I know it isn't a quick upgrade but these claims could be really impactful if you included A-B sound comparisons. Nowadays there's been so many other, good comparisons published that any similar claims without example appear much more like snake oil. Whether the tests agree or disagree, the results would be extremely meaningful (if done properly), especially coming form a source like yourselves.
Only way to do it correctly is to mix in Realtime and record a live band straight to DSD. Then put that recording on SACD, never touching pcm. Takes time to mix live but it's the right way.
Michael Jackson’s ‘Thriller’ Celebrates 40th Anniversary With Double-CD Set Mastering the anniversary package from the original analog master tapes, Mobile Fidelity will also make the original Thriller available as a 33rpm LP, limited to 40,000 numbered copies, as well as a hybrid super audio CD (SACD). An UltraDisc two-LP set of Thriller will be released at a later date.
It sounds like whatever the conversion "process" (from Sony) for DSD to DXD is the problem. If this Zephyr ...filter is essentially doing the same thing (DSD -> DXD) but at better quality, then that brings into question how Sony is doing the conversion. Is it some magic proprietary chip that only Sony can see into (a la SACD DAC chips in SACD players)? "DXD" isn't *the* problem; it's just the medium. It's all PCM in the end: Sony ouputted DXD or Zephyr outputted DXD. As Paul eludes to, the D to A / A to D conversion absolutely depends on the quality of the hardware (or software). Same goes for the process to convert DSD to PCM. And I get that if Pyramix is/was the only game in town, then that's all you have. That doesn't mean it's the best way to do things, though.
If the befits are mainly while recording and it's possible to keep PCM in the rest of the chain maintaining the higher quality - I hope you understand that Zepher filter is a golden egg. I'm sure all the big guys in the recording industry like Apogee, Antelope, Lynx, RME and Avid will get in line to get a license.
The greatest-sounding albums in my high-resolution collection are my DSD256 and DXD albums. Now, are they equal? That depends on the listener, in my humble opinion.
I agree with most of what Paul has to say. I can't comment on the Zepher because we don't have one. I completely agree that recording to DSD and converting to DXD is a sonic dissappointment. I very much agree that the DA or DAC that is playing the audio makes a huge difference in the sonic quality of any file. There is a process called "rendering" that has not been discussed. RENDERING a file is much different than converting a file. If editing is required for a DSD file in the Pyramix, the file is "converted" to DXD to do the editing. But here's the thing.... think of the DXD editing as a plotting a roadmap to get to your destination. After you've made your edits (or whatever) in DXD, you RENDER the file back to DSD... it's not a conversion. RENDERING applies changes in only the places needed - like small edits for splicing in a new take of a song. This is how it's been explained to me by the crew at Merging Technologies (the company behind Pyramix). It's pretty obvious that when the file is RENDERED back to DSD it sounds WAY better than its DXD counterpart. You don't want to listen to the DXD after the DSD has been rendered from it. Again, at Blue Coast Records, we MIX (add efx, volume changes, pan, etc) in analog, but we do very small edits at the front and end to avoid popping your speakers. On occasion we'll make small fixes with editiing between two takes of songs. The RENDER from DXD to DSD keeps the DSD the same and only converts the very small milliseconds of the edit. The resulting RENDERED DSD256 sounds WAY better than the DXD. At Blue Coast Records, the source file and best sounding we sell is the DSD256. All other files we sell are made from that source file. Blue Coast Collection 4 is a good example of recording to DSD256 and working in this way. bluecoastmusic.com/blue-coast-collection-4 Enjoy your music!
Hi Cookie! I'm a film composer and have been working through the kinks of how to enable a DSD workflow for our studio's film score soundtracks. We've got a ton of questions, but one I have about Pyramix is if you record DSD files into Pyramix but then need to MIX, does Pyramix automatically convert the project to DXD? Or is that a step that you manually have to do? Is it for the whole project or just specific tracks that you're doing MIX edits on? I'm really curious how Blue Coast Records handles the process and will try to reach out via email to hopefully chat with you!
