Lavry anyone?

[mojobone]

"A recording made accurately at 44.1 kHz will always sound better (or more "analog") than an inaccurate recording made at 96 or 192 kHz. In many cases this would be mostly due to jitter in the AD and DA conversion, as versus the sample rate. But even in these cases, the degradation is almost certain to be caused by more than just this one source of inaccuracy."

I know there are tons of people who disagree & maybe they know more than than Lavry, just thought it something to consider if you're planning on breaking the bank to get to 192.

Because I'm poor I like his viewpoint!

Having recorded in the digital world longer than most any one else in this business I have dealt with that idea for a long time. My first recordings made in 1889 were sampled at 100K. I disagree with the twice over theory. If you change or eliminate a high frequency it will effect the lower ones. Any one who rolls of highs and hears the sound change knows that. Therefore the same is true for frequencies above 40K. And because the curve is exponential if you take away frequencies like at 40K it will effect everything below it. So the higher you can sample the better. But 192 is way up there. I would only do it if I had the power and the money.

That's undoubtedly a Lavry rep and not Mr. Lavry, because jitter affects playback only; Bob Katz' book, Mastering Audio, explains why. Dwayne is correct about EQ; one of the reasons a Baxandall EQ works the way it does is because the shelving/corner frequencies are outside the human hearing range-but you can definitely hear it working. Even linear phase digital and look-ahead EQs exhibit this behavior to some extent.

The Nyquist frequency (half the sampling rate) is important to audio capture but the filter you use to remove frequencies above the Nyquist is probably the most important component for optimal sound quality in a digital system, arguably far more important than the sample rate. I believe this is what the Lavry rep was trying to say.

[mojobone]

We can rest assured that mics DON'T produce anything over 20khz or the manufacturers would be all about blowing their horn in the spec sheets. a u87 is rated at 20khz but I've heard engineers claim that it really topped out at a little over 19khz in their test, so a 20khz spec might be a rounded up example used just to match the range of human hearing. So if you're using a mic it's a reasonable assumption that there's nothing over 20khz unless it's a buildup of noise.

Earthworks' mics can capture 40kHz and beyond: http://www.earthworksaudio.com/our-microphones/

Just because only dogs can hear it, it doesn't mean it isn't there. The first rule of mastering is that anything you do to any part of the spectrum affects every other part; why should ultrasonics be any different?

So Dwayne's right, Nyquist was wrong. It might be helpful to take a closer look at what those mic specs really mean; when they say 20hz-20kHz, those are the points at which the mic's response is 3dB from flat, NOT that they don't capture stuff beyond human hearing range.

[benjet]

I am once again humbled by the amount of knowledge you guys possess. Thanks for sharing ....

[Dwayne Russell]

The Nyquist frequency (double the sampling rate) is important to audio capture but the filter you use to remove frequencies above the Nyquist is probably the most important component for optimal sound quality in a digital system, arguably far more important than the sample rate. I believe this is what the Lavry rep was trying to say.

There are several filters used in the A/D D/A process. To my knowledge they are not filters for frequency although in a particular system these days there might be. They are for noise and other things that the term filter is not a good word for it. I would say the most important part is all of it. In particular the sync of the process. That is why I use the Black Lion Word clock with a $50 BNC cable. I'm clocking at 4 picoseconds! No jitter! Crystal clear!

But absolutely the sample and hold process is the most important part for capturing to the disk what the mic heard. Newbies to recording don't take this into account. Their first priority goes to the mic and pre.

[mojobone]

Newbies to recording don't take this into account. Their first priority goes to the mic and pre.

True, but that's as it should be, or close enough. The mic and preamp aren't part of the digital system, and filter ripple, Nyquist frequencies and jitter all have to do with playback and how well you can hear and therefore mix the music, once captured. A significant improvement on the digital side likely won't happen without spending an additional $2000-$4000 per channel; a good reason to have your tracks mixed by someone who already spent the money. The best thing for a newb to focus on is learning about mic placement and room acoustics/treatment. Once you have a great space and a great instrument and player to record, then look at the mic and the pre, is my opinion. Great converters let you hear better, and that's always a good thing, but odds are, the end user won't have anything near that good.

[Dwayne Russell]

True, but that's as it should be, or close enough. The mic and preamp aren't part of the digital system, and filter ripple, Nyquist frequencies and jitter all have to do with playback and how well you can hear and therefore mix the music, once captured. A significant improvement on the digital side likely won't happen without spending an additional $2000-$4000 per channel; a good reason to have your tracks mixed by someone who already spent the money. The best thing for a newb to focus on is learning about mic placement and room acoustics/treatment. Once you have a great space and a great instrument and player to record, then look at the mic and the pre, is my opinion. Great converters let you hear better, and that's always a good thing, but odds are, the end user won't have anything near that good.

Well I dont know what you mean by nyquist frequencies. But jitter can occur in recording and playback. Jitter is timing errors in the D/A and A/D process whether it be in coming or out going. So, yeah, it matters for anyone recording to a hard disk. There is no studio that can save a poor recorded source after it has been recorded.

So, yeah, the newbies put the concern in the wrong place usually. And no, it wont cost $2000 per channel to get a good clock. I had my fireface modded for $400 and added a killer clock and BNC cable for an additional $500. My A/D and D/A is as good as any lavry or all the other ones at that price point.

[mojobone]

Strongly recommend Dr. Katz' book. I'm pretty sure I understand it well enough to explain it, but I don't have time to write one. (I know it seems counter-intuitive; it took him a couple of chapters to convince me, heh) Even in a perfect, theoretical world, jitter on the uptake does not matter, (except maybe at very low sample rates) and if you're recording in a noisy, untreated space with poor technique, all the Lavrys in the world cannot help you sound better. Besides, plenty of great, amazing-sounding recordings have been made in crap rooms with crap gear, though I wouldn't necessarily characterize them as "hi-fi".

We've jacked this thread far enough off-course, but I reckon the subject is worthy of further discussion, even if I think accuracy is overrated. Meet me in Technology, Recording & Production on the Nyquist thread.

[Dwayne Russell]

Strongly recommend Dr. Katz' book. I'm pretty sure I understand it well enough to explain it, but I don't have time to write one. (I know it seems counter-intuitive; it took him a couple of chapters to convince me, heh) Even in a perfect, theoretical world, jitter on the uptake does not matter, (except maybe at very low sample rates) and if you're recording in a noisy, untreated space with poor technique, all the Lavrys in the world cannot help you sound better. Besides, plenty of great, amazing-sounding recordings have been made in crap rooms with crap gear, though I wouldn't necessarily characterize them as "hi-fi".

We've jacked this thread far enough off-course, but I reckon the subject is worthy of further discussion, even if I think accuracy is overrated. Meet me in Technology, Recording & Production on the Nyquist thread.

Katz used to come to my office in West LA every now an then regarding our digital recording system. I have spoke to him quite a few times about digital recording. That was my business at that time.

[Len911]

Here's an informative review from PAR. Lavry, Mytek, Benchmark and Prism.

http://www.proaudioreview.com/article/26252

[mojobone]

My favorite sentence of that review? This one:

Although meaningful, the differences between models were not as profound as microphone choices, monitoring choices, and room acoustics.

Also, I found it interesting that the mix and mastering engineers tended to agree on the conclusions in blind tests.