the peak.

My rationale is that IF a cd at 16 bit 44.1khz took samples at different points on each master then the Better tranfer would benefit. It it is the same then it wouldn't matter. Does this happen. Here is a illistration

This is a sound wave tt tttt t
t t t t t
t t t t t
t t t t t
t t t t
ttt tttt

PCM (Low Q) samples ] ] ] ] ] ]
DSD ] ] ] ] ] ] ] ] ] ] ]

cd sample same ] ] ] ] ] ]
cd sample hypothetical ] ] ] ] ] ]

In other words, you lose some samples, but what you gain is actually more accurate when it is rounded out as opposed to the first cd sample.

Mon May 19, 2003 5:32 am

You see, this sucks!!! I cannot prove my point. oh well, maybe you will understand what I mean greg.

Mon May 19, 2003 6:00 pm

Actually anybody know this answer? Is the sampling of a cd arbitrary?

Samples and stuff

Mon May 19, 2003 7:06 pm


CD sampling is not arbitrary, there is a measurement taken every 1/44100 seconds. What is arbitrary is where on a particular recording it will be sampled if you make multiple attempts at the recording. So If I did a digital transfer of a song, then went back and did it again, because I probably couldn't start the playback machine and the digital transfer machine at exactly the same time, of course the samples on the second transfer wouldn't happen in exactly the same place in the music as those in the first transfer.

As frightening as this sounds, it is OK for this reason: There is a thing called the Nhyquist Theorem. Nhyquist says that to reconstruct a wave with an acceptable accuracy, you must sample at twice the highest frequency you are wanting to reconstruct. Back in the days leading up to the Red Book Standard, they noted:

1. The average Human's ears have a frequency response of 20Hz-20KHz
2. If we add 2.05KHz for a good margin of error to the highest range humans can hear, then double it to satisfy the Nhyquist theorem, then we can sample at 44.1KHz and properly encode the music in terms of frequency response.
3. 16 bits has enough discrete values (65536) to represent all of the amplitude variations the average human can hear.

So, in theory at least, no matter how your samples are lined up, you will always get a good encoding of your music when you do it at 16/44.1, at least all those frequencies average humans can hear. Maybe your dog will know the difference due to his excellent perception of high frequencies compared to yours!

However, audiophiles have never accepted this. They don't believe in nhyquist. They don't believe that sampling a wave at twice its frequency, as Mr. Nhyquist said, can possible allow you to rebuild that wave accurately. Makes sense. And audiophiles don't think we should filter everything above 22.05KHZ, so it was bound to happen that someday people would encode at much higher rates, but Red Book necessitates downsampling back to 16/44.1. The audiophiles say further that records are so much better because it is a continuous sample, which is true, but the audiophiles never mention how the vinyl record they hold in their hand is at best a 7th generation copy of the master, how it has been limited, phase adjusted and filtered, and how it gets more and more damaged, as well as the needle on the playback equipment, each time you use it. Also never mentioned is the fact that there is very little musical information that happens above the range of human hearing. A quick look at the specs of most of the famous microphones will reveal that they don't capture high frequencies anyway. Lastly, I am not sure how the downsampling tools in the common audio editing software works. Downsampling could accurately be called a lossy compression algorithm like an MP3 or JPG converter, so not knowing how this software decided which information to discard in order to meet Red Book, I'd rather just encode at Red Book standard and leave it at that.

None of this should indicate I believe totally in Red Book or Nhyquist. I am simply providing some of the theories I have heard from proponents and opponents of digital audio. I have had audiophiles say that removing all the frequencies we can't hear (above 20KHz) degrades the music, but spectral analysis of records doesn't show anything in that range anyway, so I do not know what to think. I do know that if you find the 0 generation master tape, encode at Red Book Standard at the highest volume possible without overdriving the encoder, and don't tamper with it by using EQ, compression, or reverb, you will end up with a CD that, to most human ears, sounds every bit as good as the original master tape.

Did this help any?


Mon May 19, 2003 7:20 pm

I want to get this completely straight. So I don't understand. If you have a digital transfer PCM or DSD, then the cd could sample at different points. What I mean is this. yes it is every 1/44.1 seconds, but wouldn't that start at the same time? How could it change? Say you took 2 different transfers of the same material. How could the cd master sample at different points. In other words, isn't it just numbers? It seems like you know what I am talking about, I just want the why?


Mon May 19, 2003 8:01 pm


Think about it like this. The machine doing the digital encoding takes a snapshot of the amplitude of the wave every so often, in this case it is every 1/44100 seconds. anything could happen in between each sample, so if something happened between two samples and didn't get registered during one encoding attempt, if you made another attempt at encoding, and because people and machines aren't perfect, the sampling routine could actually be lined up to capture this event the next time you encoded that particular piece of music. However, there are at least 3 things to consider here:

1. If you believe Nhyquist, anything that happens between samples like that would be representative of one of those frequencies beyond human hearing so it doesn't matter.
2. Any sonic event that humans could actually hear would have components spread over at least two samples, so it couldn't possibly happen between samples and be lost of ot was something important.
3. In practice, quality A-D converters will take a look at what happens between samples and pass the sampling component the average of what went on since the last sample, resulting in more accurate reconstruction.

Does this help any?


Mon May 19, 2003 8:17 pm


Thanks for all this explanation. and for you patience.


Mon May 19, 2003 8:34 pm

<Thanks for all this explanation. and for you patience.

I am trying to make up for how I got Genesim upset the other day :D

Mon May 19, 2003 8:37 pm

NO Greg, you are missing my point. Would it CONSISTENTLY be off. In other words, could it be mastered that way. I am talking about the cd only here.

In other words. Lets say that tranfer #1 has cd sampling starting at point A, B, etc. etc etc in integrals of 16 bit/44.1 (this is of course correspoding to the other 15 bits wich would cover more area....but I am expaining for simplicity)

Lets say we have the same latitiude for tranfer #2 starting at A 1/100th, B 1/100th , C and 1/100th etc. Again same lattitiude 16bit/44.1

Is this possible or actually likely? My question is when a cd samples is it a set number or can it actually deiviate consistently(like never changing again). Does it have to be A, B, C, or can it actually change from cd to cd?

Mon May 19, 2003 8:41 pm

I don't mean this in a rude way. But please, don't think what I mean! Just answer the question. I really really want to know. I am figuring the answer is yes, but I want the reasoning behind it!

Mon May 19, 2003 8:51 pm

I hate the fact, that this is difficult to describe.

It is my understanding the actual capture of a the music on the cd is a series of samples(16bit/44.1khz) of the transfer!!

What I want to know is this. The actual template for all cd's produced, how is that sampled.

IS IT POSSIBLE, for each master tape that is used for the template(or whatever you call it), IS IT SAMPLED IN THE SAME PLACE IN SUCCESSION.

DO they all go sample AAAAAAAAAAAAAAAA

or does it sometimes meander for that and get like A.1 A.1 up to 16 and B.1 B.1 up to 16

Is it exact for every round of cd pressings or do cd templates(there are all the same) actually sample different depending on the transfer used.

WHen I say different, I dont' mean a change in sampling rate 16/46.1 BUT where it samples and set up the integrals.

Mon May 19, 2003 9:15 pm

Greg you still around? I swear, I just need this one answer, then I am done. It is tough to find it on the internet.


Mon May 19, 2003 9:26 pm


I think I am having a hard time understanding what it is that you want to know. It's cool, we'll figure it out somehow. I didn't follow your discussion about latitudes and a,b,c.

All I am really saying, is that when you are encoding something at 16/44.1, then every 1/44100 seconds, the converter takes a look at the voltage level coming in and assigns a 16 bit number to to signify the voltage level at that point. That 16 bit number can be either negative or positive depending on the voltage level. Then, this same operation is repeated again exactly every 1/44100 seconds. So on a CD, if you could view the actual digital binary data for one second of music, you would see 44,100 entries, and each entry would be some 16 bit binary number representing the voltage level the converter detected at that time. For example, you might see something like this for the first 5 samples of a song:






This might be the biginning of a song. The first sample has all zeroes, which would indicate no voltage or silence. The next sample has a 1, which would indicate a rising voltage or a sound getting louder. The next 3 are voltages going higher still, or an increasingly louder sound.

Now, let's say you encode the same piece of music again, but you start your sampling at a slightly different time. You may get something that looks like this:






Notice everything starts the same, with all zeroes. The second one is 1 just like the preceeding example. Cool, right? Then, we get another 1. How can this be? Well, remember earlier I said good converters take an average? Well, in this case, because the encoder didn't start at the same time, the averages gave us different data. But these are only 5 samples, around a 10,000th of a second, so the fact that the data looks a little different isn't a big deal. In the end, you'll never know the difference, one digital transfer should sound the same as another as long as the setup is the same.

When you play back the resulting CD, the D-A converter will grab a packet of data corresponding to the first sample, figure out what the voltage level should be as indicated by that 16 bit number, then tell the analog output section to generate that voltage level. Then it will perform that operation over and over again every 1/44100 seconds. This output voltage is amplified and sent to your speakers and it is music....

If you had an analog tape machine that could be started in exactly the same place every time, or perhaps had an electronic tone on the tape that could be used as a cue to tell a digital encoder to start encoding, I guess you could arrive at identical digital information if you encoded the same piece of music numerous times. This is neither likely or necessary. As I tried to say before, the subtle differences that one would detect when analyzing several different encodings of the same source would only be visible in the most extreme close up view of the waveform and would not impact the sound at all, at least not to average human ears.

