sounds like a weird plugin to me

 

Not really. In a plugin the electronic components are replaced by mathematical functions, equations and variables, but the logic remains the same.
At the inputstage the incoming samples are measured and defined to a value between 0.xxxxxxx and 1. Then these samples are multiplied by an inputgain positive or negative. Behind the gain there is a crossover which splits the signal into audio path and sidechain. In the sidechain there is a filter and a (peak-/RMS-)detector, which recalculates the incoming samples. This calculation is passed on to a gate - the so-called threshold - which splits the signal once more and passes the plus-part of it on to the [...] ratio circuit,etc... until the sidechain signal finally lands on a vca which multiplies the samples in the audio path negatively by the ratio/knee output.

 

Last edited: Feb 6, 2020

absolutely not, because in a fixed-threshold setup, at least in the analog sector, components are used that generate overtones and changes the tonal balance as the gain increases. the gain also increases the noise level.

yes, because it's something completely different. a threshold doesn't make sound.

gain = multiplier
threshold = potential-gate

 

Last edited: Feb 6, 2020

Technically that's correct. It's just a doorkeeper who decides who can pass to be pushed down (compressed) and who not, so to say.
I think what tun means it that the following compression can also alter the sound (add harmonics), if I got him right, that is.

But we're off-topic for pages now but if Gyro is fine with it, ok, but I'm still missing a proof ('AS approval batch' ) for a plugin for which this calibration is really important.

 

It's not off-topic. @tun claim is false, but it describes the topic well in itself.

When we talk about the "sweet spot" of plugins, we're talking about a potential, the threshold (variable or fixed).
crossing this threshold doesn't produce any tonal changes by itself!
so, questions that remain are:

what happens after the threshold of a particular plugin?
are the tonal changes triggered by increasing the input gain (multiplication),
by driving the input-gain into a threshold or
by crossing a threshold before the input gain (Input signal level)?

so to speak, it's absolutely in topic!

 

Last edited: Feb 6, 2020

oh boy...i thought you read the threads so proper...

i did bring it up cause kwissbeats brought up the the fixed threshold thing on a 1176

 

sure

 

disclaimer: I'm quite lazy with typing so forgive me if what I say isn't super clear or I've omitted something;) Just ask me to elaborate further.
There's a lot of misunderstanding going on where 'standards' are concerned, and we haven't even mention 'chlorinated chicken' yet;)

Do we agree that at the very least we all use D/A conversion to monitor our work and make decisions? I believe there's agreement to that.
So, lets get to the point...
Which is that we need a reference point to work with. It's difficult to explain anything audio related when using different standards, isn't it?
So I should define the ones I use...

Sweet spot = Nominal level.
A nominal level implies a 'normal' or typical level in equipment. Nominal operating level of a piece of equipment is thought of as the typical signal level with which it operates. It's used in audio to specify a signal level. On equipment with +4 dBu inputs and outputs the nominal operating level is said to be +4 dBu.
When I refer to nominal levels in audio equipment I'm generally referring to zero reference levels.
(Where my definition of audio equipment includes modelled plugins, the A/D and D/A converters, pre-amps, a DAW etc, basically anything that makes and/or processes audio)

Unity Gain = A gain of factor 1 (equivalent to 0 dB)
When both input and output are at the same voltage and impedance it's known as unity gain. Establishing unity gain allows you to have control over what is being boosted or cut and by what amount, instead of audio via random circumstance.

RTFM = I mostly care about one thing in any manual and that is the 'nominal level', usually found in the specifications' section.
For example, I'm mixing a couple of projects on a Focusrite Saffire Pro24 DSP. I currently don't use any outboard gear, all mixing is ITB so what do I need from the manual? I need the Saffire's Maximum Output Level (0dBFS) which is +16dBu. Anything above that will clip its converters.
What do I do with this info? I use it to calibrate my VU meter so I can do my gain staging ITB and work at Unity Gain throughout, in order to avoid certain pitfalls and be able to make unbiased decisions (eg louder sounds better trap)
Why a VU? Because...

Did someone say 'use your ears'? Well...

Again...
Sweet spot = Nominal level = 0VU
Which in the example with the Saffire, since it's got balanced outputs, it's 0VU = +4dBu = 1.228 volts RMS
So my VU gets calibrated to -16DBFS = 0VU and that's my 'Reference Point', throughout, and that tells me that I have a headroom of 16dB to work with. Stick over 100 channels through the master and you'll soon find out it can easily get eaten up.
But there's another problem. VU meters don't represent percussive sounds well so I also need a Peak Meter to see how far in my 'headroom' the transients go. Also my speakers are calibrated to K-14 SPL levels @Unity Gain.

The -18DBFS story is mostly an estimate which will get most people in the ball park, to be able to work with Unity, preserve Dynamic Range and have enough Headroom to avoid clipping their converters (or inputs of analog modelled plugins like eg Waves or UAD)

Conclusion...
RTFM! There's no "one size fits all". It all depends on the medium we're delivering to (be it a CD, music for film, broadcasting, Spotify etc) and the equipment we're working with at any given time. I choose a standard reference to work with on those two reasons and it's a (not 'the') correct way to go about it.
Without a 'reference' there's only 'darkness'.

Cheers.

 

Last edited: Feb 6, 2020

Valid point. But I'm not sure if even the companies like Waves or Plugin Alliance know this. The programmers, probably, but the companies?

 

Fudsey Plange you are WRONG......NO AVENGER is RIGHT. This is the sad reality for you.

 

but here it seems to me that there's some smart guy who's giving out valuable and pertinent info with the real matter of soud engineriing, and not the fake youtube gurus you're talking about, so maybe you should be thanking this opportunity instead of throwing up a banal "just stop it!"

 

Well said guy, well said..

 

.del

 

Last edited by a moderator: Feb 7, 2020

Are you talking about 'binary' with those numbers?

 

It shows -21dB RMS because 'RMS +3' haven't been used like it should;)

'This mode is available via Settings': He's talking about the settings of the DIGICheck, you'll find 'RMS +3dB' setting in Voxengo's Span too etc.

 

Last edited: Feb 7, 2020

thats similar, but not the same as the way i have built them. i certainly wouldnt use a crossover to split the SC signal and the audio signal, that would cause either a LP output with HP SC or vice versa.
for the design i have in mind the signal is rectified and converted to a unipolar measurement before it hits the threshold, then the addition/subtraction happens to the signal before it hits its fixed threshold. the addition and subtraction stage here is what i am saying you can somewhat fake. i was a little excited saying its the same, but it is close enough to fake it. the result would be similar to the ratio changing along with the threshold, so it would sound different, but good enough to make changes if youre stuck with a fixed threshold.

 

a gain control doesnt make sound either

 

sinus was referring to floating point values.

 

I don't have the right term in english...splitter maybe? it is not a filter crossover ;)

 

yea splitter. like the hairs we are splitting XD