Does digital audio work like digital images...ie more bits for highend?

Hi Barry,

I was raising a new point which occured to me after thinking about your analogy.

Warmly,
Kevin

 

And each will swear they are right.

 

This is the only answer. It's the real thing played in your room upon command. listen to a recording of it here. http://members.aol.com/tgcnc/weary.htm I'll stick with my CD's

 

I actually visited a collection of mechanical orchestras at House on A Rock in Spring Green, Wisconsin. They have more artifacts than MP3 at its lowest bitrate.

I'm fine with CD as a format, too, Robert.

Warmly,
Kevin

 

That one would probably be at home in a Tim Burton movie.

 

If you want to have some fun, you can find a track to compare all kinds of sample rates between 44.1 and 192k in 16 and 24 bits. The sample track is recorded from an LP ( late fifties recording with a quite high-end record playing chain), to a Korg MR-1000 using 128x DSD then converted to other PCM bitrates using Korg's AudioGate software and the original DSD recording. You can download the original DSD file as well.

http://www.penna-media.hu/hires_jazz_lp_sampler/

 

Hi Barry,

I am approaching the exercise from a much more simplistic point of view. I have a high quality analogue input source and I am simply determing if making a digital transcription of it changes the sound perceptibly to my ears in any way. If it does, then that makes it inaccurate to a greater or lesser extent.

Digtal theory states that the copy should be audibly indistinguishable from the original. But it isn't - at any resolution available for my testing. As soon as I go PCM at any resolution the sound hardens, it brightens and instruments - especially violins above all else - lose the subtleties of their timbres. It should not matter whether I have a live mic feed, am recording from an open reel recorder or phono amp. It is still a comparison of an analogue source to a digitalisation of that analogue source. If the digital processing changes the sound then it is no longer transparent and that means 24-192 is not good enough. Sure, analogue changes the sound too, but it does not "damage" it in a way that that digital does.

A lot of people will argue based on blind tests, for example, that inserting even a 16-44 ADC and DAC process is completely inaudible when listening to a vinyl source. My own tests indicate that not even a 24-192 ADC-DAC processing chain is inaudible, let alone lower resolutions.

I'm not trying to argue that analogue is more accurate than high res digital. It is pretty obvious to me that both recording technologies colour the original sound and analogue colours it much more technically speaking. I'm simply saying based on my experience that I never in my life heard a digital recording of a violin that sounds accurate. I have, however, heard analogue recordings of violins which sound accurate.

From my point of view, it never has really been a matter of what I prefer or not, so much as seeking a method of digitalisation which is completely transparent to a high quality analogue source, regardless of origin.

As I intimated earlier, when you get to such high fidelity reproduction as top end analogue versus 24-192 PCM in any case, there are far more important things in the recording chain than the technology which will determine which approach I prefer. there are terrible analogue recordings and terrible digital recordings.

I would much rather have someone who choses and hangs up the mics and who really knmows what acoustic instruments sound like, and who is "hampered" by a "mere" 24-48 recorder, than I would some engineer who can't tell the sound of a Strad from the sound of a cheap factory imitation.

To my way of thinking, sound engineers are far bigger a "problem" than technology ever was. Take the high res 24-96 Linn downloads for example. Hardly any of them sound remotely like a real orchestra, but the downloads are popular because people think they sound good. But unless they have a lifetime of experience with the real thing and have extremely critical hearing, I don't know how they would really know. Then take the low resolution live recordings of the Phildelphia Orchestra. I can tell you without hesitation which ones sound far more accurate (even though neither of them are), and it has nothing at all to do with the recording technology itself. It's all about the engineers, the mics and the way they use them.

I would love to hear the original masters of those Phildelphia downloads. I'm not sure what res they are, but I think they are 24-96. Whatever the case, I would wager these are possibly the most accurate orchestral digital recordings in history if I take into account everything that I know is lost when one makes a low res copy.

 

The original recording has a lot to do with it. And I might argue that microphone placement and quality is a significant part of it.

 

Hi JonP,

What I'm saying is that one A-D converter does not represent all A-D converters and is not a good basic from which to draw any conclusions (at least to me it isn't). Also, your Benchmark is not the only thing involved in your digital transcriptions. What are you using as a recording device and what sort of connection are you using between the device and the converter?

