RV 2000 convolution reverb will this kill all future Re: reverb extensions

[doinky]

I'm surprised I don't see very many people talking about this with more enthusiasm.

To me it seems like a big big deal in terms of all racks extensions that are reverbs. Convolution kind of kills all that doesn't it?

[Exowildebeest]

Uh, no.

The RV7000 is a quite basic convolution reverb.

Plenty of room for a more advanced convolution/hybrid reverbs as soon as the SDK allows sample loading.

[maketunes]

Uh, no.

The RV7000 is a quite basic convolution reverb.

Plenty of room for a more advanced convolution/hybrid reverbs as soon as the SDK allows sample loading.

Totally agree

[manisnotabird]

Not if they have any kind of modulation. You can slap a chorus or unison before the RV7000 (if it is a send effect) but that is not exactly that same as having internal modulation on a reverb.

[Exowildebeest]

Of course this obviously isn't a good time to release your new reverb RE in the shop

[Gaja]

Also to my understanding convolution reverb is always a static image of a space. A fixed position for the signal source and the "listener". That means you can not accurately describe movement through that space realisticly. So in that way algorithms can do much more.
Butapart from that, the concept of convolution is pretty nice, with the ability to imitate different types of hardware and all kinds of sonic awesomeness.

[Exowildebeest]

Also to my understanding convolution reverb is always a static image of a space. A fixed position for the signal source and the "listener". That means you can not accurately describe movement through that space realisticly. So in that way algorithms can do much more. Butapart from that, the concept of convolution is pretty nice, with the ability to imitate different types of hardware and all kinds of sonic awesomeness.

Therefore: hybrid convolution-algorithmic reverbs are welcome

[Gaja]

Therefore: hybrid convolution-algorithmic reverbs are welcome

Most welcome indeed.

[ArcoZ]

That's why I've never bought any reverb except for TSAR.
I was waiting for sample upload RE update but they did it the other way.

[Ostermilk]

Therefore: hybrid convolution-algorithmic reverbs are welcome

Now that we have the convolution engine component even if you don't own any RE's you've now got plenty of hybrid reverb options too.

Even the venerable old RV7 is no slouch when it comes to generating early reflections.

The fact that the MkII's convolution engine is quite basic isn't a drawback in the Rack environment, the only reason that convolution will never totally replace algorithmic reverbs is in its static nature as there aren't any major limitations in the convolution engine implementation here.

Stack 'em, modulate 'em, filter 'em, hybridise them you can do what you like now such a wholesome ingredient has been added to the kitchen cupboard.

The biggest, simplest and most obvious reason that it will likely not prevent any new reverb devices appearing though is that most people (including myself) want a brand new toy from time to time.

[Gaja]

Therefore: hybrid convolution-algorithmic reverbs are welcome

Most welcome indeed.

 
Yah well, I have a funny feeling we wont be seeing Altiverb any time soon.

True, but I'd prefer Ircam tools, which is probably just as likely to happen

[normen]

Btw, heres the opinion of the Lexicon 480 creator Michael Carnes on impulse responses made from Lexicon reverbs and impulse response in general (around the 26:30 mark)

A very interesting interview all in all, where he also talks about the supposed "magic" of certain AD/DA converters etc. and debunks most of these myths.

https://www.box.com/shared/static/pbdqy ... masldl.mp3

[Ostermilk]

Btw, heres the opinion of the Lexicon 480 creator Michael Carnes on impulse responses made from Lexicon reverbs and impulse response in general (around the 26:30 mark)

A very interesting interview all in all, where he also talks about the supposed "magic" of certain AD/DA converters etc. and debunks most of these myths.

https://www.box.com/shared/static/pbdqy ... masldl.mp3

Yes the comments on convolution are spot on.  i.e. it's a great way of affecting audio but don't read too much into the science of what makes a useful impulse response.

Which is kind of what I've been saying to you all along when you've often majored on somebodies revered career as being a validation of the impulse responses they produce and the price they charge for them.

