|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
24.167.222.192
A supposed solution to the massive compression used on recordings.
I thought it might be of interest. I didn't have time to study it, and it sounds like it might make the recording muddy.
What do you think?
Follow Ups:
Parallel compression has been around for a while. It works by splitting a signal and running it through two paths in parallel, one uncompressed (or lightly compressed) and the other heavily compressed, and then mixing the two. The simplistic explanation is that a traditional compressor works by squashing the peaks to bring their level down closer to the average, while parallel compression mostly leaves the peaks alone and raises the low level stuff.
I think it was traditionally used for percussion. I don't know how commonly it's used in mastering, i.e. on whole mixes. It's no panacea.
It may be sonically more innocuous, but you're still losing dynamic range. You're still raising the average level which will cause listeners to turn the volume down and thus reduce the peaks.
The article references Michael Brauer and Andrew Scheps. Michael Brauer is known for elaborate compression schemes using multiple parallel paths with different levels/types of compression, and for using parallel compression for artistic effect on vocals.
My wife and I have a few albums that Michael Brauer is known for as a mixing engineer, including the Coldplay and John Mayer albums that won Grammys. The Coldplay material is typically over-compressed loudness wars throwaway stuff. The John Mayer material is better, but I discovered that the best sounding tracks (IMO) were the ones mixed by Manny Marroquin and not Michael Brauer.
I wasn't really familiar with Andrew Scheps, but Googling his work shows that he's one of the chief villains of the loudness wars, having worked on many of the most compressed, worst sounding albums of the last 15 years including the poster child, Metallica's Death Magnetic. Check his Wikipedia page.
Unfortunately, the brief article you linked is still written in the loudness wars mentality. Quote from the first sentence:
While the "loudness wars" seem to thankfully be winding down, that doesn't mean you don't want the level of your mixes to be as high as possible while still retaining natural dynamics.
No, you don't want the level of your mixes to be as high as possible. Haven't we learned this yet?
There is another factor afoot. Even if you have a really powerful and efficient system, you really cannot reproduce a live concert. They have banks of speakers just for the bass drum sometimes. You might reproduce a rock concert because it is all going through amps and that is compressed to prevent clipping. But a full blown orchestra ? Forget it. You might come close, but if you been to one live you always know when you are not there.
Now, this PC (which I will use henceforth to save keystrokes), when you mix the uncompressed with the compressed it is different than just changing the compression "ratio". What matters here, other than that is the timing.. The rise and fall times of the compression. If this is manipulated properly we might have some apparent gain. And it is high fidelity because we are trying to restore the sound to studio or hall quality. Some people shun equalizers, OK, then they can spend fifty grand to have an addition put on to their house.
The PC has to adapt somehow, it needs a way to adapt to different kinds of music, Led Zepplin is different than Thelonius Monk. They are compressed differently. Really, I am pretty sure the software does that.
But there is no reason there can't be a factor included that changes the gain of the DAC. Or maybe they already do. The original CDs had preemph and deemph that was either on or off. It switched so fast and transient free you never knew it.
But I digress. The point is that most people cannot reproduce at the levels required for faithful reproduction of music. Therefore there is compression. What's more, even if you CAN reproduce it at live levels, do you really want to ? In many places the cops come. Neighbors get pissed off.
I think they should give you a choice. Four channels, two compressed and two not, the same thing. You get your own little mixing box and play with it and see what sounds good to you, and on your system. They could sell those too, a little four channel DAC that runs off their files you can download or get on a DVD.
I see a marketing opportunity here but I do not know the right people to do anything like that, or even have that much of an idea how to do it. But the OP posted this for a reason, and it could go somewhere in audiophilia. This is another step in the same direction, like look at iTunes, they make money. Download FLACs ? It is pretty hard to get those for free from what I hear.
So we get someone like Lewin Edwards to write some code to invent a new type of DAC the mixes two and two channels. Drivers for all the neat shit, MAC and everything. You sell the files that are derived right from the recording studios.
And you know what ? These files are pretty much secure. Know why ? Because the two compressed channels are REALLY compressed, I am mean like AM radio compressed. Played without mixing with the other channels it sounds like shit. Now if they want to pirate the uncompressed channels now they need thousands of watts. I mean, if you buy a totally uncompressed copy and then compress it, just where did you go to school ?
But the main thing about my idea of being able to mix it on your own is that you can take it to the limit of your system. People have different power levels, amps that have different clipping headroom and dynamic headroom. Even speakers that start to behave differently at higher power levels.
Give the customer a choice. You give him a box with a pot on it that says "COMPRESSED" on one end and at the other "UNCOMPRESSED". Well we'll have to abbreviate that but you should see the idea. Plugs into their USB port and they run the disk, or maybe embed the download site for the software, but that would incur. With USB the way it is, there is enough memory to have many versions for different OSes and even samples, if you can get them.
Problem is the recording industry has to cooperate and they might not see it as good for them.
Hi, Jeff,
It does seem like a plausible solution, as long as the timing is exactly the same. Otherwise I think there'd be some smearing of micro-details. And unless there was a smooth transition between the compressed version and the original, the zone along the peaks could sound unnatural.
An analogy in digital image editing (digital darkroom) is the use of image blending, whereby you mix the original image with an image that highlights certain attributes you're after. The result can be impressive, but if done poorly, the blend can look very artificial.
Regards,
Tom
Post a Followup:
FAQ |
Post a Message! |
Forgot Password? |
|
||||||||||||||
|
This post is made possible by the generous support of people like you and our sponsors: