Please comment on Lavry/CraneSong DAC

[Chris Kress]

Originally Posted by MattGray

It's amazing how many ME's claim to have "transparent" or "flat" mastering rooms & yet they all sound completely different from each other. Sure some are going to be more accurate than others, but a lot of it is still down to budget & personal taste.

The main thing is that you know your room & monitoring intimately & that your work translates well outside the room. I doubt there is such a thing as the 'perfect' room.

Matt

I was waiting for this guy to say that he works in an anechoic chamber : ) I overlooked his previous test recommendation which does look flawed to me anyway. Otherwise, for me, this issue is resolved.

I would like to say thank you all very much for the excellent help I received. Here is the point that was confidently re-enforced: When the technical excellence of a system rises to a fairly high level, the combination of personal preference and tuning of the entire chain with various components, cables, acoutic treatments etc., will bring the system to completion.

For anyone who responds to the thread title, here is where I ended up:
My main monitors are fed by the lavry which is connected directly to a hafler amp with about 3' of zoalla cable. I am pretty sure my hafler is ready for a re-capping but I am not touching it. It softens transients in a range that I have problems dealing with. I'll admit it, I use a higher end MB quart speaker and I can not let them go either. As far as my room: see the link below..

My secondary monitors are controlled by the coleman, fed by a second lavry. Funny enough, the bryston is here feeding avantones. I like the bryston best on nearfield monitors so it will stay in this secondary position as I occasionally switch out nearfields....

I am back to work. Best wishes to all..

If you are interested, here is and example of good help I received in the acoustics forum and pictures of my room coming together..

Ideal control room shape- II continuing

[audio ergo sum]

Originally Posted by MattGray

It's amazing how many ME's claim to have "transparent" or "flat" mastering rooms & yet they all sound completely different from each other. Sure some are going to be more accurate than others, but a lot of it is still down to budget & personal taste.

The main thing is that you know your room & monitoring intimately & that your work translates well outside the room. I doubt there is such a thing as the 'perfect' room.

Matt

So true.
There is nothing more humbling to see than the measured frequency response or RT60 graph of a state of the art listening room without more than a 1/3rd octave smoothing. It really puts things in perspective...

More power (and budgets) to acousticians...

[kjg]

Originally Posted by Chris Kress

I was waiting for this guy to say that he works in an anechoic chamber : )

Or headphones...

[bob katz]

Originally Posted by Chris Kress

I was waiting for this guy to say that he works in an anechoic chamber : ) I overlooked his previous test recommendation which does look flawed to me anyway. Otherwise, for me, this issue is resolved.

I would like to say thank you all very much for the excellent help I received. Here is the point that was confidently re-enforced: When the technical excellence of a system rises to a fairly high level, the combination of personal preference and tuning of the entire chain with various components, cables, acoutic treatments etc., will bring the system to completion.

By the way, Chris, I can guess probably why you are preferring the direct connection of the DAC to your monitors than going through the Avocet! As I had a similar experience. I used to have a pair of Reference 3As in Studio A, supplemented by subs, and now I ask myself how I could possibly have worked with those. They are considered to be excellent speakers, but their degree of coloration became quite obvious over time. They are designed to be audiophile speakers with a "reasonable" degree of accuracy and you can hear most inner details in them.

In the end, how many misjudgments did I make during the period when I had the 3As? Actually, not very many. Because I applied all the science and art and skill at my disposal to linearize that room, components and speaker system in an integrated way! And then to learn the remaining defects and learn how that system translated to the outside world. During that period of time, a minimalist route to the loudspeakers and a direct connection from my DAC actually did sound best!

[Fred]

But Philip would be dismayed to know that Forssell's opamps have only about 20-40 dB of open loop gain and under full negative feedback still exhibit meaningful amounts of harmonic distortion compared to a 5534

Actually Bob that is only true for my older vacuum tube opamps which have distortion in the 0.006% range. My JFET-993 opamps (which I've been using for about 5 years now) have 75- 80 dB open loop gain and THD+N figures in the 0.0002% range.