A chance encounter with a weird space with a field recorder may actually yeild a far more useful IR than a series of concert halls given the treatment by bunch of audio engineers.

It's strength is to be able create an impression and its effectiveness is governed to a greater extent by who is using it rather than who is charging the most per impulse response whilst telling you they are giving you the most advanced re-creation of a real space.  If convolution worked that way we'd also have some great EQ based room 'correction' solutions too.  The truth is though we haven't but audio convolution through another signal does indeed yeild an interesting effect which is dependent on both the source and the IR.

As a creative effect then it is every bit as valid as any other form of modulation, as a means of representing reality however, it is about as effective as using a family photo to represent that family in it's full richness of quirks and foibles.

Now in Reason we can play with that stuff for fun without having the proprietary cash hawks circling overhead trying to exploit the unwary out of substantial amounts of cash for what is basically snake oil.

Gotta love democracy.

[selig]

Btw, heres the opinion of the Lexicon 480 creator Michael Carnes on impulse responses made from Lexicon reverbs and impulse response in general (around the 26:30 mark)

A very interesting interview all in all, where he also talks about the supposed "magic" of certain AD/DA converters etc. and debunks most of these myths.

https://www.box.com/shared/static/pbdqy ... masldl.mp3

Michael started work on the PCM-90 and later IIRC. He lives near me in Utah and though I've only conversed with him via email I found him to be a very nice and accessible guy!

Current convolution technology is similar to early sampling tech - one sample used to represent the entire space/instrument. When we can "multi-sample" spaces and especially non-linear FX like amps etc then things will get interesting.

Also, I'm a fan of Reverberate's modulation features and had hoped to help bring that tech to Reason at one point - still waiting on the ability to load user IRs to make it useful!

[Ostermilk]

Btw, heres the opinion of the Lexicon 480 creator Michael Carnes on impulse responses made from Lexicon reverbs and impulse response in general (around the 26:30 mark)

A very interesting interview all in all, where he also talks about the supposed "magic" of certain AD/DA converters etc. and debunks most of these myths.

https://www.box.com/shared/static/pbdqy ... masldl.mp3

Michael started work on the PCM-90 and later IIRC. He lives near me in Utah and though I've only conversed with him via email I found him to be a very nice and accessible guy!

Current convolution technology is similar to early sampling tech - one sample used to represent the entire space/instrument. When we can "multi-sample" spaces and especially non-linear FX like amps etc then things will get interesting.

Also, I'm a fan of Reverberate's modulation features and had hoped to help bring that tech to Reason at one point - still waiting on the ability to load user IRs to make it useful!

What do you find that's ground breaking about Reverberate?

Sure it's a nicely thought out package for the job things like the ER and Tail sections getting indiviidual treatment, with some EQ, Filters, Chorus and Delay geared up for use with convolution, there's nothing I can see however that's lacking now the convolution engine is in place for putting those elements in a combinator.  Unlless of course I'm missing something...

Sure it's a nice tool with many features and at a great price but man when it comes to working with it 'Convoluted' is the key word, personally I'd rather stick those extras on as seperate devices as wading through all those screens on it is less fun than maintaining a stamp album ...

[normen]

Current convolution technology is similar to early sampling tech - one sample used to represent the entire space/instrument. When we can "multi-sample" spaces and especially non-linear FX like amps etc then things will get interesting.

Well the response of a static room with a static mic and a static sound source in it is the same no matter what volume the signal is - so for that exact purpose an IR actually captures all the info.

[selig]

Current convolution technology is similar to early sampling tech - one sample used to represent the entire space/instrument. When we can "multi-sample" spaces and especially non-linear FX like amps etc then things will get interesting.

Well the response of a static room with a static mic and a static sound source in it is the same no matter what volume the signal is - so for that exact purpose an IR actually captures all the info.

