|
Post by Casey Leedom on Nov 13, 2016 6:05:15 GMT -5
I stumbled over a note in the XMC-1 product brief on emotive.com stating that it supported DSD. So I downloaded the manual and it says that the XMC-1 only supports DSD over HDMI (presumably from a Disc Play of some sort). But these days we can buy DSD audio directly (DSD Audio Files, often as ".dsf" files) and serve them up over Ethernet or USB.
Certainly I can buy an out-board DAC like the Teac NT-503 to handle that duty and connect it via balanced cables, but there's a perfectly good DAC in the XMC-1 (and presumably a better one in the upcoming RMC-1). Is there any possibility that the XMC-1/RMC-1 will eventually support DSD via Ethernet (presumably via DLNA, ROON, etc.) or directly via USB? Or even possibly as DSD over PCM ("DoP")?
Casey
|
|
|
Post by wilburthegoose on Nov 13, 2016 9:48:54 GMT -5
Isn't DSD dead?
|
|
|
Post by Casey Leedom on Nov 13, 2016 11:40:14 GMT -5
Nope, not dead at all. I think that it's the main format used for studio recording and mastering. And now there are all sorts of sites which sell the mastered DSD data.
I have a friend who's gone nuts buying DSDs and adding them to our combined Digital Music collection. He has a Teac NT-503 which he's using to play them. I was wondering if anyone at Emotive was keeping track of that.
And we still have the awkward problem of having to get the Digital Music data to the XMC-1/RMC-1. Currently I'm using an aging Squeeze Box Touch and feeding the data to my old DMC-1 via Optical S/PDIF but that maxes out at 192kHz/24bit, so the DSD data needs to be down sampled in any case. It would be much nicer if I could just feed it into the XMC-1/RMC-1 directly via Ethernet to some kind of DLNA[1] or ROON agent. ROON seems to be taking off but it seems to be a bit ... fresh ...
Casey
[1] Note that my friend was unable to find any DLNA software that A. would serve DSD unconverted to PCM and B. had a decent User Interface, so he was forced to move his computer next to the stereo and use USB with Audio Nirvana as the software control solution. And the Audio Nirvana solution is still more clumsy than the Squeeze Server + iOS Peng solution we're using for our Squeeze Boxes.
|
|
|
Post by audiosyndrome on Nov 13, 2016 15:47:26 GMT -5
Isn't DSD dead? No. Not by a long shot. Russ
|
|
|
Post by rhale64 on Nov 13, 2016 21:59:16 GMT -5
Nope and it is gaining in steam. Even though technically inferior. It just sounds better to me. And a bunch of other people too.
MQA although I have not heard it yet sounds like it may sound similar to dsd. But possibly even better when used with the proper equipment.
|
|
|
Post by Casey Leedom on Nov 14, 2016 0:13:38 GMT -5
So I'm curious, why is it technically inferior? And this isn't baiting you, I'm really curious and references would be great. I'm curious as to why it's being used in studio recording and mastering if it's technically inferior.
Casey
|
|
|
Post by rhale64 on Nov 14, 2016 5:21:08 GMT -5
I am getting ready to head to work. After work I will try to find the articles I read on the subject. Something about the signal being dessimated. I will find it.
|
|
|
Post by garbulky on Nov 14, 2016 10:37:17 GMT -5
I don't think DSD is the main format used for studio mastering. Though people do use it in the studio, it is less widely used because it is more restrictive in terms of editing.
|
|
|
Post by Casey Leedom on Nov 14, 2016 11:34:11 GMT -5
I know that this is heading suspiciously close to needing to reclassify this into the Digital Music forum threads, but it still does involve XMC-1/RMC-1 capabilities. So has anyone seen anything about the Sonore microRendu[1]? From the Computer Audiophile review[2,3] it appears to offer an Ethernet->USB solution which offers Squeeze Server, Roon and several other service end-point solutions to get USB-based audio across a network to a USB-DAC. It's a bit pricey at $640 but at least that would reduce the XMC-1/RMC-1 problem down to just needing to handle the DSD formats on their USB inputs. It would be nicer to have all of this service end-point functionality in the XMC-1/RMC-1 themselves but this seems like a start. Casey [1] Sonore microRendu www.sonore.us/microRendu.html[2] Computer Audiophile Sonore microRendu Review, Part 1 www.computeraudiophile.com/content/698-sonore-microrendu-review-part-1/[3] Computer Audiophile Sonore microRendu Review, Part 2 www.computeraudiophile.com/content/705-sonore-microrendu-review-part-2/
|
|
|
Post by cwt on Nov 14, 2016 11:45:18 GMT -5
So I'm curious, why is it technically inferior? And this isn't baiting you, I'm really curious and references would be great. I'm curious as to why it's being used in studio recording and mastering if it's technically inferior. Casey Depends on what type of dsd your looking at Casey ; the studio ones may be a higher resolution than whats available at home . Good article here showing why higher sampling dsd exists to shift noise out of the ears frequency range ; and why complicated filters are required to eradicate the nasty ultrasonic noise dsd has . PCM has a much simpler low pass filter .. Not to say pcm audio with 96khz or 192 makes any sense when it should be a multiple of 44.1; bringing quantization noise . Note this comes from a company that sells dacs but a lot of articles have similar thoughts www.mojo-audio.com/blog/dsd-vs-pcm-myth-vs-truth/Here a post from Amir ; the designer of hd dvd www.whatsbestforum.com/showthread.php?19882-Multi-bit-DSD-versus-PCM&p=374999&viewfull=1#post374999and why he doesn't like 1BIT systems ; YMMV www.aes.org/e-lib/browse.cfm?elib=9903
|
|
KeithL
Administrator
Posts: 10,276
|
Post by KeithL on Nov 14, 2016 12:33:59 GMT -5
The fact isn't so much that DSD is inferior as that it has some serious limitations and no clear advantages to balance them out. The sound quality of SACDs (DSDx1), in terms of all the normally accepted metrics, is about equivalent to 24/88k PCM. This does give it a very slight edge over CDs (which are 16/44k). This could be considered important back in the day when 16/44k PCM CDs were the only alternative to SACD. Beyond that, the technical arguments in its favor are, to be quite blunt, largely incorrect and largely irrelevant. For example, DSD has a frequency response up to 100 kHz, and can be converted back into analog audio using a relatively simple conversion filter. These would seem to be significant benefits.... HOWEVER, the noise floor at high frequencies is very high, which means that you actually need to add a very sharp, and quite complex, filter to remove the high frequency noise and make the signal usable. (And that filter is just as complicated, and just as detrimental to the sound quality - or not - as the one used for PCM.) The other issue is that DSD is VERY difficult to edit (it's just a consequence of how the coding actually works). This means that most editing suites convert the DSD original into PCM for editing, then convert it back afterwards. The remaining few are quite expensive and complex (so, while a DSD recorder can be had for a reasonable price, very few studios are set up to mix or edit DSD). This means that many DSD recordings are the equivalent of "direct-to-disc recordings" - with little post editing. This can be an advantage for simple recordings of small ensembles, but pretty much precludes mastering large multi-track productions using DSD. (Well, to be honest, it doesn't preclude it so much as simply making it less practical than doing it the other way.) It's also quite possible to master in PCM and avoid excessive editing - it just doesn't happen very often. The reality is that, according to any legitimate measurement, including things like S/N and frequency response, single rate DSD downloads are about equal to 24/96k PCM. Some people anecdotally insist that "it sounds better" - but, because any format conversion changes the sound slightly, and virtually all SACDs have been remastered from the PCM version, very little apples-to-apples comparison has been done. (It's not at all surprising that a remastered "audiophile version" of an album might sound different than other versions.) The reality is that Sony, who originated the format, has abandoned it as a commercial product - with good reason. It is now a niche product, beloved of a few people for mostly non-technical reasons. It has simply become "the latest fad", which is far from a actual benefit. It would be interesting to see the actual statistics of how many albums are MASTERED AND THEN DISTRIBUTED IN DSD.
(Eliminating those that were edited in PCM, or were mastered on analog tape, and then converted to DSD for distribution.) I think you will find that there are very few, and virtually all of them are "niche labels". (QUite a few major albums have been "remastered to DSD" - but VERY few were originally mastered ON DSD.) So I'm curious, why is it technically inferior? And this isn't baiting you, I'm really curious and references would be great. I'm curious as to why it's being used in studio recording and mastering if it's technically inferior. Casey
|
|
KeithL
Administrator
Posts: 10,276
|
Post by KeithL on Nov 14, 2016 12:36:14 GMT -5
To answer the question, however........
The RMC-1 will support DSD playback over HDMI (probably at least 2xDSD). We also expect to include DOP support via USB. We do NOT expect to be including audio playback over Ethernet.
(no firm details yet).
|
|
|
Post by Casey Leedom on Nov 14, 2016 13:46:27 GMT -5
Thanks for the very comprehensive answers Keith (as per usual). The microRendu that I referenced earlier might become a solution and the server would just have to convert DSDs into PCM for the RMC. Or it may make sense to get something like my friend's Teac NT-503 ... but I'd rather use the RMC-1's own DACs and keep everything simple and the analog signal paths short. Or maybe I'll just stick with the SqueezeBox Touch I've got till it fails ...
