Topic: Transfer from Analog to Digital

[hotpin#3]

I was wondering what is the best method of doing this. What equipment would best this done ? I would prefer to go CD's utilizing 44.1 redbook or  24 bit 96hz.

I know all this has to be transfered in real time, until it's in the digital format. Anyone out there making a 100 min. CD blank ?

thank you for your input on this matter.

[adrianf74]

I was wondering what is the best method of doing this. What equipment would best this done ? I would prefer to go CD's utilizing 44.1 redbook or  24 bit 96hz.

I know all this has to be transfered in real time, until it's in the digital format. Anyone out there making a 100 min. CD blank ?

thank you for your input on this matter.

I'm guessing here that you're transferring old analog masters (or vinyl) in hopes of preserving them in some sort of digital format. 

I don't know the nature of your source material.  If it's analog tape, I've never really been able to wrap my head around why anybody would use 96KHz/24-Bit for it as, sonically, there's not enough data on that source for such a data-rate.  Others would say go for the largest rate you can safely use because you're archiving.

In your case, you're saying you want to put everything on CD.  If this is the case, then you might want to consider using 44.1KHz/24-bit and then downsample to 16-bit for CD and archive the transferred .WAV (or better yet, in FLAC) to a DVD (or better yet, an External Hard Drive). 

90-99 min blank CD's exist but only a couple of manufacturers make them.  I've tried them a few times to find mixed results - and from an integrity standpoint, they don't adhere to redbook or any variant that I know of, and they can tend to flake on you.  For archiving purposes, I'd use TY (Japan) if you can find them and stick with 80 minute discs at the maximum.

[Church-Audio]

I was wondering what is the best method of doing this. What equipment would best this done ? I would prefer to go CD's utilizing 44.1 redbook or  24 bit 96hz.

I know all this has to be transfered in real time, until it's in the digital format. Anyone out there making a 100 min. CD blank ?

thank you for your input on this matter.

I'm guessing here that you're transferring old analog masters (or vinyl) in hopes of preserving them in some sort of digital format. 

I don't know the nature of your source material.  If it's analog tape, I've never really been able to wrap my head around why anybody would use 96KHz/24-Bit for it as, sonically, there's not enough data on that source for such a data-rate.  Others would say go for the largest rate you can safely use because you're archiving.

In your case, you're saying you want to put everything on CD.  If this is the case, then you might want to consider using 44.1KHz/24-bit and then downsample to 16-bit for CD and archive the transferred .WAV (or better yet, in FLAC) to a DVD (or better yet, an External Hard Drive). 

90-99 min blank CD's exist but only a couple of manufacturers make them.  I've tried them a few times to find mixed results - and from an integrity standpoint, they don't adhere to redbook or any variant that I know of, and they can tend to flake on you.  For archiving purposes, I'd use TY (Japan) if you can find them and stick with 80 minute discs at the maximum.

I say when your converting analog to digital you go with the largest bit rate you can and you make sure your signal to noise ratio is good. In the end the more bits you have and the higher the sampling rate the more truncated your waveform becomes and the more accurate it is. Now that depends on the converter but if it was me I would crank the converter to 11 BUT this also depends on what your going to be playing it back on if your never going to be playing it back at 192 then go 48k

[sunjan]

I was wondering what is the best method of doing this. What equipment would best this done ? I would prefer to go CD's utilizing 44.1 redbook or  24 bit 96hz.

I know all this has to be transfered in real time, until it's in the digital format. Anyone out there making a 100 min. CD blank ?

if you ever see yourself doing any postprocessing or cleanup, I'm with Church. Go for the highest possible bitrate, which gives you room for restoration later on.

And archive the untouched WAVs on DVD, whether you want to use that medium for playing the music or not. You can easily resample and create playable 16/44.1 CDs, but you will at least have archival backups for future generations.
If you have backup DVDs, feel free to experiment with 90min CDs, but I'd stay away from them.
Search the forum here for a discussion on manufacturer ID on blank media, and why the maker of a DVD isn't the same as the brand.

What are your analog sources? Cassettes, vinyl, reel to reel?

Also, which sound card and type of optical jack are you intending to use? I leart the hard way that some sound cards (read Soundblaster!) with optical-in actually resample the signal internally, and therefore aren't suitable for this kind of work.

