Apodizing Filters
default -- Mon, 09/07/2009 - 10:46
Much has been made of these new filters for CD players with RH declaring the Meridian 808 the most significant product in the CD era in large part because of this new filter. As of now, I am aware of only three high-end players with this new filter (Ayre C-5xeMP, Meridian 808 & Esoteric SA-50). If this filter is as significant as RH, and others, espouse do I conclude that any player without it is technologically dated?
JLeeMD
All content, design, and layout are Copyright © 1999 - 2011 NextScreen. All Rights Reserved.
Reproduction in whole or part in any form or medium without specific written permission is prohibited.
Reproduction in whole or part in any form or medium without specific written permission is prohibited.
The Ayre people told me it was significant. A reviewer I respect compared it from going from an outstanding speaker to a great one - although the diminishing returns have kicked in, it's still substantial. If you own the Ayre , $250! A no brainer. If you own the original Meridian or the Esoteric and have to sell the old one to upgrade, it's $3-$8K! Much tougher decision, then. (Makes you wonder why others are not like Ayre and take their customers for fools.)
I don't know if its necessarily taking them as a fool. I think within a couple of years if a new update is there then it would be nice to get the upgrade for a small fee. But stuff thats 3 or 5 years or older is just old. Its a lot of losses for a company to take. Do you ever hear that if someone buys a Sony TV then all upgrades will be from small fees? generally you use that same TV for 5 to 10 years.....somepeople even more then you just give it away and buy a new one with the new technology. The other problem is that Digital player technology has been changing the most rapidly in audio over the last many years, while speakers, amplifiers, cables remain relatively stable in general. For companies to keep giving massive discounts on their cutting edge products to owners of previous models does not make economic sense buisness wise. As a customer it SUCKS because you just spent $5K or $10K not too long ago...... but such is technology with phones, computers, cars, all expensive but loose value very quickly and technology running real fast. Hovland couldn't hold up the competition and with audio being such a small NON mass market comunity that these guys would not make any money if they gave $250 upgrades for cutting edge new Meridian and Esoteric technology. I do think that Ayre is a good bet between the 3 for the customers. I also think that as the above poster implies, its not a total loss, The Esoteric and Maridians of not too long ago that don't have these new filters may not be cutting edge and the latest designs but they still are among the best. The desire to keep having the best/latest at every instant is difficult to maintain. Driving a new BMW 5 series out of the lot will immediately drop its financial value but it doesn't make it outdated or useless. You still get years of great service and pleasure out of it even though the new model is just around the corner. It still remains a great product.
Sam,
What you are saying makes a lot of sense. I am not sure what I would do if I were running the business - I guess it would depend on the circumstances. Ayre must be doing fairly well while others are cooking the leather from their shoes not to starve to be able to do this.
Also, Robert Harley supposedly never heard the new Ayre. The Ayre could be as good or even better than the Meridian. It's only $6K new. It's no doubt better than the old Meridian. Meridian and Esoteric owners who own the previous model must be pissed. If I own the old Meridian, it could be a $10K hit for me! That's a lot of money for most people!!!
Maybe the Ayre guys are the fools. Or maybe they just know how to build the value of their brand.
I bought a Sony SCD-777ES about 7-8 years ago. Just after I bought it, all the new CD players offered upsampling, which is pretty standard issue these days. Your BMW analogy is correct in that there is always this sort of new technology coming down the pike. However, judging from RH's comments, this one is a game changer.
Anyone aware of other CD players other than the 3 mentioned above with the apodizing digital filter?
