DMG EQuilibrium

[Breval]

What's the cpu usage compared to EQuality when using the EQuality or mixing presets for general channel / tracking eq?

[A.M. Gold]

Dave, did you thoroughly state space model the hardware EQ's represented in the plug-in like in the Sonalksis EQ?

Sorry if this has been covered here already. I haven't read the whole thread but I didn't see this specifically mentioned on your website, only mention of specific hardware curves.

That's a bit of a marketing phrase. And not a term I'd ever use (it came from Sonalksis, before my time).

These are careful circuit models. Schematics, plots, etc. all the science. It's fun!

Dave.
Edit: the correct phrase would be a model that was implemented via a state-space representation. "State space model" conflates two different things: what it's doing (the model) and how it's doing it (the state space representation). State space representations are a real thing, from linear system theory. It is by definition identical to any other topology, because that's what "representation" means here- a different way of storing your numbers. There was some argument that for float, it provided better noise performance, but actually using double precision floats (as all my stuff does internally, start to end), is not only better still, it also uses a lot less CPU! The state space representation uses a lot more multiplies per sample, and I can't find a measurement whereby its an improvement on what I use.)

Oh, ok, thanks for clarifying. I didn't realize you weren't a part of Sonalksis when they used the state space technique. But it sounds like you did some thorough modeling.

[Seaside Music]

Ok, wow! So i guess it should be even possible to setup a preset which has the frequency bands of the 550 pre set.
Do each of the bands of the hw 550 react the same or have u had to model different ones for some of the freq bands?.

I was hoping to do this as well, but I'm not sure it's easily possible using a minimum of bandstrips since the frequency knob is always infinitely sweepable from 10hz to 22k.

From what I can tell there's no way to set up a single bandstrip freq knob that would snap to multiple individual frequency settings. For example the lows on a 550: 50,100,200,300,400.

I think you would have to set up 15 bandstrips with all of the available settings of a 550a already preset but it wouldn't be as elegant as being able to do it in just 3 strips.

[Amon1973]

And once againg, stunning GUI Krzysztof: I love your work.

Aww, thanks man, I loved making it! It's taught me a lot and I've more to learn from it - like a good track.

Drop Dave a mail and ask him for the 1.02 candidate because you've got the popping bug. Give that a whirl, see if it fixes your issues - it's all looking good here, but that's not a guarantee of success for all!

If anyone has a bug at the moment that hasn't been addressed, e-mail Dave now! He is oiled and poised for action!

Will do ASAP. But now I'm going to buy it

[PatchAdamz]

Just used the demo on a few mixes.
Its a very well conceived equalizer with grade A sound quality.
Excellent job!

[Krzysztof Oktalski]

What's the cpu usage compared to EQuality when using the EQuality or mixing presets for general channel / tracking eq?

It's really good in Digital mode, we had cinema guys on the beta that have over a 1000 instances in a session! It's roughly the same as EQuality, maybe slightly heavier depending on what settings you've selected.

[Krzysztof Oktalski]

I was hoping to do this as well, but I'm not sure it's easily possible using a minimum of bandstrips since the frequency knob is always infinitely sweepable from 10hz to 22k.

It's true, the bands don't snap - this was discussed at great length during the beta. Dave and I have no idea at all why you'd want arbitrary frequencies selected for you, that are nearly certainly wrong for your music. The people that made those EQ's never heard your track, didn't know what key it was in and so gave a few broad brush strokes for you. Seems a bit of a silly thing to keep doing now that you've got a transparent framework where you can see the answers. After a lot of debate, it was agreed that snapping band frequencies served no function in a modern EQ, and there are some facets of past EQ design that should be left in the past.

[kenobi77]

Ok, wow! So i guess it should be even possible to setup a preset which has the frequency bands of the 550 pre set.

Yes, I think I saw a preset like this - check the demo! If not, it's not going to take you long to make one.

Do each of the bands of the hw 550 react the same or have u had to model different ones for some of the freq bands?

The shelves and peaks are modelled with their respective interactions.

Thanks for your prompt replies, no doubt will demo soon. As an aside I know such a product is rated on its sound and usefulness at the end of the day, but i am so impressed with the concept, the blue sky thinking of what an eq should deliver in this current age.