And now we'll just wait for dsatz's post where he explains why 12 or 13 bit is sufficient, and anything above that is a total waste ;-)

[adrianf74]

And now we'll just wait for dsatz's post where he explains why 12 or 13 bit is sufficient, and anything above that is a total waste ;-)

ROTFLMAO.

Well stated, I think all three of us are looking at this from a similar vantage point.  In a perfect world, always record at the largest data-rate you can afford to.  However, depending on the source (r-to-r or tape), there may be little benefit to going beyond 44.1/24. 

[boojum]

I am with DSatz when he says 12 or 13 bit is sufficient as the dynamic range of analog is not great enough to require anything more than that.  You can xfer it at 24/192 but that is a lot of wasted space.  Suit yourself.  You may even find it "sounds better."  I am not convinced there is any reason for it to "sound better" however. 

Bottom line: do what pleases yourself.     

[adrianf74]

I am with DSatz when he says 12 or 13 bit is sufficient as the dynamic range of analog is not great enough to require anything more than that.  You can xfer it at 24/192 but that is a lot of wasted space.  Suit yourself.  You may even find it "sounds better."  I am not convinced there is any reason for it to "sound better" however. 

Bottom line: do what pleases yourself.     

Knew this was going to be said; that's why I said 44.1/24-bit (esp. if the person wants to downsample to 16-bit for CD) is a good way to go.  It's still overkill without getting stupid.

[Church-Audio]

I am with DSatz when he says 12 or 13 bit is sufficient as the dynamic range of analog is not great enough to require anything more than that.  You can xfer it at 24/192 but that is a lot of wasted space.  Suit yourself.  You may even find it "sounds better."  I am not convinced there is any reason for it to "sound better" however. 

Bottom line: do what pleases yourself.     

Knew this was going to be said; that's why I said 44.1/24-bit (esp. if the person wants to downsample to 16-bit for CD) is a good way to go.  It's still overkill without getting stupid.

Yes but in todays world of cheap storage why not go as big as you can with the files? I could see if blank cd's cost $10 each like they once did I remember the day when I had my 2x cdr recorder from sony it was only $2k what a deal... I say you store it in a high sample rate format then down sample if you need to. With the cost of storage being so cheap why not?

[adrianf74]

With the cost of storage being so cheap why not?

I remember the days of $15 blanks, so, yes, I know where you're coming from. My first 2x PIONEER SCSI recorder was $900 (!!) but I think there is a point where you can safely say it's overkill and a waste of space.

Just because you can doesn't mean you always should...

[Church-Audio]

With the cost of storage being so cheap why not?

I remember the days of $15 blanks, so, yes, I know where you're coming from. My first 2x PIONEER SCSI recorder was $900 (!!) but I think there is a point where you can safely say it's overkill and a waste of space.

Just because you can doesn't mean you always should...

This is the way I think of digital conversion.. When you look at a simple sine wave.. The higher your sampling rate is the more your dissecting that sine wave. So the more dissections of the sine wave or your source. The more accurate the recreation of that waveform will be. On a complex waveform such as music one could technically argue that 12 bit or 44.1 is all you need well I say if that was the case why does 192k sound so much better? I think the more you can dissect a waveform the better your chances are of it being accurate when you convert it back to analog. Even when your talking about reproducing an old tape you have to remember the waveform is much more complex then 44.1 can ever capture with out any of the parts of the waveform being left out. Converters do a good job of "guessing" the bits that are missing. But why leave it up to guesswork? when you can make sure you accurately capture at least as good as is possible the waveform your trying to recreate. Remember the sine wave its a simple waveform its duration and amplitude are very basic compared to music. Its very easy to predict a 1k tone. But when you look at music there are bits and pieces between the sampling rate that are just left to guesswork. That guesswork IMO is whats wrong with digital in the first place. So I say crank up the sampling rate. But if your converter sucks then going higher might not be a good thing. It all depends on hardware and noise floor.



Chris

[tilomagnet]

Bit depths and sample rates are overrated. 

Get yourself a good quality cassette deck that lets you adjust the playback azimuth and a decent ADC that doesn't add much noise. Then you're good to go. If you want to do post production (i.e. hiss reduction, EQ'ing etc.) I'd go 24 bit. Also masters or rare 1st gens are generally worth it to be transfered in 24 bit as well IMHO.

Otherwise 16/44 is fine.

[Roving Sign]

With the cost of storage being so cheap why not?

I remember the days of $15 blanks, so, yes, I know where you're coming from. My first 2x PIONEER SCSI recorder was $900 (!!) but I think there is a point where you can safely say it's overkill and a waste of space.

Just because you can doesn't mean you always should...

This is the way I think of digital conversion.. When you look at a simple sine wave.. The higher your sampling rate is the more your dissecting that sine wave. So the more dissections of the sine wave or your source. The more accurate the recreation of that waveform will be.

And thats just the point that DSatz seems to be repeatedly refuting...

And believe me - that's exacty the way I have thought of it...it's taking me a while to get past it - but his explanation makes a bit sense.

Good post here Chris...

http://taperssection.com/index.php/topic,103553.msg1395204.html#msg1395204

[DSatz]

Church-Audio, you write as if there were general agreement that higher sampling rates "sound better." I know a lot of people in professional audio, and what I find instead is that people who believe as you do about "dissection" (a very good word for the theory which you follow) expect a higher sampling rate to bring increased waveform fidelity, and this expectation influences what they perceive.

If you were to make fair, direct comparisons when you didn't know in advance what sampling rate you were listening to, I think you would find that it is nearly impossible to hear differences except with specially concocted test signals--and then (for most people) only with some ear training in advance. Those test signals have to be oriented to the specific characteristics of the particular filters used in the A/Ds and D/As; music and speech might hit those specific bit patterns once in a gazillion years, or they might not.

I could tell you quite a few stories about situations in which very prominent classical producers and engineers--guys who think they can tell when silver vs. copper wire is used in the light bulbs over their heads--didn't realize that they were monitoring through absolutely conventional 16-bit 44.1 kHz A/D and D/A converters. Under real-world recording conditions when you don't set yourself up to hear what you expect to hear, the better ones can be essentially audibly transparent on program material nearly all the time.

Finally, the expectations to which I refer are based on a very basic misunderstanding about digital audio, since the sampling and reconstruction process in fact does not dissect audio waveforms in the way that you seem to mean. All the in-between moments of the varying analog signal are encompassed--not just the discrete moments at which samples are recorded. That can't be seen in the visual model which most people have of the process, which shows the effects of sampling, but not the effects of the analog signal's subsequent reconstruction. However, it can be shown both mathematically and empirically (with analog oscilloscope traces and through unbiased "blind" listening experiments) that it is so.

If it were not so, then the signal-to-noise ratio of the CD medium would be worse than that of an analog cassette. Think about it: If the system really worked the way you imagine, there would be no signal energy during playback in the moments between the samples, and the power in the "pulses" or "spikes" would need to be distributed among the much greater expanses of time in which there was no signal energy, like an "area under the curve" problem. That would greatly diminish the dynamic range of the system; if the pulses were (say) 1% as wide as the intervals between them, 40 dB of dynamic range would be lost and so on.

You could say that the extent to which a PCM recording system can approach its theoretical maximum signal-to-noise ratio is the extent to which its recorded samples actually cover the entire sampling interval which they each describe. Since the actual limit is usually set by the associated analog electronics, that should tell you something pretty important about your theory.

--best regards

[Church-Audio]

Church-Audio, you write as if there were general agreement that higher sampling rates "sound better"--what evidence do you really have of such a general agreement? I know a lot of people in professional audio, and what I find instead is that people who believe as you do about "dissection" (a very good word for the theory which you follow) expect a higher sampling rate to bring increased waveform fidelity, and this expectation influences what they perceive.

If you were to make fair, direct comparisons when you didn't know in advance what sampling rate you were listening to, I think you would find that it is nearly impossible to hear differences except with specially concocted test signals--and then (for most people) only with some ear training in advance. Those test signals have to be oriented to the specific characteristics of the particular filters used in the A/Ds and D/As; music and speech might hit those specific bit patterns once in a gazillion years, or they might not.

I could tell you quite a few stories about situations in which very prominent classical producers and engineers--guys who think they can tell when silver vs. copper wire is used in the light bulbs over their heads--didn't realize that they were monitoring through absolutely conventional 16-bit 44.1 kHz A/D and D/A converters. Under real-world recording conditions when you don't set yourself up to hear what you expect to hear, the better ones can be essentially audibly transparent on program material nearly all the time.

Finally, the expectations to which I refer are based on a very basic misunderstanding about digital audio, since the sampling and reconstruction process in fact does not dissect audio waveforms in the way that you seem to mean. All the in-between moments of the varying analog signal are encompassed--not just the discrete moments at which samples are recorded. That can't be seen in the visual model which most people have of the process, which shows the effects of sampling, but not the effects of the analog signal's subsequent reconstruction. However, it can be shown both mathematically and empirically (with analog oscilloscope traces and through unbiased "blind" listening experiments) that it is so.

If it were not so, then the signal-to-noise ratio of the CD medium would be worse than that of an analog cassette. Think about it: If the system really worked the way you imagine, there would be no signal energy during playback in the moments between the samples, and the power in the "pulses" or "spikes" would need to be distributed among the much greater expanses of time in which there was no signal energy, like an "area under the curve" problem. That would greatly diminish the dynamic range of the system; if the pulses were (say) 1% as wide as the intervals between them, 40 dB of dynamic range would be lost and so on.

You could say that the extent to which a PCM recording system can approach its theoretical maximum signal-to-noise ratio is the extent to which its recorded samples actually cover the entire sampling interval which they each describe. Since the actual limit is usually set by the associated analog electronics, that should tell you something pretty important about your theory.

--best regards

I think it depends on your source. There are some sources where you could never hear a difference. But to my ears I can tell the difference in a source I am familiar with between 44.1 and 192k big time. Should we record everything in 192k? No but my point is if your archiving why not record in the best sample rate possible if the method of storage is cheap. You can always down sample later on. There are sometimes when going to 192 shows defects in the recording that 44.1 does not show and sometimes these defects make the music sound better. As I am sure a professional such as your self knows there is no perfect solution for every situation. Only best "guesses" I dont think 192 is for everything.. but if you dont have the years of experience that both you and I possess sometimes its better to just dump everything at 192 and then subjectively try different sample rates when time allows at a later date.



For me its about data the more data you have the better your recreation of a waveform will be. Its pretty simple. But again there are some sources that actually get hurt by increasing the "focus" if you will of the converter by increasing the bit rate and sample rate. For me there is no absolutes in audio there are only things that work for some and things that work for others.

Anyone that says this is how it is, imo is usually wrong. I have heard 44.1 sound amazing I have heard 192 sound amazing I think when we consider the source we then can find the best sample rate for the given material.

I was only generalizing because unless your going to "audition" different sample rates and find the best one that works for your source then it might be better to sample high now and downsample later on. I am certainly not a digital expert. But I know what I like to hear and my ears are still in very good shape.

I appreciate your experience here and your input into all subjects. Audio is subjective sample rates for me are also something that can be subject to opinion. I think that's why we have more then one sample rate to pick from. I also feel as I stated before that high sample rates on shitty converters = shitty sound. But a high sample rate on a great converter with a very good slew rate is something to listen to IMO.

BTW my theory of how a-d converters work is correct. Here is a picture that best illustrates the sampling and chopping up of a complex waveform into voltage values and then into 0's and 1's. It would make more sense that the more conversion points of voltage that you have them more likely you are to capture the real source but again I did also state that high sample rates are not for every source. The other factor that is rarely talked about is how fast a converter can do this math and if there are any delays in processing what effect does that have on the bit stream. There are many facets to the digital world sample rate and bit rate are only a few.

Chris

[adrianf74]

Chris, I think the point that's trying to be stressed here is that going beyond 44.1/24 for an analog-based source makes NO difference what-so-ever as there is no sonic benefit from a science standard.

It may "sound" better at 192, but is it?  Scientifically, it isn't.  If the data isn't there to begin with you're just wasting space.

The initial person mentioned they wanted to burn stuff to CD, that's the other reason I stressed 44.1/24.  24 would have to be downsampled but at least it's in a better than CD format.  Nothing is gained by the larger file size; except for wasted space.

[Church-Audio]

Chris, I think the point that's trying to be stressed here is that going beyond 44.1/24 for an analog-based source makes NO difference what-so-ever as there is no sonic benefit from a science standard.

