As a new concert season is coming up with BSO opening night next week I was sprucing up my recording studio. The machine is 3MHz with 2G memory – plenty run SoundLab5. The SoundLab just record and play, I do absolutely nothing else in this software, no editing of any kind. The machine is completely fan-less with external power supply. It has no sound card but juts Lynx-16 digits interface that is high quality 110Ohm digital in and out. I changed the stock Linx-16’s proprietary AES/EBU cable to a custom –way more convenient to use cable, though I did not detect sonic differences between them.

The major consideration this time was (and still is) about the AD and DA processors. I use external possessors: analog-to-digital is Lavry Gold AD-122 and digital-to-analog is Lavry’s multibit Gold DA-924

http://www.lavryengineering.com/index_html.html

The objectives were to have play a FM broadcast directly from tuner and do not hear any differences when AD-DA chains is introduced in a signal path. Larry Gold processors do it. I would not make the same claim if I use tape, microphones or records as sources (I did not do it at the level as I would accept it convincing and methodologically correct) but in the context of somehow restricted FM broadcasters Lavry Gold converters chain are absolutely transparent.

Over the course of the last 8-9 years I was searching an opportunity to play with sound of Pacific Microsonics converters. I would not go into details – whoever know what the Pacifics were would know what I’m talking about. Recently I found a very good unit - a very last Model One production of 1998 and software update in 2000. I was contemplating to get Model Two but since I am not going to go higher then 88/24 I decided do not spend more money then I will be able to use. The Model Two has 96kHz and 192kHz, single wire operation option, extras post-production DD operation modes and more sophisticated choices for extra external clocking. Pretty much anything that that Model Two does is unnecessary (for me) except a theoretical ability to subordinate D/A clock from the A/D clock. From another side I think that Model two, since it has multibit AD (!!!) and runs 192kHz then it would have more powerful processing power and theoretically more opportunities for different noises. Why would I need if I stay with 88 kHz?

The Pacific Microsonics (Euphonix) unit is huge with externals power supply. Lavry Gold are small but they run hot like hell and need a LOT of room around then – so both converters take a LOT of room in my rack. I was considering putting both Lavrys right next to each other in a small pro rack-type box with forced air cooling – I did not do it yet.

I run both Lavry Gold tandem and Pacific Microsonics processor in parallel (Linx-16 is 8 channels card and allow to map the channel is any way I wish) and Pacific is truly stands to it’s reputation. It is also absolutely transparent during the injection test. (Live FM vs. AD-DA). The Pacific is also remarkably good with my CD transport (it is where Lavry slips) but unfortunately it has no such a wonderful Lavry’s flexibility to detect and reset own sampling rate. So, the major question that I ask myself and possible will be asking during next few mohth would be which AD/DA combination is better: Lavry Gold or Pacific.

As now it is very obvious that the  both of the converters options are good but there are also some tiny differences between them. They are very-very small and more like “touches of sonic inflections” then truly sonic differences. It is like Lavry Gold chain has different lower bass but Pacific has different type of HF all together. They are not necessary better or worth but different – both very good quality. I would not try to make a final observation and a conclusions what “should go” until I will finish remodeling my playback and have the second 6-Chennals Super Mils done. Then it would be possible to look at fine moments of imaging, space presentation, accentuation, musical colorations and overall musicality. I need also my tweeters finalized with new dedicated amps as Pacific Microsonics has some very interesting options to inject custom dithers – something that I very much look forward to play with when my playback will be properly finished and calibrated. I will be posting my thoughts about it in a future and perhaps I should upload some files from Lavry and Pacific for others to compare and to see difference….

From Pacific Microsonics’ manual:

DAC-Monitoring System Performance Optimization

Unit has advanced DAC optimization capabilities designed to provide the highest possible fidelity with different types of monitoring electronics and power amplifiers. While DAC dither greatly improves the linearity of D/A conversion, it also adds supersonic energy to the analog output signal, which will interact differently with various types of line amplifiers and power amplifiers. Higher levels of dither generally improve DAC linearity, but also can stress electronics that have limited high frequency slewing capabilities.

This can manifest as audible Transient Intermediation Distortion, or a smearing of detail. To allow the best possible match of dither to amplifier performance, three selectable dither levels are provided for 1X conversion, and four levels are provided for 2X conversion.

The best way to determine the optimum DAC dither for a specific monitoring system is by listening to various types of source material while selecting different DAC dither curves. Optimum settings should be determined separately for 1X and 2X material. Generally, the “wider-band” or “faster” the monitoring electronics, the higher the optimum dither level will be.

OK, after much consideration, recording and listening I made a number a final concussion about the flavors of Pacific Microsonics and Lavry Gold and now I have selected which AD/DA will be serving my FM recording. I offer for you an opportunity to see yourself the difference between the processors and I would like to hear your vote.

Below are two 88/24 fragments (~33Meg and 44Meg), one recorded with Pacific another with Lavry. Both broadcasts were identically slightly compressed by FM station.  As you can hear they are different orchestras and in different halls but I still feel that those fragments do portray the deferens between the processors. Download them and play on your 88/24 playback.

You might not have bandwidth to run then from web, so right-click them “Save As” and play off you machine.

<< File # 1 was here. >>

<< File # 2 was here. >>

which has no low frequencies to speak of, very small room, very insensitive speakers, SS amp... but is the only one that can play 24/88 files, DAC1 file sounds kind of "dated", as if it were recorded in a huge concert hall with tubed electronics and vintage microphones. Bass is very woolly, spreads all over the hall and the orchestra has some kind of "linking" which creates a decent illusion of musicians playing together, though there's something about the harmonics that I don't feel completely right.
On DAC2 things are cleaner, more "a la mode", so less linking between the members of the orchestra, less sense of togeterness, tighter bass, the harmonics sound righter to me in their proportion, but probably not in the way the spread over the image of the hall.

