Lavry Engineering Paper on Hi-Res

[Talk2Me]

We all agree (I hope) that everyone has a right to their own opinion and views. But, I find reading from the professionals very educational and enlightening. However, does not mean I agree 100%.

 

Sampling Theory For Digital Audio - Lavry Engineering

www.lavryengineering.com/documents/Sampling_Theory.pdf

File Format: PDF/Adobe Acrobat - Quick View

by D Lavry - 2004 - Cited by 4 - Related articles

Sampling Theory For Digital Audio. By Dan Lavry, Lavry Engineering, Inc. Credit: Dr. Nyquist discovered the sampling theorem, one of technology's fundamental ...

 

I posted this on another thread, and was mentioned this paper deserves it's own thread. And I am smart enough to know I have a lot to learn and experience.

 

[manisandher]

Lavry obviously knows a thing or two about digital audio - although I've never heard his AD122 in person, others claim it to be one of the best ADCs available. I can well believe this - it's a 'true' R2R ADC.

 

It's been a while since I read Lavry's paper(s), but I don't recall ever seeing 'pre- and post-echo' mentioned.

 

Why do engineers believe that the frequency domain is more important than the time domain? I mean, we're listening to music... aren't we?

 

Mani.

 

[davidR]

Dan Lavry has been a big name in the pro audio world for awhile and I respect his knowledge and what he's accomplished. He builds good converters too. The only problem I have is that he's treated like a "god" in that world and whatever he says goes. I believe PeterSt and Steve Nugent of Empirical Audio have both argued with Dan on pro audio forums about D/A conversion in general and absolutely everyone mocked both Peter and Steve although the posters clearly knew little about DAC design in general. Unfortunately for Dan there are many converters out now which I and many others think sound better than even his Gold series line. Sorry Dan. Discussing theory and what's on paper only goes so far. The proof is in the pudding.

 

[Jud]

...may want to have a long talk with Dan Lavry. His DAC (the DA11) will accept S/PDIF or AES input at sample rates up to 200kHz. (USB and optical are limited to 96k.) The DAC chip he uses employs the standard "8x oversampling" ( http://www.analog.com/static/imported-files/data_sheets/AD1955.pdf ) so in fact the sample rate at the point where the D/A conversion is done is 384 or 352.8kHz, depending on input sample rate.

 

The only difference the input sample rate *may* make in Dan Lavry's DAC is that a higher rate *may* avoid one or two 2x oversampling steps. (I'm not conversant enough with electronics to know from the DAC chip info sheet. Maybe someone else can tell us if interested.)

 

[Talk2Me]

Maybe as a business person he sells what will sell? I dunno. But, what I get out of the "general message" from his thesis is 196 and above might be out of range of human hearing. It seems to me, now a days, maybe too much emphasis on the highest hi-res numbers, rather than the quality. To me, I would rather have 44.1 done state-of-the-art, than 196 from mid-fi, IMHO.

 

[Paul R]

However, I note that a lot of people in the music/sound engineering business greatly prefer to record in 24/192K or in DSD, which has a similar high sample rate.

 

I don't think you would be able to fool people who make their living from sound recording into spending a lot of money for nothing. Just my opinion there.

 

-Paul

 

 

[Jud]

But, what I get out of the "general message" from his thesis is 196 and above might be out of range of human hearing.

 

44.1 is out of the range of human hearing, too. The frequency response issue is just one aspect of the overall sound quality question. Some designers and peer-reviewed papers think 192kHz makes a difference re frequency response, others don't. But most of the designers who talk about hi-res don't tend to mention frequency response as a major consideration. Rather, they talk about filtering.

 

Now Lavry says in his paper that once you're up to about 60kHz, you've got no more worries re filtering. But here's what Keith O. Johnson of Spectral says in an interview in The Absolute Sound ( http://www.spectralaudio.com/TAS_%20Interview.pdf ):

 

How has having access to the 176kHz/24-bit high resolution files of your own recordings affected your design work on playback equipment?