It may "sound" better at 192, but is it?  Scientifically, it isn't.  If the data isn't there to begin with you're just wasting space.

The initial person mentioned they wanted to burn stuff to CD, that's the other reason I stressed 44.1/24.  24 would have to be downsampled but at least it's in a better than CD format.  Nothing is gained by the larger file size; except for wasted space.

Well not everything in sound can be scientifically explained away. I have heard the difference for my own ears. I did say that it does depend on the source. But really I think that for me I have nothing to prove I know I hear a huge difference that's all that matters at the end of the day for me. Now if you give me a record with a high noise floor would I notice the difference probably not. But with some sources YES I can hear the difference and not because I am expecting it to sound better because it does at least to my ears.

[Church-Audio]

I was wondering what is the best method of doing this. What equipment would best this done ? I would prefer to go CD's utilizing 44.1 redbook or  24 bit 96hz.

I know all this has to be transfered in real time, until it's in the digital format. Anyone out there making a 100 min. CD blank ?

thank you for your input on this matter.

I'm guessing here that you're transferring old analog masters (or vinyl) in hopes of preserving them in some sort of digital format. 

I've never really been able to wrap my head around why anybody would use 96KHz/24-Bit for it as, sonically, there's not enough data on that source for such a data-rate. 

What do you mean by data? Just wondering.

Chris

[DSatz]

Chris, when I was in high school I took a few years of German. At one point my mother decided that she wanted to try to help me with my homework, so she picked up my textbook and started trying to read a few sentences. Not knowing any German, she came across the word "die," which is simply one form of the definite article (like our word "the," to which it is closely related). But to her, this was irrevocably the verb form of "death," and she became convinced that this had been part of the problem with Germany in the Hitler era--that they threw this terrible word around so casually, they became desensitized to it.

In German, of course, the word "die" has nothing to do with death. But my point is, you can come up with some pretty wild conclusions, as my mother did, and feel that you have clear evidence for them, if you interpret something in the wrong framework of meaning.

You are doing a similar thing by interpreting sampled audio as it was still a continuous-time, analog signal. According to the behavior of continuous-time signals, everything that you say makes sense. However, a discrete-time representation of a signal is different, and must be interpreted differently. I'd like to suggest that you read a little about Shannon's sampling theorem, because I think that otherwise we will only be able to talk in circles. As I recall, Wikipedia has a pretty good article on it.

(added the next morning): I also wanted to say that some recording equipment can indeed sound different when run at different sampling rates--but when that occurs, it isn't necessarily because of those sampling rates alone. As an example, at the end of the 1980s the Sony PCM-2500 was by far the leading studio DAT recorder; it could record at 32 kHz, 44.1 kHz and 48 kHz. For many engineers it was the first piece of digital recording equipment that allowed them to compare the three sampling rates--or so they thought.

There was general consensus that there were audible differences, and that the 48 kHz rate sounded better than the 44.1 kHz rate. However, not only the sampling rates but also three (or six, for stereo) complete, separate analog anti-aliasing filters were also being compared, since the ICs which were later introduced for digital filtering didn't exist yet. And it turns out that if you bypassed the A/D and D/A converters in the deck--effectively isolating those filters and simply listening through just them and the other analog circuitry of the deck--they sounded rather different. The 44.1 kHz filter in particular could have audible levels of a kind of hard-edged distortion when it was pushed. (Apogee, the company which is well known now for its A/D converters, got its first foothold in the studio market by offering better-sounding replacement plug-in filter modules for the PCM-2500 and other similar equipment.)

Many people formed their own, honest opinions about sampling frequencies, like I think that yours are--but those opinions weren't based on what those people thought they were based on, like I think that yours probably aren't. Similar stories could be told about "tube vs. transistor" comparisons (e.g. Neumann U 67 vs. U 87) and "CD vs. vinyl" comparisons where extrinsic factors that people didn't know about had a large, hidden influence on their decisions. It's not always easy to set up listening comparisons so that you really are testing just for the one thing that you want to test for. Unfortunately for consumers and even most studio engineers who can't get into the insides of the equipment, it is sometimes quite impossible for them to do so at all.