[bob katz]

Originally Posted by Fred

Actually Bob that is only true for my older vacuum tube opamps which have distortion in the 0.006% range. My JFET-993 opamps (which I've been using for about 5 years now) have 75- 80 dB open loop gain and THD+N figures in the 0.0002% range.

Dear Fred: It's good to see you on this forum! 0.006% is fantastic for a vacuum tube anything! If "fantastic" is the right way to rate a unit just on the numbers...

Anyway, the point I was trying to make to Philip still stands, that nothing is "perfect" and that his idea that my "system integration" approach is wrong, and "to compensate for imperfect loudspeakers" is simply a complete misunderstanding of the art and the science of putting together a mastering-quality reproduction system.

One of the criteria for a high quality reproduction system is the ability to hear distortion in the sources and recognize it, and to hear and not miss inner details in the sources. There's nothing written in the technical books about that per se, the idea is that the monitor section should have low measured distortion on its own so as not to mask distortion in the source, and that is a precept. However, investigations by some designers that I know are showing that a certain amount of low level measured distortion may be desirable as it reveals depth in the sources.

When does distortion cease to be a negative and become a positive (in either direction)? That debate will go on long after I'm dead.

Distortion can be harmonic, temporal (acoustic, e.g. diffraction), inharmonic, mask in different frequency areas. It's a very complex concept and nothing, not even the most "objective" measurements, comes close to describing the holistic approach and integrated approach required to assembling a world-class monitor system.

And it's a continual struggle. The addition of some bass traps to my room improved the linearity (and definition) in the bass and other regions, but audibly removed some high frequency liveliness, dynamics and clarity. Shroeder curve measurements then confirmed that the reverberation in the upper midrange and highs had been reduced below the Shroeder recommendations that reverb time should be even in each octave band above about 125 Hz. Some diffusors were constructed and installed. Listening then confirmed the improvement from the diffusors and a further Shroeder test agreed with the positive listening. Objective and subjective, art and science live side by side, day in and day out.

Another example: Changing the power amplifier from a well-known Class A amplifier to a pair of Lipinski Class D Mono blocks improved the low frequency definition and extension in the room. The high end and upper midrange were equivalent. All of this judged by listening. But playing my shuttle launch recording, the subwoofers now bottom out (overload) at two moments in the playback where they did not before. The theory is that the Lipinskis frequency response may extend down further at the bottom end into the subsonic than the previous amplifier. This "improvement" introduced a problem that had been taken care of before by the slightly less extended bottom end response of the previous amp. Who is to say which is more or less "perfect"?

BK

[Fred]

Hi Bob, et al...

I posted my comment to correct any misconception that might be obtained from the incorrect figures that you quoted about my JFET opamps. That's all.

It's a very complex concept and nothing, not even the most "objective" measurements, comes close to describing the holistic approach and integrated approach required to assembling a world-class monitor system.

I agree with your statement here.

I've spent a TON of money on test equipment that can measure incredibly small amounts of various types distortion or do FFT plots of analog equipment with great resolution and low noise, etc. None of those measurements tell me anything about how my circuits will sound. Of course that assumes the circuit is performaning acceptably (decent freq resp, noise, etc) and is stable. Beyond that, the measurements mean very little to me as an indicator of sonic quality. They simply show me how well the circuits are performing (and that is important) but not how they will sound.

I've spent a ton of money on my monitor system too, and that is were all of my circuit designs get accepted, rejected, and/or tweaked. I love listening to music and I find that I learn more about my circuit design approaches while listening to music then stareing at test equipment.

Anyway as I said, I just wanted to correct or prevent any misconceptions about my circuit designs caused by inaccurate figures.

Sorry for the somewhat OT posting.

[philip]

Well of course I'm well aware that there is nothing such as a the perfect room, speaker or anything for that matter. Even a feather in the hat creates a comb filter if I remember correctly.
I still don't get the idea of choosing something more colored over a less colored equipment.
The mastering monitor chain is supposed to be as revealing as possible so you can do your job, not sound as good as you like.
Choosing a colored DAC over a transparent one is like putting a EQ in the monitor chain, it may sound better but it's not right.

