Skip to main content
Topic: My First HDCD disc - show as 44.1/16 in Windows Properties.... (Read 4390 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

My First HDCD disc - show as 44.1/16 in Windows Properties....

Hi all

Just popped in a disc to rip which is showing in dbPoweramp as HDCD (in Track Technical column).  Yet if I right click on one of the tracks in Windows Explorer it says it is 44.1/16.... 

Any ideas why this is the case?

Cheers

Max

My First HDCD disc - show as 44.1/16 in Windows Properties....

Reply #1
I'm sure many here know the technical reason, but my thinking is Windows Explorer wouldn't know the difference. Even dBpoweramp, by default, will rip it like any other redbook CD unless you enable its HDCD decode DSP during the ripping process, IIRC.

My First HDCD disc - show as 44.1/16 in Windows Properties....

Reply #2
Technically it's a normal CD, just processed in such a way that a HDCD decoder can supposedly extract up to 20 bits per sample.

Maybe the Windows Explorer never recognized HDCD?
"I hear it when I see it."

My First HDCD disc - show as 44.1/16 in Windows Properties....

Reply #3
I think the only to determine that a CD is HDCD is to actually read the audio data from it, which Windows Explorer probably does not do.

My First HDCD disc - show as 44.1/16 in Windows Properties....

Reply #4
Yes, you have to detect HDCD encoding by analyzing the audio stream and Windows Explorer does not do this.  It only presents a silly "file" construct using information found in the TOC which does not contain any information about HDCD encoding.

Windows Media Player supports HDCD playback.  Considering who owns HDCD, one should hope so!
Is 24-bit/192kHz good enough for your lo-fi vinyl, or do you need 32/384?

My First HDCD disc - show as 44.1/16 in Windows Properties....

Reply #5
Thanks for all the help guys. 

Greynol - that was your 10000th post!  Nice innings! 

OK, so I just want my FLAC to be 44.1/16 so presumably I can just rip this in DBPA exactly the same as i rip all my other discs...  Is that right?

My First HDCD disc - show as 44.1/16 in Windows Properties....

Reply #6
Yep, as long as the rip is bit-accurate the HDCD data will be preserved. I've got several HDCD discs I've ripped and compressed with WavPack and the HDCD plugin for FB2K works just fine with them.

My First HDCD disc - show as 44.1/16 in Windows Properties....

Reply #7
OK, so I just want my FLAC to be 44.1/16 so presumably I can just rip this in DBPA exactly the same as i rip all my other discs...  Is that right?
This thread looks pretty informative: http://forum.dbpoweramp.com/showthread.php...-Best-practices and it even references a thread back here on HA. FWIW, none of my own HDCD's have any of the special features mentioned. You could test by ripping a track normally and then again with HDCD decode. I think chances are they will be more or less identical aside from bit-depth.

EDIT: Rotareneg beat me to it. I was going to say you won't lose any information by just ripping normally (to lossless). You can always find plugins to decode during playback if you needed.

My First HDCD disc - show as 44.1/16 in Windows Properties....

Reply #8
Hmm related question I guess: Some of the CDs I ripped to flac/tak with CUETools are also detected as HDCD, stating something like
"HDCD: peak extend: yes, transient filter: none, gain: none"

- what does each of those mean?
- how do I properly play the files so that it is played as HDCD with e.g. foobar2000?
answered by the post that appeared while I was gathering the exact info

My First HDCD disc - show as 44.1/16 in Windows Properties....

Reply #9
Hmm related question I guess: Some of the CDs I ripped to flac/tak with CUETools are also detected as HDCD, stating something like
"HDCD: peak extend: yes, transient filter: none, gain: none"

- what does each of those mean?


They are playback switches that work in tandem with optional HDCD mastering switches:

Peak extend -- restores dynamic range that shouldn't have been reduced in the first place 
Transient filitering -- switching anti-imaging filters during D/A to match the multiple anti-aliasing filters used during A/D.  Why was it done in the first place?  Keith Johnson said it was necessary.
Gain: -- reverses low level signal gain applied for no good reason at the mastering stage, to again 'restore' dynamic range that should not have been reduced in the first place.



There's a lot of HDCD info here if you dig into the links:
http://www.goodwinshighend.com/hdcd.htm
as long as you ignore the audiophile hype about how much better it all sounds than CD.

