Galliardist
Major Contributor
You’re right about the Mola Mola - after posting my own response to that post I went back and checked. I guess that leaves Marantz as the main practitioners of DSD in playback.
-
Welcome to ASR. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!
DSD is better than PCM!
- Thread starter cany89
- Start date
Galliardist
Major Contributor
It's going to be very dependent on the DAC you are using and any shortcomings there. HQPlayer is a conversion toolkit, that you can use to replace some of the filtering and upscaling/conversion options built into your DAC.
You have to work with the formats offered and your DAC, and that DAC has to have shortcomings, for it to improve the sound in practice. There's no magic anywhere. So, for conversion to DSD, your DAC still has to handle DSD "better" than PCM for a conversion to DSD to be of any benefit. In what circumstance is that likely?
Well, I may well get something out of converting to DSD with my Marantz player, because its output conversion format is DSD - the internal PCM to DSD filters and noise shaping are no longer used, and if HQPlayer somehow does that conversion in an audibly better way, I might win.
But... what is the point, with most other good modern DACs, in converting to DSD first? Your player may just convert back to PCM at a lower resolution, in the worst (probably still inaudible) case.
HQPlayer can also upscale PCM. That may be useful in some cases - the one that comes to mind today is with a DAC that uses a default MQA filter. Then, you may be able to upscale to an end format that passes straight through to the converter stage and lose some of the aliasing problems.
HQPlayer can also convert DSD to PCM - that might be useful in some circumstances where a player can downscale DSD to convert it, or convert to a non-optimal PCM format, or you have DSD files and a PCM only player.
Apart from that, it's a tool that may help you learn more about digital audio.
The practical advice, if you are just after a music playing system, is just buy a good DAC in the first place. That way you don't need HQPlayer, unless you want to do a conversion that damages the sound in pursuit of "magic"- and I doubt that is in any way the intention of the developer.
You don't need to "improve" playback that already works.
DSD is obviously useless. Created by Sony to earn money, just like overpriced audiophile nonsense cables. Lost a lot of respect for Sony there like Apple. Don't bother with going out of your way with DSD unless the mastering is better. And please don't listen to the massively overcompressed recordings, because it can damage your hearing. PCM is necessary then, just in order to expand the dynamic range in post production with an audio editing program.
And high resolution PCM only makes sense to release on modern recordings that were actually recorded with high resolution. The only way you can even hear the difference from DSD is if it was recorded that way without being converted to PCM ever. Thereby no digital editing at all. You have never heard DSD, unless you have heard such a release recorded on DSD and it would still make sense for consumers to convert that release to PCM. Just because DSD has technical flaws, and therefore does not have any real aesthetics...
See for example Why 1-Bit Sigma-Delta Conversion is Unsuitable for High-Quality Applications by Stanley P. Lipshitz and John Vanderkooy, from the 110th AES Convention in 2001. They state that this Sigma-Delta conversion technique is imperfectible, because it cannot even be dithered in a decent way and takes up way too much space. So distortion and noise in the technique itself can't even be avoided. Not talking about the recordings themselves, it is the technology that doesn't work for hi-fi equipment. So PCM is good for hi-fi equipment.
The only real concern is then the AD/DA filter slope and high resolution can help there to avoid aliasing issues. 96 kHz is probably enough to handle most of that. And 24 bit is the maximum that is even possible in hardware implementations. Not talking about software there. It's the electronic components and circuitry design that make 24 bit the maximum realistic solution. And it is luckily an adequate solution anyway.
And high resolution PCM only makes sense to release on modern recordings that were actually recorded with high resolution. The only way you can even hear the difference from DSD is if it was recorded that way without being converted to PCM ever. Thereby no digital editing at all. You have never heard DSD, unless you have heard such a release recorded on DSD and it would still make sense for consumers to convert that release to PCM. Just because DSD has technical flaws, and therefore does not have any real aesthetics...
See for example Why 1-Bit Sigma-Delta Conversion is Unsuitable for High-Quality Applications by Stanley P. Lipshitz and John Vanderkooy, from the 110th AES Convention in 2001. They state that this Sigma-Delta conversion technique is imperfectible, because it cannot even be dithered in a decent way and takes up way too much space. So distortion and noise in the technique itself can't even be avoided. Not talking about the recordings themselves, it is the technology that doesn't work for hi-fi equipment. So PCM is good for hi-fi equipment.
The only real concern is then the AD/DA filter slope and high resolution can help there to avoid aliasing issues. 96 kHz is probably enough to handle most of that. And 24 bit is the maximum that is even possible in hardware implementations. Not talking about software there. It's the electronic components and circuitry design that make 24 bit the maximum realistic solution. And it is luckily an adequate solution anyway.
