There actually are quite a few relatively low priced DACs that support DSD now. iFi for example, but there are others. The input to the DAC chip is done differently for DSD, so there are some DAC chips which can support DSD, but it takes some effort on the part of the manufacturer to use this feature, so the DSD feature isn’t “turned on” by default. And since few users (especially at the lower ends of the market use DSD), use DSD the feature doesn’t get turned on by the manufacturer.
I don’t think what you are hearing has anything to do with less jitter. There probably are less conversions involved, which means more accurate conversions and less aliasing artifacts. DSD filters in the D/A conversion tend to be “softer”, so that may account for what you prefer in the result.
But it all depends on the system. My system converts DSD to PCM and it still sounds great. Better than my previous setup that played back the DSD without converting it to PCM.
DSDs just sound better on my system. I have compared DXD to DSD256s and the DSD is the hands down winner. But as always it depends on the master.
My rule is if it started out PCM, I stay there (as high a rate as I can find). If it is a pure DSD, then I stay with the DSD rate it was recorded. Some of the old analogs transfers to DSD256 are stunning.
I am well over 4TB now and those DSDs do add up fast. But disks are cheap and I have a nice array, so failures are less of a concern.
I have recently disabled the option in Roon to always convert PCM to DSD.
I noticed that some recordings while having a nicer sound on the flip side become less open ( less spacious) when they were converted to DSD. Why this is happening ? no idea.
Yes, I feel there’s less 3D dimension when converting PCM to DSD on the fly using either Roon or JRiver up-sampling. The last time, I did an experiment by comparing ‘on the fly conversion’ vs converted file and I was able to hear some difference. There’s some slight variations depending on the algorithm which they use. However, there’s a factor come into play when doing ‘On the fly’ conversion; it is more resource restraint on the CPU and this generates more noise. If the noise is not properly isolated and gets into USB-DACs then it will degrade the fidelity
I experience the exact opposite.
In my system, redbook converted to DSD sounds WAY more spacious and airy than native redbook. My native redbook sounds also very good but denser in the room, whereas the converted to DSD sounds opens up in all directions. Not only wider and deeper but also in hight.
What DAC do you use?
Maybe your DAC is more optimized for PCM?
I’m using the Aries Mini DAC.
I have an amount of DSD files that are copies of my SACD.
Playing the original via de SACD player or the DSD copy via Roon are equal in sound and openness.
That makes it a bit hard to believe that it’s my DAC.
But perhaps it’s not fair to ask from Roon to convert PCM to DSD on the fly without any trade- off ?
Perhaps the conversion proces is more delicate and it just needs much more analysis ( time) to do it right ?
Even though I think your general rule is a good one, I have recently given up general rules for a variety of reasons (or factors) because it is so difficult to judge where the limiting factors for the sound in one’s system are to be found. Probably in the combination of factors, starting not really with the electronics (and power source) but the darn room, and it’s difficult to know how to isolate any of the contributing factors. The best one can do is guess, and that’ s a dicey game. I think many of agree that the master is the most important factor, as you also point out. One can determine this more easily by listening to a recording on array of systems, and finding out if the album sounds generally excellent on a wide variety of systems, something most of us don’t have time or the inclination to do.
I would recommend DSD to anyone interested in musical source material. I have an extensive collection, even without considering my vinyl collection (the vinyl purchased largely before the price rose to $30 or more … back in the '60s and '70s with many foreign pressings on decent vinyl). But I now have much more albums in digital format, and as have added some room correction and also more sophisticated DACs and other audio gear, power conditioning, cables, preamps, and amplification, I have likewise been surprised at even the improved quality of 16bit Red Book audio, realizing that a lot of what I thought was the format’s shortcomings were not. The same holds true of DSD, which along with other formats, sounds better as other aspects of reproduction chain improved. It’s made me a lot more hesitant to blame any particular aspect of the chain like whether something is DSD64 or Flac 96/24 or Red Book is the problem. And that’s not even considering how all the items in the chain interact together. I can urge people to not rush to judgement based their love/hate relationships with Sony and the DSD format. DSD can sound really … to use your word … stunning.
Someone else in this thread mentioned NativeDSD. I also recommend them highly. No I do not have a relationship with NativeDSD.com other than the fact that I’ve been a customer for more than a few years now. and have found quite a number stunningly great albums there, and also recommend the site very highly. I have found some excellent “Pop” and “Jazz” offerings to purchase at Acoustic Sounds, but not merely as many DSD winners as I have at NativeDSD. It’s not surprising, since their focus is DSD; not everything is “native” but again, I’ve dropped the general rules altogether these days, and find my ears are telling me that some of the DSD versions to be had there are superior to the other versions of the same music that I have in other formats. If you have a good quality DAC that will play DSD, and here, I am not so much talking to William (who also seems guided by his ears) but many of you who are skeptical of DSD to buy some DSD albums and judge for yourselves whether or not DSD can be a cherished resource. I for one (a person who has always been big fan of tape and vinyl) am very pleased with the sound of most of the DSDs that I have purchased.
In that case neither myself or anyone else on this forum can answer your question.
The values that each of us hold are personal to each of us.
Only you can make a decision on the value that a DSD purchase has for you.
As an example. Lots of people think that £1000 for a season ticket to their favourite football team is a cost they are willing to pay because of the value of that ticket to them.