Your monitor system is not better then it's weakest part. By choosing the best equipment by using simple listening tests instead of choosing what sounds more pleasing to you you'll have a transparent chain in two ways. 1, you'll have a system that reveals the audio 2, you'll have a system that you easily can upgrade. Without knowing the rest of your chain is as uncolored it can be (given the factors we all deal with, budget etc) no new equipment can be added or replaced on it's own.

Example, if you buy a DAC that's sounds more pleasing to you at a given setup, even though it is less transparent, the next time you shop for new loudspeakers you'll have to compensate for the DAC. Or even worse, you don't know your DAC is colored and you will buy not the best speakers but a a speaker system that sounds good to your colored DAC.

Yes I know there is no perfect anything, but using that as an excuse for not doing your best is unprofessional.

/P

[philip]

Originally Posted by tengu

Isn't the ADC another variable which limits the affectiveness of this test?

Sure it is a high quality ADC that you are talking about but so are the DAC's and in these tests the differences that are so minimal that putting another converter in the test increases the inability to accurately isolate the differences between DAC's?

Not to mention the converter in the CD player.

Couldn't the CD DAC introduce certain jitter that already exists in one of DAC's but not in the other. This jitter would show in both DAC's after the CD output is recorded through the ADC. It could possibly disadvantage one of the DAC's in this test, the one that has better accuracy in the area where the jitter is introduced in the CD DAC.

Hopefully this makes sense to people. Basically there are 3 variables now instead of 1.

No, you are listening to a source, it doesn't matter, it could be a tape, a 24/96 high-res file or a vinyl for that matter.

The ADC is of course a potential problem but it is still a better way to judge a DAC with a reference then without.

[kjg]

Originally Posted by philip

Yes I know there is no perfect anything, but using that as an excuse for not doing your best is unprofessional.

/P

I'm really trying hard to understand you, but the last part of your post makes it hard to take you seriously.
Do you even know who you are talking to?

How about the notion that you (with you decades of professional experience on the highest level, I'm sure), might still be able to learn something from others?

Just because your point may be sensible in theory doesn't mean it actually holds up in real life. Very experienced people are trying to tell you something about that.
"My theory says (...), and therefore you are unprofessional!" ???
Really?

[philip]

Originally Posted by kjg

I'm really trying hard to understand you, but the last part of your post makes it hard to take you seriously.
Do you even know who you are talking to?

How about the notion that you (with you decades of professional experience on the highest level, I'm sure), might still be able to learn something from others?

Just because your point may be sensible in theory doesn't mean it actually holds up in real life. Very experienced people are trying to tell you something about that.
"My theory says (...), and therefore you are unprofessional!" ???
Really?

Well I don't think the idea of having a transparent monitor chain is "my theory". To me thats the first goal in mastering. Hearing whats going on in the music so you can address the potential problems in the material and not compensating for DAC, line drivers and loudspeakers.

[kjg]

Good luck then with your no compromise, perfect speakers, amp and acoustics.
How many million is your budget? And in the meantime, you cannot get anything done, because you can't trust your ears and balance a system. You need every component to measure perfectly...

Let's agree to disagree for now. Please post back here in then years, and we'll see if maybe then we can agree. You with me, or I with you

regards,
kjg

[bob katz]

Originally Posted by kjg

No, the response is: since there is no perfect component (including the acoustics), you balance out the imperfections against each other to get an accurate and comfortable chain.

Right, kjg, and here's another one that will blow Philip's mind: The sum of 3 or 4 measurably "accurate" components in series suddenly becomes inaccurate (or at least colored in a certain direction) due to accumulation of previously insignificant errors! Yes, indeed, apparent imperfections in a component that previously appeared accurate to you WILL BECOME more obvious to you both over time and in series with other components. Should you then throw out the newly-discovered imperfect component? Or perhaps do something when changing another component in the chain.

And yes, Philip, often we do discover that we have to throw out or change one component in a chain because we changed another for a more perfect one and that then revealed the weaknesses of the previous one. It is a road to perfection and we do have to live with it. Want to buy my collection of old DACs and ADCs from my garage?

