DMG Multiplicity confusion

[wintoid]

Hi,

I'm missing something (probably several things) vital about Multiplicity, so need to consult the hive mind to see if anyone can explain it to me better than the few Youtube videos I've been able to find, and the manual.

As I understand it, anything occuring during the attack phase is considered a "transient" and anything after the attack phase completes is considered a "dynamic". So let's say we set an attack time of 30ms, and we have Dyn/Tran set to Tran, we're saying Multiplicity will only affect the signal from the moment it crosses the threshold until the attack time has expired. In this circumstance, does the release phase kick in once the attack time has expired? If so, release only ever occurs during the dynamic section, which seems to be what the manual implies.

Again in this circumstance, does the knee affect the onset of compression from the moment the threshold is crossed, or does it affect the adjustment between the Transient and Dynamic stages, or both?

And how the hell does the lookahead factor into all this? Is it effectively the attack time for the Transient stage?

Hopefully someone can help me click this into place. Then once I've understood, I'll ask about Rel RMS and get even more confused

Thanks,

Simon

[n8pee]

And this is why I never use this plugin - despite the fact that it is obviously capable (and the Youtube demo video is phenomenal).

[wintoid]

Haha, well thanks for commiserating with me. At least I know I'm not alone.

I've been having some fun with the upwards expansion, bringing out some hihats which were dulled by tape on some old recordings, which means it has a use for me.

I spent several hours feeding saw waves into it, and playing with sliders and then inspecting the waveforms, and re-reading the manual. It does seem like the lookahead time is affecting the attack/release of the pre-attack-time portion of the waveform. Having said that, Multiplicity was doing things to my saws that other compressors (I used Fircomp2) were not doing, and I don't fully understand the behaviour.

[plexuss]

And this is why I never use this plugin - despite the fact that it is obviously capable (and the Youtube demo video is phenomenal).

Highly unlikely.

It's not that complicated. Spend less time reading waveforms and more time listening. The manual is detailed enough to use the tool. Trying to reverse engineer the plugin to determine the intricacies of the algorythms is going to be mostly futile. I dont think Multiplicity is the kind tool that will benefit from that approach. Are you trying to solve a specific problem, or just want to know in general? If the former, what is the problem? "Tha saws" problem? Do you mean saw waves? What is the actual use case in detail?

[wintoid]

I find if I understand, I can use the tools better, and furthermore I am more likely to start using the tools, and therefore more likely to learn them better. I guess I'm trying to understand, first of all. It's not a specific problem.

So take this section of the manual:

Transients will obey the gain calculation transfer (set with
THRESHOLD, RATIO, CEILING etc.) with an attack and release time determined by the
LOOKAHEAD control.

Will the attack and release time of the transient be equal in length, and equal to the lookahead time? Or are they half the lookahead time each? Is there any way to adjust the ratio of attack and release times during the transient stage?

[plexuss]

I understand where you are coming from. And, I am the same way. I like to learn about my tools in many difference contexts: how it sounds, what is it technically doing, etc. There are some plugins, like Multiplicity**, where there is so much going on under the hood its possible to really understand it beyond a certain point. **I think Multiplicity is one of these because I tried to break it down with test tools and got no where. Many plugins can be technically analyzed to some degree. How far do you want to expect to go? There is a limit to the granularity of the data about how a plugin functions. Is it important to know the code used for the various algorithms? It also is a personal preference, which blows it up right there.

But I get'cha. I just dont think Multiplicity can be analysed too deeply beyond the ubiquitous dynamics processor parameters a la Plugin Doctor, et al. Some developers like DMG work to move beyond the usual to offer something (competitively) different and musicially useful to justify the price. In a lot of these cases they are far too complex to analyze in detail. So you are left with the sound, which is the most important thing.

That's what I was trying to say...

[wintoid]

Thanks @plexuss actually I hadn't considered Plugin Doctor. I don't have it, but if I really can't work out Multiplicity well enough to start using it more, I might investigate.