--best regards

[Church-Audio]

Chris, when I was in high school I took a few years of German. At one point my mother decided that she wanted to try to help me with my homework, so she picked up my textbook and started trying to read a few sentences. Not knowing any German, she came across the word "die," which is simply the definite article (like our word "the," to which it is closely related) for any noun in the feminine or plural. But to her, this was irrevocably the verb form of "death," and she became convinced that this had been part of the problem with Germany in the Hitler era--that they threw this terrible word around so casually, they became desensitized to it.

In German, of course, the word "die" has nothing to do with death. But my point is, you can come up with some pretty wild conclusions, as my mother did, and feel that you have clear evidence for them, if you interpret something in the wrong framework of meaning.

You are doing a similar thing by interpreting sampled audio as it was still a continuous-time, analog signal. According to the behavior of continuous-time signals, everything that you say makes sense. However, a discrete-time representation of a signal is different, and must be interpreted differently. I'd like to suggest that you read a little about Shannon's sampling theorem, because I think that otherwise we will only be able to talk in circles. As I recall, Wikipedia has a pretty good article on it.

(added the next morning): I also wanted to say that some recording equipment can indeed sound different when run at different sampling rates--but when that occurs, it isn't necessarily because of those sampling rates alone. As an example, at the end of the 1980s the Sony PCM-2500 was by far the leading studio DAT recorder; it could record at 32 kHz, 44.1 kHz and 48 kHz. For many engineers it was the first piece of digital recording equipment that allowed them to compare the three sampling rates--or so they thought.

There was general consensus that there were audible differences, and that the 48 kHz rate sounded better than the 44.1 kHz rate. However, not only the sampling rates but also three (or six, for stereo) complete, separate analog anti-aliasing filters were also being compared, since the ICs which were later introduced for digital filtering didn't exist yet. And it turns out that if you bypassed the A/D and D/A converters in the deck--effectively isolating those filters and simply listening through just them and the other analog circuitry of the deck--they sounded rather different. The 44.1 kHz filter in particular could have audible levels of a kind of hard-edged distortion when it was pushed. (Apogee, the company which is well known now for its A/D converters, got its first foothold in the studio market by offering better-sounding replacement plug-in filter modules for the PCM-2500 and other similar equipment.)

Many people formed their own, honest opinions about sampling frequencies, like I think that yours are--but those opinions weren't based on what those people thought they were based on, like I think that yours probably aren't. Similar stories could be told about "tube vs. transistor" comparisons (e.g. Neumann U 67 vs. U 87) and "CD vs. vinyl" comparisons where extrinsic factors that people didn't know about had a large, hidden influence on their decisions. It's not always easy to set up listening comparisons so that you really are testing just for the one thing that you want to test for. Unfortunately for consumers and even most studio engineers who can't get into the insides of the equipment, it is sometimes quite impossible for them to do so at all.

--best regards

I understand perfectly about sampling rates. And I also understand perfectly that when I select a higher sampling rate when I record or when I am using a digital console it simply sounds better to my ears. That's all I will say. Increased quantization  and sampling of a waveform = increased data = a more accurate representation of the waveform. I have been working with digital since the first cd player came out. I am sure you know your theory. But I know what sounds good to my ears when I mix I trust my ears first the specs/theory second.

[Roving Sign]

Honestly Chris - you sound a bit chagrined by Mr Satz sharing of his knowledge - like somehow you've been knocked off your perch.

We respect your areas of expertise. But you are starting to sound a bit self-rightgeous, if not foolish on this one.

My understanding of digital audio is...well, WAS the same as yours - but Mr Satz has highlighted a few assumptions that we seem to make.

Im not going to pretend like "I got it" as far as Satz' explanation - but his take is eye opening, and thought provoking...

Sometimes things are not what they seem...

[Church-Audio]

Honestly Chris - you sound a bit chagrined by Mr Satz sharing of his knowledge - like somehow you've been knocked off your perch.

We respect your areas of expertise. But you are starting to sound a bit self-rightgeous, if not foolish on this one.

My understanding of digital audio is...well, WAS the same as yours - but Mr Satz has highlighted a few assumptions that we seem to make.

Im not going to pretend like "I got it" as far as Satz' explanation - but his take is eye opening, and thought provoking...

Sometimes things are not what they seem...

