[gregorio]

All continuous time analogue storage systems have major issues depending on how far you push them. Digital ones also have significant issues when used to represent and reconstruct continuous time analogue signals.

And here we are again, back to exactly where we started when I responded to you in post #58: "You're vastly confusing scale. A football and the sun can both be described as enormously big compared to an atom but of course, just because they are both "enormously big" doesn't mean that a football and the sun are roughly the same size!" In regard to an original analogue signal, digital audio does have issues at the very low, well beyond audibility, level. Analogue storage systems also have issues but very much higher, in some cases a magnitude higher but in all cases higher and well into the audible range. The equivalency between the issues of analogue and digital you are asserting is fallacious! A football and the sun are different sizes. How is it possible that you have a degree in applied physics/electronics/maths and not have any idea about relative scale?

pinnahertz I'm rather aware this could go on forever and we've both presented our present ideas and opinions.

You've presented your ideas and opinions, based on assumption, subjective observation and inapplicable facts, whereas Pinnahertz has presented actual applicable facts. A very different reality to the picture you are painting of some sort of equivalency between what you're trying to prove and what Pinnahertz is stating. For example, you could have the opinion that 1+1=2, I could have the different opinion that 1+1=5. I could demonstrate that you are wrong because for example, 1 drop of water + 1 drop of water does not equal 2 drops of water, it equals 1 drop of water (albeit a larger drop). You have proven fact to back up your opinion and I've got nothing except a dubious observation, assumption and some inapplicable facts. While we would both have an equal right to our different opinions, that does not mean our opinions are in any way equal valid!

I've asked you repeatedly to present some actual facts, some reliable, applicable evidence that digital does not result in perfect fidelity within audibility and you've repeatedly refused to do so. You either just ignore the request or you misdirect by providing details of some digital "issue" which is NOT audible. I bet you didn't get your degree by employing that tactic!

G

 

[pinnahertz]

...

Well a manufacturer of very high end digital replay gear who I spoke to in July said they add short term phase noise deliberately to the clock after LF jitter reduction....these guys make master clock boxes as well.....

I spoke to a well known manufacturer of professional digital recording and replay gear who said "We now have jitter low enough that it's inaudible".

Touché. My frustration now matches yours.

 

[SimonPac]

You don't have a reasonable analog source unless you have a live mix through an analog desk. Otherwise your analog source is corrupt, filtered, masked, modified, distorted...use whatever term you prefer.

Agreed. Yep. never said anything different.

Again....you have not made a real controlled comparison, biases have overwhelmed your results. Have you even heard a real 16/44.1 master?

Did you read what i said carefully...

Again....you have not made a real controlled comparison, biases have overwhelmed your results. Have you even heard a real 16/44.1 master?

I believe in digital enough to believe the master and my CD won't be terribly different. Why would they be?

 

[pinnahertz]

Did you read what i said carefully...

Yes, hence my comment.

I believe in digital enough to believe the master and my CD won't be terribly different. Why would they be?

Then why say "master"? You must not be aware of the current production work flow. No matter, your other comments still reveal no actual valid comparison has been done.

 

[71 dB]

LCD4's blow my HD650s right out of the water on resolution.

One would expect a $3000 headphone blowing a $300 headphone out of the water just as a $300 headphone blows a $30 headphone out of the water. The difference is that the improvement from $30 to $300 is massive whereas between $300 and $3000 much less.

 

[gregorio]

Let's actually look at your 1+1=5 opinion:

[1] To me any 44.1 has unnaturally etched transients
[2] yet paradoxically a lack of real resolution.
[3] I have no doubt mastering engineers have learned to work the format over the years.
[4] Sharp handclaps and things like maracas sound coarse.
[5] Acoustic environment ques are diminished.
[6] The whole thing sounds a little sterile and bland after 88-24 or 192-24.
[7] Those in turn, from my limited experience, sounds similarly limited compared with a reasonable analogue source if it was derived from one.

