Convolution confuses the h-ll out of me

[Shy]

Like you, I was referring to the topic itself and not anyone's personality (which I couldn't care less about). You argue that it misleads, I argue that considering your expertise on the matter, the trustworthiness of your argument should not fall short as I think it does, resulting in lack of credibility considering what has been established so far in the field on convolvers used for reverberation. All is subjective and as far as I see it. If you want to consider it "personal", too bad.

[whyterabbyt]

There's no convolver ("dynamic" or not) that provides different responses for instruments that are playing together, nor for their different and/or varying intensities.

how many reverbs of any kind are there that do?

[Hink]

It's not fine for reverbs either as far as I'm concerned.

Here's my favorite commentary about convolution used for reverb, by Michael Carnes (Lexicon reverb designer):
http://www.gearslutz.com/board/4075701-post132.html
http://www.gearslutz.com/board/4075956-post140.html

A bit of a conflict of interest issue with that commentary, I'd say.

I think convo for reverbs can run in parallel with algo and I do this sometimes. I tend to find it a bit dry and unrealistic on its own.

Weirdly enough I tend to feel the same. Thinking back - during all my year of making music, I'm not even sure if I've ever used a convolution reverb in an actual production. I always tend to find favourite algos (EOS is my favourite for the last two years) and stick with them. Bought NI Reflektor and haven't used it more than ten minutes or so.

Weird.

I'm pretty much thew same way for reverbs (IRs for speaker emulation is a different matter) but for me I go straight to variverb

[DaveClark]

Hi all,

Having worked in convolution for over ten years now, beginning with working through a text on Green's function methods (G. Barton's book) and finding that "convolution reverb" and binaural recording simulations more or less fell into my lap:

That couple of posts are misleading imoho and I explained why I think so.

I agree 100%. The many claims (not just "one fact") are false and/or extremely misleading, depending upon how you interpret terms such as "point" and "capture the space." Those two posts cited above comprise the most inaccurate depiction of convolution (and acoustics) that I have ever seen, and I've seen quite a few.

Regards,
Dave Clark

PS: Recommendations for reading/studying:

Gabriel Barton. Elements of Green's Functions and Propagation: Potentials, Diffusion, and Waves. Oxford, 1989. Reprinted 1991.

Philip M. Morse and K. Uno Ingard. Theoretical Acoustics. Princeton University Press, 1986. Originally published by McGraw-Hill, 1968.

Allan D. Pierce. Acoustics: An Introduction to Physcial Principles and Applications. AIP, 1989. Originally published by McGraw-Hill, 1981.

V. John Mathews and Giovanni L. Sicuranza. Polynomial Signal Processing. Wiley, 2000. Contains sections on multidimensional discrete FFT representations of truncated Volterra series expansions.

[Hink]

Dave, please understand my post wasn't about any of that...I simply was agreeing that I have never used a convo reverb in a song. That's only because I find variverb to be what is most intuitive for me and suits my needs, it wont be what reverb I use that makes or breaks a piece.

Like I said IRs for cabinets is a different matter, I have a few options for tube pre-amp line outs as well as the line out on my Weber attenuator that all work very nicely with IRs

[valhallasound]

Here's a question: How much CPU does it require for a dynamic convolution algorithm to emulate the time varying Lexicon algorithms designed by Michael Carnes? The Lexicon plugins are well known to be low CPU.

Michael Carnes no longer works at Lexicon, BTW. However, he designed the algorithms for a LOT of devices while he was there:

PCM91
PCM92
PCM96
Lexicon 960
PCM, LXP, MPX plugins

I think the guy has a pretty decent understanding of acoustics. As does Barry Blesser. Smiley Face!

Sean Costello

[Zaphod (giancarlo)]

Here's a question: How much CPU does it require for a dynamic convolution algorithm to emulate the time varying Lexicon algorithms designed by Michael Carnes? The Lexicon plugins are well known to be low CPU.

Michael Carnes no longer works at Lexicon, BTW. However, he designed the algorithms for a LOT of devices while he was there:

PCM91
PCM92
PCM96
Lexicon 960
PCM, LXP, MPX plugins

I think the guy has a pretty decent understanding of acoustics. As does Barry Blesser. Smiley Face!

Sean Costello

I owned and used a lot of lexicon devices, I think they are very musical, but realism is not their strongest point. There was a time they were the best (and only) thing availabe, and their particular sound was used in many productions we love. It doesn't mean they are accurate. Convolution techniques are way more accurate for a real concert hall simulation, IMHO. Try by yourself...

