Hi David. I’ve gone back and forth on that setting. Right now, it’s on “Native,” and that’s where I tend to keep it.
Cool. Have you had an opportunity to compare remodulations at different DSD sampling rates that are based on e.g. the same DXD master (such as the ones available at NativeDSD)? If so, what’s your impression on the differences between DSD64-512 (also, of course, relative to said DXD master)?
Greetings from Switzerland, David.
David, I took a run at that a while back when some 256 tracks that I cared to hear became available. After a while of switching back and forth on the MSB using a playlist to load up different versions of the same tracks), I became convinced of Teddy Roosevelt’s aphorism from a different context: “Comparison is the thief of joy.” I could find nothing consistent to guide me toward any “superior” format or conversion.
I spend whole weekends with audiophile acquaintances and friends comparing stuff, sometimes blind, sometimes not. At home, ever since I quit designing speakers and auditioning equipment for others, I try to avoid comparisons, for the exact reason you mention. Also, I listen to a lot of classical music (i.e. “time-consuming” versus bits of this and that), including historical, and can’t be bothered to listen to music only because it’s well-recorded (except, of course, once). I’m one of those audiophiles who believes the term “High Fidelity” should be taken more literally again, i.e. I want to hear what’s on the record, but I won’t allow sound quality to dictate my musical preferences. So, what I’m hearing in higher-rate DSD is a musically relevant step up from DSD64 to DSD128, after that, incremental improvements that may be nice to have, but ironically, seem to add to the enjoyment principally in poorly designed DACs, whereas great DACs manage to (almost) get out all one needs to hear from RBCD playback. I’m being my usual rational self, of course - from a purely emotional perspective, I agree high-resolution playback adds texture, realism, even a sense of flow that brings the listener closer to the experience, or in philosophical terms, keeps the listener in the present where one’s mind belongs. Having said that, I’d be curious to hear what DSD256 would sound like on a Vivaldi stack, especially given it’s an increasingly popular recording format, as I believe cutting-edge digital gear should always be designed (or in this case, updated) to play back (all) native recording formats.
Greetings from Switzerland, David.
1 Like
I do not see that the situation has changed at all from the response provided by Andrew from dCS back in January to a similar request at the beginning of this thread:
"This is the issue. The total number of titles is still extremely small and although there is demand in terms of individual customers making requests like this, the number doing so is a small minority in comparison to our total customer base.
This feature is on our radar, but we haven’t defined its position in our roadmap yet."
Bluntly dCS are not going to spend the many tens of thousands of pounds necessary to provide access to a handful of titles from mainly little known artists on three boutique labels for a fraction of dCS owners.
1 Like
Yep… except your cost estimate is off by close to an order of magnitude. Once you take into account development time, testing, verification, and support the cost is astronomical.
Thanks for that insight.
I agree with you David, 100%. Andrew and Pete, I sense some impatience with those of us making this request. I certainly don’t mean to cause distress, and I am decidedly not trying to be impertinent or disagreeable, but I am curious. Isn’t one of the benefits of the Ring technology that it is upgradeable/improvable? Aren’t we in the correct forum to be discussing this capability? So, if this feature is a “bridge too far,” maybe you can clue us in where development pounds are being spent.
Part of the reason I made the decision to upgrade my Phase One IQ3.100 digital back to the IQ4.150 was the roadmap that Phase published. Roadmaps, even sanitized ones for public consumption, can be very useful for calming the peasants.
If not 256, where is dCS headed for the next firmware upgrade? If it’s SQ, and the company cannot be more transparent than that for trade secret purposes, we all understand that. But if 256 will not happen because it is “just not worth it,” it might be time to say so. That’s different from what was said in January, and more like the tenor of the above (which sounds very much like it is “sort of on the road map but not really.” Then we irritants can try to be more constructive. If 256+ is reserved for Vivaldi’s successor, or slated for Rossini alone because Vivaldi’s hardware just can’t cut it, there’s probably a way to communicate that without beating us about the head with a broom. (That’s meant to be light-hearted.)
Other TOTL DACs have this and greater capability. I am not asking dCS to compete on a feature-set basis with DACs built around the everything-but-the-kitchen-sink DAC-chip-of-the-month. But it doesn’t seem unreasonable to want one’s TOTL DAC to be as capable as possible. Yes, the number of recordings is relatively small, but growing every day. Yes, the number of us current owners who want it might be small, but we are likely to grow, and what about the possible customers who have not yet committed to dCS? Like it or not, all consumers have checklists in their heads. It might not always be the savviest way to shop, but it is one of the ways we all shop at one time or another.