Here's a summary on one of those pages that covers the features  you ask about

Quote
HDCD Decoding and Digital Filtering

The HDCD decoding process performs precise decoding of HDCD-encoded recordings and also provides a state-of-the-art digital filter for both HDCD and standard recordings. The decoding process begins with the HDCD decoder extracting the hidden code from the LSBs of the audio data and then decoding the commands contained in the hidden code. Signal peaks limited with Peak Extend are restored and low-level gain is undone, resulting in a signal with 20 bits of dynamic range. This signal then is interpolated to 96 or 8 8.2 kHz using a filter that is complementary to the anti-alias filters used in the A/D filter-switching process. The signal can be output at 96 or 88.2 kHz or further interpolated to four or eight times over sampled frequency to drive popular 18-bit to 24-bit D/A converters.



Also of possible interest (again with the caveat to ignore the audiophile golden-ear babble*)
http://www.audioasylum.com/audio/digital/m.../15/156241.html



There's also at least one thread here on HA where HDCD technicalities were discussed in depth. 


(*No one has ever demonstrated in a fashion that we here on HA would accept as valid, that HDCD of itself sounds different from/better than plain old Redbook. )

My First HDCD disc - show as 44.1/16 in Windows Properties....

Reply #10
This thread looks pretty informative: http://forum.dbpoweramp.com/showthread.php...-Best-practices


Gosh, has it become 54 pages over there. Somewhere therein I have been arguing (and arguing well I think) against converting HDCD when using lossless formats. Long story short, the files can always be converted later but you can never undo it. Playback? On-the-fly in fb2k.

And I have CDs with some but not all CDs HDCD-encoded ...
Memento: this is Hydrogenaudio. Do not assume good faith.


My First HDCD disc - show as 44.1/16 in Windows Properties....

Reply #12
Hi

Thanks to everyone for the replies.  I've got to be honest, my head is swimming a bit with this!  Hopefully a couple of questions will help clear things up for me.  Sorry for being so slow in "getting it"...

as long as the rip is bit-accurate the HDCD data will be preserved. I've got several HDCD discs I've ripped and compressed with WavPack and the HDCD plugin for FB2K works just fine with them.


So if I rip in dbPoweramp with no DSPs or Codecs, i.e. I rip as I would any other CD, will this give a bit-accurate rip? 

And when I playback in Foobar, if I do NOT HDCD plugin will it play back as a normal ripped CD/FLAC file?

You can always find plugins to decode during playback if you needed.


Same as above really....  If I don't decode during playback, what happens? 


On another note, aside from the FLAC files I also rip to mp3 for use with an iPod.  Is there anything I need to do or think about with that in mind? 

Thanks for sticking with me...

My First HDCD disc - show as 44.1/16 in Windows Properties....

Reply #13
If you just decode it normally, its a normal CD audio track.  If you decode it using a decoder that is able to detect the HDCD information, you get marginally more dynamic range.

My First HDCD disc - show as 44.1/16 in Windows Properties....

Reply #14
So if I rip in dbPoweramp with no DSPs or Codecs, i.e. I rip as I would any other CD, will this give a bit-accurate rip?
Provided the disc isn't damaged, it should.
And when I playback in Foobar, if I do NOT HDCD plugin will it play back as a normal ripped CD/FLAC file?
Yes.
Same as above really....  If I don't decode during playback, what happens?
It will depend on how your particular disc was encoded. That's why we recommend researching it a bit. If it has peak extend for example, and you DON'T decode it as HDCD, you might get some clipped peaks maybe? It will play, in any case. In my personal collection, they're ripped no differently than any other CD.

My First HDCD disc - show as 44.1/16 in Windows Properties....

Reply #15
(*No one has ever demonstrated in a fashion that we here on HA would accept as valid, that HDCD of itself sounds different from/better than plain old Redbook. )


Interesting reading

I agree with that thread in that there are no audible differences between an HDCD decoded (20 bit) vs the decoded version dithered back to 16 bit (redbook) as you say. However when transient filter has been used for the HDCD creation, the soft limiting from the original undecoded redbook audio against the restored peaks from the decoded one may be very well ABXable. I remember after reading that thread I made my own tests and although I wasn't successful with all the samples I tried, there was at least one with distortion not that hard to notice indeed.

