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: Digital Room correction and coding transparency (Read 3161 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

Digital Room correction and coding transparency

I understand that the received wisdom is that a lossy encoded file which is transparent to  the original cannot be guaranteed to be trans codable to another lossy codec so as to remain transparent to the original.

Does the same apply where one applies DRC to the lossy encoded file? I don't find this easy to test because my dsp antimode does not generate output files (I suppose I could try recording from its S/PDIF output but I'm lazy). IN any event I might change DRC systems and am not sure whether the result would be generalisable.

I'm not bothered about the ability to transcode mp3 to anything else but this is a real issue as I would not want to be without room correction, and I also use digital crossfeed for headphones.



Re: Digital Room correction and coding transparency

Reply #1
If the room correction DSP were perfectly canceling the speakers' & room's undesirable acoustics and never introducing its own artifacts, then I think the answer would be no; it would be the ideal playback situation which the lossy codec was designed for. But such perfection is a long way off, maybe impossible, so I think the answer is always yes, there's a risk that your additional processing is, in effect, lossy. With a well-designed and well-configured system, this could well be so minimal as to be harmless, but there's a chance it could be audible, even with lossless input. No way to know without testing.

Re: Digital Room correction and coding transparency

Reply #2
I understand that the received wisdom is that a lossy encoded file which is transparent to  the original cannot be guaranteed to be trans codable to another lossy codec so as to remain transparent to the original.

Correct. This is due to the fact that lossy coding is so completely nonlinear.

Quote
Does the same apply where one applies DRC to the lossy encoded file?

In general, no.  DRC is typically applied via lossless signal processing, and is thus very different than lossy encoding.



Re: Digital Room correction and coding transparency

Reply #3
Could the room correction processing exacerbate the inaudible artifacts of the lossy encoding and make them audible?

Re: Digital Room correction and coding transparency

Reply #4
Could the room correction processing exacerbate the inaudible artifacts of the lossy encoding and make them audible?

Yes, although if the room processing is working correctly, I would expect this to be less likely than the room response itself exacerbating inaudible artifacts to make them audible.

Re: Digital Room correction and coding transparency

Reply #5
I was wondering whether signal processing might interfere with the temporal and frequency masking assumptions of the codec.

Re: Digital Room correction and coding transparency

Reply #6
I don't imagine that DRC or room reflections will serve to expose artifacts. If anything I think they will further mask them.

Wild changes in frequency response (EDIT: as delivered to the ear, referenced to the source) may be a different story.

Re: Digital Room correction and coding transparency

Reply #7
Could the room correction processing exacerbate the inaudible artifacts of the lossy encoding and make them audible?

It could, but it would be likely tp actually have the reverse effect.  IOW by correcting the subjective frequency response to smooth and balanced, it could make masking more effective than it would be if the system were allowed to have large errors in  frequency response.

Re: Digital Room correction and coding transparency

Reply #8
Could the room correction processing exacerbate the inaudible artifacts of the lossy encoding and make them audible?

It could, but it would be likely tp actually have the reverse effect.  IOW by correcting the subjective frequency response to smooth and balanced, it could make masking more effective than it would be if the system were allowed to have large errors in  frequency response.

That's a good point, DRC done correctly will flatten the response curve at the listening position, and obviously all lossy codecs assume a flat frequency response, because anything else would be silly.

No need to worry, then :-)

Re: Digital Room correction and coding transparency

Reply #9
Can room correction fully compensate for echo? I was just thinking that lossy codecs may allow for more noise at moments when it'll be masked, and then the room reflections will be smearing or otherwise be making those noises audible a few milliseconds beyond the effect of the masking. I have no experience with the theory or practice, though, so if this is an idiotic question, please forgive and maybe dumb it down for me a bit.

Re: Digital Room correction and coding transparency

Reply #10
Can room correction fully compensate for echo?
1 - I think it is fair to call echo "noise" for it is an undesired signal.

2 - and no, DRC can't correct for echo at all. 

I was just thinking that lossy codecs may allow for more noise at moments when it'll be masked, and then the room reflections will be smearing or otherwise be making those noises audible a few milliseconds beyond the effect of the masking.

So if we run with my definition of echo I think what you're asking is if a badly reflecting room (with distant enough walls) might be noisier than a less-reflecting room.  And I think that question answers itself.  So I'm wondering what I'm missing.


Creature of habit.

Re: Digital Room correction and coding transparency

Reply #11
Can room correction fully compensate for echo?

Not in an acoustic environment. Micro-second level timing would be required, and then that would correct echoes in only one location. Move your ears an inch, and the cancellation falls off quickly as frequency increases.

Quote
I was just thinking that lossy codecs may allow for more noise at moments when it'll be masked, and then the room reflections will be smearing or otherwise be making those noises audible a few milliseconds beyond the effect of the masking. I have no experience with the theory or practice, though, so if this is an idiotic question, please forgive and maybe dumb it down for me a bit.

That sort of fantasy is where thinking without reliable references and hands-on doing will get you.

Re: Digital Room correction and coding transparency

Reply #12
Thanks, guys. I will stop speculating. Any audiophiles reading this thread, just ignore my posts here.