@Cookie Marengo: I must say, I’m getting more out of your explanations than I’ve been getting out of Paul’s several attempts on this topic. Thank you too, for offering links to test my DAC; my DAC tells me only the multiple (up to 8x) of 41 kHz (or 48 kHz) of the source to which I’m listening. This information helps me know when the source is upsampling, but doesn’t tell me if I’m listening to PCM or DSD.
any difference you can hear between DSD and PCM can easily be shown on a FFT spectrum analyzer. the real question is ... what ARE those differences....?
Higher resolution and more audible concurrent soundwaves being perceived with the DSD recording, especially if the file bitrate is significantly higher than on the PCM recording (which translates into more soundwaves being clearly and distinctly reproduced with the DSD format vs. the PCM one.) In essence, the sounds become more realistic, warmer, more pleasant and impactful.
As I know that PCM is mathematically very problematic, the requirements are either not fulfilled or just difficult to fulfill. If an ultrasonic component remains in the original analog signal, it will appear as an audible sound after PCM encoding, which the FFT must show. (This is an insignificant problem with DSD.) On the playback side, PCM requires very computationally intensive digital filtering. This is only done by very exceptional devices. I'm not a professional, just a layman, I'd be happy if you could correct me if I'm wrong.
Yes, DSD is closest to analog, DXD is PCM, the sound quality of DSD can be inferior to that of DXD, and yes it depends on the recording, mastering, and digital to analog converters. I have a question regarding this process, but my question doesn’t completely focus on the topic of this video, but it is about conversion from analog to PCM to DSD and back to analog. The question is this: Could electrical issues with analog mastering equipment, along with one bit delta sigma conversion be the reason for any additional sonic artifacts heard on the final product of a recording mastered from analog for DSD? I’ve read where Mobile Fidelity Sound Lab said that additional coloration or other sonic artifacts was not their intention and that the sonic artifacts heard were only a product of the master tape. Please enlighten me on this.
@@philosophiaentis5612 I know Mobile Fidelity masters are digital, but doesn’t that make the “sonic artifacts” that “are a product of the original master tape” more noticeable?
I've never understood why mixing DSD isn't easy. Why the usual error distribution technique (noise shaping) is not enough. It would be just a dozen simple program lines. Can anyone help me?
I'm not convinced that DSD is enough of an advantage over PCM. The equipment is going to add in a decent amount of distortion. Analog introduced a lot of noise and distortion. PCM through a good converter during the recording and mixing process allows for an insane level of resolution. Even at 88.2 KHz 24 bit you can hear details at half speed that tape would never be able to translate. I like using tape *because* of those distortions, but I digress. I'm also not sold on pure audio recordings (meaning no video aspect) being at 48/96/192. Correct me if im wrong, but isn't it the case that those sample rates are more likely to introduce more audible conversion artifacts than 44.1/88.2/176.4? Maybe my ears aren't as refined or my equipment isn't as expensive but my personal experience in the studio would seem to contradict some of what is said here. I'm not saying your wrong, I'm just saying I can't seem to square this circle. Something isn't adding up.
I'm sure there are effective workarounds, maybe a phase linear low pass filter? Again I haven't personally worked with DSD converters in the studio because 16 channels of DSD is prohibitively expensive compared to just picking up some used rosettas. I'm sure Paul's studio would be a blast to record in too. The money vs. performance thing. I just can't seem to wrap my head around.
Just remember DXD originally came out in 2004, and each time it's converted from DXD over to DSD for distribution/publish which actually rises above 20kHz albeit each time it's converted back 'n' fourth during editing/production some additional noise makes it's way onto the final product. Btw, DXD is also PCM with 24bit resolution sampled at 350 as suppose to 44.1kHz. (I've worked with these formats years ago, and mainly were targetted towards niche' audience who wanted "SuperCD")
@@Wizardofgosz yeah i don't like DSD man,in theory if PCM got so high sample rate,it might beat DSD.And i doubt the DSD ability,because when you convert it into analog,the devil is there.