Doesn't work for non-linear devices as I mentioned. Also doesn't account for natural (or algorithmic) sources as Michael Carnes points out in the link you posted earlier. 

[selig]

Btw, heres the opinion of the Lexicon 480 creator Michael Carnes on impulse responses made from Lexicon reverbs and impulse response in general (around the 26:30 mark)

A very interesting interview all in all, where he also talks about the supposed "magic" of certain AD/DA converters etc. and debunks most of these myths.

https://www.box.com/shared/static/pbdqy ... masldl.mp3

Michael started work on the PCM-90 and later IIRC. He lives near me in Utah and though I've only conversed with him via email I found him to be a very nice and accessible guy!

Current convolution technology is similar to early sampling tech - one sample used to represent the entire space/instrument. When we can "multi-sample" spaces and especially non-linear FX like amps etc then things will get interesting.

Also, I'm a fan of Reverberate's modulation features and had hoped to help bring that tech to Reason at one point - still waiting on the ability to load user IRs to make it useful!

What do you find that's ground breaking about Reverberate?

Sure it's a nicely thought out package for the job things like the ER and Tail sections getting indiviidual treatment, with some EQ, Filters, Chorus and Delay geared up for use with convolution, there's nothing I can see however that's lacking now the convolution engine is in place for putting those elements in a combinator.  Unlless of course I'm missing something...

Sure it's a nice tool with many features and at a great price but man when it comes to working with it 'Convoluted' is the key word, personally I'd rather stick those extras on as seperate devices as wading through all those screens on it is less fun than maintaining a stamp album ...

The main feature is the way the EQ can sweep with the impulse is something I haven't seen on other convolution devices, though I'm not familiar with every product on the current market. 

[normen]

Doesn't work for non-linear devices as I mentioned. Also doesn't account for natural (or algorithmic) sources as Michael Carnes points out in the link you posted earlier. 

Heh, thats why I exactly outlined in what situations it actually does work. Otherwise I agree its pointless capturing most algorithmic reverbs as I keep noting in those IR threads. Having a static mic and sound source position isn't exactly unusual though.

[miscend]

Do Convolution Reverbs all sound the same? If I load the exact same impulses in Space Designer or RV 7000, will I get the same result.

[ScuzzyEye]

Heh, thats why I exactly outlined in what situations it actually does work. Otherwise I agree its pointless capturing most algorithmic reverbs as I keep noting in those IR threads. Having a static mic and sound source position isn't exactly unusual though.

The reason algorithmic reverbs internally modulate values, is because their models are more simple than the interaction of sound waves in a real room. To make their model sound more complex the room size (or some other parameter) is altered slightly as it processes a sound. The impulse response of a real space will include a lot more of those little interactions than a simple, static algorithm produces.

[joeyluck]

Oh y'all are using the RV2000 MkII? 
Wait until you get a load of the RV7000 MkII

[joeyluck]

There's always plenty of room for other reverbs.

And since there are plenty of users who say they will wait until Reason 9 is released in 2018, they won't have the MkII for a long time, so they will be buying other reverbs while waiting =P

[Ostermilk]

Do Convolution Reverbs all sound the same? If I load the exact same impulses in Space Designer or RV 7000, will I get the same result.

Pretty much, the output of a FFT is only dependant on a few variables such as the window size for example.

In fact you should be able to pretty much clone the same IR in each by deconvolving a swept sine passed through Space Designer or RV-7000's convolution engines to check that for yourself by means of differencing the two results.

I did a fair bit of deconvolution testing like this during the beta and it copes as well as any other convolution engine around these days.

If you are going to do that kind of testing however be very aware that input levels are critical in order to preserve like for like bit depths in this scenario.  In fact when using the swept sine method for capturing devices alway keep the input level as high as you can without clipping the device for best (most detailed) results, however even small amounts of distortion will ruin your IR when it comes to deconvolving as you'll generally hear elements of the swept sine itself in the finished IR.