Even Philip's search method for the perfect ADC (compare the ADC with the analog source), should it ever reveal to him the "perfect" ADC, will find over time that there are some imperfections. Of course, the ADC, the ultimate source for all the files which you are working with, especially when transferring from a precious analog tape---should be as accurate as we know how to make it. But since the ADC always has to be judged through a DAC, then the round robin always continues. And that DAC is connected to a less than perfect loudspeaker over a less than perfect amplifier. Therefore the process of assembling the best reproduction system is long and arduous. It took me nearly 10 years in this room to get my system to the "wow factor" that it has now. And it's been a very entertaining trip.

I'm reminded of a band I once recorded where the mandolinist said, "Isn't 'C' a wonderful key," whereupon the guitarist sarcastically replied, "Yes, especially when there are so many different C's". And then they tuned up. But there are many acceptable C's and I've heard some remarkable recordings where the unisons were not quite right but sounded wonderful because of it. But I digress. I'm not sure why I quoted this one...

The same round robin argument can be made for microphone judgment. Since all loudspeakers exhibit some degree of high frequency rolloff, how can you accurately judge a microphone through any loudspeaker? And what is "flat" anyway, with a microphone that exhibits proximity effect? That's why I wrote a book on the subject, because there are sensible answers to all these questions, once you understand the limitations. For microphones, there is the reciprocity method of measuring frequency response. For monitor chains, there are the methods described in my book, which I highly recommend to Philip.

BK

[kjg]

Originally Posted by philip

Well, and that's not the proper way to adress the problems. You cannot compensate a bright loudspeaker with a dark DAC, if thats what you think, I don't I have to explain why I hope.

Please do explain.
I'd like be informed how you address the problems (you mean properties...) of real life equipment, in real life acoustics, within real life budgets.

[darkwater]

I had to get rid of a harsh component from my forum monitoring system. After critical listening (reading), I found it abrasive, disrespectful, and unable to spell simple words correctly.

Philip = Ignore.



kidding....kidding....come on guys...

[dcollins]

Originally Posted by thermos

I'll also add it is much lazier to think that there is only 1 correct answer than to accept that there are many. Because if there are many correct answers, you are probably wrong. Thats hard to live with for some people.

I think there are many correct answers.


DC

[thermos]

Originally Posted by dcollins

I think there are many correct answers.


DC

Me too! I'm probably wrong though.

[philip]

Originally Posted by bob katz

Right, kjg, and here's another one that will blow Philip's mind: The sum of 3 or 4 measurably "accurate" components in series suddenly becomes inaccurate (or at least colored in a certain direction) due to accumulation of previously insignificant errors! Yes, indeed, apparent imperfections in a component that previously appeared accurate to you WILL BECOME more obvious to you both over time and in series with other components. Should you then throw out the newly-discovered imperfect component? Or perhaps do something when changing another component in the chain.

And yes, Philip, often we do discover that we have to throw out or change one component in a chain because we changed another for a more perfect one and that then revealed the weaknesses of the previous one. It is a road to perfection and we do have to live with it. Want to buy my collection of old DACs and ADCs from my garage?

Even Philip's search method for the perfect ADC (compare the ADC with the analog source), should it ever reveal to him the "perfect" ADC, will find over time that there are some imperfections. Of course, the ADC, the ultimate source for all the files which you are working with, especially when transferring from a precious analog tape---should be as accurate as we know how to make it. But since the ADC always has to be judged through a DAC, then the round robin always continues. And that DAC is connected to a less than perfect loudspeaker over a less than perfect amplifier. Therefore the process of assembling the best reproduction system is long and arduous. It took me nearly 10 years in this room to get my system to the "wow factor" that it has now. And it's been a very entertaining trip.

I'm reminded of a band I once recorded where the mandolinist said, "Isn't 'C' a wonderful key," whereupon the guitarist sarcastically replied, "Yes, especially when there are so many different C's". And then they tuned up. But there are many acceptable C's and I've heard some remarkable recordings where the unisons were not quite right but sounded wonderful because of it. But I digress. I'm not sure why I quoted this one...

The same round robin argument can be made for microphone judgment. Since all loudspeakers exhibit some degree of high frequency rolloff, how can you accurately judge a microphone through any loudspeaker? And what is "flat" anyway, with a microphone that exhibits proximity effect? That's why I wrote a book on the subject, because there are sensible answers to all these questions, once you understand the limitations. For microphones, there is the reciprocity method of measuring frequency response. For monitor chains, there are the methods described in my book, which I highly recommend to Philip.