1. No, transients are exactly the same within the range of human hearing at 44/16 or higher sample rates/bit depths.
2. No, resolution is the same.
3. No. The beauty of digital is that we don't have to work the format, the format is transparent/irrelevant. This was not the case with analogue.
4. No, they sound identical.
5. No they're not, they are rendered accurately. As has already been explained to you (but you're conveniently ignoring).
6 & 7. No, they sound identical.

My answers only take into account how digital audio actually works in practise and what is demonstrably (with reliable evidence) audible and inaudible. Obviously though, they do not take into account your personal hearing biases, incorrect assumptions and inapplicable facts but then again, neither do your answers!

I spoke to a well known manufacturer of professional digital recording and replay gear who said "We now have jitter low enough that it's inaudible".

Me too but I think he's probably dead now, it was a long time ago!

G

 

[pinnahertz]

One would expect a $3000 headphone blowing a $300 headphone out of the water just as a $300 headphone blows a $30 headphone out of the water. The difference is that the improvement from $30 to $300 is massive whereas between $300 and $3000 much less.

"One would expect..." is the key.
[Warning: /BiasMode ON] I have found incredibly poor price:quality tracking. Listened to all the exotic pricy models, the ones that cost the same as a car...etc., at Axpona a couple years ago. Largely disappointing. For me one of the winners was $500, beating the $3000 popular "gold-standard". I also have a pair of basically disposable $30 IEMs that beat some $500 IEMs, etc. And, of course, some cheap in-ear junk that everything "Beats".

/BiasMode Off

 

[SimonPac]

You guys don't seem to be reading me carefully so this is getting quite pointless and a bit repetitive. On the headphones, I was conceding that 44.1 contains information which only higher end phones will resolve.properly. So regular 44.1 is not that bad at all in very many ways. Doesn't stop even HD650 level phones benefiting from higher rate. If your amp is rubbish, you may not notice. But then I'm guessing you think they all sound the same. It's not that hard to design a bad sounding transistor amp.

I have a BSc Honours in Applied Physics, Durham UK, 1980's. I have a Postgrad Cert in Physics Education from Warwick University in the UK. I worked on the second generation PDH digital hierarchy in the 80's and early 90's. A personal friend helped define the international SDH digital hierarchy in the '90s. This was when we were using PCM for telephony and ISDN, but all that has changed is the digital medium and the resolution. And it's stereo or higher and it's linear where it was u-Law companded. A relative was until very recently a maths professor at Leeds and an ex Trinity Teaching Fellow with a speciality in Banach Space. I have discussed these things with him in the past. From what I see, you guys have very little experience in digital systems design beyond the bare essentials. I have spent enough time in the field to know roughly who knows what here on DSP and who I have something to learn from.

A lot of studio gear is much higher than 16 bit now of course, and higher sampling rate too. Yes, some say 24 for studio only, but at least some systems run faster, despite the storage overhead. Why do you think that is?

I have no problem with the fact that on a quick listen, the distortions of 44.1 16 masters are less obvious than say 15ips tape.

 

[SimonPac]

I have no problem with the fact that on a quick listen, the distortions of 44.1 16 masters are less obvious than say 15ips tape.

Not from direct experience, but I wouldn't be surprised.

 

[bigshot]

I thought it was audible, and I think I discerned which was which blind. My memory could be incorrect, and I'd be interested in repeating the experiment. Decent master tape, analogue recording chain, some acoustic instruments and/or relatively unprocessed vocals..

I've produced a lot of sound recordings and mixes over the years. The last step of any mix is to output the mix to the master and then compare the master to the original to make sure nothing was lost. The one album I recorded on 24 track magnetic tape was to be delivered on digital. (it was right at the beginning of digital when digital audio was great, but the recording and mixing tools were primitive.) As a last step before wrapping, I compared the mix to the master to make sure it was identical. It was. If I heard any difference at all, I would send the engineers back to figure out what was messed up. In all the mixes comparing 24 track tape, ADAT, 24/96 vs 16/44.1, etc... I have never encountered any digital copy or bounce down that was anything less than perfect.

