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: Restoring transients/loud parts on loud recordings, my method (Read 7401 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

Re: Restoring transients/loud parts on loud recordings, my method

Reply #25 –
Hi,
Interesting idea.  I certainly don't want to put dampeners on your enthusiasm with this project. 

I am hearing a little increase in distortion/artefacts in the processed tracks.  I am matching level for the vocal only, ignoring everything else, and have turned down your "coolerthanme" processed version by another 2dB.  The processed/unprocessed sound closer then.

A lot of the modern production techniques are not just about mastering but start really from the point after recording (and sometimes before but less so).  So much layered mangling happens now that it's a wonder there is anything left that is sonicly pleasing.  There must be at least 4 passes of dynamics processing before it gets to the mastering engineers now.

As much as I really can't be bothered with vinyl and am kinda biased against it as a medium because of it's limitations over a digital release that has been recorded and processed digitally, it would seem that some material has just been produced/mixed and obviously mastered differently for that medium.  Thus to get under the blanket of hyped mid range and brutal transient destruction, vinyl releases can sometimes provide the answer.  Of course, in an ideal world, they'd put out two digital releases, one done 'properly' with no concern for loudness as a target, and another doing whatever they think they 'must' to sell it, in an mp3 or something 'cause frankly it probably doesn't matter after they butcher it.

No intention to steer off topic. 
No attempt to start Vinyl vs. CD, or digital vs. analogue debates.

Re: Restoring transients/loud parts on loud recordings, my method

Reply #26 –
So... what would a time/frequency magnitude analysis of a lifeless CD vs a decent vinyl release look like? Would the vinyl provide hints to what dynamic processing was carried out on the CD release?

View the cd/vinyl as a sort of vocoder pair, where the cd supplies a waveform, while the vinyl supplies subband gain modulation.

-k
 

Re: Restoring transients/loud parts on loud recordings, my method

Reply #27 –
Quote from: Spanky Monkey on
I am hearing a little increase in distortion/artefacts in the processed tracks.  I am matching level for the vocal only, ignoring everything else, and have turned down your "coolerthanme" processed version by another 2dB.  The processed/unprocessed sound closer then.
There certainly are, as my processing isn't always perfect (and in this case I rushed a bit, I concede). This is the hardest and most challenging part of the exercise. Sometimes, some loud parts are detected and processed where they shouldn't be, leading to nasty, unexpected and unwanted volume shifts, even after filtering a second time. Most of the time, you can manage pretty transparent results but at times you can hear those volume shifts. You have to be very careful when doing it, especially if you target perfect transparency.

The 2dB increase is completely unwanted however. I mixed my all processed frequency bands directly on top of the -6dB instance of the song. It seems the expansion of the midrange created an overall volume increase (processing the midrange wasn't that necessary here).

Obviously, good speakers or audio set up are needed. You might think the song sounds good and leave it like that. Suddenly, you listen to it on a good headset (in my case, a Sennheiser HD598) and artefacts jump at you (it happened once on me because my expansion attempts were too strong). So in some cases, even a good band-passing isn't enough. I once thought of adding short decayed reverb to my processed frequency bands for those cases.