Galliardist
Major Contributor
Oh, come on. Do we really have to fight the format war again and again for ever? And do we have to ascribe all of the sins in audio to what is now a niche digital format?
Your first paragraph combines some sense (there's no need to go looking for DSD recordings except for better mastering) with some myths. The common compression techniques, for example, are used on PCM recordings because that's where the tools are written. Fewer SACDs (at least in the SACD layer, for hybrid discs) are" massively overcompressed" than CDs and PCM recordings. It's the one thing in their favour.
And Sony didn't invent "DSD" as a form of snake oil. It was originally an archive format, overtaken in that role by higher resolution PCM. They were the key movers in producing SACD. With the problems of other surround formats like DVD-A, discussed earlier in this thread, SACD was the only way for several years to buy a standard disc to do that - as well as one that you could use on conventional CD players. At a time when online music was downloading heavily compressed (bitrate) files, it was a potential way to allow people to upgrade over time to a superior sound format (surround). So there was some value in the format as it was introduced beyond the license fees. Because MQA is a demon and has licensing fees, no reason to presume that SACD was the same because it had them. It won that format war for a reason. The alternative format also had licensing fees, and copy protection, and you might remember that last desperate throw of the dice, DualDisc...
Sound quality? Well, every single byte of digital audio in my main system is converted to DSD. If I introduce a non-DSD, delta sigma PCM DAC of similar quality and use equivalent filters, I will get equivalent sound quality and should not be able to tell the difference blind - any more than between any other two delta sigma DACs, or if I introduce a R2R DAC of similar quality. It works. The problems are outside audibility. Let me assure you that I didn't go out and buy the player I did because I thought there was any fairy dust in DSD playback.
It's worth remembering that the Lipshitz and Vanderkooy paper was published right in the middle of the SACD vs DVD-A war. They weren't materially involved in that war, as far as I know, but the paper got a lot of one-sided publicity at the time including claims that went well beyond what it actually says. It doesn't consider the full effects of noise-shaping or further upscaling of the DSD format. As for space, I don't have that problem (most of what I play is PCM, only converted to DSD in the player anyway).
So where does that leave us? DSD does what it does. SACD is actually STILL the main way to buy surround music, until multichannel streaming takes over at some point in the future. For those of us using stereo, we don't have to go anywhere near DSD unless we want the particular mastering of a DSD stereo layer. There is neither magic nor horror to be seen here. Just accept that and move on.
Nobody wanted it, and it required special hardware beyond a basic DVD player, because it resided outside the base DVD spec. It required unique high-end spec players for an already-established consumer standard that unsurprisingly nobody bought.
Versus Bluray, where every $35 Bluray player ever made can digitally output lossless 192/24/6ch audio.
It shot itself in both feet, rather than just in one, and was wholly mooted as a format once Bluray happened.
Last edited:
That is not because a paper was published in the middle of a commercial war that it's main takeover should be ignored: the impossibility to properly dither DSD introduce distortion. The full effects of noise-shaping (?) or further upscaling won't change anything to it.
"In reality it is a little more complex, and the analogy is incomplete in that 1-bit sigma-delta converters are these days rather unusual, one reason being that a one-bit signal cannot be dithered properly: most modern sigma-delta converters are multi-bit.
"Because of the nature of sigma-delta converters, one cannot make a direct comparison between DSD and PCM. An approximation is possible, though, and would place DSD in some aspects comparable to a PCM format that has a bit depth of 24 bits and a sampling frequency of 88200 Hz.
"Because it has been extremely difficult to carry out DSP operations (for example performing EQ, balance, panning and other changes in the digital domain) in a one-bit environment, and because of the prevalence of solely PCM studio equipment such as Pro Tools, the vast majority of SACDs—especially rock and contemporary music, which rely on multitrack techniques—are in fact mixed in PCM (or mixed analog and recorded on PCM recorders) and then converted to DSD for SACD mastering."
Source:
Direct Stream Digital - Wikipedia
I am not an engineer or anything, just someone who only needs stereo. So my post was indeed "one-sided". However, you would have to publish mathematical proof or something like the AES paper to make your post fully valid.
- Joined
- Feb 23, 2016
- Messages
- 23,575
- Likes
- 44,217
Well some Sony engineers said sacd was to create a new format Sony owned as patents were expiring on CD.
tuga
Major Contributor
If I'm not mistaken Ed Meitner and Andreas Koch where leading the DSD/SACD developpment. Meitner speaks about it in this interview:
Meitner: … well, if you have it in the one bit DSD format, you can, (A) you have a pretty robust storage that way, (B) you can now convert it to any other format that may come about, PCM 96/24, whatever. So it’s a very versatile format to begin with. And don’t forget that every A to D converter that you see on the market today starts off as a DSD modulator. So then you have the DSD signal on the A to D that just goes to the PCM down sampler or decimator and gets turned into PCM, so the life of the audio in the digital world really starts off as a one-bit signal.