 

KJ It’s affected it very, very much. The high-res files set a gold standard against which you can judge CD playback. I should mention that CD can be very good, indeed. We think, “Oh my gosh, it’s 44.1kHz and 16 bits—how can that be even in the same league as something that’s got ten times more information?” It turns out if one works a CD very, very well, with the proper noise-shaping system that is accurate to about 1 part per million, then the remaining difference between high resolution and the 44.1 is time dispersion. The other factors really are not that important. And the time dispersion can be partially corrected by doing group-delay corrections.

 

Time dispersion, for the benefit of the readers, is a

smearing of transient information that occurs in digital

filters. High sampling rates relax the filter requirements

so there’s less of this spreading out of transient energy

over time.

 

KJ Exactly. That’s exactly what happens, and it’s very technical, but it’s highly audible.

 

* * *

 

So who to believe, Keith O. Johnson or Lavry? Well, besides the fact that KOJ designs the best audio electronics I've ever heard (Spectral Audio) and has won Grammys for audio engineering, Lavry's paper leaves me cold on a couple of counts:

 

1 - There's some "sleight-of-hand" going on. You're right, his general message appears to be about frequency response, but no one is really trying to argue Shannon-Nyquist. The more important issue is filtering, and on that Lavry seems more cagey than open.

 

2 - About filtering: Lavry talks about 60kHz being good enough, and later "proves" his point by illustrating a couple of filters (Butterworth and Chebyshev) with "no ringing." Except there is no such thing. The sharper a filter is, the more pre-ringing and post-ringing it has. ( http://en.wikipedia.org/wiki/Gibbs_phenomenon#Signal_processing_explanation ) This is a proven mathematical and physical reality, just like Shannon-Nyquist. Higher frequencies have the advantage of allowing gentler filtering, but since Lavry is set on saying 96kHz is good enough, he slides right past that point.

 

3 - Lavry's major objections to encoding at resolutions above 96kHz appear to be these: (a) unduly large files; (b) unduly high CPU resources; © unduly high equipment expense; (d) lower dynamic range at higher resolutions, which he presents as a "permanent" (immutable) law.

 

We can dispose of a, b and c pretty quickly. Download speeds and CPU capabilities have increased, and prices have decreased, to the point that 24/192 ADC capabilities are available in cheap consumer sound cards and downloads of high-res files take a few minutes. But wait - 24/192 consumer-level ADC? I thought Lavry said it was an immutable law that higher resolutions meant lower bit rates.

 

Look more closely, and once again Lavry appears to be playing fast and loose with the facts to try to make his point. He says at 100MHz sample rates only 8 bits of dynamic range are possible. So who exactly is holding out for a 100mHz sample rate? He then says 16 bits are possible at 1mHz. Again, who is asking for such a dynamic range/sample rate combination? (SACD/DSD is recorded at 2.8/5.6mHz with "1-bit" dynamic range. Do you notice the designers here saying this is inadequate for decent sound?)

 

Surely he will discuss the sampling frequencies of interest, 176.4kHz and 192kHz, right? Errrmm, no. He drops right down to 50-60Hz as his next example. So in a 27 page paper supposedly devoted to showing why 96kHz is better than 176.4 or 192kHz, he never actually points to anything objectionable about the latter two sampling rates. (Clearly 24 bit/192kHz parts are possible - as I noted above, one can find them in consumer sound cards for quite reasonable prices these days, to say nothing of professional recording equipment.)

 

So thanks, but as between KOJ and Dan Lavry, I'll take the guy who *doesn't* seem to be trying to obfuscate.

 

[Talk2Me]

Spectral, as good as it's reputation and quality is, leaves me cold, un-emotional, if you will. That is why I am trying to end each post with "IMHO". All I know, and all I feel, is what effects me emotionally when I listen to music. That is the only criteria that I know and understand. If I can, I would a t-shirt that says, "It is not the size of the bits, it is how those bits work their magic."