[Funkybot's Evil Twin]

I was hoping to do this as well, but I'm not sure it's easily possible using a minimum of bandstrips since the frequency knob is always infinitely sweepable from 10hz to 22k.

It's true, the bands don't snap - this was discussed at great length during the beta. Dave and I have no idea at all why you'd want arbitrary frequencies selected for you, that are nearly certainly wrong for your music. The people that made those EQ's never heard your track, didn't know what key it was in and so gave a few broad brush strokes for you. Seems a bit of a silly thing to keep doing now that you've got a transparent framework where you can see the answers. After a lot of debate, it was agreed that snapping band frequencies served no function in a modern EQ, and there are some facets of past EQ design that should be left in the past.

I strongly disagree with the above and hope you guys reconsider. The argument for allowing presets to snap to certain frequencies is that:

1) It's fast. Someone has already chosen some key frequencies to work with, which should sound good on a wealth of material. Will those preset frequencies be perfect on my tracks? Probably not, but they'd get the gold medal in the "get me the best sound in the fastest possible amount of time with the least amount of thought possible" category!

2) It's a big part of what made the EQ's famous in the first place. How do you separate the shape of the EQ from the frequency selection? Would it really be a Neve 1073 without the fixed hishelf at 12k? Would it be a Pultec without the attenuation at 5, 10, and 20k? Would the 500A "sound great on electric guitars" without those preset frequencies?

3) The overlap of certain frequency combinations in those units are also a huge part of the sound? The Pultec low trick is one example, even with something like a 1073 (why is there no 1073 in EQuilibrium?) boosting with a lowshelf at 160 while cutting mids at 220 has a nice interaction that's just part of the unit's sound.

4) It's easy. Noob's and pro's alike prefer things simple. This is why the LA-2A is so popular, and why many mixing engineers prefer console EQ's to all in wonder boxes. It's easy to think of, I have a low shelf, a low peak, a mid peak, and a high shelf, especially when combined with stepped frequencies. Again, fast, easy mixing by limiting choices.

5) It would make EQuilibrium more desirable if you could sell it as: build your own 550A with stepped EQ ranges and all!

Anyway, I'll probably be buying EQuilibrium before the demo ends, but don't try and tell me stepped bands are a bad thing!!! It's a feature, not a design limitation!

[PeterL]

What's the correct usage of Pultec style bands related to the original ?
1) Use 2 bands (1boost/1cut) at same frequency, leaving Gain at 0 and set just Boost/Cut amounts (I think so).
2) Use just 1 band, set Gain up to boost and amount of Cut to cut.

I just want to know if it is the same to increase Gain or to increase Boost.

[maxxxter]

DELETED

[bishop666]

+1

[Funkybot's Evil Twin]

I was hoping to do this as well, but I'm not sure it's easily possible using a minimum of bandstrips since the frequency knob is always infinitely sweepable from 10hz to 22k.

It's true, the bands don't snap - this was discussed at great length during the beta. Dave and I have no idea at all why you'd want arbitrary frequencies selected for you, that are nearly certainly wrong for your music. The people that made those EQ's never heard your track, didn't know what key it was in and so gave a few broad brush strokes for you. Seems a bit of a silly thing to keep doing now that you've got a transparent framework where you can see the answers. After a lot of debate, it was agreed that snapping band frequencies served no function in a modern EQ, and there are some facets of past EQ design that should be left in the past.

Unlike others here, I STRONGLY AGREE with your assesment. A fixed selection of frequencies is absurd.

As absurd as not having threshold, ratio, and attack and release times on a compressor? To me they're comparable limitations, and yet a ton of people love the LA-2A which was has two fixed ratios and no attack/release controls. I also don't hear complaints about the 1176 having a fixed threshold either, or about not being able to dial in a 1.25:1 ratio with it. Stepped frequencies on an EQ are the exact same kind of limitation.

I understand if the DMG team just don't want to do that because they like options, it's completely cool if they want to be the every possibility under the sun company. I just see there being value in having stepped frequencies.

[bishop666]

Simply ignore the new possibilities.


There is no problem at all.

[maxxxter]

DELETED