Upsampling the way to go? ?


As if we didn't have enough to decide with the format wars, the latest issue of Stereophile implies upsampling is the magic to make cds as good as sacd. ARC however, disagrees. Has anyone actually listened to the ARC CD3 vs the MF NU Vista 3D,Cary, EMC 1,or other comparably priced players with upsampling?
tonyp54
Audio Note as well as 47 Labs etc also disagree. After extensive listening here I prefer no upsampling. That is my humble opinion. (as well as many customers who have come over armed with every DAC imaginable).Others will of course have different views.
I haven't heard any of the players you list long enough to comment on them, but I have been listening to upconverted CD for years through the dCS Purcell and Elgar, and have heard other comparable upconversion systemsas well (e.g., the Accuphase). I have never heard any CD player without upconversion that can come close, although I believe that when I heard the Linn there was something wrong in the system, because LP didn't sound very good there either. I heard an Audio Aero Capitole player recently, and it also sounded fabulous.
I used to feel that way also. If you are ever in the area bring your DCS over and we can have some fun.:)
I have upsampling on my Integra Research RDC-7 . I like it alot. I do not have SACD yet as I am waiting for a worthy universal player. I do however have DVDA. I got the DVDA before the Integra. When I first heard the DVDA , on certain well recorded discs , I was very impressed with the resolution. When I got the Integra , I got used to listening to my better recorded cd's with upsampling. Now the resolution difference is very close. CD's are a very bright format. So I got a Classe amp. Now Im in sonic heaven. Mix upsampling with a Classe or a tube amp if you are so inclined, and you will be a happy camper. That does not mean Im anti SACD. I will get a player and some discs. But when you have a massive CD collection, and a limit on the number of SACD's that your actually interested in, Upsampling done well can be very satisfying
I recently upgraded my Leveinson #39 to the new #390s. The #39 was an excellent sounding player, but the 390s is a big improvement. The 390s increases the sample rate to 352.8 kHz or 384 kHz and is the main reason for the improvement.
I think that both 16 bit and upsampled 24 bit can sound very good. Most of the "problems" that we hear are not with either format, but with the amount, type and how the filtering is used in each design. I've tried it both ways and i prefer upsampling. Then again, my upsampler doesn't have any filtering so that may be part of the reason that i like it so much. The non-oversampling ( 1X ) Audio Note's also don't have any filtering and that may be why so many people think that they sound so good also. Sean
>
The upsampling is only as good as the implementation. If the implementation isn't done with care the added circuitry can easily overwhelm any benefits of upsampling. Better to buy a well-executed non-upsampling DAC than a mediocre upsampling DAC. Also, it's important to understand that upsampling is not analogous to interpolation where the word length is increased to 24 bits or so--they are two separate things and should not be lumped together. I'm drunk and going to sleep now--night night.

Tim
Tim, I thought interpolation *was* "upsampling", whereas increasing word length was accomplished by adding randomly generated dither below the signal level, so it sounds to me like *you* might be confusing two separate things. But then again, what the heck do I really know about this stuff?! Not a lot, let me tell you!!

To address the main question, I just yesterday received a new-to-me MSB Gold Link upsampling DAC bought here on the 'Gon, to compare with my reference Theta DSPro Basic IIIa (Theta also seems to disagree with the theory). I've just started auditioning, and will post my findings when I've had enough time to make a meaningful decision about which unit I will keep. But I can tell you right now that the MSB with upsampling engaged definitely sounds different not only from the Theta, but from itself with the upsampling turned off, so it ain't doing nothing. Finding an upsampling-capable unit and auditioning it against your reference may well prove to be an educational experience - I think it will be for me.
Zaikesman--I'm not an engineer and I could be wrong, but upsampling is upsampling and is totally separate from increasing the word length, which I believe is done via an interpolation algorithm. I was told this by Rick Schultz at EVS who has designed his own DAC and has modified a few in his time. Many DACs claim to upsample to 24/96 or even 24/192 but do not really interpolate to 24 bits. The Crystal chip does do both and may be what is used in the MSB DAC--I'm not sure. Hey, if it's working it's working, whatever it is. Better systems are used by Audio Aero(STARS) and dcs, but they obviously come at a premium.