I almost bought Meitner for this very reason. I am ultimately glad I did not, but part of the reason I did not was that upgradeability quality. I have a whopping three albums recorded in DSD256. Yes, it’s a tiny number. but they are something else, and it’s unlikely there won’t be more. But I have dozens in 256 and a growing number in 512DSD. The Upsampler itself is testament to the attractiveness of certain methods of upsampling. In a sense, dCS is itself one of the factors that makes some of your customers desire the expansion of this capability. Are we cutting edge? Maybe. Shouldn’t dCS want such customers?
I’m going back to beautiful music. I thoroughly enjoy my Vivaldi stack, and I love the music it makes in my system. I hope everyone here is safe and healthy, and I truly appreciate the participation by dCS team members and each of our forum members.
Anupc privately reminded me of his Feb 23 post above which, if correct, certainly lends support to the magnitude of the challenge. And I would never want dCS to think that this customer thinks the difficulty is trivial.
Greg, Andrew’s point is not about technical ability but the economics. As you are new to the dCS world I don’t know if you are aware that when dCS make a firmware upgrade available this is provided cost free to users.
If there is sufficient demand ( which Andrew indicates is not currently the case) then there might be such an upgrade. Such a level of demand may, however, not emerge or feasibly may only become a reality by the time of Vivaldi’s successor, the Monteverdi or Corelli ( at least I hope it will be called either of those if in the tradition of great Italian composers 
) which I am sure is in development.
One other thing, what do you really want? Although you say that you have “dozens” of DSD 256 files you also mention that you have some DSD 512 as well. If you got an upgrade to 256 would 512 then be an issue for you? If so should dCS wait until not only is there sufficient demand for 256 but also 512?
Actually, thinking further about it this would be a huge upgrade as it would not just affect the DAC but presumably also upsampler and clock and, I suspect, there would need to be Rossini and Bartok versions as dCS could hardly withhold access to the popular DSD 512 format except to customers of its flagship line.
1 Like
Hi Pete. I understand the demand/economics issue. And while new to the dCS community, I am familiar with the excellent history of upgrades to products. It’s well-documented in the reviews and here in the forum, and my dealer highlighted it as well. However, if dCS were to say: “Greater than 128 is going to require customers to pay $xxxx” (like the previous Upsampler hardware upgrade), that would not necessarily surprise me. As for your question about “is 256 enough or will you want 512,” I guess the answer is “yes and yes.” 128 was good enough for me to buy. Any upgrade will be welcome, and more will always be desired. 
I feel like this discussion may have become unnecessarily contentious—and I helped with that. My regrets. To answer your question most directly: specifically as a very happy Vivaldi owner, I would like to understand better the Vivaldi end game. This particular component “group” is about 8 years old, and because I do understand the economics issue, I completely understand that sinking more significant funds into Vivaldi’s technical capabilities may make little sense. And this causes me no consternation. If this is as good as Vivaldi gets, I am not unhappy; I will simply know that I may be waiting for something else in the dCS pipeline to hear how well dCS technology can play these files. My dealer and I discussed this very possibility when I made the purchase switch from Rossini to Vivaldi.
For the sake of clarification, I am able to play all my >128DSD files through the Vivaldi stack courtesy of Roon downsampling. And there is other server/streamer software besides Roon that will accomplish the same for anyone who owns any dCS equipment. I don’t own any music that I cannot listen to on my system. Trying to compare how some of the native 256 files sound between my headphone system (which can play these files at their native resolution) and the Vivaldi stack speaker system, is an unilluminating exercise; they sound different, no matter what. And they sound superb.
I fully believe we will continue to get a growing, though admittedly still small, catalog of native higher DSD recordings out of Blue Coast, NativeDSD, and perhaps others. Apparently, their overall catalogs are sufficiently successful to enable this expansion of their technical offerings. For clarity, I am not implying a comparison between their business model and dCS’s business; I doubt there is much in common. But I do think these recordings will continue to trickle out, and that components from companies like dCS, Meitner, MSB, TotalDAC, and others will be judged by whether they reveal the best SQ of those recordings. It may not be necessary to play those files at their native DSD in order to reveal their beauty. It may even be true that most listeners cannot hear the difference either way. But I would love to find out.
1 Like
Bartók and Debussy are much less Italian than other Italian composers 
but it is correct that until now their flagship gear were named after Italian composers.
Possibly the greater sound quality improvement might not come from the digital domain, but from either analog domain or power management…I guess we can trust dCS for continuous improvement of their products.
Being an owner of a Vivaldi Dac (+ Network Bridge) I would prefer dCS showing up one day with a Vivaldi 3.0…
1 Like
Except ( putting on my pedant’s cap) for the first which was Elgar. I guess they thought that having the next flagship called Vaughan Williams, Bax or Holst ( or even Byrd or Gibbons) might have sounded a bit odd 
As for a Vivaldi 3, it’s possible in theory were it to be only a firmware upgrade as the range changes have so far come about only when the existing hardware platform can no longer be updated.