The PS-Audio PWT-D
Yes you are right. This is a game changer! And yes it's expensive for those who already bought something else. I almost bought a different player but by luck this thing got released at the right time. I only know 3 players that use custom software of this type and that's the new spectral cd player, the Berkeley DAC, and the new maridian cd player. From what I understood only maridian claims to do apodising filters and pre ringing fix. I believe the other players also address these issues and have written their own software but I'm not sure about the exact specifics. The makers of spectral and Berkeley are the same guys who made the pacificmicrosonics model 2 and therefore their products sound similar. I bought the Berkeley because I want to add a music server to my system. I feel for the folks who spent lots of money recently on cd players and now we have this significant new change in cd technology. I also feel that other companies will follow this trend making it more affordable for many
The Ayre player has 2 settings: Measure and Listen. When set to "measure", the filter is an apodising filter, same as Meridian. When set to listen, it's Ayre's own MP filter, derived from in house listening tests.
PS Audio uses apodizing filters...and has a great article on their website on how they work. They say that (older) linear-phase filters were *causing* the distortion artifacts we heard before. They don't claim it "corrects errors" at the ADC...which is a laughable statement by Meridian Audio and reported by Robert Harley.
That is, they don't claim that "apodizing" filters correct errors......
My understanding of Meridian's filter comes from Dr. Peter Craven's Audio Engineering Society papers "Anti-Alias Filters and System Transient Response at High-Sample Rates" and "Controlled Pre-Response Anti-Alias Filters for Use at 96kHz and 192kHz" as well as an extended discussion with Bob Stuart about the filter.
The term "apodising" is Meridian's creation, but has been co-opted by other manufacturers to mean any minimum-phase digital filter. The term "apodising" comes from a technique used in radio astronomy, a field studied by, not coincidentally, Dr. Peter Craven (see his AES paper for the parallel between radio telescopes and digital audio filters).
A minimum-phase filter doesn't add pre-ringing to the signal. Period. Meridian's filter doesn't add pre-ringing either, but it also removes pre-ringing introduced earlier in the chain (see Dr. Craven's explaination of "convolution" to understand how this can be possible). My understanding is that Meridian's filter is unique in the world, other manufacturers' use of the term "apodising" not withstanding.
I have heard other digital sources that use minimum-phase filters and they don't sound like the 808.2. In my experience, the 808.2 is a singular achivement, particularly when playing poor-sounding CDs.
If you claim Dr. Peter Craven's assertion is "laughable" I suggest that you present a technical argument to show that it is false.
What about the Meridian G08...does it not have the same filter?
From what I understand, NO. Its the New Flagship 808 series players that have this unique apodizing filter. I don't know if Ayre C5XeMP has been directly compared to this player, although in Stereophile quite a few reviewers use the Ayre as a reference and rate it with the best, regardless of what filter it uses. The reviewers in Stereophile seem more than happy with the Ayre player and some reviews that I have read of the super CD players, the reviewers claim that yes the megabuck player is slightly or more better than the Ayre but for real world audiophiles and themselves they keep saying Ayre, Ayre, Ayre is hard to beat.......I would definately give it a listen if price is an issue and see if a comparison test is available or you can audition both.
It doesn't take much to see that linear-phase filters were causing the problem. Even PS Audio says this. They had to be, if we are now *post* ringing.
All we can do in playback is reveal what's on the recording - now matter what Craven claims. One man is not enough for (me) to convince....
"All we can do in playback is reveal what's on the recording"
That just isn't as literally true as you paint it. If we can characterize a systematic distortion which happens while the recording is made, that distortion can be removed during playback. Two examples are Dolby noise reduction and lossless compression. Both rely on knowledge of the system on the creation side to design compensating adjustments on the playback side. And these examples come from just one conceptual approach. Dr. Craven may be talking about some other method that I haven't thought of. But it seems to me you are laughing prematurely.
Noise reduction does not correct errors.
Anyway, it seems that I'm right so far. Yes, we have Craven's claims...but if we stop and think about it, it's a false assertion. And RH simply went along with it, without questioning its merits. (RH does this a lot, BTW, in promoting the products he reviews).