I still maintain that an upsampling DAC does not necessarily sound better than a non-upsampling unit--it depends more on how the whole DAC is designed and implemented, and there are a LOT of variables there(i.e. Sean's point about filtering, damping, shielding, parts quality, power supply, circuit design, etc.). I'd say just pick the DAC/CD player that sounds best to you--if it happens to upsample great, if not that's fine too. Personally I think Sam Tellig has gone way overboard on this upsampling thing and is misleading many to think it's the second coming when it's only one small link in a very long chain.

Tim
I have a Cary 303/200, and just recently began to use the upsampling feature, because I wanted to get a feel for how the unit sounded in the standard mode. To me, there is only a subtle difference; and I am not sure that I like it. There seems to be more extension and "air" to the treble, but at the expense of losing some fullness in the mids. Am I too quick to judge--Does the upsampling circuitry have to be "broken in" as well?
Yesterday(6/29).I listened to the MF DAC unit that Sam Tellig raved on about recently.It was hooked up to Mark Levinson & B&W gear. Upsampling was different but not all that great. warmth? no I'd say more like fuzz,like a badly recorded live club perfomance.CD in question was solo piano on Naxos--- it sounds just fine on my MF A3 player.
Not to prolong the point, Tim, but why would an interpolation process be needed to generate an extra 8 bits of *random* dither? As I understand it, this dither acts as sort of a "digital bias", and is not related to the actual signal. I think the increase in word length simply helps the DAC remain linear deeper into the noise floor. It does seem to me, however, that interpolation would be appropriate to aid in generating the "extra" samples needed to raise the sampling rate from 44k to 96k, or whatever higher rate is chosen. With only a finite frequency of voltage samples available off the disc, some sort of algorithim that interpolates new sample voltages in between the recorded ones would have to be implemented to raise that frequency, no? So I agree that upsampling is separate from increasing word length, but wonder if you might be confusing which of the two processes incorporates interpolation algorithims. (I already know that *I'm* confused!) And since I'm certainly no digital engineer either, I hope someone who is can come along in this thread, and soon!
Zaikesman--I certainly don't mind this point being prolonged because I think there are a lot of CD players and DACs out there that claim to upsample to 24/96 or 24/192 and really don't increase word length in the more strict and proper sense(whatever that is--help). I too would appreciate anyone who knows more about the process by which upsampling is done vs. increasing word length, and maybe even give us a way to identify the "pretenders" from those that truly do increase word length along with upsampling--there seems to be a lot of misunderstanding out there regarding this point.

My sense is that upsampling by itself will not make a huge difference and could do harm if not implemented with care due to the added circuitry involved(some of the above comments seem to bear this out to some degree), but that combining it with a longer word length can bring about even more dramatic changes if done properly. Where's an engineer when you need one?

Tim
Tim is correct ... upsampling refers to increasing the sampling frequency from 44 kHz to a higher number, and using the extra samples to interpolate.

The benefit is that the anti-aliasing rolloff filter is much simpler to implement since it can rolloff more gradually (since the sampling frequency is now much much greater than the maximum frequency recommended ... it's all to do with Nyquist's theory of sampled signals).

At the same time one can also increase the word length of each sample by adding dithered noise. Not extra information .. noise. Dithering is a wierd concept where adding some noise increases the perceived clarity. I'm not so familiar with it ... but as an experiment try looking at a pixelated image and then blur your vision and notice how the image becomes much more recognisable.

Neither upsampling nor adding dither and extra bits can add any information that was not originally present in the recording, neither can they increase the theoretical maximum dynamic range ... instead I think they help to mitigate some of the real-world problems associated with extracting the signal (the roll off filter being the biggest problem.)

Of course a well implemented non-upsampler will outperform a bad upsampler ... probably as much to do with the electronic design, analog output stage, noise from digital bleeding into analog .... it's just good design vs bad design. However since there are plenty of people out there who just have to own the latest buzz words upsampling will sell. Whether a good upsampler will outperform a good non-upsampler ... I really don't know ... probably down to individual taste.

