KVR Audio

antto · Post by **antto** » Tue May 22, 2012 7:00 pm

i think i *DO* understand the idea behind making the 1st order LP delayless..

i've also been thinking about that long before the first thread about "delayless" spawned.. but i couldn't figure it out, i guess i was thinking in the wrong context (about a feedback compressor, rather than filter)
but i had the feeling that it is so simple that it should be possible to precalculate the "next output"

but if you have 4 such delayless 1st order LPs, you can't just chain them and take the output of the last one and feed the first one to make a ladder filter, right?

as far as i understand.. the so called "TPT" ladder filter is no longer made out of individual peices of code (aka 1st order LP filters) but rather the whole thing becomes one big fat "mess" of problems.. and my brain overheats instantaneously

AdmiralQuality · Post by **AdmiralQuality** » Tue May 22, 2012 7:40 pm

Z1202 wrote:Hi all

Here it is: The Art of VA Filter Design. It is based around the TPT (a.k.a. zero-delay feedback) filter design method. No preliminary filter design knowledge required.

The book has been actually written in the last week of April, but there were some unforseen delays getting it published. In the meantime I replaced the originally used DF2 integrators with TDF2 integrators (thanks to mystran ). It's remarkable how at the very same time the same topic got extensively discussed on KVR

The book text hasn't been extensively proofread, so if you spot any mistakes, the feedback is welcome. Equally, I would appreciate the mirroring of the file, in which case please drop me a note, so that I can inform you of the updates. Quite probably the book will be also hosted on NI's website in the near future.

I have also been recently thinking about the Acknowledgements section of the book. While I have acknowledged everything which I consciously used in the book's writing, there's at least one uncovered area: acknowledgments of the pioneering works, even though not explicitly used during the book's writing. E.g. the book describes, among other things, Antti's ladder filter model, IIRC the diode ladder filter model has been described in some DAFX article etc. But then, even the lowpass RC filter must have had some inventor, and as such should deserve the acknowledgement as well. So I'm a little bit unsure how to address this issue and would appreciate the advice of the community.

Enjoy!

Regards,
Vadim

Thanks for sharing! Will take a look ASAP.

Ichad.c · Post by **Ichad.c** » Tue May 22, 2012 8:24 pm

antto wrote: but if you have 4 such delayless 1st order LPs, you can't just chain them and take the output of the last one and feed the first one to make a ladder filter, right?

Nope.

antto wrote: as far as i understand.. the so called "TPT" ladder filter is no longer made out of individual peices of code (aka 1st order LP filters) but rather the whole thing becomes one big fat "mess" of problems.. and my brain overheats instantaneously

Yip, I've spent hours with Maxima now with zero results. If you don't know/understand the math behind it - it's futile IMHO. Unless someone writes a step by step Maxima or Octave tutorial (from a pre-schooler's perspective) -> Delayless filters are not obtainable by us mere men.

Andrew

antto · Post by **antto** » Tue May 22, 2012 8:46 pm

well, i don't know how to use neither of the two things you mentioned
i only once tryied Scilab (if that is even the same sort of app??) but i was like in front of an alien interface

mystran · Post by **mystran** » Tue May 22, 2012 9:49 pm

Ichad.c wrote: Yip, I've spent hours with Maxima now with zero results. If you don't know/understand the math behind it - it's futile IMHO. Unless someone writes a step by step Maxima or Octave tutorial (from a pre-schooler's perspective) -> Delayless filters are not obtainable by us mere men.

Ok, so we basically need:

1. integration (needs to be understood conceptually)

2. trapezoidal integration as a digital filter; how does it approximate integration (conceptual level should be fine), and how do we implement it

3. solving systems of equations

If you don't understand one of these, maybe we can try to give some pointers. Please tell which point you have trouble with, and I (or someone else) can try to help.

Ichad.c · Post by **Ichad.c** » Tue May 22, 2012 9:54 pm

antto wrote: Scilab (if that is even the same sort of app??)

Yeah, Maxima and Scilab are the same - I prefer Maxima though - much easier interface for a person like me

to understand. Mainly use it for its "simplify" command, and taylor/pade aproximations. You can use it to solve for delayless feedback. Mystran gives a short example in his cheap zero-delay thread(page2), though you still need to understand the math behind it to do anything fancy.

antto · Post by **antto** » Tue May 22, 2012 10:03 pm

nah, i give up.. i'd better concentrate on something that i *do* understand
altho, dealing with dsp without understanding filters doesn't fit together

trimph1 · Post by **trimph1** » Tue May 22, 2012 10:48 pm

Great read here...thanks for the new form of headache!!!

bmrzycki · Post by **bmrzycki** » Wed May 23, 2012 12:30 am

Vadim, is it possible to get the book in a portable reader format like epub and mobi? I'd love to read this on the go with my kindle and while it supports PDF the feature-set is limited. If it's too much effort then don't bother but it never hurts to ask.

Many thanks for this! I definitely will read it over time, regardless of format.

Unaspected · Post by **Unaspected** » Wed May 23, 2012 12:37 am

Great post, Vadim. It's downloaded and I shall make myself understand this! Eventually.

kuniklo · Post by **kuniklo** » Wed May 23, 2012 2:09 am

Thank you very much for putting this together and for so generously putting it out for free!

Until now all this information was scattered across DSP literature on generic filter design and lots of short papers on the black art of making musical filters.

Z1202 · Post by **Z1202** » Wed May 23, 2012 4:48 am

I'd like to thank everybody for the initial warm reception. Let's see how well the book works in practice

As for the desire for less math, the math background is already very much stripped down here compared to what it could have been, should I have attempted a full explanation of the way the filters work

Nevertheless, the math needed in the majority of the practical aspects is actually restricted to the basic school algebra rules! Essentially, it's just about solving the equations of the form x = ax+b

As for putting out a MOBI or EPUB version of the book, I guess one needs something like a PDF to EPUB converter, not sure which ones are out there and how well can they work for this kind of PDF, but you could try them out. Maybe later I'll check the options of building EPUB directly from latex, no promises though as to when or whether, as I'm pretty sure this will require at least some hand-tuning if being possible at all. I also wonder how well would this kind of text scale across different reader devices.

Regards,
Vadim

PS. Google celebrates Bob Moog's birthday today (didn't know it's his birthday). Nice: the book's release was just on the previous day, as if honoring his contributions. And he gets the most important acknowledgement in the book

V@dim · Post by **V@dim** » Wed May 23, 2012 9:07 am

Thanks for the book! Have been waiting for such a document a long time ago! Will give it a try read

whyterabbyt · Post by **whyterabbyt** » Wed May 23, 2012 9:12 am

bmrzycki wrote:Vadim, is it possible to get the book in a portable reader format like epub and mobi?

Try this:

http://calibre-ebook.com/

cheppner · Post by **cheppner** » Wed May 23, 2012 10:05 am

Wow, that's a very nice book and covers a well defined area of special interest (not all the bulk around it).

Thanks!

Book: The Art of VA Filter Design 2.1.2