Skip to main content

Notice

Please note that most of the software linked on this forum is likely to be safe to use. If you are unsure, feel free to ask in the relevant topics, or send a private message to an administrator or moderator. To help curb the problems of false positives, or in the event that you do find actual malware, you can contribute through the article linked here.
Topic: HDCD Decoder (Read 249269 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

Re: HDCD Decoder

Reply #500
Hello, everyone! It's been a while.

So, while the channel decoding bug I had reported here has been fixed, I am still having trouble with the 12 Rounds release "Pleasant Smell" that I am referring to in the aforementioned post. The extended peaks are still INCREDIBLY loud and I can hear crackling/clipping noises even if I decode to 24-bit with halving the volume (I checked both with "Always" and "Only when PE is enabled"). ffmpeg and the latest version of libhdcd produce the same exact errors. Interestingly, foo_hdcd's scanner says that the maximum gain is 0.0db, while libhdcd's hdcd_detect.exe says that the max_gain_adj is -4.0. Both say that the PE is "enabled permanently". TF is disabled and there are no detectable errors. Also, hdcd_detect calls it "B:5806", whatever that means. I haven't had any similar problems with the countless PE-enabled HDCD's I've decoded.

Could I trouble someone with a PM so they can have a look at the files? Thanks in advance.

I haven't checked "For the Masses" yet; I will report back if I find similar errors.

Re: HDCD Decoder

Reply #501
You will need to re-scan your files' ReplayGain levels any time you mess with the HDCD decoder settings.

Re: HDCD Decoder

Reply #502
You will need to re-scan your files' ReplayGain levels any time you mess with the HDCD decoder settings.

I tried that on a track, but it doesn't do anything, since ReplayGain is applied after the HDCD postprocessing and the result is already clipped before it reaches the RG. (for the record, when Halve Volume is set to "Never", the RG is -6.69db, while when it's set to "Always" or "Only when...", it's -0.67db). No matter if the volume is turned down by the HDCD decoder or with applied RG, it still sounds horrible even with the additional headroom. The only time I'm getting acceptable results is when I decode with Halve set to "Never" in 16- or 24-bit (NOT 32-bit!) and then turn it down manually in a DAW. It's definitely the weirdest HDCD ever.






Re: HDCD Decoder

Reply #503
I'll need a PM of an example track from this album, or I'll have to find it myself. RG difference of 6 dB makes sense for RG halving the volume on its own, and the filter should be applying both the peak extension and the gain reduction features in an unclipped fashion. Gain doesn't really need it, since it's only reducing the gain, but peak extension does need it, since it's effectively doubling the volume level.

Re: HDCD Decoder

Reply #504
Cool, I didn't know it was possible to screw up an HDCD. This appears to be a screwed up HDCD. It is full of masses of hard clipped peaks, and the enablement of peak extension does nothing but accentuate that clipping and make it quite audible. This CD should have the "HDCD=no" tag set on it to prevent this from being decoded at all. Blame the producer for pumping the volume so damn much.

Re: HDCD Decoder

Reply #505
Sadly, it's something that happened more than once. I can think of at least one great song on an HDCD with clipping problems.
Think millionaire, but with cannons.

 

Re: HDCD Decoder

Reply #506
I see. Thank you all for the info!

 
SimplePortal 1.0.0 RC1 © 2008-2021