This reminds me of the bitstream vs multibit arguments of about 10 years ago. Both can be done well .. both can be done badly.

All this said (and I have neither heard an upsampler nor SACD) I cannot imagine that upsampling can beat SACD since SACD encodes more of the original information and upsampling appears to "fake it". Perhaps SACD real world implementation has not yet quite caught up with the theoretical possibilities ?
I am biased as I own the Delius and Purcell however dCS does have some good information on their web stie.

dCS White Papers

Cheers,
Dan
Sean T., sorry to second-guess, but your first sentence strikes me as being just a *little* bit confused itself! :-) Tim and I don't disagree about what upsampling is (as opposed to increasing word length), we are just, again, a *little* confused about how it happens!

You say the upsampling "...uses the extra samples to interpolate." That seems backwards to me. Wouldn't upsampling have to *use* interpolation to *arrive* at those "extra" samples? Then the DAC would use *all* the samples, including the "added" ones, to reconstruct the wave? Or, yet again, do I "misunderestimate" this process completely? My new MSB Gold Link DAC mentions in the manual that "oscillators" are used to upsample the signal. My dictionary defines "algorithm" as any mechanical or recursive computational procedure. "Recursive" is defined as descriptive of a mathematical expression like a polynomial, each term of which is determined by applying a formula to preceding terms, or of the formula that generates the successive terms of such an expression. The definition of "mechanical" that would seem to apply in the case of an oscillator is "governed by mechanics". Sounds simple, doesn't it?

The Gold Link comes with two oscillators to choose between, and seems to sound better on CD's with the upsampling set to use the one at 132KHz instead of 96k, which is recommended by MSB since 132k is an even multiple of 44k. The math has got to come out better this way in a "mechanical" process, no? Still confused, Z.
Zaikesman ... I think you're being a bit pedantic. Increasing the sampling rate is the goal, and it is necessary to have some sort of interpolation in order to be able to assign a value to those new samples. One cannot interpolate without increased samples, one cannot increase the samples without interpolation of some sort.

Word length is a completely separate thing.

Using a multiple of 44kHz would seem to me to make the job of the upsampler easiest, especially with linear interpolation, since the original samples are preserved ... but I think a non-multiple could also be made to work, albeit with a bit more complexity.

Of course oscillators are used ... an oscillator is no more than a clock ... so they are saying they have extra clocks within the DAC ... I'm guessing from this that they are stating that they generate their own clocks at the higher sampling frequency instead of deriving them from the (potentially less accurate) digital source signal. Any reclocking DAC does the same.

I try to describe things as clearly as I can ... your response seems to be worded in an deliberately obscure way. Is there are reason for this ?

Actually I think Sean's (the other one) original post was spot on. The format is not the problem .. it's the implementation. That said SACD is a superior format, if not yet a superior implementation.
No, Sean T. - pedantic is *all* I'm being! Sorry if my writing seems obscure, it's not deliberate. I just tend to write densely (or is it that I'm just a dense writer?!).

Or could it be that maybe you're just not closely reading my posts (certainly your perogitive)? Your first three paragraphs above essentially reiterate things I believe I've already stipulated throughout my posts on this thread. Thanks for trying to be of help, though. :-)
Zaikesman ... your accusations are quite valid ... I usually don't have much time to do a thorough read. Re-reading the thread we are in close agreement.

Thanks also to dan2112 ... this white paper was very interesting. Perhaps if I get the time and money I'll have a listen to an upsampler. However much SACD is theoretically better I'm certainly not in the mood to replace several hundred CDs, especially since many were (unfortunately) digitally recorded, probably at 44kHz !