 

[Jud]

Spectral, as good as it's reputation and quality is, leaves me cold, un-emotional, if you will.

 

That's cool. There are many folks with the budget to choose between Spectral and other components who choose something else.

 

But KOJ has done a hell of a lot of research into digital recording and playback, and his discussion of filtering and transient response reflects what I've read elsewhere. Mani mentioned the lack of this in Lavry's paper, and I went into more detail about it, including quoting KOJ re why it is important.

 

[thesurfingalien]

Hi Paul,

 

AFAIK, recordings benefit from a higher sample-rate and bit-depth because of the editing that takes place afterwards.

 

I am nowhere near a recording engineer, but at times I do make recordings of my wife's performances, and sometimes for others. The advice I always got from more knowledgeable people was to record at best possible quality, do the editing (in my case very little/simple) and then down-sample for CD.

 

Peter

 

 

 

 

[bdiament]

Hi Peter,

 

**"...AFAIK, recordings benefit from a higher sample-rate and bit-depth because of the editing that takes place afterwards...."**

 

In my experience, there is always a benefit to longer word length when digital processing (not necessarily editing) is going to occur.

 

The reasons are numerous but among them is the fact that even the slightest process (say a gain adjustment of 1/4 dB) is going to lengthen the digital word. The word length of the processed file must exceed the word length of the target or low level information will be lost. This is one reason why mastering at 24-bits, even for a 16-bit CD target will result in a better CD. (Much of the software I use for mastering uses 48, 64 or 80-bit data paths.) With many processes, there is also benefit to high sample rates.

 

However, all that said, in my view, the main reason recordings benefit from higher sample rate and word length has to do not with any benefits for subsequent processing but simply with how they sound on playback, even when no processing is involved.

 

Different folks will hear different things of course and different gear will reproduce it differently. And different designers are going to have different ideas. As an engineer, what gets me is that with the better high res gear (I'm talking about 4x rates like 176.4 and 192k), for the very first time in my decades of recording, I'm getting back the sound of my mic feed. I never got this before from any analog or digital recorder, regardless of format or price. Even the best digital gear doesn't get close at 2x rates (88.2 and 96k), which I find closer to CD standard than they are to properly done 4x. But at 24/192 (my preferred rate), the sound coming back is the same as the sound going in. I never thought I'd hear it and at one time I would have bet that if I did, it wasn't going to be from a digital device.

 

To be clear, there is a lot of gear that sounds "very good". But if I'm seeking a non-editorialized capture of the input, "very good" is ultimately very bad. I want gear without a sound, rather than a "good" sound. (This is my personal preference. I understand some folks like certain colors and would not argue with what brings a person listening pleasure.)

 

My only regret in this regard is that so very much of the "192" spec'd gear out there seems to fall far short of the potential for the medium. As I've said elsewhere, I attribute this to clocking that is not up to the task and analog stages that aren't performing at these bandwidths. Unfortunately, as long as the opportunity for commerce exists, there will be folks planting the number 192 on devices that are already huffing and puffing at 96. (This doesn't say much for some of the devices that tout even higher numbers.)

 

Brings back the words of the poet, Paul Haines:

"Better a lot

of what's wrong,

than a little

of what's right."

 

As always, just my perspective.

 

Best regards,

Barry

www.soundkeeperrecordings.com

www.barrydiamentaudio.com

 

 

 

[Jud]

If you've already answered this elsewhere, I can go look for it: What would you say are the main ways in which 2x rates through good equipment fall short of your mic feed?

 

[esldude]

Regardless of any theoretical perfection questions, well done 96/24 would be one huge amount better than most recordings show us for any number of other reasons. For that matter, a great majority of the time one can say that about 44.1/16.

 

[Mark Powell]

One of his basic points, near the beginning, is that you don't get anywhere near a 24-bit word length due to inherent inaccuracies until you have a sample rate as low as 50-60 Hz.

 

But several people here are totally ignoring this and talking abou 24/192.

 

So do you think he is just plain wrong on this?