Pappas: And so this is just a natural extension of that.
Meitner: That’s right. Yes. So now if you have an audio chain, you have one bit coming out of the A to D and going into a storage medium, i.e., hard disk or the AIT. And your D to A converter again is relatively simple because you don’t have to do any conversion from PCM to bitstream. And most converters out today are some form of bitstream. So we just cut out the PCM part.
Pappas: So in effect, DSD is simpler because you’re not doing the kind of conversions using a decimation filter to make it into PCM. You’re just eliminating all that extra stuff.
The DSD/SACD Revolution, Part II: PF Interviews Digital Designer Ed Meitner
https://positive-feedback.com/pfbackissues/0802/pappas.Meitner.rev.8n2.html
Meitner: … well, if you have it in the one bit DSD format, you can, (A) you have a pretty robust storage that way, (B) you can now convert it to any other format that may come about, PCM 96/24, whatever. So it’s a very versatile format to begin with. And don’t forget that every A to D converter that you see on the market today starts off as a DSD modulator. So then you have the DSD signal on the A to D that just goes to the PCM down sampler or decimator and gets turned into PCM, so the life of the audio in the digital world really starts off as a one-bit signal.
Pappas: And so this is just a natural extension of that.
Meitner: That’s right. Yes. So now if you have an audio chain, you have one bit coming out of the A to D and going into a storage medium, i.e., hard disk or the AIT. And your D to A converter again is relatively simple because you don’t have to do any conversion from PCM to bitstream. And most converters out today are some form of bitstream. So we just cut out the PCM part.
Pappas: So in effect, DSD is simpler because you’re not doing the kind of conversions using a decimation filter to make it into PCM. You’re just eliminating all that extra stuff.
The DSD/SACD Revolution, Part II: PF Interviews Digital Designer Ed Meitner
https://positive-feedback.com/pfbackissues/0802/pappas.Meitner.rev.8n2.html
- Joined
- Feb 23, 2016
- Messages
- 23,575
- Likes
- 44,217
Of course those things are not true anymore. Sigma-delta DACs use multi-bit not one bit. Those comments are from circa 2002.
DSD is such a pain in the rear in every other way. It is inefficient requiring huge files, and you can't edit it. And what are the benefits? Somewhere between non-existent and things that don't appear to show up in the audible range. Let's give it a big benefit of the doubt and say 24/96 is 99.8% of what is possible and good DSD is 100%. You aren't getting much for all the trouble, and for the great majority of music the recording isn't up to showing any difference. And remember that is me giving the idea it is better which hasn't really ever been shown to be the case.
Last edited:
tuga
Major Contributor
He was talking about ADCs.
tuga
Major Contributor
I agree. DSD makes sense for digitalising analogue masters, either of old recordings or new ones made with analogue equipment (e.g. Cookie Marenco/Bluecoast Records). Or perhaps "direct to master" un-edited recordings (e.g. PlayClassics, although Mario would not be able to "calibrate" the mic feeds).
If 99% of ADCs and DACs use a ΣΔ modulator then DSD is probably better (at least in theory).
How many of these are truly single bit? How many of the state of art designs are truly single bit?
- Joined
- Feb 23, 2016
- Messages
- 23,575
- Likes
- 44,217
ADCs aren't single bit these days either. Haven't been for over a decade.
I don't think it makes sense for any use case I can find. I think the whole idea a plague on digital audio.
tuga
Major Contributor
I am not qualified to answer that question. Maybe this post by @Miska will help:
Important thing to understand is that in SDM converter there are various stages with different number of "bits" and the meaning of "bit" changes on each stage. For that reason in SDM context we usually talk about number of levels instead of bits. Especially because in most cases it is convenient to have odd number of levels and that's not even close to any power of two. So a typical number of output levels could be 5 which is around 2.5-bits in terms of power-of-two binary number.
https://audiophilestyle.com/forums/...bit-dsd-debate/?do=findComment&comment=273800
None.
Galliardist
Major Contributor
I'm not denying that that was part of the story.
- Joined
- Feb 23, 2016
- Messages
- 23,575
- Likes
- 44,217
What I'd like to know is where the heavy, deep, abiding faith in the superior nature of DSD comes from no matter what evidence is presented.
Faith, by definition, doesn't need to come from anywhere.
Similar threads