BK

yea that really blow my mind,

Yes it is fun buying gear for your studio but it takes time and costs money.

You may know from your psykoacostic class that hearing is very adaptive. You will pretty much go for anything that sounds as an reference as long as you don't have anything to compare it against.
So by comparing equipment to the truth you can listen to the difference, kind of usefull instead of comparing to nothing.
This is not really any different from a ABX test, if you don't know about ABX I think there are books that covers that subject. Or maybe you have a friend how is interested in science who can help you understand?

So... instead of more or less randomly throwing in different colors hoping the net result will be transparent you can sort the colors [you don't fancy] out. Now you even can choose from the colors you for some reason want. So it's a very usefull way yo test equipment even of you don't want to/can understand the principles of a transparent monitor chain.

/P

[bob katz]

Originally Posted by philip

yea that really blow my mind,

So... instead of more or less randomly throwing in different colors hoping the net result will be transparent you can sort the colors [you don't fancy] out.

/P

I think you should write a book about that, Philip! By the way, ABX is so passé. A B C Hidden Reference has much more resolution than ABX. You can learn about ABC Hidden reference from one of my friends, a scientist who helped draft the ITU recommendation BS.1116, which is in my opinion, the definitive method for an objective listening test.

Here's a paper on one of the tests that he supervised and conducted, evaluating codecs:
http://www.telos-systems.com/techtalk/00222.pdf

When you examine the careful, far from casual methods that these scientists used to perform the codec tests, perhaps you can see why the casual ABX tests often arrive at null results.

BK

[Waltz Mastering]

Originally Posted by MattGray

The main thing is that you know your room & monitoring intimately & that your work translates well outside the room. I doubt there is such a thing as the 'perfect' room.

I agree.

There is no "perfect room". Even an anechoic chamber is not a perfect listening environment.

There are rooms and systems that work perfectly for the individual using it and that is all highly subjective.

The converters are just one piece of the puzzle as everyone has been saying it's a highly individual choice. Vanilla or Chocolate?

TW

[philip]

Originally Posted by Waltz Mastering

I agree.

There is no "perfect room". Even an anechoic chamber is not a perfect listening environment.

There are rooms and systems that work perfectly for the individual using it and that is all highly subjective.

The converters are just one piece of the puzzle as everyone has been saying it's a highly individual choice. Vanilla or Chocolate?

TW

Yes you are right, there is no perfect anything, but is that a reason to stop quantifying transparency?

[thermos]

Hey, how this for bringing the thread back ot.

I read this over at Brad's forum about the Lavry DA-11 that no one has heard yet.

The DAC in the DA-11 is better than in the DA-10, according to Dan. Also, if your Lavry "goes to -11," you get crystal synch (similar to CrystalLock™) even up at 200 kHz Fs. Furthermore, the headphones still work when the main output is muted, and has power click muting. It's probably best to use this pro-sumer DAC for mastering unless you can spring for the DA924 and don't also need to convert from 4x LPCM to analog. My hunch is that it sounds better than the blue, since the old black was so close to the blue in
"goodness," with only the faintest inkling of DSP from the de-jitter circuit's SRC process compared to the blue's RAM buffer approach to reclocking, which is even more evolved in the Gold DAC (which, of course, features an on-board oven for keeping the temps relatively constant - sheesh).



Andrew Hamilton


I know the op liked the Lavry Black better than the Hedd/Avocet, but I just didn't jive with it (for processing, maybe for monitoring its perfect), but maybe this is even better.

[kjg]

Originally Posted by thermos

Hey, how this for bringing the thread back ot.

I read this over at Brad's forum about the Lavry DA-11 that no one has heard yet.

The DAC in the DA-11 is better than in the DA-10, according to Dan. Also, if your Lavry "goes to -11," you get crystal synch (similar to CrystalLock™) even up at 200 kHz Fs. Furthermore, the headphones still work when the main output is muted, and has power click muting.

Any release date/price info? Will the DA-11 replace the DA-10?