 

[Talk2Me]

No, I think we are wrong in thinking that the more bits, the more horsepower, the more resolution, the more "more", is the key to hi-fi nirvana. I am going on the record by saying it is not the greatest amount of resolution one can obtain, but rather the quality of what 44.1 represents. Do that right first, then you can move forward, no?

 

[thesurfingalien]

Hi Barry,

 

I am obviously just an amateur parroting what others told me, but you are, even more obvious, completely correct :-) I should have said processing...

 

If you don't mind I will do some "hijacking" to seek your advice on what I do because I need all the advice I can get :-)

 

Anyway, I use VERY simple gear: 2 Oktava MK-012 mics, an Edirol UA-25EX USB AudioCapture (24/96 max) and Audacity.

 

I always record without any limiter or compressor, and during rehearsals I set the recording level to about -18 dB. My limited experience is that the actual performance is always louder.

 

After recording I usually need (?) to apply a steep high-pass filter @ 25 Hz. to reduce rumble from passing heavy traffic, air conditioners kicking in etc. Then, depending on recording-location I do some equalizing, but not much.

 

Last, I maximize the tracks to -3dB. and downsample to 16/44.1.

 

Any advice is welcome!

 

Peter

 

 

 

 

 

 

 

 

 

 

[Talk2Me]

Peter, what do you own ears tell you?

 

[thesurfingalien]

Hey Talk2Me,

 

In general it sounds pretty good, but I believe there is room for improvement on recording technique.

 

But as for a difference between the 24/96 and 16/44.1 final result versions, no, I am not able to distinguish between them.

 

But keep in mind that my recording equipment is quite moderate, and so are my recording skills.

 

Peter

 

[bdiament]

Hi Jud,

 

**"...What would you say are the main ways in which 2x rates through good equipment fall short of your mic feed?..."**

 

I don't believe I've posted elsewhere about this.

Strange how I've been pondering how to answer your question since I first read it last night. If you were in my studio (or at a recording session), I could demonstrate the differences in seconds but articulating them in words is not so easy.

 

As a preface, I'll say that for my ears, the 2x rates show just how drearily lacking the 1x rates are. At least a cello sounds like a cello at the 2x rates. While it might be recognizable as such at 1x rates, a direct comparison will show the 1x more than a little bit reminiscent of a kazoo (this impression, which I have mentioned elsewhere, dates back to when I first heard 2x rates).

 

But the cello in the mic feed (and well done 4x) is still different. Even with the best 2x I've heard, there is a diminution of harmonic complexity. Nowhere near as drastic as with 1x but it still won't be confused with the mic signal (or the event itself).

 

Spatial cues too are just a bit "misty" and vague when compared with what comes from the mics. Not necessarily positioning but the space between and around the players sounds to me, like just a bit of focus has been lost. While much truer than 1x, I find there is still some difficulty in determining the size and materials of the space in which the event is taking place - something that is instant and obvious with the mic feed (and the best 4x).

 

Lastly, though not at all least, the bottom end sounds "very good", which is exactly what I find wrong with it. It is pleasing, extended, dynamic, etc. But it doesn't sound the same as what comes from the mics.

 

It is in these areas, particularly the bottom, that I find properly done 4x (much more rare than one might expect) to cross a threshold. Tonality and space sound like what I hear from the mic feed (again, this is something I've wished to hear for decades) and perhaps most amazingly to me, the bottom, for the first time in my experience, sounds like bass does in real life, with a speed and definition of pitch I've not experienced from any recording device before now.

 

I find the best 4x to offer so many sonic advantages; once that threshold is crossed, it really is magical. Hence, it always strikes me as odd when folks say 2x (or 1x !) is enough. It may be enough for them and that's fine. They may see 4x as overkill and that's fine too, though I admit to reading such statements as I would read one that insisted there are no colors in a rainbow. As long as I'm free to record at and to listen at 4x.

 

To be clear, this is just my perspective, how I hear it.

