Parallel processing question.

This forum is for discussing Reason. Questions, answers, ideas, and opinions... all apply.
User avatar
mcatalao
Competition Winner
Posts: 1824
Joined: 17 Jan 2015

25 Nov 2015

Stranger, i really am not seeing the practicality on this, but in your limiter example, i was able to almost null the sounds adding 0,07 db to the limiter channel. I couldn't be more precise, but i might try to get it to 0,08 db or 0,06 db with my controllers at home, but i doubt it.

My point is, is 0,07 db difference between both channels at unity that important? That's probably simply the limiter working, it has to do something. You actually can't "dry" that device, unless you disable the limiter and the Knee, witch ultimately will be the same as bypassing the device.
The limiter is working, it is doing something. So it seems at least it attenuates any sound you feed it. By 0,07 (or 0.08 or 0.06) db.

STILL, as Selig said before i don't see where less than 0,1 db is important where this would only happen in completely in phase, same frequency, things that never happen in a normal mix. Or better said, the same sound reversed. And ideally (unless you're searching some particular effect) it's not usually done even on Paralel Channels, IMHO. Same for the usage of VMG-01, it is put and acts on one of the channels, but you never invert the phase.

IMHO, and i mean no disrespect, we got the difference in volume at less than 0,1 db in 120+ dB headroom digital environments, this seems like the audioholic mambo jambo that gets people buiyng wooden knobs for amplifiers and magic stones for high frequency dampening... :)

User avatar
Voyager
Posts: 535
Joined: 21 Dec 2015

18 Jan 2018

selig wrote:
27 Oct 2015

For adding loudness/intensity via parallel processing I find it key to use a fast compressor and often also saturation.


For that matter do you prefer to have the saturation before the compressor or after ?

User avatar
selig
RE Developer
Posts: 11685
Joined: 15 Jan 2015
Location: The NorthWoods, CT, USA

18 Jan 2018

Voyager wrote:
18 Jan 2018
selig wrote:
27 Oct 2015

For adding loudness/intensity via parallel processing I find it key to use a fast compressor and often also saturation.


For that matter do you prefer to have the saturation before the compressor or after ?
There is no 'preferred' processing order that works for every situation. In most cases I try both and see which one I prefer!

That being said there are a few issues to consider, maybe even some I'm overlooking here.

The first is the transient response or crest factor of the original track. If the track is transient heavy (high crest factor), there can be a big difference between sat/comp vs comp/sat. If the crest factor is low, there may be little for the compressor to act on and the processing order may not matter as much (read below about when processing order matters).

With saturation first in such a scenario it's more the peaks that will saturate. With saturation, the higher the level the more the saturation, and peaks/transients represent the highest level. So in this case the saturated peaks are more what you would pass on to the compressor.

With compression first, it will depend on how fast you compress the transients. If you use a fast attack, the transients will be more reduced (depending on ratio), and so your crest factor will be lowered and you'll have a higher average/rms level. This will mean more of the signal will be saturated, and not just the transients, since more of the signal will be higher in level now and again, with saturation the higher the level the more saturation. With slower attack times, more transients will pass and the resultant sound may be more like the previous example. Additional variables here include saturation drive amount and compression makeup gain.

Also consider the mechanics of saturation, which actually does TWO things to audio signals: adds harmonics, and adds sustain (reduces peak level). Saturation adds sustain (as guitar players know) because it's basically clipping/limiting the audio signal. It adds harmonics because of the waveform distortion of the clipping. It's the "limiting" effect of saturation which will most affect the processing order with a compressor, since "limiting" changes the envelope of the sound, and compression ALSO changes the envelope of the sound.

BTW, the higher harmonics will have a less obvious (if any) change on the processing order, since they are lower in level and will more likely fall below the overall threshold - meaning, they won't change the overall envelope of the sound themselves, so it won't affect the compression that follows.

As a general rule, processing order is more important (more obviously different sounding) when BOTH processors are doing something obvious to the sound. When only one is doing a lot, the order is less important because the differences between the two processes will be minimal. Example: if you are saturating VERY subtly and compressing heavily, then the processing order will probably not matter much (as long as you're saturation effect is not adding/subtracting gain). Same for using very light compression and heavy saturation. But since both of these devices are nonlinear with respect to level (meaning: sensitive to level changes), even if one is simply changing the overall gain significantly, the processing order could result in a dramatically different sounding end product!

Again, best advice is to listen to both orders, and choose the one you like best. Over time, you'll probably begin to be able to guess which one you'll prefer just by paying attention to what each process is adding or not adding, based on past experience.
Selig Audio, LLC

User avatar
Voyager
Posts: 535
Joined: 21 Dec 2015

18 Jan 2018

Thanks for the insight, i'll try to switch time to time and trust my ears.

househoppin09
Posts: 536
Joined: 03 Aug 2016

18 Jan 2018

Seriously, if all of selig's posts in these threads were collated by subject and published, the resulting book would be an instant classic. Hmm, now there's an idea... ;)

Post Reply
  • Information
  • Who is online

    Users browsing this forum: No registered users and 31 guests