Looper's Delight Archive Top (Search)
Date Index
Thread Index
Author Index
Looper's Delight Home
Mailing List Info

[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Date Index][Thread Index][Author Index]

Re: MAX/MSP Conversion Complete...



Jeffrey Larson wrote:
> Yes I understand.  My point is that you are ultimately limited by the
>  Max intrinsic objects.  In the case of filters, my "MSP Filters" 
> menu shows 14 objects: allPass~, biquad~, etc.

Mine has 16 objects but not all are filters... ;-)
If you know how digital filters work, you probably could do most of
these "analog emulatins" Most filter designs rely on a simple biquad
formula. The secret is the parameters. Recently there was a flag added
to filtergraph~ (an UI object to control a biquad~) called "Imitation
Analog Flavor". The object which does the actual filtering is still the
same as before, it just gets a different set of parameters to do its
magic...

And you are not limited to the Max intrinsic objects. Especially for
filters there are a lot of 3rd party externals and patches which deliver
you filters of all flavours.

You can also create brickwall filters with FFT techniques, not available
for any analog flavor way of thinking (Its not part of the filter object
list). One of my favorites is the "forbidden planet" patch which comes
as example with Max

> can we say for certain that with these 14 objects we can exactly 
> emulate the filters on something like an Access Virus? I don't mean 
> just close, I mean near sample accurate emulation?

If you know the exact algorithm you could do it sample accurate. But
they won't give you that information and more important there is no
point in doing so. As soon you understand enough of the matter to be
able to redo it, you certainly would do it much more personal and 
different.
And there is ALWAYS more than 14 objects involved... Thats the diference
to Reaktor, there you think in "Filter" and thats it. In Max you embed
the filter into a bigger picture to actually control it. And the way you
control your sounds are more important to the sound itself than any
technology involved.

> Is there one and only one possible implementation of hilbert~ ?

As its a mathematically defined treatment, the implementation might
differ, the result must be the same.

> maybe it just takes a lot of work with hilbert~, phaseshift~, and 
> comb~ to get the same "warmth" as some Reaktor patches.

These three I'd never associate with "warmth" ;-)

> I don't know, but what I'm wondering is if these fundamental 
> algorithms color the sound enough so that people can tell the 
> difference between the "Max sound" and the "Reaktor sound".

There is a similar issue with studios, you can hear if a peace is made 
at IRCAM. I consider this a weakness not a strength. If it sounds like 
Reaktor, its done with the presets which come with it. If somebode gets 
into it an tweaks the sounds, it stops sounding like Reaktor, the same 
applies to Max or Kyma or even Nordlead versus Virus... You'll love it 
if the presets fit your taste...

> Maybe it is simply a matter of Reaktor having a better default object
> library that sounds warmer without a lot of tweaking.

Thats the main difference. Reaktor is filled with sounds average people
don't need to tweak to get the expected (average) sound.

Enough, I better watch the sunset on my flight to Oslo...

Stefan

-- 
Stefan Tiedje------------x-------
--_____-----------|--------------
--(_|_ ----|\-----|-----()-------
-- _|_)----|-----()--------------
----------()--------www.ccmix.com