http://www.dcsltd.co.uk/papers/aes97ny.pdf
Anybody interested in reading about the results of my auditioning of my new upsampling DAC against my reference that I referred to above, should be able to see it shortly on the forum in a new post titled Upsampling Put To The Test (hopefully it'll show up there later today). It's a long article though, don't say you haven't been warned... :-)
The Theta III would not be the one I would use as a comparison as to upsampling and non upsampling devices. I have always found them to have great bass and dynamics but not much in the musicality dept. We have compared MANY upsamplers to the Audio Note DAC's and it has always been no contest. We have done many shootouts here with customers and it doesn't take but 5 minutes to know that at this point anyway, upsampling is a great marketing ploy. We have extended the offer numerous times to allow any Audiogon member in the Bay Area CA to bring their upsamplers over and compare to non. No one ever takes us up on it. It seems people like to throw technical mumbo jumbo around as to benefits of one over the other. Or they compare a new upsampler to some antiquated DAC and say wow what a difference. Bring it on:)
I will state the rules again: Be open minded. I will be. Lets have fun and you can report back here since my views will be taken with a grain of salt as a dealer. Who by the way has sold and heard many a DAC.
After reading your test I applaud your initiative. And even though I didn't think the Theta would slay the upsampler I was proven wrong on that assertion. Good job and I agree completely with your results. Good job!
I actually had the opportunity to listen to the Musical Fidelity Nu Vista cd w/the intergrated MF,(sorry can't recall the model #,along with the AR CD3 W/LS16 ARC PRE&100.2 ARC AMP.Both systems were running w/ the Vandersteen 3a Sig's and Volcano AQ speaker cable.First off my initial impresssion was that the MF system was really warm and a bit too bloomy in the bass.In fact it was so bloomy I felt as if I were in a fog .Details in the mid's and highs were really masked ,needless to say I didn't care for the sound.When the dealer switched over to the ARC setup the detail became more pronounced and the bass had much more dynamics and overall punch.To my ears it sounded much better and less fatiguing .Now given the right equipment I'm sure you could get a better result w/the MF setup,I think w/ARC, Vandersteen and AQ wiring the syngery is really strong,hence the better presentationIMHO.
I actually had the opportunity to listen to the Musical Fidelity Nu Vista cd w/the intergrated MF,(sorry can't recall the model #,along with the AR CD3 W/LS16 ARC PRE&100.2 ARC AMP.Both systems were running w/ the Vandersteen 3a Sig's and Volcano AQ speaker cable.First off my initial impresssion was that the MF system was really warm and a bit too bloomy in the bass.In fact it was so bloomy I felt as if I were in a fog .Details in the mid's and highs were really masked ,needless to say I didn't care for the sound.When the dealer switched over to the ARC setup the detail became more pronounced and the bass had much more dynamics and overall punch.To my ears it sounded much better and less fatiguing .Now given the right equipment I'm sure you could get a better result w/the MF setup,I think w/ARC, Vandersteen and AQ wiring the syngery is really strong,hence the better presentationIMHO.
JC - Thanks for the kudos and backup. As I took pains to point out in my new thread, what I wrote wasn't supposed to focus on which of these DAC's 'slayed' the other, although of course for my own purposes this determination was important (and I also didn't mention the fact that the Basic prevents me from eventually using my DAC with formats other than 16/44 CD, which the Link handles). Rather, I wanted to examine in some kind of sensible way how it was that I could have experimented with an upsampling DAC that has been well-thought of and reviewed, and yet still come to the conclusion, both subjectively and objectively, that the upsampling process, now matter how subjectively pleasing it might have been on certain material in the short run, was basically degrading in its result, and seemed somewhat spurious and arbitrary in its implementation.

The fact is, I have always agreed with you about my Theta, believing that it had many discrete 'hi-fi' virtues, but was not the most 'musical' DAC in any overt way. In my new post, I deliberately omitted my lengthy listening impressions so that I could zero in on the issues I wanted to raise (I actually went on for several paragraphs about what I heard from the Gold Link and the Basic playing music, and all their strengths and weaknesses, in my original version, but wound up erasing all that, as it took too long, and wasn't really pertinent to my main points).