Not at all. Really I totally respect his knowledge. But I also respect my ears. I guess we will agree to disagree. I have no issues what so ever with being knocked off my "perch" for me its not all about who knows more its about what my ears tell me. Because at the end of the day that's all that matters to me. I totaly have nothing but respect for DSATZ! He is a very smart man. But we will simply have to agree to disagree. That's not me saying I think he is 100% wrong that's me saying I trust my ears. I am not so full of my self that I dont think there are others around here that know a thing or two more about audio then I do, because that's just silly. I really am just responding to his comments that were directed towards me. I respect his point of view but I dont share it. That does not mean I cant respect the guy who has a different point of view. I dont agree with the points of view of some professors that I know and have known. Does that make them wrong? No does it make me wrong? Maybe not maybe yes. Its all cool with me I really dont care one way or the other. I was just stating my opinion. And he was stating his. I felt in a very respectful manor. I guess that's the beauty of text it can be taken many different ways.

But rest assured I did not mean for my replies to be taken as anything but JUST MY POINT OF VIEW. I have learned alot of things here big time and continue to learn from you guys. Don't ever think that I am above being proven wrong. I am not that's how I learn. But for me in this case my ears tell me a better sampling rate sounds better so that's why I am so firm on this point at least from my perspective and the perspective of all of the engineers I know.

Chris

[DSatz]

Chris, I'm not even trying to change what you think, let alone knock you off of any perch. (Or haddock.) Your beliefs are your business, and whether you change your mind or not probably won't have much effect on my life. I'm just having a good time here, and am hoping that the same is true for you.

It's odd about science--there are conclusions that people have reached after long, hard study and research and experimentation, and meanwhile everybody else who didn't make those experiments can only judge those conclusions on whatever their general plausibility seems to be. Often, people have very little choice but to take scientific claims on faith, or else not do so. What may be perceived as "obvious" in a given time and place is extremely subject to change--sometimes drastic and rapid change. (I'll spare you the usual list of examples.)

Ultimately it comes down to the predictive ability of a theory. If you want to know whether a theory is valid, try making a squirm-proof prediction based on it. Put it all on the line. What I said earlier about THD+N is an example--I strongly agree that you have no reason to believe what I said (that adding more bits can help reduce THD+N, but that increasing the sampling rate does not help at all for frequencies below 1/2 the sampling rate) until you've actually made this experiment. Go ahead; try it yourself and see what you get. Then try to explain the result in terms of what you are so sure is the way things work ("Increased quantization  and sampling of a waveform = increased data = a more accurate representation of the waveform").

As I said early on, when you say "increased data" in this way (meaning data rate, i.e. total bits per second), you blur the distinction between sampling precision and sampling frequency. To use a single combined figure for "data rate" or "bit rate" makes some sense in the MP3 world, where the data reduction itself is the main limiting factor in sound quality. But in linear (not data-reduced or "compressed") PCM as used in CDs and DATs, etc., those two dimensions mean two different things and follow different rules. You can't sample at half the rate but use twice as many bits per sample and still get the same quality, nor vice versa. I think most people understand that very well.

Rather, the sampling needs to be precise enough to capture the full dynamic range of the signal without adding significant noise or distortion, and the sampling frequency (sampling rate) has to be more than twice the highest frequency in the signal to be recorded. Once it's high enough, though, it can't be made "higher enough" or "high enougher"--it's still just "high enough."

--best regards

[boyacrobat]

ears always the first language to the brain
theory has no sound, just thought .

both have reason to exist

g

[indietaperwloo]

I actually convert analog sources to digital for a living and my advice is this:

Worry about your playback source, analog stage and the quality of your ADCs more than what bit depth and sample rates you're going to use.  You can convert your old vinyl collection to DSD using a $20,000 delta-sigma ADC with analog filters for all I care but if you use a junk playback stage, the archive is going to sound like crap.

Also, consider the S/N ratio of your analog gear.  The typical noise floor of a vinyl record (12 or 7 inch at either 33 or 45 rpm) even with a decent signal chain between the turntable and the ADC it bottoms out at around anywhere from -45 to -50 dBFS at silence points with the ADCs on my system bottom out with no signal coming in at all at around -83 dBFS (faders with the connected Mackie 1402 VLZ Pro at inf and all channels muted and all master section monitoring options turned off).  So considering that, if you're transferring a vinyl record at 16/44.1 LPCM to a .wav or .aiff file, half your bits are taken up by surface noise from the vinyl and probably noise from the analog stage thus greatly reducing headroom.  Therefore, recording at 24 bit with those kinds of numbers doesn't really make much sense since you're increasing the word length with noise (from -144 to -45) and if you're sending it to CD you'd just have to trunciate the word length down to 16 anyways and add possibly more noise with a dithering process (I personally use the Apogee UV22 plugin included with Wavelab if I have to resort to such measures).