Some, `Regular,' CDs, sound much better than others. On, `average,' quality players. Some major labels, over the years, have been, `all over the place.' Even back in the `80's, some guys just seemed to have a talent to convert audio tapes, including vintage ones, to obtain excellent sound. Others, it appears, simply went through the motions, and at times achieved poor results. There are quite a few, haven't you heard of, `Remasters?' Quite a few! PCM is simply easier, more convenient!
Steve Hoffman is an excellent, very talented Mastering Engineer. There's an interview with PS Audio's Ted Smith. Rock On! Steve, `thinks out of the box!'
I'm so confused. We're talking about 1's and 0's. Whether they are captured 1 bit at a time or 24 bits at a time they are still just 1's and 0's so the issues in sound quality has to be in the conversion of those 1's and 0's back to what we hear. I know this is simplistic but think about it.
@@drunkenhearted7256 As Paul has pointed out you can't do math to a 1 Bit DSD stream which is why they have to convert to PCM or DXD to do the mixing and level changes. The actual storage of the music though is 1's and 0's that get converted to an analog signal for us to listen to. Why the different methods of storing the music sound different must lie in that conversion back to analog.
Timing my boy timing. Think of it like gas, when you go to the pump and you pump gas It comes out of a constant rate, now imagine for a moment that you could affect the quality of the gas coming out of the nozzle simply by changing the rate of flow. Also imagine that different pumps will have different flow rates and so by extension will have different qualities of gas. That in a nutshell is timing and by and large is a big part of the difference in digital audio sound quality, and of course let's not forget jitter which can and does affect sonic characteristics.
@@Audiorevue I prefer the analogy of interlaced vs. progressive scan TV. Both systems depend on fooling the human senses with the speed of how they produce a picture. Few people argue that progressive scan is superior but single bit recording systems depend on fooling your hearing with the speed of forming each word for the sample rate. It is just a conundrum I have never reconciled because with most things in high end audio the differences tend to be small and depend on what the listener hears and likes.
All 1's and 0's are not created equal. They are different languages that uses a different algorithm thus the meaning of the 1's and 0's are different with different results. For example; the capital letter 'A' is represented by 01000001 and 11000001 in ASCII and EBCDIC code sets respectively. DSD and PCM capture and represent audio in completely different manner which also has an audible difference.
Omg Paul you’re so not wrong or right! I love the explanation… but… I think people are looking for resolution their systems don’t resolve. Trying to look for sounds they can’t get. I have a $5k+system British and am mooned with aptX cd quality sounds. I won’t lie and say I’m an expert finding online lossless music. I don’t see paying for it. Not enough content. The day will come. I’ll wait for it.
Any lossless file will sound pleasant on a decent system. However, you are correct about people seeking resolutions in music files they can't even play. At the present moment, I am able to play back PCM resolution up to 32/384 and DSD from 64 to 256, with my Fiio M17. I haven't come around to owning a nice HiFi audio system yet. My M17 will suffice for the time being, as it sits in my home office - I am able to work or relax with a pair of high-resolution "cans" over my ears listening to tunes.
As I understand it, DSD is easier to convert to analog as literally all you need is a low pass filter, even just a good balancing transformer. Where PCM requires some fairly complicated integration to get even close. That is why the DAC is so critical in PCM and can completely make or break success. You talk about not hearing a difference? But you get to use the sam e DACs. Some guy at home as whatever in a DAC. So wouldn't the possibility of better playback be with DSD because of removing the variability of a PCM DAC? Just the big sonic difference between a generic RtoR and Sigma Delta.
yeah it's pulse density modulation or pulse width modulation kind of similar. I don't understand this technology much,if it works,it works.PCM is more mathematical friendly to my brain to understand. If this technology being used in the future,old PCM,wav files will be forgotten.It is so sad.
i don't know if this type of modulation is actually superior to the traditional PCM or not.Maybe only mathematicians can answer,But i doubt that because analog is still a realm of unpredictable aspects cause by unmatched components or noises,who makes sure that the system would be more fidelity?
@@lanmichaelmix2818 The original promise for digital audio was consistent prefect playback. If that was the case there would not be competition in DACs. So everything you ascribe to analog is obviously just as true in digital. And added to analog.