Here's a little discussion I brought up at Audio Asylum:
http://www.audioasylum.com/cgi/vt.mpl?f=digital&m=148151
First, from the standpoint of the source file, the recording side aspects of noise reduction are an error. What makes it removable is its systematic structure. That's how error correction works, and many other schemes for processing signals to be more like the source signal. The issue raised by your "laughable" comment is whether it is ever possible to remove a distortion (error) created during the recording process. The answer is: yes it is possible.
You assert that you're right, and suggest we think about things, but you don't supply any logic (what normally passes for thinking) or math. I've given logic about why your sweeping assertion is simply wrong.
None of this demonstrates that the apodising filter used by Meridian works as claimed. It simply points out that it isn't logically impossible for the type of claim made by Meridian and Dr. Craven to be true.
Along those lines, the audioasylum thread primarily consists of Charlie Hansen (of Ayre, I believe) suggesting that true apodising filters do not sound better than very good non-apodising filters. That seems entirely possible, but it is another subject in turn. Almost any signal processing will have side effects, and it seems to me quite possible that the apodising filter has some drawbacks. If I remember correctly, RH described some of these in his Meridian review (though it is probably not possible for him to know if the issues relate to apodising or something else).
We're not talking about analog !! You're sweeping assertions don't hold true for digital - once embedded with an error, it's permanently encoded in the signal. PERIOD. Playback will simply reveal what's on the recording.
Error correction replaces bits lost in *playback* - not in recording !! The Nova Physics (memory playback) people are very big on this one.
Any pinhead can see that it was *playback* doing the damage. If guys like Charlie Hansen say that we're not "correcting" errors, then I'll believe them, not the inventor-promoter of the technique.
Not sure what axe you're grinding, but let's remain calm.
Perhaps I'm mistaken, but I don't see why the digital encoding of an analog signal couldn't have systematic artifacts that could be removed during playback. The decoding of compressed files seems to do this. Another example would be encoding a pre-emphasis which one later removes. These are not sweeping assertions, they are specific examples of systematic and predicatable errors that can be reversed. Again, I don't know if ADC pre-ringing is systematic and predictable in a way that can be reversed. I'm only giving examples that I think show it isn't logically impossible as you perhaps casually claimed.
I can imagine that playback has been problematic. I don't seen any reason why recording mightn't have been problematic too. Maybe that's because I'm not a pinhead?
As for Charlie, who I assume knows a lot more about this than either of us (though he probably isn't omniscient), he says:
"It is this apodizing filter that is *supposed* to filter out problems created by the ADC."
"So whatever problems may or may not exist during the ADC process, I don't think that the apodizing filter cures many (if any) of them." His support for the latter is based on listening tests which show that the benefits of apodising are less than the benefits of the Ayre filters. If we think about what he says, he certainly doesn't say it is impossible to correct problems in the ADC process (if he allows, as he does, that apodising filters might correct some problems in the ADC process then it obviously isn't impossible in his mind). He does say earlier in the audioasylum post:
"But there is no claim by anybody that minimum-phase filters by themselves can "fix" anything." But that is a statement about minimum-phase filters, which I believe are intended to avoid introducing pre-ringing. Apodising filters are another matter as I read it.
I'm not sure why you're so insistent on defending an unproven claim......
Digital recording is different from analog - you need to understand this. *Anything* encoded into the signal (inc. jitter) is permanently part of it. PERIOD. But Hansen is not convinced that "ringing" is added to the signal at the ADC - this is purely conjecture. Even RH was hinting (in his review of the 808.2) that playback filtering might be doing the damage. Actually, his statements were not clear here - check out his review to see what I mean.
Digital is an incredibly accurate, zero distortion, zero noise recording system that has capturing rates (since the late 1970's) that cover the range of a large orchestra and/or human hearing. Whatever problems they had with it, they took care of by the early 1990's. PLAYBACK was the killer !!!!
I'm not sure why you're so insistent on defending an unproven claim......
Actually, he seems to be keeping an open mind, whereas you seem to have rejected it outright without any proof.