Are you using USB? (Hopefully not as USB, while fine for MP3s, is inadequate for high quality digital audio.)

Actually, I think this is more a statement from digital marketing. ;-}
(and audio folks on the internet who bought it)

Best regards,
Barry
www.soundkeeperrecordings.com
www.barrydiamentaudio.com

 

I agree with Barry that A/D converters vary wildly. And a quality microphone cable is important.

 

Yes, I agree they vary wildly, but so what? Are we now saying in order for me to enjoy everything 24-192 has to offer then I have to have a system far more expensive than my existing analogue system with which I have no complaints? What is the point then? What is the point of me "needing" something comparable to Barry's clandestine world-beating hardware to cut the mustard when I already have perfectly good analogue source for only $1000 and which makes music actually sound like music? That's like Jeremy Clarkson saying a Honda is a hopeless car, but then it would be given what he actually drives himself.

The benefits of high resolution digital to a consumer becomes a pointless argument in a practical sense if we are all going to have to raid underground laboratories to get stuff to make it sound good.

 

No you can get good sound for less. I think we are saying that with quality mics you get very close to that natural violin sound you seek.

 

Yes - quality mics and quality engineering and high resolution ADC. Unfortunately those three things come together extremely rarely.

 

Hi JonP,

The point is, if they vary wildly and one has a somewhat limited experience with only one or two, broad conclusions about all don't make sense as there is no logical basis to support such a perspective.

I wonder how much of your experience with this might be the result of what the Benchmark is connected to or if USB part of your auditions. And I wonder about the monitoring involved, since what you hear can only be as honest as your monitoring allows. (For example, a rough edged monitoring system will "prefer" softer than neutral source material. I'm not saying that is the case here but in the absence of better information, it is a reasonable possibility.)

Or it could just be that digital is not your sonic cup of tea. That is another possibility, based on your posts and as I said earlier, I can fully understand.

Best regards,
Barry
www.soundkeeperrecordings.com
www.barrydiamentaudio.com

 

The reality is that people that do ** have another point after they have done it. Falloutboy only say it like IT IS and if anyone don't believe this, well, show this in a scientific way and change the facts that are "wrong".

 

What does "**" mean?

 

Hi Barry,

There is no USB involvement in my recording / listening chain. The benchmark connects via coax to the SPDIF input of an ASUS Xonar card at 24-192 and the Xonar with the ASIO bitperfect drivers together with the hard drive acts as a "transport".

In any case I have finally had a chance to hear what a miked recording using 24-192 sounds like. Acousence records are now selling silver disks containing 24-192 FLAC files. They are live orchestral recordings. I can now say that I have never heard reproduced sound this accurate. It is a step up (signficant) from anything analogue I have ever heard and the violins sound pretty darned accurate to me. Whilst I still can't close my eyes and be convinced I am at a live performance it probably gets around 95% the way there. If I were sitting next to someone in the same type of chair I sit in at a real concert and a bit of nearby perfume wafted by I could well suspend disbelief.

I hope this company continues to put more recordings out. They absolutely blow away anything I have ever heard before. But I don't think it is purely the 24-192 doing it. The 24-96 is at least as good as the best analogue I have ever heard as well, so I think this s a case of an engineer who has great equipment but is one of the extremely small number of people who knows how to maximise it's potential.

http://www.acousence.de/

 

Hi JonP,

I'm glad you have heard that 24/192 can indeed deliver quite excellent results (e.g. where the violins actually sound like violins, etc.). After having worked at 24/96 for many years, the solo classical guitar recording I did the other day at 24/192 has raised the bar for me on how much of the feeling of being in the presence of the performance can be captured and brought home.

What I've found is that only a minority of recordings achieve the full potential of the format on which they are delivered, whether it is vinyl LP, analog tape, CD or some form of high res. And one must be fortunate to find these recordings in order to discover the usually hidden potential in each of the formats.

Best regards,
Barry
www.soundkeeperrecordings.com
www.barrydiamentaudio.com

 

Something we cannot talk about here