 

Best regards,

Barry

www.soundkeeperrecordings.com

www.barrydiamentaudio.com

 

 

 

[bdiament]

Hi Peter,

 

With apologies to our fellow CA members, a short answer here. In order to keep this thread on topic, we can continue off line (use the "Contact" page at one of the sites in my signature).

 

**"...I always record without any limiter or compressor, and during rehearsals I set the recording level to about -18 dB. My limited experience is that the actual performance is always louder...."**

 

I'm not sure what you mean by "-18 dB". Is that max peak level?

Are you recording at 24-bits? (If so, are you sure your hardware and software are able to? Regardless of what their spec sheets say?)

 

I always leave lots of headroom when recording and yes, the real performance always seems to hit higher peaks than the rehearsal. In my experience, A-D converters perform at their lowest distortion when the maximum peak is no higher than -6 dBFS. (In the old days - and still today for a few who don't realize it - folks thought they should see 0 dBFS max peaks for the highest resolution.)

 

As long as the format is truly 24-bit (hardware and software), I'd rather have a max peak at -20 than have it at -2. I find the end result is cleaner.

 

 

**"...After recording I usually need (?) to apply a steep high-pass filter @ 25 Hz. to reduce rumble from passing heavy traffic, air conditioners kicking in etc. Then, depending on recording-location I do some equalizing, but not much...."**

 

I know too many folks who believe applying a high-pass filter "cleans up" their recordings. Among the many (to me mistaken) assumptions, is that this will not affect the music. But in every single case I've heard over the years, the music ends up being thinned down.

 

What I've found is that all instruments, even an orchestral triangle, need that bottom. There must be a low end component to all of them because when you take it away and they no longer sound natural.

This, aside from the time smear that filtering will engender on top of the thinning.

 

**"...Last, I maximize the tracks to -3dB. and downsample to 16/44.1...."**

 

Questions I would as are:

Why -3dB? What are you using to effect the gain change?

What downsampler? Which dither algorithm?

 

Hope this helps - or at least provides some food for audio thought.

But out of respect for this thread, let us take this offline.

 

Best regards,

Barry

www.soundkeeperrecordings.com

www.barrydiamentaudio.com

 

 

[thesurfingalien]

Thanks so far Barry! I will send you a PM shortly.

 

Regards,

Peter

 

[Talk2Me]

Barry,

I read your very interesting and informative threat where you where trying to get higher res fro a blu ray player, and did you find a better sounding player (please no Oppo). Sorry for the OT people.

 

[bdiament]

Hi Talk2Me,

 

**"...I read your very interesting and informative threat where you where trying to get higher res fro a blu ray player, and did you find a better sounding player (please no Oppo). Sorry for the OT people..."**

 

I sure hope that's a typo. I would hope I never posted (or uttered) a "threat" to anyone.

 

It wasn't simply the players that turned me off to Blu-ray for audio, it was HDMI.

 

I want to be able to get the 24/192 audio directly to the DAC of my choice and not the one built into the player or into any A/V receiver.

 

I know some folks have spoken of third-party interface boxes that may strip the audio from HDMI but my feeling is this is not how most folks would get to hear the music. (Besides, in my view, S/PDIF isn't exactly state of the art either.)

 

In any event, our own testing tells us that listening to the program material as .aif or .wav files from a computer (particularly using Firewire as the interface protocol to the DAC, rather than S/PDIF) gets the listener much closer to the original than any disc in any player or transport does, so I abandoned the idea of Blu-ray for our high res audio releases.

 

Best regards,

Barry

www.soundkeeperrecordings.com

www.barrydiamentaudio.com

 

 

 

[Talk2Me]

LOL. Typo. I am typing in bed...Yes alone and in pjs. i noticed you did not mention usb.

 

[bdiament]

Hi Talk2Me,

 

I do use it for connection to the hard drive housing the music library - because I'm using Firewire for the connection to the DAC. ;-}

 

Best regards,

Barry

www.soundkeeperrecordings.com

www.barrydiamentaudio.com