Of course the Vivaldi is so amazing that I cannot even conceive of anything better. I have no doubt that I will be proved wrong in due course
2 Likes
^^^^^^^^^^ This.
Once dCS markets a Beethoven, I’ll really want to hear that!
Greetings from Switzerland, David.
1 Like
I hope they would not call it Beethoven , especially in this anniversary year. I can think of several audio products over the years called " Beethoven" so that it is now a bit of a cliché. But whatever it is called, it will be coming sooner or later.
1 Like
Giving this topic a well-deserved bump. It was first raised 3 years ago, and last commented on 2 years ago. That is a long time in technology and high-end audio.
Content variety in DSD256 (and above) has grown considerably. NativeDSD’s database indicates about 1,000 records in DSD256, about 600 in DSD512 and about a dozen in DSD1024 format even (but let’s not get carried away) and counting.
Internet bandwidth has improved over the years, downloading a couple GB for a single album no longer breaks a sweat (about an hour max for ~35GB across 8 albums last night in my case), neither does cost of storage. A DSD256 album cost me USD28, and I could have saved just USD6 (21%) for buying ‘only’ a DSD128 of the same recording. Not that it would matter to the typical dCS customer, the point is merely to illustrate that DSD256+ is going mainstream. The ‘just a dozen albums’ when this thread was created 3 years ago has grown to a thousand, and I reasonably assume growth accelerates.
DACs continue to evolve and dCS to all our delight remains at the forefront, generally. The same database linked above has a tab with DAC offerings in detail, showing their various levels of DSD support. There are about 600 DACs listed, across all spectrums of the market. If I look just at those with a list price of let’s say USD10k and above for argument’s sake then I see quite a number that offer DSD256 (and above), perhaps about half of the models listed. I won’t bother listing the brands and models here which, according to this list, support DSD256/512/1024. If of interest, you can read and study that at your leisure.
So, what is the point of all this? I love my ‘gateway drug’ dCS Bartók, in fact I recently added a dCS Rossini clock. One of the considerations is to select a company that is renowned for innovation, continuous progress, and customer centricity. I look forward to dCS native support of improved content formats such as DSD256.
6 Likes
While I’m all for supporting a format like DSD256, a closer look at current DSD256 material available on the market is quite revealing (at least based on the DSD256 albums I’ve purchased from both NativeDSD and HDTT).
Even among the best sounding DSD256 transfers, a look at the spectrum of the DSF files reveals that there’s only DSD128 bandwidth (or less) worth of actual musical content in them! The rest is just pure shaped noise (and in some cases noise from other elements of the transfer process).
Take Cannonball Adderley “Somethin’ Else” album DSD256 transfer off HDTT for example, arguably among the best sounding digital transfer anywhere.
Notice both the DSD128 (top image) and the DSD256 have no musical content above about 28kHZ. So even on a DAC that natively supports DSD256, the two will sound no different.
Why would dCS justify the R&D investment required for DSD256?
5 Likes
But, but, but Anup. I know I can hear the difference! And downsampling just ruins it! 
Thanks for posting this. For me, my reasons are more mundane: I like supporting the various vendors who are in fact employing higher rates in the production process. It seems to me that use of the technology has potential, and I personally seem to find DSD, and specifically higher-res DSD, less fatiguing. Even if many of the transfers of older recordings aren’t delivering more information. I like how Cookie Marenco explained why it’s worth supporting higher-res—and by extension, even higher-res—music here. “If you build it, they will come.”
5 Likes
I’m actually with you, Greg. Given a choice, I’d rather not have to downsample anything off-box regardless, and if required, personally, I’d rather see dCS downsample DSD256 inbox, than to have Roon do it.
99% of DACs which support DSD256 are doing so because the latest range of ESS and AKM chips support it, and so it comes “free” with virtually no R&D cost involved at all on the part of the DAC vendor. Among the FPGA DAC vendors (like Chord), there’s zero evidence of improved sound quality from DSD256 files vs. DSD128 (I’ve personally tested it on a Chord Dave when I owned one).
So, if improved sound quality is the prime directive to support DSD256, and R&D cost is involved, then I don’t think too many vendors (dCS included), would want to spend the money at this point in time.
By the way, on one of the “Pure DSD256” albums I analysed, there was what looked like a metronome modulated onto a higher frequency captured on the track; you could see the tick-tock-tick-tock in the realtime spectrum analysis . I’ve also seen DSD256 files generated from upsampled 96/24 content.
Basically, DSD256 is still “all over the map” in terms of quality 
5 Likes