Why DSD over DXD?

Why not record in DXD (352kHz PCM) in the first place?

[NoEggu]

That information was helpful. Good stuff!

[twitchbook-]

Have a good and thanks for all the advices

[noliverslol]

Good explanation, however, is is true than that de the dsd is a marketing format, since the hi re pcm (dxd) was ' second original" format.. you capture in dsd, but mixing etc is always done in pcm. Or are there dsd recordings which have been edited in dsd too? would that be better? yeah, you got me confused now.

[johnmarchington]

Do you think someone will ever invent a sort of inverse Zephyr filter able to convert the Zephyr-PCM back to DSD without loss - or is that crystal-ball gazing?

[chipsnmydip]

Very intrigued about the zephyr filter, but I see a real problem that if the PCM DAC used has a typical steep/brickwall filter, its still going to sound like PCM since only the first half of the conversion processes, the A-D, is correct. Makes sense then to convert it back to DSD OR use with a DSD upsampling DAC.

[thirtysixverts]

Did some Google searching but came up empty - is this Zephyr filter software? Hardware? Something that happens in Pyramix?

[KallusGarnet]

DSD is an amazing format for playback, but I need to try it out for capturing sound sources, what audio interface/recorder can do this and what software?

[carlstineman]

When you convert DSD to analog for mixing purposes is that done "on the fly" somehow during the mixing process or is there an analog recording made that is then used for mixing? If the latter, doesn't that introduce all of the issues associated with analog recordings like dynamic range limitations etc.?

[jozno]

Paul, I believe the reason you continue to get questions surrounding the Octave choice in recording format, is due to almost 100% of the listeners of the music having no experience with hearing the live recording. Your hearing musicians performances going into a board, is quite different from someone comparing formats while sitting in their listening room. The product is vastly different from their vinyl/CD/download of the same songs.

[charlesnr]

When played back on PS audio equipment, does it filter out all the noise above 20K from the DSD file or the PCM file.

[Also_Ran]

Paul keeps mentioning this Zephyr filter or codec... Is this a Telos Zephyr unit he's talking about? That company's been around a long time and are standard in broadcast facilities. It would make sense if they have some kind of DSD to PCM. But I can't find any other reference to a "zephyr filter" in and of itself.

[danielraju]

Great marketing pitch! So I'm assuming based on his conclusion, engineers using Pyramix and recording in DXD have a problem, since it's in PCM format? Perhaps Paul should/may have listen(ed) to Lyndberg Sound and choose to believe if he could ever achieve the sonority of sound as on those recordings with DSD. I'm guessing Paul's audience is an audiophile community and not an engineering community.

[dripslobber]

Where did you get the picture behind you? Thanks for all your videos.

[vinylrules]

I would still like Paul to let us know what A/D D/A converter(s) he used when comparing to DSD. Not questioning it sounds better, just want to know his reference for converters.

[ThinkingBetter]

ANY digital mastering of music produces a PCM master track as output and ANY transcoding of this PCM output to DSD will ALWAYS be lossy as quantization errors and ultrasonic noise will be added to the signal. Thus, for distribution of audio, DSD makes really only sense for analog mastered music. Nowadays digital mastering can be done in up to 768kHz PCM (Sound Forge Pro) thus digital mastering can be quite excellent. Capturing the analog input through sigma delta ADC to DSD is great but I guess Paul's concern with PCM has to do with some case of poor math involved in transcoding DSD to high resolution PCM (e.g. DXD) per track within the mastering process.

[joesmith]

What’s a Zephyr filter for PCM? How do you get one?

[MrMftech]

More information about DSD-DxD is available here:

[JonAnderhub]

Well, we can't really pretend that DSD is even close to the perfect capture given the need to overdrive the Delta-Sigma converter, the need to add a large quantity of noise to keep DSD quantized, and then the need to attempt to filter out that noise although the noise extends into the audible range.
This is just the first disaster to occur!
Now add in the harmonic distortion and quantization errors that occur in the DSD to PCM conversion, the fact more noise, but different noise has to be added to the PCM for quantization purposes and the nyquist filtering required for PCM.
If that isn't enough, yet more distortion, more quantization errors, and even more noise has to be added to the signal, and yet another Delta-Sigma converter distortion due to being overdriven again.

So surely the DSD to analog, to DSD conversion must be better right?
Sorry all you Audiophiles that pretend DSD is so much better sounding, it is not.
Besides the already mentioned distortion caused by the overdriven Delta Sigma converter, the high levels of noise added for dithering, and the subsequent need to use a low pass filter in the digital to analog conversion that will either leave residual noise in the upper audible frequencies or will filter out the upper frequencies.
Now add in the considerably higher noise level of an analog signal path, along with analog phase shift at every part of the "editing" process, the increased harmonic distortion, and loss of 30 dB or more of dynamic range.
That loss of dynamic range and increased noise floor typically requires the addition of compression, read additional distortion, phase shift, and noise.
If all that distortion and noise isn't enough then consider this.
To go back to DSD (PDM) there is yet more Delta-Sigma converter overdrive distortion, the noise and phase shift from the analog source is recorded as well, and even more, large quantities of noise are added once again to quantize the DSD (PDM) signal.
What you Audiophiles are mistaking for warmth is signal distortion, and what you are mistaking for "air", or "space" is the added noise.
For older people like Paul, whose hearing is probably deteriorated due to his age, and who was sold a bill of goods when he was younger by old school record companies selling vinyl as "high fidelity", it is understandable that they might "pretend that DSD is the perfect capture".
But scientifically a pure PCM signal path using quality converters is far superior to DSD in terms of noise, dynamic range, converter to converter distortion, and frequency response.