In my opinion mixing and equalizing in digital domain doesn’t sound good. Analogue mixer if it’s good quality sounds way better. But good analogue mixing console is not cheap, its very expensive. Maybe I’m not an expert, but I always choose AAD early CDs and pure DSD SACD.
Roon is very good. Audirvana and J River offer good solutions. You'll need a DSD DAC to hear DSD256 downloads. Many are available from $300-$30,000 and more. Paul sells a wonderful DAC. We use Roon in our studio with Windows.
Hey man with all due respect your comment comes off really entitled and in poor taste. They really do not have to justify their preferences to you for any reason. If you don’t agree with them that’s cool but try to avoid demanding someone present you with scientific data to support their own opinions…
I assume your a guy so let's sit around and stroke each other with the knowledge that what we know is the truth. And then after we're done with that we can intimately copulate on the basis of what it is we think he's not telling us and how that juxtaposes against our supposed truth.
@@drunkenhearted7256 But Paul clearly states as a fact repeatedly in his videos that DSD is superior to PCM but offers no evidence for his claim. It is niether unreasonable, or condescending to ask a person to back up their claim with facts.
Ah, there we go again, measurements! There is so much we cannot or do not yet know how to measure, and often what we do measure has nothing to do with sound. Trust your ears.
So the bottom line is that anybody buying a DSD (PDM) SACD is actually just getting a PDM conversion of a PCM file. It's funny how you lament the DSD recordings fidelity, but in the end, you slip in that there is some loss in the PCM to DSD (PDM) conversion and as a result, people who purchase your SACDs are not getting the same DSD (PDM) quality that you are touting in the recording process.
Yeah, I don’t see how it’s funny. He says recording in DSD sounds much better than recording in PCM. Then you have to convert to PCM to mix, than you can get the file in which format you like.
Not all DSD recording engineers use Paul's methods. So no, not all DSD recordings are converted from PCM. I agree with a lot that Paul says, but it happens at Blue Coast Records use a different method that avoids conversions from PCM
I suggest, please listen again, perhaps with a little more care. May be a little mixed up, confusing, explanation, `goes back & forth:' 1 - DSD = SACD; 2 - DSD is not: PCM. I don't recall him mentioning: PDM. Pulse Density Modulation? While, `your ears, my ears, are not Paul's ears,' I appreciate how he considers the results, very carefully, & so have spent a bunch of $, plus, for a complete PS Audio system, + recommended connectors; through other quality speakers, everything purchased before his were available. Perhaps: Someday!
Funny enough that DXD was originally developed by Philips & Merging Technologies to make high resolution recordings in 2004!, and sell "Super CD's". Altho DXD is PCM with 24bit resolution and sampled at ~350.0kHz which is 8x more then standard CD. You can also record/edit in "Digital eXtreme Definition", albeit during the converstion to DSD for publishing it will add 20kHz or more noise to the signal each time it's converted back back 'n' fourth during editing so should keep that to minimal!
DoP is DSD -- it's DSD encapsulated in a PCM to be transported within a PCM system to an endpoint that can deal with DSD. i.e. A computer audio subsystem is build on PCM thus to transport a DSD signal through the subsystem it is chucked up into PCM packets, transmitted via the subsystem to your DAC, and then unpacked back into its original DSD stream for playback.
I still don't understand why after it's in PCM, you'd convert back to DSD. Just because you want it in DSD? If you're going for absolute clarity, you'd just keep it in PCM. It has to be a business decision, not a quality of clarity decision.
When editing in DXD in the Pyramix.. it's not a conversion back to DSD256, it's a "RENDER". Sonicaly, rendering back to DSD256 is much, much better sounding than DXD. Again, it's not a conversion.. rendering the edits back to DSD only affects where edits were made and can be less than a 1/4 of a second. You hear the better sound immediately rendering back to DSD. I work with DSD daily in the studio and it's a personal decision to provide our customers the best sound I can... which is DSD256 in our case.
Hey Paul… please don’t discontinue the PerfectWave and DirectStream before i can pick ‘em up… or at least have sales contact me before you do. Thanks for investing so heavily in High Quality Digital