Ringing at the ADC has *never* neen proven. That's a problem to start with.....
I should say *audible* ringing at the ADC....
While I completely agree with you that many people improperly assign characteristics of analog recording to digital (for example, making claims that two CDs with the same lands and pits, representing the same binary words, can somehow sound different), I think I have to disagree with your assertion that ANYTHING encoded into the signal is permanently part of it. It may be permanently part of the original digital stream, but it can be "recognized" after or while being converted to analog and then "rewritten" back to digital, even if that's in live time.
This was proven to me recently by listening to a BBC documentary about the Beatles remasters. In the documentary, they took single tracks from the multitrack (for example a single track that contained drums, guitar and bass) and were able, using digital techniques, to re-isolate the instruments, even instruments that occupied the same frequency range. I'm an ex-recording engineer and I do software development and even I didn't think this was possible other than using narrow EQ filtering, which doesn't do a very good job. But it apparently is. If one can do that, it seems to me that techniques could be developed that would "recognize" harmonic distortion and other artifacts and correct for it.
I've heard Robert talk about this on TAS conference calls. My clear impression is that RH thinks that minimum phase D/A or other techniques that remove pre-ringing are an important step forward. That's why the Spectral, Berkeley, Ayre and others with this feature are getting such good reviews. His point about the Meridian is that it does something in addition. Since we've had two steps forward in D/A at about the same time, it is important to note that the wonders of apodising filters shouldn't take away from the benefits of minimum phase filters and the like. Apodising is icing on a recently improved cake. And, as we know from other technologies, there may be tradeoffs in going the additional step that Meridian takes.
AVGuide webmaster and general drudge
I don't think Berkeley or Spectral are using minimum phase filters.......
Maybe not. Berkeley doesn't seem to make this claim, but then again they don't exactly explain what their approach is. Spectral doesn't seem to describe their filter approach in detail either. But, RH reports that the filters are designed to eliminate pre-ringing; hence my somewhat casual grouping.
AVGuide webmaster and general drudge
Alright, I'm confused. If a new technology such as the apodising filter is a "game changer," shouldn't those products without it pale in comparison to those with it. For example, the Meridian G08.2 which has the new filter received a good but unspectacular review (by Messenger or Gold) yet some of the best recent products (Berkeley DAC, Esoteric X-05) without the new filter are described in rave reviews as having closed the gap between analog and digital. Should the correct conclusion be, then, that the new apodising filter is a significant but incremental advance and that the Meridian 808.2 is merely a singular achievement, not the most significant product in the CD era?
Sorry for the confusion. RH can wade in with his view, but I think I can shed some light on the issue. RH said in his review in TAS that the 808.2 was a singular acheivement in regard to its wholesale transformation of the CD experience via its increased sense of three dimensionality and space between instruments. Reading this entire statement we see that he doesn't say the 808.2 is better at everything. He reinforces that point when he critiques the 808.2's bass and dynamics.
A second area of confusion may stem from the general caveat that any review we print has to be read with the caveat "in my experience" applied. Obviously, RH hasn't heard every competing product, in all possible systems, perceived through the sensory apparatus of all listeners. So, when RH says the 808.2 is a singular acheivement, I believe he must mean that in the context of "in my (extensive but not unlimited) experience". I think Gadgetman is allowing for that necessary truth above.
Thirdly, and sorry if this adds rather than subtracts from the confusion, since the 808.2 addresses both D/A pre-ringing and A/D pre-ringing it is hard to sort out exactly which technology is doing what. RH reports on Meridian's theory of why the 808.2 is different and then correlates that with the sound he hears. Really, he doesn't know if the apodising technology described by Meridian is the cause of the sound. So, it is possible for example, that Meridian's approach to D/A pre-ringing is causing all of the benefits RH heard. Let's just assume that the benefits of apodising beyond that are nil. If this were the case, then someone else might have an equally good approach to pre-ringing, without apodising, and could accomplish the same or greater benefits as Meridian. Robert has made this comparison of the Meridian with the Berkely and Spectral products (which address D/A pre-ringing) and has some evidence that the Meridian produces results that these D/A converters don't. But that still doesn't prove that apodising is the difference, it simply supports the idea.
