Downward vs Upward Compression

[VibraSound]

Downward Compression
Loud sounds over the threshold are reduced in level while quiet sounds remain unchanged.

Upward Compression
Quiet sounds below the threshold are amplified while the louder sounds remain unchanged.



What type of compression do you prefer and why ?

[futurefields]

Does NY compression qualify as downward or upward compression?

What I like to do is take the original signal, compress that heavily with something like Stillwell Rocket, and then use the parrallel compression knob starting on 100% dry, i will begin to blend in the wet/compressed signal, until i feel like i've added enough body/sustain. I find this is a good method of compression that can achieve very punchy, transparent results, and with a plugin like Stillwell Rocket or any of the others with a build in mix knob, this is really easy to do. You can use this on any source where you dont want to mess with the transient dynamics, but want to add in some compressed signal to even out the response and add some body/sustain.

[Dionysos]

Does NY compression qualify as downward or upward compression?

I'd say, depends if it's downward of upward compression that you're adding to the dry signal.
NY compression doesn't change the type of compression that's applied, it just weakens its effect (compared to the 100% wet compressed signal).

[Lenticular]

Depends on the material. Mainly downward & for certain orchestral instruments, tracks or stems I'll try upward first.

[VibraSound]

I prefer Upward Compression for Mastering

[annode]

It depends whether i'm doing my yoga headstand exercise while listening to music.
But seriously,like lenticular sez...all depends on the material content.
If you have a track of a female vocal take who was feeling a little shy singing, and kept moving away from the mic at the wrong moments...i'd use upward compression without question. Downward comp would be a waste of time and ineffective.

You want the vocal notes to" hold" as long as she intended them to be held. But since she isn't a seasoned singer her notes don't hold long enough and just fall off seemingly abruptly. The moving around the mic creates similar problems.
Upwards comp corrects that so well, you don't even notice there was ever a problem.

[VibraSound]

Please correct me if I am wrong.


Downward Compression - To Shape the Transient (Attack).

Upward Compression - To Bring Up the Body (Sustain).

[annode]

Please correct me if I am wrong.


Downward Compression - To Shape the Transient (Attack).

Upward Compression - To Bring Up the Body (Sustain).

You are correct sir. Why are you suddenly in doubt?

UPWARD COMPRESSION

If you've used the L1 or L2 Ultramaximizers for peak limiting to make your mixes louder, you've undoubtedly discovered that using peak limiting to reach for competitive level often entails a considerable sacrifice in dynamics. Here's one way, called low-level or upward compression, to raise the volume of the softer sounds in the mix without getting the squashed sound of too much peak limiting. This technique has long been popular in classical music, in which wide dynamic range is common and peak limiting is rarely used.

For upward compression, use the Linear Phase Multiband EQ. Start with the factory default settings, then use the Master controls to set all the bands to these settings: Threshold, -50 dB; Range, -6 dB; Gain, 6 dB (see Fig. 2). Those settings will increase the level of soft passages while leaving the louder passages untouched.

The audio just below the Threshold value will be compressed upward a maximum of 6 dB. The high-level signals (the ones significantly above the threshold) will have no gain change, because at high levels the Range and Gain settings, which are set to complementary values, combine to equal unity gain. You can see what's happening by observing the yellow Dynamic Line display. You'll see a significant difference in the curve when the input signal is low compared with when it's high.

Experiment with the threshold and crossover points in the individual bands to fine-tune the compression to suit your program material. Always make sure the Range (negative) and Gain (positive) controls are set to exactly offsetting values, and that Makeup is set to Manual. Fine-tuning the Master Threshold control will allow you to lift the lower dynamic sections of the mix without sitting on the peaks.

As a final touch, use an L2 placed after the Linear Multiband. Set it for no more than 3 dB of gain reduction. That provides competitive level with punch that sounds more musical and interesting.

Taken from;
http://emusician.com/mag/emusic_mastering_waves/

[futurefields]

Does NY compression qualify as downward or upward compression?

I'd say, depends if it's downward of upward compression that you're adding to the dry signal.
NY compression doesn't change the type of compression that's applied, it just weakens its effect (compared to the 100% wet compressed signal).

I believe it's still technically upward compression. Because, the top peaks are the original ones, the compressed signal is more of a squashed one, used to blend in more body and sustain. so you get the mix of the highly dynamic, uncompressed signal creating the peaks, and the squashed signal adding tons of body. I like this method a lot.

[AL321]

For me, upwards compression is always the best option, unless i actually want the effect of having the top squashed for creative reasons. If you use a compressor that lets you draw the curve (i use melda plugins for this), you can actually specify how many decibels, and in what area, you want to raise, keeping the curve smooth so it still sounds natural.
After learning this, i see normal compressors as clumsy, and often avoid them when working on important upfront sounds.

Does NY compression qualify as downward or upward compression?

Yeah, if the compressed signal is hard compressed, it's not really that much different to upwards compression, but there is one difference that makes me prefer upwards compression every time. With NY compression, you are mixing two signals together, so you are changing the volume of the loudest transients, even if you aren't squashing them. But with upwards compression, they remain untouched, so if you're drums are peaking at -2db before compression, they will still be -2db after compression. That's why i would always choose real upwards compression for the master bus, rather than NY style. You can easily A/B changes without the compressor causing any volume changes to screw up the comparisons (or screw up the settings of any other master plugins you have active).

[futurefields]

That makes sense. Thanks for the tip!

[VibraSound]

If you use a compressor that lets you draw the curve (i use melda plugins for this), you can actually specify how many decibels, and in what area, you want to raise, keeping the curve smooth so it still sounds natural.
After learning this, i see normal compressors as clumsy, and often avoid them when working on important upfront sounds.

Same thing for me, I much prefer compressors that lets me draw the curve

[S-N-S]

Same thing for me, I much prefer compressors that lets me draw the curve

what other software comps will let you draw in curves??

[VibraSound]

what other software comps will let you draw in curves??

Maximus
http://www.image-line.com/documents/maximus.html

[Dionysos]

NY compression doesn't change the type of compression that's applied, it just weakens its effect (compared to the 100% wet compressed signal).

I believe it's still technically upward compression.

But it isn't. If you downward compress a signal and then add that to the original signal, you're still adding downward compression to the signal, nothing else.

If there's a difference between upward and downward compression, it's how the gain envelope is created, and this doesn't change only because you mix dry and wet.

Btw, upward compression will also always affect and shape peaks/transients. I don't think the two types of compression are actually that different if you appropriately set the attack/release times.