But here I will briefly say that the Basic IIIa has always impressed me as being a DAC that features a sound I would describe as having great clarity, power, and fullness. It shows a large soundstage with solid objects that are well-separated. It has a high degree of tonal neutrality where nothing predominates, but does possess notable air and weight. Dynamics at both ends are fully expressed. Transparency and detail are both of a high degree. It is not obviously smooth in its presentation, and adds no warmth. Tonal colors have strong intensity. Transients, with a clean digital feed, are very concisely portrayed, with no smear or overhang. I can hear very little in the way of phase or timing anomolies. No haziness or flatness. So what's not to like?

Well, as I mention in the new post, the Basic has always struck me as being slighty cool and dry in character. Maybe a tiny bit bright, with a little emphasis on leading edges of sounds that can seem somewhat clinical. There is a very slight background texture of a fine-grained 'digital burr', that sort of rides over the top of music like a light scrim or a fingerprint on glass, which is not really obscuring, but is there if you listen for it (which, to be fair, is true of most CD sound I have ever heard, unless some greater form of distortion or omission covers it up). It is not in any way romantic or intimate in its sound, but very matter-of-fact. It sounds correct, but not terribly personal.

One of the important things my controlled bypass test seems to have shown me, however, is that most of what I assumed were the somewhat less than completely captivating aspects of its character, are in reality probably an accurate representation of what's on the CDs. These caveats basically do not make an appearance when doing the controlled real-time comparision. They are not, I now suspect, intrinsic to the DAC, but rather to the majority of available disks. If another player could successfully render disks in a subjectively more sensuous way and still score as highly on the objective bypass test, then that would be an advance, and I concede the possibility - probably for much more money though. And I still fully sympathize with the subjectivists' wanting to love their CDs' sound, not just respect it. That desire may cloud one's judgement, however. I can't love anything (or anyone!) that isn't essentially honest with me, because that's the only way I can place my trust in something. I'm not blindly partial to my Basic, but I haven't ever been let down by it either. (And I can afford it!)

P.S. - To anyone investigating one of these used, you'll need a quality transport, jitter-box, balanced power conditioning, and digital interconnects to hear its best, in my experience.
The link below will connect you to an article on exactly this topic from hi-fi news and record review. It is informative.....at the very least.......

http://www.aslgroup.com/dcs/upandover.htm
Why go upsampling when you can have 24/96 native resolution in a up and coming format? Check out the HiFi Farm Newswire in this months Stereophile for some great commentary by the leader in hi-rez recording. Some upsamplers for your redbooks work well,some don't . It totally depends on who's it happens to be. The future is on the horizon!
When comparing upsampling, which I was championing, to one bit sampling, I was astounded how much information was lost with the upsampling equipment auditioned. The very information which provides the layering of the sound stage and air about the performers.
Some of these go back a ways: MBS Full Nelson, Millenium #1, Dodson DA-217 MK IID, 47 Labs, Audio Logic, Kora Hermes, and Audio Note and at least a couple of others. BTW, the Audio Note, the cheapest DAC 1.1, SOUNDED the best. I have the Audio Note 3.1 Signature. Jim.
Up and oversampling ala GamuT CD-1 Danish CD Player- better then twice DSD sample rate. $2950, see review in The Absolute Sound. I just got mine, and am loving it!!!!! Huge 3 D sound stage.
With respect that's nothing but a number. I believe the earliest bitstream DACS had a sampling rate of 11.2 Mhz, 10 years ago.

I've also heard it said that soundstage is simply a product of frequency response, re/up/over/whateversamplers thus simply alter the tonal balance to achieve a "better" (according to the tastes of the beholder) sound.
Anyone interested in reading a little more about the process of upsampling, oversampling, interpolation, dither, etc. and how it can benefit(or hurt) the musical signal may find this interesting from the Perpetual Technology folks(see link below--sorry if someone else already referred to this). I found the Level III explanation(see link at the bottom of that page) to be the most interesting and it doesn't take all that much time/effort to read. What I took away was that it seems like it is the companies that go to the trouble of creating proprietary interpolation algorithms to meaningfully increase word length(i.e. not simply just adding dither) in addition to the upsampling that are the ones that are really taking CDs to a higher level with this concept(obviously a lot of non-upsampling players sound fantastic too). In addition to the PT P1A, the Audio Aero Capitole(STARS system) and I assume the dCS Purcell do this to great effect--anyone know what Electrocompaniet does with the EMC1(24/192)? This makes sense to me, but it also again makes the important point that all upsamplers are not created equal--not by a longshot.

http://www.av123.com/pt_interpolate1.html

Tim
This has been interesting reading, but no one has answered the question if they have heard a ARC CD3 compared to one of the upsamplers. To me the truth is still in the listening, not the stats.
Tony

I directly A/B'd my upsampling 24/192 EVS Millenium II to an ARC DAC3 and CD2 and felt that the Millenium showed much more detail, low level resolution and presence.