What is the point of this whole numbers argument?  Well, to me, a CD whose source is a vinyl record sounds no more different than a 24/96 wave file from the same source.  The only difference to me is that it's just a bigger file.  The numbers I think justify what my ears hear.  What I DO notice is what's happening in the analog stage.  If I don't do any kind of noise reduction or restoration of any kind, I notice the characteristics of the signal chain between the source media and the ADC (in the case of vinyl...turntable (cartridge/tonearm) > phono preamp > mixer > ADC).  These are things you really have to take into account when you're archiving audio and you want to do a decent job.

[tilomagnet]

I actually convert analog sources to digital for a living and my advice is this:

Worry about your playback source, analog stage and the quality of your ADCs more than what bit depth and sample rates you're going to use. 

QFT.

And from these (PB source, preamp & ADC) the playback device is by far the most important part of the signal chain.

When using different decks for playback I do usually notice the different sound characteristics of each, however I've never been able to tell the difference between different ADCs, let alone bit depths and sample rates. And I've tried 16/44, 24/96 and DSD.

Also, when working w/ Dolby encoded tapes, the quality of Dolby decoding makes 10x the difference of an ADC upgrade.

[Church-Audio]

Chris, I'm not even trying to change what you think, let alone knock you off of any perch. (Or haddock.) Your beliefs are your business, and whether you change your mind or not probably won't have much effect on my life. I'm just having a good time here, and am hoping that the same is true for you.

It's odd about science--there are conclusions that people have reached after long, hard study and research and experimentation, and meanwhile everybody else who didn't make those experiments can only judge those conclusions on whatever their general plausibility seems to be. Often, people have very little choice but to take scientific claims on faith, or else not do so. What may be perceived as "obvious" in a given time and place is extremely subject to change--sometimes drastic and rapid change. (I'll spare you the usual list of examples.)

Ultimately it comes down to the predictive ability of a theory. If you want to know whether a theory is valid, try making a squirm-proof prediction based on it. Put it all on the line. What I said earlier about THD+N is an example--I strongly agree that you have no reason to believe what I said (that adding more bits can help reduce THD+N, but that increasing the sampling rate does not help at all for frequencies below 1/2 the sampling rate) until you've actually made this experiment. Go ahead; try it yourself and see what you get. Then try to explain the result in terms of what you are so sure is the way things work ("Increased quantization  and sampling of a waveform = increased data = a more accurate representation of the waveform").

As I said early on, when you say "increased data" in this way (meaning data rate, i.e. total bits per second), you blur the distinction between sampling precision and sampling frequency. To use a single combined figure for "data rate" or "bit rate" makes some sense in the MP3 world, where the data reduction itself is the main limiting factor in sound quality. But in linear (not data-reduced or "compressed") PCM as used in CDs and DATs, etc., those two dimensions mean two different things and follow different rules. You can't sample at half the rate but use twice as many bits per sample and still get the same quality, nor vice versa. I think most people understand that very well.

Rather, the sampling needs to be precise enough to capture the full dynamic range of the signal without adding significant noise or distortion, and the sampling frequency (sampling rate) has to be more than twice the highest frequency in the signal to be recorded. Once it's high enough, though, it can't be made "higher enough" or "high enougher"--it's still just "high enough."

--best regards

My perch lol... I hear the difference plain and simple. I know what I can hear and what I cant. I dont want to argue with you on this point. With certain sources I can hear the difference between lower sampling rates and higher sampling rates. I have nothing but respect for your knowledge, but that does not mean you are above being me disagreeing with you. You can have respect and still not agree on every point. My main point is when I use say a digital console and I set the converters to 44.1 and then I set them to 96k there is a world of difference for my ears. When I am using a real source like kick drum or anything else. So to my ears there is a difference between sampling rates and sound quality. That is the basis of my argument.


No perch here trust me.

Chris