I take the same approach to my home equipment. Whenever I get a new piece of gear, I rack it up next to my main reference stuff and compare it. I've never found any digital audio product that sounded different from any other. If I did, I would probably send it back as defective. I expect digital audio to be perfect and I double check to make sure it is. I recognize that ' bias, so I carefully set up controlled conditions where I would definitely be able to detect a difference if I one existed because I don't want to be stuck with out of spec equipment. If I believed that every format or piece of equipment sounds different, I'd probably try to convince myself I was hearing differences and my bias would make me believe I actually was hearing a difference. So I'd set up very controlled ways of comparing to avoid that. I'd also try to establish a goal coloration so I'd get the sound I really want, not some random roll of the dice. The funny thing is, I never see any person who believes everything has its own sound establishing a benchmark for comparison. If I was them, I wouldn't waste my time trying to describe differences with flowery poetry. I'd be trying to figure out a target response curve and desirable degrees of specific types of euphonious distortion. I'd come up with a test to check equipment against my benchmark so I wasn't churning through expensive equipment randomly.

There's a proper application of science and an improper application. Science used to help you define and achieve a goal is great. Science used to create theoretical problems that you can't even prove is audible isn't so great.

All digital filters are a compromise, and fail differently and probably not that differently from each other if reconstruction using 44.1 is your aim. Marketing claims are often or even generally exaggerated, yes. You don't have the pre-anti aliasing signal to listen to; that's the point.

When I compare 24/96 out of the mixing board to a 16/44.1 bounce down I am. And I usually have four or five very trained sets of ears in the room focused on making the comparison.

I've worked in analogue too as I said. The compromises there are several orders of magnitude greater than digital. Generation loss, tape hiss, noise floors, dropouts, distortion... If you want compromise, try recording on 24 track tape. I don't know of a single sound house here in LA that records to tape like that any more. Each house has a 24 track in the machine room to transfer existing masters to digital for editing. That's it. There's a reason for that.

I have a BSc Honours in Applied Physics, Durham UK, 1980's.

I have no doubt that you know more about digital theory than I do. That's a given. But I have experience in applying digital technology to the recording and mixing of sound intended for home audio. I know what is important for that purpose and what isn't important. I also have spent as much time studying the thresholds of human hearing as I have on digital theory. So I can have a very good idea what can be heard and can't be heard just by looking at the size of the artifact being described. The specs of the human ear are every bit as important as the specs of audio equipment. If you don't have any ballpark understanding of what we can and can't hear, it's really easy to go down audiophile rabbit holes, chasing inaudible bugaboos like jitter and errors in reconstructing microscopic pulses that don't even exist in the real world.

The ears are where the rubber meets the road. Human ears are the final stage of any system. Whatever you hear depends on them more than it does digital reconstruction filters. It all has to be perceptible to affect sound quality.

 

[SimonPac]

You must not be aware of the current production work flow.

Correct. Please fill my in. Why is the 2 * 44.1 -16 data stream, minus alignment/error correction etc, different to the master?

 

[bigshot]

Is it audibly different? How did we determine that?

 

[gregorio]

[1] You guys don't seem to be reading me carefully so this is getting quite pointless and a bit repetitive.
[2] On the headphones, I was conceding that 44.1 contains information which only higher end phones will resolve.properly.
[2a] So regular 44.1 is not that bad at all in very many ways.
[2b] Doesn't stop even HD650 level phones benefiting from higher rate.
[2c] It's not that hard to design a bad sounding transistor amp.
[3] I have a BSc Honours in Applied Physics, Durham UK, 1980's. I have a Postgrad Cert in Physics Education from Warwick University in the UK. ...
[4] From what I see, you guys have very little experience in digital systems design beyond the bare essentials.... I have spent enough time in the field to know roughly who knows what here on DSP and who I have something to learn from.
[4a] A lot of studio gear is much higher than 16 bit now of course, and higher sampling rate too. ... Why do you think that is?