Haven't heard the CD3, but my results were amazing compared to those two ARC products, which sounded flat compared to the EVS (and I own ARC amp and pre)

Tom
The only comparison I've seen between an upsampling CD player and the CD3 is the one below from audioreview.com comparing it to the Electrocompaniet EMC1(24/192 version)--for what it's worth. I know the Electrocompaniet EMC1 is highly thought of and the fact that the CD3 stays with it in this comparison is a good indication that it will hang with well-executed upsampling players, but in this case it seems the CD3's higher price was not justified for this person given the two units similar level of performance. Anyway, here it is:

"My unit is the 24/192 version and all I can say is the EMC-1 has everything that I am looking for; smooth, transparent, deep & wide sounstage, and very, very analogue! I compare this unit to Audio Research's latest CD3 side by side and my conclusion is I like the performance of both units, but not the price. CD3 is double the price of EMC-1 (and very ugly, too)...so I end up with the EMC-1.
BTW, please use the XLR outputs to achieve a better sonic performance than the unbalance outputs. I could not believe my ears when I switched to the balanced output."
Soix posted a link to the papers written by Perpetual Technologies on upsampling vs interpolation. The papers describe their "P-1A Digital Correction Engine" which was plagued with problems from the start that were never resolved after repeated promises by the manufacturer.

If you do an internet search on google.com for Perpetual Technologies, you'll find a handful of product reviews and in more than one review from different magazines, you'll read about how the reviewers P-1A didn't work quite right. A search on Audio Asylum will also reveal several owner complaints about the Perpetual P-1A. Not only was this product plagued with problems from the beginning, the company failed to deliver on promised updates and software revisions.

Their white papers may sound good in theory but the P-1A which they write about, failed miserably in reality. Is the company still in business?
I have also heard there have been some issues with the Perpetual Technologies P1A, and I don't know if it is an ongoing problem or if PT has fixed the problems. In any event, I posted the link purely for informational purposes and not to endorse any particular product--just to be clear. Although PT's implementation of the technology may leave something to be desired(although a friend of mine was quite happy with his P1A and there have been some very favorable reviews of the product as well), the general overview still provides useful insights into the basic components of the upsampling process(i.e. not specific to Perpetual Technologies), which is what I thought would be interesting given the many questions and misconceptions on upsampling that seem to repeatedly appear here.

Tim
Tom: I find your results comparing the Mill II to the DAC 3 pretty interesting. Like you, i was also interested in that piece. I found the Mill II to sound much more "organic" i.e. more musical, natural and fuller sounding. In comparison, the DAC 3 sounded somewhat "bleached", lacking detail and lacked the warmth, body and bottom end extension that the Mill II is capable of. I'm sure that i could've gotten better results with the ARC if i really tinkered with it, but i didn't feel the need. Sean
>
Is it true that the Audio Note has no filtering ??

This could not be good for amplifiers and tweeters.

I thought it was a requirement that the aliasing frequencies must be filtered out of the analog output.

Can anyone comment on this?
As far as I know, an unfiltered DAC's analog output will only affect the speakers negatively if it affects the amp negatively. In other words, it's the amp that will potentially damage the speakers if it is made unstable by the unfiltered ultrasonics. The resulting amp distortion artifacts would cause any speaker damage; the speaker is actually a filter in and of itself, and would not be damaged the ultrasonics alone. Fortunately, I think few amplifiers would be disturbed to this degree, but the possibility always exists (presumably, AN electronics would not be among these).