[aciddose]

There's no convolver ("dynamic" or not) that provides different responses for instruments that are playing together, nor for their different and/or varying intensities.

how many reverbs of any kind are there that do?

i would think anything include any non-linearity whether made up of clipping, limiting or compression, waveshaping or saturation would have such an effect.

so you'd have to assume that any proper plate reverb would.

air is certainly non-linear with respect to the amplitude to frequency relationship. (waves should interact and stretch one-another.)

although this really isn't a A vs. B type discussion, in any case this simply isn't possible with convolution. whether or not it's been done in any particular algorithmic reverb doesn't really matter.

[Zaphod (giancarlo)]

There's no convolver ("dynamic" or not) that provides different responses for instruments that are playing together, nor for their different and/or varying intensities.

how many reverbs of any kind are there that do?

i would think anything include any non-linearity whether made up of clipping, limiting or compression, waveshaping or saturation would have such an effect.

so you'd have to assume that any proper plate reverb would.

air is certainly non-linear with respect to the amplitude to frequency relationship. (waves should interact and stretch one-another.)

although this really isn't a A vs. B type discussion, in any case this simply isn't possible with convolution. whether or not it's been done in any particular algorithmic reverb doesn't really matter.

we can, so it's possible (for example vnxt library). Whether or not it's done in other convolution based products doesn't really matter

[aciddose]

you can produce hysteresis combined with an impulse? link the patent.

[AudioGuy720]

Both algo and convo reverb have their uses. I find GOOD convo IR's (Altiverb or Waves) to sound the most natural but algo is way more flexible.

[whyterabbyt]

There's no convolver ("dynamic" or not) that provides different responses for instruments that are playing together, nor for their different and/or varying intensities.

how many reverbs of any kind are there that do?

i would think anything include any non-linearity whether made up of clipping, limiting or compression, waveshaping or saturation would have such an effect.

so you'd have to assume that any proper plate reverb would.

I thought real plate reverbs had at most two inputs?

Im talking about the 'different responses for instruments that are playing together'. That kind of dictates a discrete input per instrument, surely?

[aciddose]

sometimes i don't know if you're just being ridiculous or not, this time i'll go out on a limb and assume you really just had a brain-fart and didn't realize that any non-linear device will react differently when you input more than one component of a mixed signal. (for example, two guitar strings vs. one.)

for example, a distortion.

reverbs that take multiple inputs though? i've seen many of these which allow you to position both the multiples of inputs and outputs and use various virtual mic configurations. i'm not sure of any particulars because i've never used them though, maybe someone else can chime in on that.

[Zaphod (giancarlo)]

you can produce hysteresis combined with an impulse? link the patent.

there are pubblications around, google them, and check the technical page on our website, something was not pubblicated but you could guess it from there

http://www.acustica-audio.com/index.php ... Itemid=133

[whyterabbyt]

sometimes i don't know if you're just being ridiculous or not, this time i'll go out on a limb and assume you really just had a brain-fart and didn't realize that any non-linear device will react differently when you input more than one component of a mixed signal. (for example, two guitar strings vs. one.)[/quote[

sometimes i don't know if you're just being ridiculous or not, this time i'll go out on a limb and assume you really just had a brain-fart and forgot the actual context raised by the poster I was addressing, who happened not to be you. To wit:

As I said before, a real orchestra is spread out in space. So is the audience. The impulse response from the second clarinet to seat 7C is different than the impulse response from the first clarinet to the same seat. A convolver effectively squishes the entire orchestra into a couple of points, quite accurately mimicking descent into a black hole.

You might have started going on about 'non-linearity', but that's not what I had any interest in. Witter on about it to your heart's content, if you want, but Im asking something else of someone else, so perhaps you could stop pontificating to me about something irrelevant to me as though i missed your point.

Im interested in knowing which reverbs are actually capable of managing the soundspace of a full orchesta, instrument by instrument. And his reply made me think that Shy was talking about that too. So perhaps you'll let him speak for himself, instead of interjecting yourself in the middle of a question I asked him.

reverbs that take multiple inputs though? i've seen many of these which allow you to position both the multiples of inputs and outputs and use various virtual mic configurations. i'm not sure of any particulars because i've never used them though, maybe someone else can chime in on that.

Next time, maybe just say nothing if you dont have an appropriate answer, then.