Casey
|
|
|
Post by rhale64 on Nov 14, 2016 14:10:07 GMT -5
The mojo article is the main article I read on the subject. Also a few others. Thanks for finding that and posting for me. Saved me a bunch of digging. It was all about the filters that have to be used.
|
|
|
Post by Loop 7 on Nov 14, 2016 14:13:23 GMT -5
I listen to a lot of orchestral recordings, many of which are available in DSD, but it seems about half the actual recordings were made using PCM. I assume for ease of editing.
I always understood DSD was a superior recording format with many people saying it has a lot of analog's best qualities without the drawbacks.
|
|
|
Post by audiosyndrome on Nov 14, 2016 14:38:30 GMT -5
One advantage of DSD not discussed above is the abundance of surround recordings which I find in general to better stereo recordings of the same piece. The stereo layer of an SACD sounds great in my system. Listening to the surround layer almost always sounds even better than the stereo layer. BUT, going immediately back to the stereo layer now sounds flat, lifeless, restricted, etc.. Energize the whole room: you won't be disappointed. IMHO. Russ
|
|
KeithL
Administrator
Posts: 10,276
|
Post by KeithL on Nov 14, 2016 14:59:32 GMT -5
Unfortunately there's often a big gap between what "people understand" and what's actually TRUE. Back when Sony rolled out the SACD format, they spent a lot of PR effort convincing people that it was "more like analog than CDs". (The pulses in a DSD pulse train get further apart and closer together "in time to the musical signal", while our brains don't easily read digital numbers. This means that an oscilloscope picture of a DSD signal really does "look" more like an analog waveform - but that's just an artifact. You might as well say that one car "looks more like an electric car than another".) Both DSD and PCM are PURELY DIGITAL FORMATS. Of course, the reality is that you rarely if ever get to do an apples-to-apples comparison between DSD and PCM versions of the same content. In fact, it's more often true that both were mastered separately, and each adjusted to appeal to its intended audience. (And the engineer who mastered the SACD mixed it to appeal to a target audience of audiophiles who "like a more analog sound".) In short, many SACDs do in fact sound more like analog than their CD brethren - but it's simply because that's the way they were mastered. If you ever get the chance, listen to some DSD files, and 24/96k versions of the exact same files (converted using a high quality format converter). It doesn't matter which is the original. You will almost certainly find that they sound a tiny bit different due to the slight but unavoidable effects of the filtering that's part of the conversion process. But I doubt that, in a double blind test, you'll be able to reliably tell which is which, or prefer one over the other. I listen to a lot of orchestral recordings, many of which are available in DSD, but it seems about half the actual recordings were made using PCM. I assume for ease of editing. I always understood DSD was a superior recording format with many people saying it has a lot of analog's best qualities without the drawbacks.
|
|
KeithL
Administrator
Posts: 10,276
|
Post by KeithL on Nov 14, 2016 15:10:10 GMT -5
Good point..... IF you happen to like surround sound, and IF you're comparing SACDs to CDs. However, FLAC files support the same surround sound formats as DSD files. (So anything you can get as a surround sound DSD file can also be packaged as a surround sound FLAC file.) It is true that, as far as disc-based releases, MOST SACDs include a surround sound version, while most others discs do not. (Actually surround sound content was often available on DVD-A as well.) However, this is a holdover from the lack of support for surround sound on CDs. There's no reason for any surround sound recording that is released as files for download not to be released in both DSD and FLAC. Note that the music industry liked the SACD disc format because it included very effective copy protection. The format only died, or came close to it, because of the scarcity of compatible players. DSD file downloads DO NOT have the same copy protection that made SACD DISCS so difficult to copy. One advantage of DSD not discussed above is the abundance of surround recordings which I find in general to better stereo recordings of the same piece. The stereo layer of an SACD sounds great in my system. Listening to the surround layer almost always sounds even better than the stereo layer. BUT, going immediately back to the stereo layer now sounds flat, lifeless, restricted, etc.. Energize the whole room: you won't be disappointed. IMHO. Russ
|
|
|
Post by audiosyndrome on Nov 14, 2016 18:43:24 GMT -5
I was comparing the stereo SACD layer to the surround SACD layer. Good point about the copy protection.
Russ
|
|
|
Post by Casey Leedom on Nov 15, 2016 1:33:40 GMT -5
Thanks again Keith. I think I'll just let the DSD recordings that my friend Bill keeps adding get down sampled to 96kHz/24bit and call it a day. And even Bill says he's mostly sticking to 96kHz/24bit when he buys stuff now.
Casey
|
|