I don't know why other reviewers didn't hear what RH and I heard. Maybe they didn't do the comparisons he did. Maybe there are drawbacks to apodising (see point the first above) which dampened their enthusiasm.
Final thought, which is speculative on my part. When RH says the 808.2 is the most significant player in the history of the CD, I think he is speaking of the technological direction it embodies (and for which it seems to be the practical origin). The 808.2 technology holds out the promise that two major distortions, once thought endemic to red book CD, can be very successfully addressed. If further understanding reveals that apodising (beyond removing D/A pre-ringing) is contributing nothing to what we hear from the 808.2, then the "most significant" statement should be transferred to whatever specific technology proves to be the cause.
Hope that helps.
CEO and Editorial Director, Nextscreen LLC
JA says that the Boulder disc player is *very* close to the 808.2. It's probably the D/A ringing.....
The Spectral and Berkeley filters are both minimum-phase types—see my interview with Keith Johnson in TAS Issue 190 for his comments about the Spectral filter.
Incidentally, another example of correcting on playback a distortion introduced during recording is Keith Johnson's "conjugate" filter in earlier Spectral players (the SDR-1000 and SDR-2000) that correct for the phase shift introduced by the A/D's anti-aliasing filter.
I disagree with the JR-1's assertion that "Digital is an incredibly accurate, zero distortion, zero noise recording system that has capturing rates (since the late 1970's) that cover the range of a large orchestra and/or human hearing."
Well, digital *is* zero distortion and zero noise - Malcolm Hawksford says this (Stereophile March, '96 issue). And it does cover the range of a symphony orchestra - how does it not ? It covers human hearing, too. No study I've seen says we can hear above 20kHz.
They were aware of (all) of this in 1978, save the actual distortion numbers (which are probably lower than they thought). CD is not based on "1970's technology" - it is based on the fundamentals of sound and hearing. We did need a little more room in production - that was unforseen...but solved in the early 1990's.
I want to purchase digital front-end in the 5-10k range. However, I am now "paralyzed" with uncertainty...the last thing I want to do is to purchase an expensive player (i.e., Esoteric X-05) and then discover shortly thereafter that the new digital filters are "game changers." Aaarrrggghh.
Why don't you audition the XO5, c5xemp,and alpha DAC and decide on what you like most? All are 5 to 6k range. Othewise you will be waiting for a long time. And that's fine if you like that. Seems like u r looking for the last player ever. And that's hard to do since these filters are kind of new. And I think more of it will be coming out. I can tell u this much that the alpha is jaw dropping I couldn't believe how good it is after hearing it. It also has a preamp with it. Flexibility to add music server, iPod, cd transpot. Quite a steal! I would audition and choose between these 3. If u can go between 15 to 20k audition the meridian or spectral. But don't expect they will stop researching and not improve over the current models and filters
Sam,
I don't have the option to to hear or demo any one the three. I also understand that technology will continue to evolve resulting in better players, but I fear that based on RH's review of the 808.2 that these new filters address a fundamental distortion...wouldn't I be making a mistake buying a 5-10k digital player without this new technology that led RH to proclaim the 808.2 the most sigificant product of the CD era?
JR-1's claims seem to be self limiting and make no sense at all except that he thinks what he has understood is right and everyone else is wrong. Not an uncommon thing. You don't even have to try state of art players 10 years ago vs these newones to tell how far cd has come along. I have used some very expensive and cheap players over the last 20 years and the differences are beyond obvious. The players RH mentions are jaw drawping good. I wonder if JR1 has auditioned them. Further more the players are made by companies that are at cutting edge of CD technology and that too at it for 20+ years. Now would I trust the sound and interview of maridian guy or some random self expert jr1. You be the judge. I hope we can expand more on the details of these filters and the post topic rather than "I say so that's why it's correct". And I'm not trying to pick up fights. I just find jr1's assertions silly!
