Help - Search - Members - Calendar
Full Version: Extract HDCD
Hydrogenaudio Forums > CD-R and Audio Hardware > CD Hardware/Software
Pages: 1, 2, 3
ProtectYaNeck36
QUOTE(SebastianG @ Jan 28 2005, 11:59) *

QUOTE(adlai @ Jan 28 2005, 09:00 AM)
I was always under the impression that HDCD was simply a better dithering algorithm.
*



HDCD is:
1) dithering + noise shaping
2) peak compression (kind of a reversible dynamic compressor)
3) hiding commands in some least significant bits of the samples to tell an HDCD decoder which anti-alias lowpass filter it has to use for upsampling

...whereas most of the dynamic range "increase" comes from (1).
("increase" ? yes, it depends on how you measure it. If you perceptually weight the quantization noise the noise power will be lower compared to dithering without noise shaping)


SebastianG


I just came across this post and I had a question for. #2 threw me off when you said "reversible dynamic compressor." are the signal peaks actually compressed (or limited?) or is it a compressor acting as an expander (the "reversible" part is what lead me to believe you were describing a compressor reversed)?
krabapple
QUOTE(ProtectYaNeck36 @ Oct 20 2007, 12:41) *

I just came across this post and I had a question for. #2 threw me off when you said "reversible dynamic compressor." are the signal peaks actually compressed (or limited?) or is it a compressor acting as an expander (the "reversible" part is what lead me to believe you were describing a compressor reversed)?



Compression during encoding (in the studio), expansion during decoding (in the player) -- it's enabled by the 'peak extension' option available to the mastering engineer.
MRC01
QUOTE(ProtectYaNeck36 @ Oct 20 2007, 08:41) *
are the signal peaks actually compressed (or limited?)

When playing an HDCD encoded CD on a normal (non-HDCD) player, yes - you will usually get dynamic compression (if the HDCD encoding used this feature).

The $64k question is: why? They're not using the full 90+ dB of dynamic range that CD offers, and even after expanding the compression, most recordings still don't use that 90+ dB of dynamic range. In other words, the compression (or "peak extension" as they call it) is pointless because standard redbook CD already offers more dynamic range than they are using.

My cynical side says it's a way to make money by licensing a technology to make recordings "sound better". I use the words "sound better" loosely, since what they're really doing is making the redbook CD version sound worse so when played through the licensed HDCD decoder it sounds better. But ultimately, when played on an HDCD decoder, it should sound the same as it would have sounded if they had never used HDCD or dynamic compression at all.
Wombat
This tool is pretty interesting. I havenīt much HDs but one of them "Yim Hok-Man - Poems Of Thunder, The Master Chinese Percussionist" i bought for 7.95 at Amazon is a HD showing no sign of it on the box. Only a small info in the booklet about mastering equipment using it.
In the first song are some strong Kudo drumhits that are real distorted.
Decoding it with the peak extension found makes these drumhits absolut clear!
Looking in an editor shows the non decoded one as totally smashed.
I replagained both and did an abx in foobar with no problem.
skamp
QUOTE(bryant @ Aug 31 2007, 05:36) *
I also tried this with 3 of my HDCDs, and found that for tracks encoded using "peak extension" you can gain considerable compression improvement by using very small block sizes. I found that --blocksize=4410 worked nicely, and on some tracks --blocksize=2205 was even better. And I'm talking like 5-10% better compression!

These are the results I get for the Joni Mitchell "Blue" HDCD, with Monkey's Audio 3.99-u4-b5, wavpack 4.41.0, flac 1.2.1 (all under linux):

16-bit WAVs (not HDCD decoded)
CODE
176.9 MiB    mac -c5000 (Monkey's Audio insane)
185.4 MiB    wavpack -hhx6
191.1 MiB    flac --best

As expected, here Monkey's Audio compresses much better than WavPack, which itself is substantially better than FLAC.

