HDCD DACS

[manisandher]

Yeah, good question. I've spent the last 20 minutes reading through the dBpoweramp forum link I mentioned earlier, plus a HydrogenAudio forum thread. Wow, some heavy reading, and my head is spinning, but I found two particularly interesting posts, which might cast some light on the subject -
https://forum.dbpoweramp.com/showth...Best-practices&p=127272&viewfull=1#post127272
https://hydrogenaud.io/index.php/topic,79427.msg737725.html#msg737725

Yep, I agree with both of these.

If the mastering engineer went to the trouble of using a PM Model One/Two, irrespective of which HDCD processes were engaged during encoding, the sound will likely be good. I don't think I've come across a bad-sounding HDCD yet (from a purely subjective POV, of course). I listen to something like the HDCD version of Brian Ferry's Avalon, and am amazed at how good it still sounds.

Mani.

 

[manisandher]

Wow, some heavy reading, and my head is spinning...

Try reading the PM Model One/Two user manual. Hours of fun .

Mani.

 

[tcli]

With my Model Two, I could disengage all HDCD processing, and yet the HDCD flag would remain in the 16-bit file. Or, I could engage all (four?) of the HDCD processes. So, how does HDCD.exe know which to decode?

It's encoded in the HDCD flags, so the decoder know what to do.
HDCD.exe decode peak extend (PE) and low level extension. It will detect filter changes but don't do anything with it, probably like PMD100 (don't konw for PMD200).
But what is the fourth HDCD process ? , I know about 3 only.

 

[Sokel]

I f you have a PC or laptop with CD-DVD drive just install this on foobar rip them and play them.
I have 5-4 of them but I only ripped Mark Knopfler's Sailing to Philadelphia (with the nice Speedway to Nazareth).

 

[manisandher]

But what is the fourth HDCD process ?

I couldn't remember how many, hence the "?". But from the manual:

"During HDCD 16-bit A/D conversion, or after HDCD 24-bit or 20-bit A/D conversion and 24- bit or 20-bit digital processing, the Model Two’s HDCD 16-bit amplitude encoding options, Peak Extension, HDCD Limiting, Low Level Extension, and HDCD 16-bit Dithers, are used to produce the best possible 16-bit HDCD encoded release masters from an analog or 24-bit or 20-bit source."

Mani.

 

[AnalogSteph]

The EAD DACs that I use are the DSP 7000 EAD models that were state of the art at the time they were made. Obviously, the clocking is no where near as advanced as a modern DAC, and the 20-bit Burr Brown chip is old by modern standards. I am trying to ascertain, how much of a problem this is, or even whether or not is a problem.

For one thing, the trusty PCM63 arguably is good for the best results among all classic multibit DACs, as it can be trimmed out to a very good degree of linearity. I would want to verify that the EAD can accept and decode 24-bit input properly (or at least the first 20 bits, which would be good enough), as you are much more likely to find straight 24-bit output than special treatment for 16-bit equipment these days.

That being said, you can probably find performance as good as this or better for $200 in a modern-day DAC.

I haven't found anything within the computer realm that I liked for doing the HDCD decoding prior to entering a DAC.

Why not? I see no indications for anything relevant not being implemented.

High Definition Compatible Digital - Hydrogenaudio Knowledgebase

wiki.hydrogenaud.io

Also, that method requires a 24 bit file, and I am keeping things at 16 bits.

Why? It's 2023. Any potato can play 24-bit content. Do you have that many HDCDs that actually make use of HDCD features (most notably, peak extension) for any difference in file size to matter? (There are various ways of scanning the files and identifying features used, e.g. you get a context menu option after installing foo_hdcd in Foobar2000.)

For the handful of HDCDs that have found their way into my own collection, it's pretty much been a "software decode them, dither to 16 bits and forget about it" affair here. This is about all the fuss the format deserves nowadays. I wouldn't want to be dragging a 25-year-old solution behind me like a dead deer.

(Not like I'm a stranger to things like this. My mom always wonders why I won't ditch my 2001 art grade A3 scanner - with autofocus no less - and 2005 laser printer for an MFP with ADF and all that jazz. It would be more practical, but I'm not really keen on a cheap CIS scanner.)

 

[AudioStudies]

Any potato can play 24-bit content.

I have a great many devices that I use, both old and new, some made for home audio, some for the pro-audio market. All my music is in uncompressed WAV files, and for both memory storage reasons and compatibility with all my DACs, both old and new, I want to keep everything in 16 bit files.

 

[AudioStudies]

My now retired Mark Levinson No 360S decodes HDCD. You can see it measured here: https://www.audiosciencereview.com/forum/index.php?threads/mark-levinson-no-360s-dac-review.10842/

101 dB SINAD so not bad!

I will sell it for you but seems like they have kept their value well. So won't be very cheap.

Curious to know your asking price.

 

[zeropoint]

Some suggestions. The Denon DVD-A1UD (and DVD-A1XV, DVD-A11, DVD-A1, and other optical disc players) decodes HDCD, IIRC. The old Musical Fidelity X-DAC also decodes HDCD with a PMD100 (PCM69AP, PMD100, YM3623B).