What specifically is silly ?
JR-1, are you saying that quantization noise doesn't exist? What happens when the value of the analog waveform happens to fall midway between quantization steps? The quantizer selects the closest digital value. The differerence between the original analog value and the value assigned to that quantization step is manifested as quantization noise.
You also make a fundamental mistake in saying that 44.1kHz sampling is fast enough because we can't hear above 20kHz. Although we can't hear sinewaves above 20kHz, musical signals have a steepness in transient attack that 20kHz bandwidth can't adequately capture. Read the HDCD patent application for an exposition on this subject and talk to Keith Johnson about the sonic effect of band-limiting the signal to 20kHz. Moreover, higher sample rates improve the system's time-domain performance—read the AES papers Mike Storey (founder of DCS) presented in the early 1990s.
The CD specifications were set in the late 1970s and were dictated by the limits of the technology at the time.
As for JLeeMD's quandary, I just received for review the new Esoteric SA-50, a $6000 CD/SACD player with a minimum-phase filter.
Quantization noise exists only in certain recordings (w/ wide dynamic range) and only then after signal processing. Bob Katz is on record for stating that 16/44 is "seemingly enough" for original-source material (before intermediate calculations are done). If you don't believe me, e-mail him at his website. Bob is one of the most cordial persons I've corresponded with. And why is Peter McGrath "hard pressed" to hear beyond 44.1kHz. for his masters ?
Like I stated before, most of the noise (pun) for higher sampling rates comes from Keith Johnson. Yes, there was quantization noise...but problem solved years ago when they lifted the capturing rate. Do you get this ? This absorbed whatever problems they had !!
16/44, as a consumer-disc format, *must* be enough if the 808.2 is showing that CD can surpass SACD and equal 24/88 in sound. How close, RH, was the 808.2 to the Hi-Rez Berkeley ? I think you're afraid to answer. Was the 808.2 better than SACD ? Nah, you won't answer that one either....
CD's specs, as far as humans can hear and orchestras produce, are Hi-Rez. This, as long as they record a little higher for large dynamic-range music. The 808.2 proves it. Get over it......
I eagerly await your impressions. I am most curious to know how this new filter improves the sound and musicality of an Esoteric SACD/CD player. Any word on when they will implement the MP filters into the rest of their line?
Always something new will come a long. My dealer told me the 808.2 is very good but in his opening my Marantz SA-7S1 is just fine and he suggested money could be spent better else where in my system. Filter 3 on my Marantz has short pre and post ring filters. I still enjoy filter 1 though, 3 is more detailed though and on some discs quite enjoyable.
In the end like LP's it gets back to the mastering. Todays cd's are sounding better than my LP's in some cases. They got the mastering down pat and they now understand how to use digital. Early days they were just using LP eq's, they found out qucik that was a big no no.
I am also amazed at how standard CD's sound on my Marantz to the point where SACD don't really standout as much. I just listened to a pure DSD recording (not SACD) on Telarc and it just don't get much better than that and I been collecting music from the time I was 8 years old and I know my music.
So much can be done with digital to the point where you just may lose the intent of the recording with all the processing, upsampling that is possible to do in this digtial day and age. If that makes you enjoy the music more than it is a good thing I guess.
I'll post my early impressions of the new Esoteric SA-50 after I have some time with it. I wouldn't be surprised if Esoteric (and other manufacturers) move toward minimum-phase filters throughout their lines.
As for JR-1, I'm finished debating him on this subject. If he continues to believe that 44.1kHz/16-bit is is perfect, that's his choice, and no amount of evidence or argument will persuade him otherwise. In the meantime, I'll be enjoying the spectacular sound of 176.4kHz/24-bit HRx files played from my fan-less, drive-less music server through the Berkeley Alpha DAC. This is the state-of-the-art in digital audio, not CD.