It's very difficult to me to make any valuable judgment and even more deciding which of two AD is keeping better the TU-X1's sound. If both performances were broadcasted from the very same venue and using the same equipment, then for tone and harmonic structure maybe I'd choose the DAC2, but for the capacity to portray a believable sense of space, then DAC1. Knowing the TU-X1 qualities and what makes it a very special piece of gear, in my opinion DAC1 preserves them better.
It will be very illustrative knowing Romy's take.

Regards,
A.

op.9 wrote:

Romy, is there a problem with the DAC1 file? After 16' there is an audible click then a lowering of level and a much more bleached out sound.

op.9 wrote:

(Also the channels are reversed)…

jessie.dazzle wrote:

Though I have a nice TU 9900, I'm still kicking myself for having let what looked like a really mint, one-owner TUX-1 get away for a ridiculous price (something like $900) on eBay about a week ago (discovered it after the sale).

drdna wrote:

If it was not clear from my oblique comments before, I agree with what all have said, DAC1 seems to be techinically more correct but at the expense of some information loss.  DAC2, which I prefer, has more distortions but also has more complete musical information.

Antonio J. wrote:

…to make any valuable judgment and even more deciding which of two AD is keeping better the TU-X1's sound.

For whatever reason people expressed a lot of interest to Pacific vs. Lavry A/D and I recently got an relatively a large number of emails, including from the people I do not even know, asking me to compare identical analog tracjs and post the founding.

I could do better than this. Below are two files of the same music converted by Pacific Microsonics A/D and Lavry Gold AD122 MkIII. The files are 88.2/24 around 90Meg each. No processing of any kind was used, no dithers, no noise redaction, no EQ or gain change.  The regular commercial PHIL L5137 4-track tape at 7.5 IPS was played on Stellavox SU-8 and then via Purist Proteus cable was outputted in one AD processor and then to another. The Stella’s heads were cleaned and demagnetized. Stella run from batteries, the AD and DAW right from the wall (Screw you the lazy PurePower APS people, where are you???!!!!)

I picked music that has relatively diverted combination of low lever passages and high level passages, good instrumentation and combination of groups and solo. It is opening thyme of the last movement of Brahms’ Double Concerto recorded in 1972 with Haitink and Royal Concertgebouw. The leads played by two giants: Henryk Szeryng and Janos Starker. Unfortunately the entire performance is not up to the lever that might be expected from those participants – something did not “clicked” in there – I think it was the Bernard Haitink… Still, it is a good pieces of music played by the maters who do know own business.

Peter Foster. If you read it can you processes one of tracks (that you like better  - they do sound differently) with your magic “K-Stereo” machine and send me the file. I will repost it.

Other falks, if you feel to comment on the differences between the Pacific Microsonics vs. Lavry Gold then be my guests… To those of you who have well-imaging installations I would be specifically interested to hear your thoughts about the difference in imaging between A/D processors. (I have compromised one channel as now and my view on nowadays are very handicapped)

88.2/24 arond 90Meg each file:

<< File # 1 was here. >>

<< File # 2 was here. >>

Rgs, the Cat

drdna wrote:

It is funny since they are identical digitally recorded tracks you would think they are the same.  I would very much be interested to know if the binary code is identical or not for the two tracks.

I do suggest to re-read my post above where I stated that Pacific has remarkably similar sound to Bidat. In Bidat there is one quality that sets it apart from any other converters out there – the Bidat’s dynamic bursts are orders of magnitude less annoying then in any other DACs I have  heard – they kind of blossoming with a completely different lisening feeling – the one that I very much appreciate and cherish…

So, how big my surprise was that Pacific have absolutely the same type of dynamic bursts – and I never have seen any DAC besides Bidat ever did it!!!

“At the heart of the Museatex Bidat is a unique and proprietary digital filter algorithm …. this algorithm examines the digital audio data and makes an intelligent determination on how best to up sample this data …. The Bidat algorithm analyzes the digital audio samples for relative accelerations and then chooses the type of filter most appropriate for that segment of data... More important, it produces a uniquely pure impulse response without Gibb's ringing. For example, a square wave will have clean rectangular transitions and no ringing. The Museatex Bidat can claim transient response demonstrably MORE faithful to the recorded signal than the best analog playback systems for vinyl LP's and open-reel tapes. The undistorted transient response of the Bidat means the algorithm contributes no coloration of its own to the processed audio. This performance eliminates the "digital washout" of sonic subtleties and the harshness heard in listening to a conventional interpolating algorithm. “

I was under impression the Bidat is the only DAC that has the adoptive filters. However, here is from one of the designers of the Pacific Microsonics:

“…We slightly delay …. signal, not enough to cause any sync problems but enough that we can do a continuous Fast Fourier Transform. The resultant information is digitally analyzed in real time by an algorithm that determines, based upon a model of the mechanics of hearing and psychoacoustics, what is perceptually dominant in the signal from instant to instant. And that information is used to optimize the decimation filter. One moment you might have a sudden sharp transient, so it uses a filter with minimum time dispersion to pass the transient cleanly. The next instant, there might be a cymbal crash, so it uses a filter that minimizes alias distortion. All the filters are the same length, so you are not getting a phase shift as this is going on."