BTW, I am always talking about peak extension because I have honestly no idea if anyone has ever been able to ABX an HDCD which uses LLE only. Not me, that's for sure.

If you just decode it normally, its a normal CD audio track.  If you decode it using a decoder that is able to detect the HDCD information, you get marginally more dynamic range.


I must agree here with the small differences in dynamic range between HDCD encoded an decoded audio. If I recall correctly the specification says that the maximun gain for peak extension in HDCDs is 6dB which are barely archieved as far as I can tell. 3, 2 or even only 1dB are more usual values in my experience. However, I think that difference shouldn't be ignored or underestimate without being put in some context.

This is just a guess with not enough statistical base at all from my side (my HDCD collection in very limited), but I am under the impression that HDCDs released with peak extension, have actually greater dynamic range than the average in CDs from the same music genre/release year thanks to those extra dB. Perhaps it is not that noticeable if we look at the overall album gains, but particulary at certain tracks. I recall in particular a pop song from a very mainstream singer in Japan which reached DR12 from DR9 after decoding in an EP she released in 2003. And it wasn't her only release as HDCD during those years with around DR10 in most of her albums.

Such things are quite suprising to find for me to be honest, although as I say, I do not have too many CDs for testing to draw strong conclusions. It would be something interesting to investigate deeper, though.

My First HDCD disc - show as 44.1/16 in Windows Properties....

Reply #16
Interesting reading
Hm, my first impression after installing the HDCD decoder yesterday was similar to what Wombat posted in that thread - the clipping seems to have/gone away with restored peaks. I didn't do an ABX test though.


Quote
I must agree here with the small differences in dynamic range between HDCD encoded an decoded audio. If I recall correctly the specification says that the maximun gain for peak extension in HDCDs is 6dB which are barely archieved as far as I can tell. 3, 2 or even only 1dB are more usual values in my experience. However, I think that difference shouldn't be ignored or underestimate without being put in some context.
On the 2-3 CDs that gets identified as HDCD with peak extension, the RG value changed by the full 6dB after scanning them with HDCD decoder active. Is that not the full range?

My First HDCD disc - show as 44.1/16 in Windows Properties....

Reply #17
Quote
On the 2-3 CDs that gets identified as HDCD with peak extension, the RG value changed by the full 6dB after scanning them with HDCD decoder active. Is that not the full range?

Are you confusing RG and DR?

My First HDCD disc - show as 44.1/16 in Windows Properties....

Reply #18
Quote
On the 2-3 CDs that gets identified as HDCD with peak extension, the RG value changed by the full 6dB after scanning them with HDCD decoder active. Is that not the full range?

If the peaks remain the same (0dB?) and ReplayGain tells you the overall loudness has been reduced by 6dB, I'd say yes!



P.S
You may not hear a 6dB change in dynamics...  i,e. A few short-term +6dB peaks here and there won't change your perception of the sound.    If there are enough boosted peaks in a row for your brain to detect an increase in loudness, then you'll hear it.

My First HDCD disc - show as 44.1/16 in Windows Properties....

Reply #19
Are you confusing RG and DR?


Should be "RG", not "DR" except by coincidence; a HDCD decoder will reduce by 6 dB first, in order to make room for the peak extension; then it will stretch what was the top 3 dB (which are now the -6 to -9 range) to filling the 0 to -9 dB range. Most likely I  have oversimplified.

As for comparing the effect of HDCD:
I suspect that some talk about the effect of decoding an encoded CD (which is what they as consumers have at hand) and some will be referring to whether the process of encoding and decoding HDCD is transparent. Maybe people could be a bit clearer ...?
Memento: this is Hydrogenaudio. Do not assume good faith.

My First HDCD disc - show as 44.1/16 in Windows Properties....

Reply #20
This is pretty confusing...

Without hdcd plugin, one sample track has
- track peak 0.973480
- DR5

With hdcd plugin, the same track has
- track peak 0.677700
- DR7


My First HDCD disc - show as 44.1/16 in Windows Properties....

Reply #21
The decoding does not increase the peaks - it can't because they are already at maximum. Instead it lowers everything else.

 

My First HDCD disc - show as 44.1/16 in Windows Properties....

Reply #22
Like a downward-expander or upward-expander with attenuation (enough to ensure that you don't get clipping even in the worst case).
"I hear it when I see it."

 
SimplePortal 1.0.0 RC1 © 2008-2018