Stereo Tool

I've never been able to make coupling work as expected, so I haven't tried in a while. However, I would not recommend unchecking the "Norm" checkbox. The idea is that if you have percentages which total more than 100%, you can force a band to create "artificial" amounts of gain reduction. For most use cases, that would be undesirable. At most, we want a band to have *as much* gain reduction as some adjacent band, but almost never do we want it to have more, just because of coupling.

I may play with this again, but some of what I've read on this thread makes me less excited about the feature. I'd prefer Orban-styled coupling. Compression is the only thing that ought to be coupled. Limiting is super fast, and should not cause an adjacent band to duck. Hopefully, I misread that, but I don't believe I did.

Here's how it works. In this situation I noticed that the 1st band was often going down much further than band 2. So, I increased the 2nd slider in the top line to make band 1 respond to band 1 and 2 equally. Which means that in case of a deep bass kick band 1 doesn't get lowered as much anymore.

I'll play with it some more. It seems that I did understand the UI correctly. I'd want that effect between band 1 and band 2. I'd also want the inverse effect between bands 6 and 7 (band 7 could have no less GR than band 6, in this case). In either case, I still want the limiters to act completely independently of the compressors. If band 2 had a sudden spike causing limiting in band 2, I don't want band 1 to take on additional GR. Is this the case?

True. What I'm seeing is that if there's not much limiting going on, the effect is pretty small. But if you limit a lot it's big. And - worse - the limiting doesn't affect the band that's causing the problem enough. So I guess I need to change this.

And yes, Norm should almost always (probably really always) be off.

Well, I understand it a bit better now. It seems that even 100% coupling isn't *quite* 100%. I have bands 1 and 2 coupled together as you do, but I consistently observe that when there isn't much band 1 energy, it will still have about 1 or 2 db less GR than band 2. That isn't such a huge problem. In fact, in the old analog days, many engineers would couple a little less than 100% anyway. As long as it is consistent, I don't mind if 100% isn't precisely 100%.

I *would* fix the limiting issue, though.

Hm. Ok. So that's a completely different type of coupling...

First, the limiters are NOT behaving as you want. And that's indeed bad.

Second, if you couple band 7 to band 6, the most you can make it do is reduce the "freedom" of band 7 to move, but you cannot pull it down to at most band 6 level. Which might actually be a very good idea...

So....... Please tell me exactly what you expect band coupling to do......... the 'band coupling for dummies' version, since I seem to misunderstand what it should do.

OK, I'll do my best to explain.

I'll explain from my background. Orban offers Band 2 > Band 1 coupling. Band 3 > Band 4 coupling, and Band 4 > 5 coupling settings. This is what I'm used to so it's what I expect. Here's how it works.

When a band is coupled TO (1, 4, or 5), that band can never have less gain reduction than the band it is coupled FROM. But it *is* allowed to have MORE gain reduction. The idea in the case of Band 4 > Band 5 is to basically turn bands 4 and 5 into a unified band with the additional feature of a sibilance reduction feature in band 5. It is rare to have a song which would have much more energy in bands 1 and 5 than 2 and 4, respectively. But if that happens, we want 1 and 5 to do as much extra GR as necessary. The inverse is NOT true. Because bands 1 and 5 represent the extreme ends of the spectrum, we don't really want them to get any more dense than they already are. In particular, a very dense band 5 is bad on FM, and a dense band 1 gives a nasty rumble and puts a lot of extra energy into the final clipper.

Now, that doesn't mean you have to do it exactly the same way. If there is a better approach, that's fine. You asked what I expected, so I'm telling you.

I should note than in the Optimod 8200, what I described isn't exactly true (but it is in all subsequent Optimods). In the 8200, the gains of bands 4 and 5 were precisely identical with the exception of limiting. The other bands' coupling worked as I described above.

In no case, is there ever interaction between coupling and limiters for their respective bands (that I am aware of).

Could be, although in my opinion there is no such thing as 'wrong use'. If it shapes the sound in a way you like, then there's nothing wrong with that, isn't it? I (usually) work with very slow attack/release times and very little compression. This just added a bit of extra energy in the low-end.

Meantime I adjusted the settings according to Hans' explanation. let's see how the low reacts now.
Thanks for the explanation Hans!

Ok. This sounds very logical but it's completely different from the matrix that I've just added. Maybe I should just remove it again... and add what you described later (soon). By the way does this mean that coupling can only be on or off??

So, if no-one finds an improvement with this I'll remove it.

I do like it Hans! Or better: I would like to have control over coupling. If Weskeene's option is the better one, that is of course fine with me as well.