But as an owner of the Meridian 808.2 who *also* plays HDTracks via DCS decoder, I don't hear much of a difference !! It's an extremely close call with the Hi-Rez edging out CD in lower-midrange bloom. But it's so close, some folks don't hear a difference (here) were I do.
You could have compared the two as well, but chose not to...OR...not tell us. And I understand - with CD sounding *virtually* as good as Hi-Rez, you're boxed-in and don't know where to go. And with more and more CD players sounding like this (read Atkinson on Boulder), you'll have to cave in at some point in the future.
Again, I'm not hearing how the CD standard was "compromised" or what "1970s technology" means. Like so many of your other statements, you're vague. As far as they knew in 1978, Red Book covered the entire range of sound - as dictated by measurements of a symphony orchestra and human hearing. These parameters have *not* changed....
RH-
What software and hardware (sound card or other device) do you use as interface between you server hard drive and the Berkley?
I'm asking as I also use a server and am starting to buy hi-res files
RH:
I apologize if this question is "Digital Audio 101." I have a Sony SCD-777ES SACD/CD (single-ended version of SCD-1) player which I suspect would serve as a good transport. Are you saying that were I to purchase a Berkeley DAC instead of, say, an Esoteric X-05, and used 176/24 files, this set-up would sound better than an SACD? And are there as many titles available c/w SACD? How much should I budget for a Music Server?
JleeMD where do u live?
Unfortunately, there are very few true high-resolution titles available. Many of them touted as "high-res" have simply been upconverted from 44.1/16 masters. Once the signal has been in the 44.1/16 domain, the damage has been done (despite JR-1's belief that 44.1/16 is transparent). I would buy the playback system that sounds best on the largest percentage of your music catalog, which I suspect is CD, followed by SACD. I would not buy a music server just for the few titles that are available. I bought the music server because it is an invaluable reference for judging the performance of downstream gear---DACs, preamps, amplifiers, and loudspeakers. To have a source with such high resolution, timbral purity, and dynamics is helpful in judging the quality of the other gear.
Does this answer your question?
I wrote my reply to JLeeMD rather hurriedly (my wife was waiting for me to take her to dinner) and later thought of some things he should consider.
If you buy the Berkeley DAC and a music server, you get the benefit of the Berkeley's terrific sound, high-res capability, and ability to drive a power amplifier directly (if that works in your system). You also get the convenience of a music server. On the other hand, this system lacks SACD compatibility. The music-server function can be a liability is you don't like messing with computers to play music. If playing SACDs is important, I'd go with the Esoteric. I should mention that the SA-50 was designed with music servers in mind---it has multiple digital inputs, high-res capability (192kHz/24-bit), a USB input (48kHz/16-bit), and can drive a power amplifier directly.
I'll have some comments soon on the SA-50's sound with CD and SACD. I have on hand for SACD comparison a Sony SCD-XA9000ES.
Hi!
If you'd like a pro-grade DAC (the same DAC used at George Lucas' Skywalker Sound Recording Studio), look at our Benchmark DAC1 HDR. It has all of the features mentioned above, with a much better USB input (96 kHz, 24 bit). This is also the same DAC that is used by Grammy-Award winning mastering engineer Gavin Lurssen, who T-Bone Burnett has master all of his productions, including Album of the Year "Raising Sand" from Robert Plant and Allison Kraus.
Robert Green reviewed the DAC1 PRE for The Absolute Sound recently (read the review here), saying it was "excellent compared to anything at any price point".
All the best!!
-Elias
There can't be much "damage" to 16/44 if Bob Stuart himself said (in his most recent TAS interview) that we CAN'T HEAR A DIFFERENCE between 16/44 and 24/96 - on playback. That is, comparing high-res to the 808.2 CD player. I agree !!!