CDs that are pressed in a factory should all be 100% identical to one another since they aren't recorded, they are stamped from a master disc just like records. This should also be true for CDRs burned on home PCs, provided good software is used. Some of the cheap/free CD programs out there have better error detection than others.....

Sorry I don't totally understand your question. I am really trying.....


Mon May 19, 2003 9:28 pm


I was busy typing a response when I messed your other replies. If that doesn't clear it up let me know.


Mon May 19, 2003 9:43 pm

Thats what I am talking about!!! You have proved what I knew all along.

I agree the difference is MINIMAL, and probably not detetectable to the human ear....BUT it is there.

Ok lets get to it.

Ok you agree the more samples that are taken, the closer it is to the original sound wave right?

So now we have DSD that is capable of 100khz 64 times as much as PCM STANDARD that was used on the original master.

NOW we get down to the controversy. The downsampling.

Ok if a sample was taken at the exact point on each digital master(glass master?), then the results would be identical. BUT as you say, this is not always the case, and actually not likely.

Now going back to the original arguement. MOre samples equals more accracy for EVERY point RELATIVE to the other.

The best case scenario is this: Simplifying matters quite a bit.

Lets say you have a sound wave ON the original master. There is a sharp peak.

Ok lets say PCM has 2 samples One at the point right before the peak and then the second at the point right after peak.

Now lets say you have a high quality PCM or DSD and they in effect double the samples(actually it is more, but I am simplyifiying here!). Ok lets say you get a sample right at the bottom of the peak, then one at the same point as the first sample on the first example, then you get a sample at the TOP of the peak, then you get a sample the same as the second sample of the first example. This makes 4 samples for high quality PCM or DSD.

Now lets say that the CD comes along and takes 2 samples.
Since samples can start at different places depending upon the process.

Lets say (this is equal distant 16 bit/44.1 khz. No difference) that both samples are taken from the first example. But on the second transfer it takes the first sample and then the TOP of the peak. Even when rounding some of the characteristics of the soundwave is lost(the peak) and thus is less accurate. Now stretch that out across the whole cd and my theory is that some difference could occur. Barring one has supeman hearing, and the theories about spacial separation apply!!


Mon May 19, 2003 10:06 pm

Bump for KIWIALAN, OHNONO, VINYJUNKIE, VINYLMAN, and all the other non believers.

Heart And Soul COULD sound better because of DSD technology. Like my Calculus teacher said "it is mathematically possible!"

You may not agree gred, but I do hope you at least see my point.

Mon May 19, 2003 11:22 pm

<You may not agree gred, but I do hope you at least see my point.

I see your point. However, if someone were to determine that Heart & Soul sounded better, how do you know it was because of mastering at 24/96? Maybe it's because they used different masters they found at Iron Mountain or someplace. Maybe it's because they have better A-D converters this time around. Maybe it's for any number of different reasons. With the exact same source, exact equipment, and all variables the same except sampling rate, I don't think there is going to be much improvement in a source encoded at 16/44.1 redbook versus 24/96 downsampled to redbook. Heard and Soul isn't the best way to tell because lord knows what else they did to the remaster. It does sound fine, though.

You mentioned before not to ignore the outliers. That's always good advice. Along those lines, if one source is better sounding than another, never assume it is for the obvious reason. Take the US version of the MSG show for example. It is dull and lifeless, and even harsh. The Japanese Paper sleeve version is smooth, pleasant and lively. So it must be the 24 bit mastering, right? Well, not necessarily. It turns out that is from a different source than the US CD, and mastered much better. 24/96 probably hasn't as much to do with it as the care taken to do it the right way does. This is kind of like those news stories you hear about from time to time. I heard one the other day that said tea drinkers live longer, so people should drink more caffiene. I question whether it is the caffeine in the tea that helps you live longer, perhaps it is something else in the tea. If they are right, then coffee drinkers should live longer too. I don't know if they do, but if you want to study caffeine, don't use tea, use pure caffeine. And if you want to find out whether 24/96 downsampled is better than redbook, all you can do to know for sure is to do an experiment maintaining all variables except sampling rate, and do a spectral analysis on the finished product to see which is really better. Listening to the old versus new RCA CDs won't help because you don't know what else they have done.......



Tue May 20, 2003 1:13 am

genesim wrote:Bump for KIWIALAN, OHNONO, VINYJUNKIE, VINYLMAN, and all the other non believers.

Heart And Soul COULD sound better because of DSD technology. Like my Calculus teacher said "it is mathematically possible!"

You may not agree gred, but I do hope you at least see my point.