[jdg]

its out now:
Lavry Engineering

[Audiop]

Originally Posted by kjg

nobody spoke about putting together a monitoring chain the audiophool way.

Hi!

Reading this thread that's exactly the impression I got.

it was never about "randomly throwing in different colors".
it was about balancing out known properties of devices as part of a chain, as to obtain a certain neutrality and pleasantness within a budget, in an actual room.

Which does not make sense IMO for the transmission links (and I view the AD-DA together as a transmission link even though they are converters between domains). Pleasantness belong to the music, not the gear. Neutrality is a wise goal though but the big question is how do we test for that? :-)

is there another approach if you are not backed by a research institute or independently wealthy?

what would you think is a wiser way?

Of course, various loop-back tests and bypass tests where you compare the signal into the DUT to the signal out of the DUT.


/Peter

[kjg]

I'm not going to work all day on something that is too bright to my taste. Even if it measures perfectly.
As long as work translates well, what is wrong with having a speaker that is down a little in the top end if that is more pleasant?

Pleasant is not only about music, pleasant is also about productivity, being able to work comfortably etc.

Making it translate is what matters.

So, yes, personal preference is perfectly valid in a chain, IMO.

I agree with you in theory Peter, utmost transparency is the goal. But in real life there is nothing wrong with being pragmatic. If it works for you, if you can make it translate well, AND you can be productive and comfortable because you are not straining yourself for one reason or another, then all is well. If you manage that withing a budget, excellent. That is not audio phoolery and has nothing to do with being a "hippie" either.

It is definitely good to know how devices measure, I think we all agree on that. maybe I'll reread the thread tomorrow, but I did not get the "wooden volume knob" vibe you seem to get from it..

regards,
Klaas-Jan

[bob katz]

Originally Posted by Audiop

Of course, various loop-back tests and bypass tests where you compare the signal into the DUT to the signal out of the DUT.


/Peter

I'd gone to lurking but can't resist making this point:

There are ways to quantify an ADC by listening and a DAC by listening. There are even (very difficult) ways of isolating which of the two is the probable cause of the deviation from neutrality. But never without considering the ADC and DAC working together. You cannot ignore the fact that we must make judgments of an ADC by listening through a DAC. That determining an absolute for an ADC is EXTREMELY difficult. That as the state of the art in either ADCs or DACs evolve, our judgments on what was previously neutral have to change. Just as we cannot ignore the fact that we make judgments of microphone quality by listening through a loudspeaker system.

I'd love to write an essay on how we as mastering engineers reconcile and deal with these issues. I think it's a very valid and useful topic. But not in a thread where one participant has been nasty, disrespectful, and worse, ignored all other arguments. I suggest that anyone who doubts the truth of this and the number of examples that have been given to support the views should carefully read every post, in order.

Sorry. I'm gone again.

[dcollins]

Originally Posted by kjg

I'm not going to work all day on something that is too bright to my taste. Even if it measures perfectly.
As long as work translates well, what is wrong with having a speaker that is down a little in the top end if that is more pleasant?

Pleasant is not only about music, pleasant is also about productivity, being able to work comfortably etc.

Making it translate is what matters.

So, yes, personal preference is perfectly valid in a chain, IMO.

Agreed completely. And when I've gone to some other mastering rooms and heard music I was very familiar with, sometimes it sounds completely foreign! Yet they seem to get by alright.

For instance there is a brand of loudspeaker that you hear a lot about, and the few times I've heard it, I thought it to be nearly unusable for mastering. Perhaps the owners would say that about my Questeds, but again I've been able to get by.

I agree with you in theory Peter, utmost transparency is the goal. But in real life there is nothing wrong with being pragmatic. If it works for you, if you can make it translate well, AND you can be productive and comfortable because you are not straining yourself for one reason or another, then all is well. If you manage that withing a budget, excellent. That is not audio phoolery and has nothing to do with being a "hippie" either.

I've always been curious when audiophile (lover of sound) became audiophool (lover of stupidity). Maybe Wikipedia has something.

It is definitely good to know how devices measure, I think we all agree on that. maybe I'll reread the thread tomorrow, but I did not get the "wooden volume knob" vibe you seem to get from it..

The customer has no idea how beautiful the FFT looked.


DC