Roomworks / Reverbs / Send Effects / CPU

[cshieldsx]

Hello all,
I've recently started using Cubase SX3 and I'm getting alright at using it (so I should cope relatively okay with any tech jargon). I have roomworks reverb, and It's very CPU heavy and if I run more than 2 or 3 instances of it my computer can't cope.

If I use the 2 or 3 roomworks that my PC can cope with as send effects, and route 10 tracks through them - will that still eat CPU ? Or is that the whole point of send effects that It wont eat CPU - it will just be the same as having 3 tracks going into 3 roomworks ?

Also, does anyone rate roomworks as a good reverb plugin ? Or isit rubbish ? Which ones do you recommend ? Anyone know of any good free reverb plugins ?

[.jon]

Sends are made for that sort of thing, yes. One instance only eats 1 x CPU, no matter how many tracks you put through it.

It also helps in mixing, if the tracks have the same reverb applied on them, they exist in the same virtual space then.

I use "Glaceverb" a lot for mixing, it's quite low CPU and sounds smooth. "Ambience" is another good one, lots of control, donationware, not CPU friendly. Third reverb I use is Anwida DX Reverb light, it's good for artificial sounding long reverb washes.

.jon

[aciddose]

if the reverb is linear which it should be then you'll get exactly the same result sending 8 individual channels through 8 individual reverbs as you will sending 8 channels through one reverb with the same settings. this also applies to most filters like eq and etc. nonlinear effects include distortion, compression, etc.

[Cordelia]

What.jon said.

[whyterabbyt]

if the reverb is linear which it should be

Actually, a lot of reverb algorithms are non-linear, in that they use modulation on certain parameters to simulate a larger number of reflections and prevent resonances. So no, there's no reason a reverb 'should' be linear at all.

then you'll get exactly the same result sending 8 individual channels through 8 individual reverbs as you will sending 8 channels through one reverb with the same settings.

No, you wont, necessarily.

[kmonkey]

And like it or not bur roomworks is pretty bad coded. I mean today you have so much better reverbs to choose with much lower CPU idle. Take a shot on Waves Masterverb 5 or if you need better take a look at ArtsAcoustic Reverb or Classik Studio Reverb (also named CSR). You can find some info here at kvr. Just search through effect tree in right corner

[.jon]

In real life situation (as described above) those differences coming from internal parameter modulation are meaningless. The results will not phase-cancel each other out, but you will not be able to tell what is what, even less what is "better". It's just very, very slightly different.

[mike@manike]

I use roomworks. I think its pretty good for certain types of stuff. I use it side by side with a lexicon mpx 1. I shouldnt worry about using it.

Arent Ambience and Roomworks related?

Why do you say its badly coded?

There are great reverbs out there however which I would like to get my hands on.

Mike

[whyterabbyt]

In real life situation (as described above) those differences coming from internal parameter modulation are meaningless. The results will not phase-cancel each other out, but you will not be able to tell what is what, even less what is "better". It's just very, very slightly different.

True. However its incorrect that reverbs 'should' be linear.

[martian]

And like it or not bur roomworks is pretty bad coded. I mean today you have so much better reverbs to choose with much lower CPU idle. Take a shot on Waves Masterverb 5 or if you need better take a look at ArtsAcoustic Reverb or Classik Studio Reverb (also named CSR). You can find some info here at kvr. Just search through effect tree in right corner

i use roomworks (ambience) and lexi impluses over AA personally. AA is amazing, versatile and unique but roomworks gets me closer to the halls i like.

[kmonkey]

And like it or not bur roomworks is pretty bad coded. I mean today you have so much better reverbs to choose with much lower CPU idle. Take a shot on Waves Masterverb 5 or if you need better take a look at ArtsAcoustic Reverb or Classik Studio Reverb (also named CSR). You can find some info here at kvr. Just search through effect tree in right corner

i use roomworks (ambience) and lexi impluses over AA personally. AA is amazing, versatile and unique but roomworks gets me closer to the halls i like.

bad coded = use too much CPU cycles for what you can get with....

[.jon]

Roomworks = Ambience?

cool, yeah it eats CPUs but IMHO not a bad verb at all, I skip it's own eq for critical things and use PLParEq instead. All the freebie native verbs have issues, but there's always SIR for final mixes <3

[aciddose]

white; actually parameter modulation does not make a filter nonlinear. if you modulate reverb A the same way you modulate reverb B, still A+B == A|B. i've never seen a nonlinear "plain reverb" effect and i would be highly surprised to see one demonstrated. flanger, chorus, delays, phasers, those are all also linear. an example of a nonlinear filter is: n^2.

perhaps a tape delay or spring reverb emulation may contain nonlinear elements. a linear system can be represented entirely by an impulse responce and therefore reproduced perfectly by convolution. a system containing nonlinear elements can not be represented or reproduced in this way.

[michu]

aciddose
modulated flanger or phaser cannot be acurately reproduced by single impulse response
as far as modulating reverb parameters go, random (non-cyclic) modulation is often used for this, so the same input run twice thru that kind of reverb won't necessarily cancel out with itself

[whyterabbyt]

white; actually parameter modulation does not make a filter nonlinear.

Depends how the parameters are modulated, I suspect.

if you modulate reverb A the same way you modulate reverb B, still A+B == A|B.

Except in the realilty of reverbs, the parameter modulation includes randomness, so the practical results will contain differences.

i've never seen a nonlinear "plain reverb" effect and i would be highly surprised to see one demonstrated.

Real room reverb is non-linear. Many reverb manufacturers specifically state that their reverbs are nonlinear.
Do a Google search for 'nonlinear reverb'.

perhaps a tape delay or spring reverb emulation may contain nonlinear elements.

Any complex physical system is likely to be nonlinear.

a linear system can be represented entirely by an impulse responce and therefore reproduced perfectly by convolution.

Which is exactly why convolution reverbs are not considered a 100% replacement for complex high-end reverbs; because those reverbs

a system containing nonlinear elements can not be represented or reproduced in this way.