24-bit WAVs (HDCD decoded), default blocksizes
CODE
338.9 MiB    mac -c5000 (Monkey's Audio insane)
210.6 MiB    wavpack -hhx6
198.8 MiB    flac --best (default blocksize: 4096)

I thought there was something wrong with my setup, until I saw Walrusbonzo's results, confirming mine. I can't believe how bad Monkey's Audio fared.

24-bit WAVs (HDCD decoded), custom blocksizes
CODE
196.6 MiB    wavpack -hhx6 --blocksize=4410
197.1 MiB    wavpack -hhx6 --blocksize=2205
196.1 MiB    mix of the above (smallest encodes from each)
197.1 MiB    flac -l 12 -b 2048 -m -e -r 6 (equivalent to --best)

Now we're talking! Overall, wavpack with a blocksize of 4410 fares better than with a blocksize of 2205, although some tracks (4, 6, and 7) are indeed a bit smaller with the latter.
Note: Monkey's Audio doesn't provide the ability to change blocksizes.
foo_lover
I was thinking of having HDCD decoder as foobar2000 plugin and came to conclusion that it won't fit in player's model... It would break replaygain, for example. So, the only reasonable implementation of HDCD decoder would be as a part of player's core.. IMHO, of course.
kremb
Here is what I did. I'm not sure if I notice any appreciable difference in quality but what the heck.

EAC to .wav files

hdcd.exe .wav files to 24 bit .wav files

import 24 bit .wav files into itunes at 320kbps AAC

Burn itunes playlist to CD


*****

here are the size of the various wave files I burned to CD

original EAC Rip - 23 870 KB - lame mp3 via foobar - 3816 KB
24 bit .wav via hdcd.exe - 35 804 KB - lame mp3 via foobar - 3754 KB
.wav from itunes burned cd via EAC - 23 870 KB - lame mp3 via foobar - 3704 KB
.wav from nero burning 24bit .wav to cd - 23 870 KB - lame mp3 via foobar - 3762 KB

*****

I think I have answer my own question, I didn't think the AAC was going to be so lossy. Looks like simply burning the 24 bit wav files using nero audio selection is the best choice if a person is determined to try and include all 20 bits of data in their NON-hdcd CD backup.

*****

Feel free to tell me where my logic is flawed and should just settle with the backwards compatible hdcd playback.

Does anyone have another suggestion to try and use the most of the 20 bits of data on a regular CD? Is this a job for soundforge? Or into some sort of DVD format that will decode ALL 20 bits?

*****

I've decided to author a dvd with still frame photos for video and the 24 bit wav files for the audio. I'm using pretty simple authoring software called TMPGEnc DVD Author. I think it's going to work out pretty well, other than I'll have to play the tracks on a DVD player.

Hope my long winded process helps someone else.

Enjoy your tunes!

*****

easier said than done. dvd author program insists on reencoding some black box method to AC3.

Thus ....

I decided to compress to 24 bit flac and listen to them via my Ipod using Rockbox. I THINK i can notice a difference with my cans on vs a good little cd player I have. We'll have to see how much of an audiophile I really am as I only have 2GB on my ipod.
Nubben
dBpoweramp R13 (just released) now contains an option to rip these into 24-bit wav/aiff files.

Nubben
odyssey
QUOTE(foo_lover @ Mar 25 2008, 09:29) *

I was thinking of having HDCD decoder as foobar2000 plugin and came to conclusion that it won't fit in player's model... It would break replaygain, for example. So, the only reasonable implementation of HDCD decoder would be as a part of player's core.. IMHO, of course.

Did you see the latest changelog? They implemented some replaygain stuff for 3rd party components. What exactly it is or how it works, i have no idea about, but maybe it might come in handy?
This is a "lo-fi" version of our main content. To view the full version with more information, formatting and images, please click here.
Invision Power Board © 2001-2008 Invision Power Services, Inc.