For me, that season ticket has negative value. I would have to be paid many thousands of pounds to go and watch a number of football matches.
I paid over £1000 for a guitar though… If you asked me if it was worth it I’d say… for me… definitely. I’ve had a great deal of pleasure from it. It has high value to me.
If you asked me what the value is to you of a £1000 guitar… I have no idea, but I can tell you attributes about it which make it valuable to me, one of which… funnily enough… is sound quality.
So… you can maybe see why members of the community started discussing sound quality as one of the attributes of DSD when you asked about the ‘value’ of them.
Anyway… I’m rambling… have you had a listen to DSD yet? What do you think? Does it have higher value than CD to you?
I am completely confused that you would try to determine value without sound quality in the equation. Would you like to explain why that makes sense to you? It makes no sense at all to attempt to separate that out of the equation.
I conclude that It’s all about the complete chain
When I bought my aural Aries DAC I was able to stream DSD files
It sounded so much better compared to my PCM files
When I heard about the impact of galvanic isolation in the chain I also gave that a try.
I isolated the ethernet input on my DAC using a optical media converter and the sound changed for the better. it had minor impact on DSD but made a huge impact on PCM.
PCM and DSD now sounded equal.
I recently added the Schiit modi multibit DAC to my chain
I use my Auralic aries as ethernet input stage and its SPIF output to feed my Schiit DAC
The Schiit multibit DAC is a only 16 bit 96KHz . Its not build around a chip but it’s a ladder DAC ( R2R)
Here comes the interesting part. I used Roon to convert my DSD files into PCM (24bit 96KHz)
The Auralic aries moves the PCM stream over the SPDIF to the Schiit DAC.
DSD files sound again much better compared to PCM.
Technically I don’t understand why but using room for DSD to PCM conversion + the Schiit multiunit DAC improved the chain.
BTW I at first doubted my concluding ( I’m a big fan of DSD )and therefore asked someone to verify my conclusion. Within seconds he agreed DSD outperformed PCM
Try having Roon convert DSD to 176.4 PCM and feed it to the Schiit Modi Multibit. The Schiit will indeed play up to 24/192, however, anything above 96 will be fed through the DAC in NOS mode and not through the Schiit Burrito filter. I have liked NOS mode on it more than I thought I would.
Thanks Daniel, I thought that it always had to go via their filter.
I will make that comparison. interesting.
What about my comments that DSD even though converted to PCC sounds better compared to a true PCM (flac) file? Do you have the same experience ?
if you do… any thoughts why it sounds better ?
Writing this feedback I’m thinking about something, can it be that it’s the flac format ? if I play a flac file via Roon. does Roon convert flac to PCM before sending it to my dac ?
I find it depends on the end DAC and its DSD implementation. On the Pono players which has Ayre chip, DSD.
I find it depends on the source. A lot of DSD material are actually different masters and what you might be noticing is a difference between the PCM and DSD source. But, i have converted DSD to PCM with Roon and it has sounded great.
Flac is really a container format. What is in the flac file is compressed PCM. You could store other things in a Flac file. That is how DoP opertates, the DSD is encapsulated in a Flac wrapper and sent to the DAC.
I think with DSD, both in mastering and in playback, there are a lot of factors in-play.
The first is the quality of the master/original source. Assuming that the analogue master is of a good quality (ideally, the original/first generation), then the DSD conversion process is very straightforward, assuming there is no further mixing in the digital domain that is needed. Done correctly with the correct DSD A/D conversion, the DSD process should produce an accurate digital ‘mirror’ of the source. This is especially pertinent when considering analogue sources/masters, and anyone who has heard the RCA ‘Living Stereo’ SACD series, mastered in DSD ‘straight’ from the original three-channel masters, can probably testify that the conversion from analogue to DSD is absolutely sublime. I have all the original RCA SACD issues, and there’s something ‘right’ about their sound via my SACD player.
Recordings originally recorded in DSD can be equally beguiling, but don’t forget that it’s impossible to mix a recording in DSD - it has to be converted to PCM first. The series of earlier SACD releases from the jazz pianist Hiromi illustrate just how lovely a DSD recording can be, especially in multichannel
With DSD file playback (and not SACD playback), it’s a case of slightly different considerations. Assuming one has a DSD-capable DAC, what of the material? The DSD material I have personally come across has been more-or-less first rate, probably derived directly from the original DSF files. It sounds sublime in my system, and is a clear-cut above other digital formats. I suppose I could be regarded as lucky, as my source (now Roon ‘endpoint’) can play native DSD/DFF, and in general if I compare DSD to PCM, I generally prefer DSD. Also, because Roon can upsample PCM to DSD (which my Lumin A1 can also do, ‘on-the-fly’), then I can appreciate how upsampling from PCM to DSD sounds. And in general, I like this conversion. I do think that in my system at least, upsampling to DSD produces a slightly wider soundstage, and a more ‘holographic’ stereo image. But I think I’m blessed with a good DAC/network player, that can reproduce DSD very well. Other DAC’s might not be so capable, and in these particular circumstances, PCM might be preferred.
So in summary, I’m a DSD supporter. How long it’s going to be around, with all this talk of MQA etc, time will only tell. But I know one thing: when it goes, I shall miss it