I was wondering if I could rip a HDCD encoded CD into ALAC or FLAC or even m4a lossy and play it at the full resolution through a amplifier/ receiver with HDCD decoder chip. On the Microsoft web site ( they own HDCD now) it says you could use a non-hdcd cd player with a digital output to an amp with HDCD decoding capabilities and it would work. I don't have the amp now ,but if anyone can verify that it works I'd love to get one. I've read in other post here that all the HDCD info should be copied if you rip to lossless. I have listened to my HDCD discs in windows media player and I can really tell the difference (through headphones) over a non HDCD sound. It would be awesome to be able to rip the discs to a hard drive player with digital outs. Too bad my iPod doesn't have digital output. But anyway , do you guys think this would work or know it does?

Lossless yes. lossy no. And lossless only if sent digitally to this amplifier.

The concept works using the less significant bits to store data instead of audio content. (the actual implementation might vary). This data is then used to reconstruct audio, similarly to how SBR or PS work for AAC. I don't have any hdcd disc, so i can't comment on it being a small or big change on the audio. Common sense say it shouldn't be too much.

Now that we're on the topic of HDCD discs...

I'm aware of one HDCD disc in my collection: "Lateralus" by Tool. When I listened to the CD using Windows Media Player (with the HDCD logo visible, so the HDCD data was being used) I didn't really notice a difference.

However, I'm thinking of doing some ABX testing on it. Now my question is: has anybody of you ever heard a difference between HDCD enabled and HDCD disabled on this particular CD? Can anybody of you recommend me a specific passage on the disc where the difference should be audible?

I have both Lateralus and Beck's Midnight Vultures in HDCD. I'm not much of an ABXer though... should be easy enough though to do at least a single-blind test using WMP... just have a friend go into the device settings for your speakers within WMP and enable/disable 24-bit output for HDCD to work (you will need to perform this test on a 24-bit soundcard).

Note that this is single-blind because the tester (your friend) knows which setting he is using.

I thought HDCDs were only regular CDs (16-bit audio) with noise shaped dithering / eg they sacrifice sample rate to increase SnR...

I noticed quite a difference with Beck Mutations and Sea-change also by beck. I plugged my er4p and shure e4s directly into the soundcard. I never did an abx either ,but I thought it sounded a lot better than regular cda. I don't know that I could do a very good job explaining how technically. It felt like I was more involved with the music. Like I was in the studio as opposed to being farther away? If that makes sense. I Think Sea Change sounded the best, I'll have to try them out again.
But Thanks for verifying that it would work. I'm Thinking about some hardware I'll need to get.

Never underestimate the power of placebo. You'd really be surprised. I suggest you just do a simple ABX to see if you really can tell the difference.

http://en.wikipedia.org/wiki/HDCD

It seems HDCDs really do have the full 20 bit range, and only on specific players. However, aside from the aliasing filter, I doubt the 4 extra bits really make any difference to the mere mortal.

Even if they do, ABXing 16 verses 20 bits for typical music is hardly a given. Even if they have the full 20 bits resolution with no downsides, I'd be pretty surprised if this made any difference.

You guys might be right ,I was excited to find out some of my favorite albums had this feature. Its not easy to do an abx with WMPlayer. That is the only decoder I have now. I ,d need to have somone switch the tracks for me ,so I won't know what the x is. I'll let you know if I do that. Thanks for your posts

Shade[ST]: I didn't claim that HDCD didn't give 20-bit audio, I was just saying that the difference he heard may just be placebo. I could be wrong though!

jas1612: I guess the best you can do is a single blind test, since the only software HDCD decoder is WMP. If you had a hardware decoder, you could do a double blind test easily, but you said you don't have any other decoders...

Has anyone actually done a test to see if HDCD makes a measurable difference? Or if anyone's reverse engineering it to figure out how it works?

I suspect that an HDCD disc decoded on HDCD equipment will sound better than an HDCD disc decoded without HDCD. HOWEVER, this is NOT a fair comparison that will indicate HDCD is better than a regular CD - all it indicates is that HDCDs should be played on the correct equipment.

For a fair comparison you must take a perfectly mastered CD (that is one that is NOT mastered with HDCD, as this would cause some compression (one of the HDCD features is a compander) and prevent the optimal choice on whether/how to dither the least significant bit) and pit it against a perfectly mastered HDCD of the same content. Even with the proviso that you could perhaps replace perfectly mastered with really well mastered, this isn't going to happen. There will always be doubts over how well the standard CD was mastered, especially of late.

If it means anything (take this with a grain of salt) Chip Davis of Mannheim Steamroller fame really likes HDCD. Chip Davis always has always been known to be on the cutting edge of hi quality audio recordings.

In any event I thought this subject had been covered before on the boards here.

I agree with you, Martin, about the fair way of testing, because it is vital to retain the dynamics present on the properly decoded HDCD if it is doing any volume adjustments based on the HDCD sub-coded data.

If someone has a virtual soundcard or SPDIF to record via loopback or any other means of digitally capturing the 20 or 24-bit output from Windows Media Player operating in HDCD mode then this 24-bit decoded HDCD output could be saved into a lossless PCM file at 24-bit depth.

That 24-bit 44.1 kHz PCM file could then be converted using foobar2000 (I'd suggest flat dither to give HDCD the maximum chance to win so there can be no accusations against the noise-shaping algorithm) into a lossless format 44.1 kHz 16-bit PCM file (standard CD format). This CD mastered file could be padded out to 24-bit to ensure identical format.

The two files could then be subject to ABX testing by anyone with 24-bit soundcards.

If peak extension is used (it's an optio nduring HDCD encoding) then decoded playback will tend to be several dB lower (RMS avg) than nondecoded. I'd imagine that would be audible. (From what I recall, CD/DVD players licensed to decode HDCD used to be required -- by Pacific Microsonics/Microsoft-- to lower nondecoded HDCD and standard CD output in order to 'level the field -- so people wouldn't say CD sounded better than HDCD due to the psychoacoustics of loudness difference)

Here's an example of an HDCD that appears to have used peak extension

http://www.hydrogenaudio.org/forums/index....mp;#entry475231


THat thread;s also got good posts about HDCD.



I'd be surprised if any sources for the HDCD had a dynamic range large enough to sound different in 16 vs 20 bit. Dynamic range available from dithered CD (>95 dB) is already well beyond the dyn amic range of analog tape sources. THis suggests to me that plain old CD could have captured the full dynamic range of any of those King Crimson, Yes, Tool, Beck etc source tapes, without HDCD.

Apart from 'more bits', though, the other touted 'advantage' of HDCD is supposed to be "Precision digital interpolation filtering with multiple modes of operation, which can reduce alias distortion and temporal smearing, resulting in a more natural, open, and accurate sound reproduction." (to quote from the rather credulous Wikipedia entry cited above).




One possible way to do the ABX: digitally record the output of the nondecoded and decoded versions @24 bits and [insert some silly high sample rate that your 24-bit soundcard supports here], then ABX the two files with foobar.