We see your point but all logical reasoning leaves us with a different answer. The limitation of the original tapes are very important. And the fundamental understanding of the theory behind digital engineering is a very important factor.

I think Greg has tried as good as possible to bring this to a level that can be read without too much technical background. But it also leaves the problem that it can be too simplified.

I have followed this discussion for a long time. To me it seems that the discussion is actually about, which frequencies we can hear. As in principal we can recreate a square wave from a row of sinusoidal waveforms. But to do this from a CD-DA is impossible. And that is the problem for many who do not know the theory behind the Nhyquist theorem and fourier analyzis. Any waveforms can be represented by a row of sinusoidal waveforms. But a CD-DA representation will not have all the sinuses needed (limited frequency response) to recreate the original waveform. So when gensim tries to calculate his samples and compare he is actually entering a level above 22.05kHz. But below this frequency it's no problem to reproduce the waveform using CD-DA as a standard. Even mastered from a DSD tape the CD-DA format has the same limitations. All extra frequencies that have been in the DSD master will be lost during the transfer to CD-DA.

How come we still is discussing this?: When genesim finds that a waveform cannot be sampled correctly, this is what we can call a transient. And that is correct, genesim, but I am not sure you know that it is such extremes that you describe. Not from your writing I have never seen that you refer to such. Such transients are not too relevant in this dicussion, as those was lost when Elvis was recorded to a tape. The Elvis master tapes has bigger limitations than the CD-DA standard. And that is fact as long as we speak of tapes dating from 1954-1977. If we can agree on that I think we are getting a bit further in this discussion.

Tue May 20, 2003 2:51 am

ok, Greg I agree with you totally. I like the old saying "you see hoofprints think horses NOT zebras". I love the studies where they say that smoking has gone up...Alzheimers has gone duh Smoking lessens your chance of getting Alzheimers.

As you know from are last debate that I don't believe in absolutes. Once again, I might not be as full of crap as you first thought. I go by logical reasoning. YES YES, I agree that it could very well be mastering. Time, care all that are factors. The limitations of the human ear cannot be ignored. All these things I have never denied. Again, it comes down to not possible to maybe possible.

Now certain other people on this board ignore the evidencel. I admit that I am not a sound engineer, BUT I can use logic and reasoning to form a theory. I took the sound curve the same as doing a Integral problem. The more you partition a function the closer you get to the actual wave form. Further more, you break up that same representation further but at a different place you have more to gain.

Vinylman I don't think you read my post, cause you are still spouting the same stuff. THERE IS A GAIN FROM A SUPERIOR TRANFER EVEN WHEN DOWNSAMPLED. I may not be the best writer, but I illustrated the point 3 times!!!! Even Greg agrees with me.

The question is not WHY it sounds better, it is the question of COULD it sound better because of DSD mastering. It is all in the numbers. Just reread my post and reason it out. Then comeback and disprove what I say. Specifics that is and I will be happy to explain it further!

Now you go on to say that the equipment could not reproduce the sound well check out this website then get back to me on that.

It comes down to the equipment. Elvis had some of the best...and I got news for you, actually some microphones are better then todays!!! Analog equipment hasn't changed that much. That is not to say there isn't some better equipment now, but seriously. If telephone wire can produce high frequency's BEYOND what the human ear can hear...then does what you say make any sense? How fat were those master tapes? Do you really believe that it is impossible? When was the last time you actually had a master in your hand? We are talking about up to 24 tracks here. Do you have any idea how much information is there?

Lastly, when you say a CD can't produce a frequency that is misleading. It can produce any frequency it is programmed to do. The problem is the standard!!! It is down to space. You could fill up a whole Fuuckin cd with one sound!!!! No offense, but I don't think you understand. The key here is the accuracy of the sample. Wherever the cd samples, it is going to be closer to the original sound curve then a lower quality transfer

Tue May 20, 2003 3:01 am

If you look closely...I mean study it Vinylman. I did not change the Frequency. Any sound peak can be hit, as long as it is in the 96 db range. This has NOTHING to do with the frequency of the sample. When sampled at a different is again has the same SAMPLING RATE...but taking a more accurate part of the curve. In other words, the cd can hit any part of the sound curve, just not all of them at the same time as on a 100khz transfer. Get it! What it does sample is going to be more accurate!!! There is a limit and it comes from the medium and how it is programed.

Re: More Stereo Thoughts....UGH!

Tue May 20, 2003 3:04 am

Greg wrote:
I think the reason that the 24/96 recordings sound so good is because in addition to mastering at the superior bit rate, people are taking the time to find better masters, use more conservative noise reduction and EQ, and A/D converters have come a long way since the early CD releases.


This is avery significant point - those mastering at higher rates will by their very nature take more care and time throughout the entire process.

Whether the extra time - thus cost - is viable financially is another matter.