1. Nice one, accuse us of exactly what you're guilty of. In all fairness, I think you probably are reading it, it's just that you're apparently incapable of understanding it?
2. No. Not even the best phones get near the resolution of 44/16.
2a. Correct, if by "not that bad" you mean audibly perfect.
2b. No, no headphones benefit from higher rates, they can't even get near to 44/16!
2c. And it's not expensive to buy an audibly perfect DAC chip, about $1.50 at bulk trade prices.
3. A pissing contest isn't going to get you far here, for one thing, it's a contest you'd loose!
4. And there's your problems! What you "see" is an extremely limited view of digital systems design, you apparently have little/no understanding or experience of how all the technical details in your head actually apply to digital audio practice and therefore, virtually no concept of scale, all of which results ridiculous assertions based on inapplicable facts! ...
4a. Case in point! I don't need to "think why that is", I know exactly why it is because I was there, working in some of the world's top studios when the change over happened and I discussed the matter in detail with some of the world's leading experts, top engineers and several of the most major manufacturers of pro audio equipment. It's obvious that you've jumped to incorrect conclusions as to "why that is".

I can understand and appreciate why you're so proud of your BSc Honours but from my perspective, that's just a starting point, it would even be just a starting point if it were in a subject more directly concerned with music recording technology. Instead of relying upon your theoretical knowledge, adding it to your subjective observations and coming out with erroneous assertions, why are you not prepared to look beyond your BSc in Physics/Maths? Admittedly, you probably won't learn anything more about physics/maths but you'll sure learn a huge amount about how digital recording and reproduction work in practise because it's painfully obvious you currently know very little.

We are starting to be more and more harsh with you because you appear stuck in this vicious circle of subjective opinion + inapplicable facts = ridiculous assertions. Would this not be a good opportunity to reboot the situation, to ask questions and try to understand why your assertions are so erroneous, rather than just continuing to argue, thinking your assertions must be correct because you have a BSc in physics? If you're not prepared to do this in a reasonably polite manner, we will be forced into some subjective conclusions of our own, such as; you're a troll or are some sort of shill, and that would be a shame because if your stated education/experience is actually true, some of your knowledge would be useful in this forum. Please, think about it!

G

 

[71 dB]

I believe a lot of the bias for "hi res" formats is the desire for it to be better. It's not that easy to accept that 16/44.1 is it, that there is nothing beyond it for our ears. I have gone through that acceptance process myself years ago. It is kind of loosing your faith. Well, not as hard I guess, but a process nevertheless. For audiophiles it's so important that there is always something better waiting for us, be it a new "better" headphone model or perhaps a better audio format, but at some point we reach practical perfection in technology. 16/44.1 has been perfected and we have reached the destination on that issue. It should be celebrated I think.

 

[bigshot]

I think the problem is that people tend to apply things that are valid concerns in analogue formats to digital. When I was first starting out to build a sound system, I would go shopping for an amp and see a wide range of different kinds of things to choose from. I remember buying a receiver where the manufacturer tested the specs of each unit and stuck a printout to the box. I dug through the pile of boxes to find one with distortion levels under 1%. It was even worse with turntables. There were compromises in a dozen different directions... stylus shape, tracking weight, belt vs direct drive, tonearm resonance, etc. You would be constantly tacking back and forth between options trying to find the perfect answer.

Today, CD players and amps have become standardized. Almost all of them are transparent regardless of price. That leaves audiophiles without anything to noodle. So they ape the ideas from the old days of "separates" sounding better that all in one units, and they try to finesse specs they can't even hear because back in the day, distortion and response variation certainly were audible. The funny thing is, they also resist things that actually would help them- like clean digital EQ- just because back in the 70s analogue equalizers were noisy and had tons of spill.

There's still plenty to finesse in the areas of transducers and room treatment and equalization, but I guess that isn't as fun as getting fancy black boxes that weigh a ton and get hot when you run them too long. It doesn't have to be that way though, because none of that has anything to do with audible sound.