 

[zeropoint]

Here is an extensive list of DAC converters, among which can be found many devices using the PMD100 and 200 decoder/filters:

The complete d/a DAC converter list - DutchAudioClassics.nl

The complete d/a DAC converter list. In this list you can find all equipment and see what d/a converter and cdmechanism it uses.

www.dutchaudioclassics.nl

 

[pkane]

Yeah, good question. I've spent the last 20 minutes reading through the dBpoweramp forum link I mentioned earlier, plus a HydrogenAudio forum thread. Wow, some heavy reading, and my head is spinning, but I found two particularly interesting posts, which might cast some light on the subject -
https://forum.dbpoweramp.com/showth...Best-practices&p=127272&viewfull=1#post127272
https://hydrogenaud.io/index.php/topic,79427.msg737725.html#msg737725

I've ripped a number of HDCDs using dBPoweramp. The one I have handy (Mike Garson's Oxnards Sessions Vol 2) shows that at least the low-level and peak extensions are being processed. Peaks are taller, and troughs are deeper.

Peak extension:

Low-level extensions:

 

[AudioStudies]

I know in some ways this is like a group of guys in their 20s, were one of them goes crazy after a certain girl and the other can't quite figure it out. Well my audio girl is HDCD, particularly now that I have moved away from vinyl and for the most part moved away from tubes also (tubes remain in only one system). I confess to love messing around with a lot of vintage hardware also and none more so than that with HDCD, since all my turntables are being sold. I am in the modern world with my Aurender servers and have approximately 1 terrabyte of music all in 16 bit 44.1k uncompressed WAV files. Some of these files are duplicates, because not only is the music separated by genre, but I have some folders separated by technology also, such as HDCD, Super Bit Mapping (SBM), XRCD, and audiophile niche labels such as Mobile Fidelity and labels that were known for good recording techniques such as Nimbus. So I have a large folder dedicated solely to HDCD files. I must confess the EAD dacs hold their own pretty damn well in this modern world, if not state of the art by measurement, perhaps pretty darn good enough for my ears. But I want to experiment more . . . as always . . .

 

[pkane]

I know in some ways this is like a group of guys in their 20s, were one of them goes crazy after a certain girl and the other can't quite figure it out. Well my audio girl is HDCD, particularly now that I have moved away from vinyl and for the most part moved away from tubes also (tubes remain in only one system). I confess to love messing around with a lot of vintage hardware also and none more so than that with HDCD, since all my turntables are being sold. I am in the modern world with my Aurender servers and have approximately 1 terrabyte of music all in 16 bit 44.1k uncompressed WAV files. Some of these files are duplicates, because not only is the music separated by genre, but I have some folders separated by technology also, such as HDCD, Super Bit Mapping (SBM), XRCD, and audiophile niche labels such as Mobile Fidelity and labels that were known for good recording techniques such as Nimbus. So I have a large folder dedicated solely to HDCD files. I must confess the EAD dacs hold their own pretty damn well in this modern world, if not state of the art by measurement, perhaps pretty darn good enough for my ears. But I want to experiment more . . . as always . . .

I still have a DAC with HDCD decoding somewhere buried in the basement. I did like the sound of HDCDs when they first came out, although not much content was available then. When 16-bit CD was the only digital format, HDCD was an improvement, although not a major one, in sound quality.

 

[SoundsGood64]

When playing an HDCD (or streaming the standard CD rip of that HDCD) on Sony UBP-x800m2 Blu-ray player connected to Denon AVR-3808 via HDMI, the "HDCD" indicator on the AVR-3808 lights up. AVR-3808 can be used as a DAC, but more of course, and would be a relatively inexpensive solution.

LOL at AnalogSteph's metaphor & simile of potatos and deer in post 26 above. Somewhat older than AVR-3808, I still use an HP laser printer from 24 years ago that gets data from a parallel port. Slow but simple & reliable.

 

[AudioStudies]

The California Audio Labs HDCD player arrived today and I hooked it up a little while ago. Very happy with it, as it sounds great as a stand alone and also as a transport feeding one of my EAD DSP 7000s.

 

[linuxfan]

This discussion prompted me to search through my digital music collection to find which albums originated from a CD with the HDCD control signal - it turns out I have just four; Boz Scaggs - Dig and Speak Low, and Stevie Wonder - Talking Book and Songs in the Key of Life (Audio Fidelity remasters). At whatever point in time that I ripped these CD's I remember that I checked them for the HDCD marker, and having found it, I then used hdcd.exe to decode/convert them to 24bit 44.1kHz WAV, then losslessly compressed these to FLAC.

I just re-ripped these four CD's now, using Exact Audio Copy. Then I opened these WAV's in foobar2000, and scanned them using the foo_hdcd plugin.
For the two Boz Scaggs CD's the result is - Peak extension: Disabled LLE: 0.0 dB Transient filter: Disabled
For the two Stevie Wonder CD's the result is - Peak extension: Disabled LLE: 0.0 dB Transient filter: Intermittent

So it appears that in my situation, HDCD is a big nothingburger!

 

[manisandher]

So it appears that in my situation, HDCD is a big nothingburger!

But you can rest assured that they were mastered using the best equipment available at the time. And you can therefore assume that everything was probably done with care.

Mani.

 

[linuxfan]

Sure.
I'm just disappointed that my HDCD decode process achieved nothing ... well actually, I just checked now, and I see that the levels were dropped by 6dB.
There's no point keeping my old 24bit "decoded" versions. I will replace with the 16bit unmodified rips.

 

[Ultrasonic]

I have a Tag McLaren DAC20 that decodes HDCD, currently sitting unused in a box somewhere. The few HDCDs I had all sounded good but it was impossible to tell if this was purely because they were simply well-recorded or due specifically to them being HDCDs. My hunch was always the former though.

 

[nerdstrike]

Ah, that's what the light on my RCD01 is for! I've never seen it lit.