Support

If you have a problem or need to report a bug please email : support@dsprobotics.com

There are 3 sections to this support area:

DOWNLOADS: access to product manuals, support files and drivers

HELP & INFORMATION: tutorials and example files for learning or finding pre-made modules for your projects

USER FORUMS: meet with other users and exchange ideas, you can also get help and assistance here

NEW REGISTRATIONS - please contact us if you wish to register on the forum

Lin to freq

DSP related issues, mathematics, processing and techniques

Re: Lin to freq

Postby Drnkhobo » Wed Jan 09, 2013 11:34 pm

I should have mentioned, they are The Multimode Biquad Filters from the SM Wiki

http://www.synthmaker.co.uk/dokuwiki/doku.php?id=user_creations:effects:multi_mode_bi-quad:multi_mode_bi-quad

I still dont understand what you mean with the knob Trog. . .maybe you could trow together a basic example?

Much appreciated if you do ;)

Dont get me wrong about the results of my little test, they are "around" the freq I said. +- 10 - 45hz
Drnkhobo
 
Posts: 312
Joined: Sun Aug 19, 2012 7:13 pm
Location: ZA

Re: Lin to freq

Postby trogluddite » Thu Jan 10, 2013 12:47 am

Sorry, think I may have confused you there - I was going from memory and though that the Pitch-Freq output Hz, but I had a peek in FS, and of course it outputs 0...1, so I got all the conversions back-to-front! :oops:

This is what I really meant...
Scale Freq Knb.fsm
(125.5 KiB) Downloaded 1266 times

I've set the knob scale to give roughly 20 - 20k Hz, but the note numbers (i.e. knob min/max) can be floats, so a bit of tweaking after the decimal point would be able to get that a little more accurate.
In fact, with most filters, you want to avoid getting too close to zero or one anyway, as it will often cause the filter to become very unstable
All schematics/modules I post are free for all to use - but a credit is always polite!
Don't stagnate, mutate to create!
User avatar
trogluddite
 
Posts: 1727
Joined: Fri Oct 22, 2010 12:46 am
Location: Yorkshire, UK

Re: Lin to freq

Postby Drnkhobo » Thu Jan 10, 2013 12:24 pm

Ahh, no worries Trog! :lol:

Thanks for the example, Ive tested and it still seems to be off around the lower registers...
Either the calc is wrong or Ozone is lying, but when I listen to it, it does seem to be off quite a bit. . . hmmm

I did a quick control test and it does appear off. . .

Test2:
Im running a kick drum (nice and beefy) thru a HP, the knob is set like your example and I send the Lin(0..1) to the filters cutoff and Hertz to my label....

Only around 116 (on the knob) do I START to hear attenuation of the lower freq . . . so basically i get a 'range' of +-18 on the knob. You get me?
(The last 116-134 values actually do something, the rest dont)


Thats saying only the last 1/7th turn of my knob attenuates . . . and its not my system, im using a good pair of headphones (no cheapies) and my experience in music prod means I can hear when frequencies are not there
(Lower freqs are easier,its not hard)

Im gonna play around today some more ;)
Drnkhobo
 
Posts: 312
Joined: Sun Aug 19, 2012 7:13 pm
Location: ZA

Re: Lin to freq

Postby Nubeat7 » Thu Jan 10, 2013 2:28 pm

how about using an array with 127 fixed values instead..
User avatar
Nubeat7
 
Posts: 1347
Joined: Sat Apr 14, 2012 9:59 am
Location: Vienna

Re: Lin to freq

Postby Drnkhobo » Thu Jan 10, 2013 6:12 pm

Wouldnt all this be easier if somehow I could extract the current cutoff freq from the actual filter - -
Because its the filter response that im looking to get (yeah in hz)
Drnkhobo
 
Posts: 312
Joined: Sun Aug 19, 2012 7:13 pm
Location: ZA

Re: Lin to freq

Postby trogluddite » Thu Jan 10, 2013 8:55 pm

It's a nice idea, but in practice is wouldn't help very much.
Within the filter the frequency you input is transformed into a matrix of values that are pretty complex to work out, using sin/cos etc. That spits out usually five values that then are used for processing the audio.
In principle, you could take those five actual values, reverse all the maths, and find out from them what frequency was really being used. But the maths in reverse would be just as inaccurate as the maths going forwards, so you wouldn't find out much that you don't already know.
I think at this point, we really need to see the schematic to get to the bottom of what is going on - there are too many possibilities to be able to run through them all in forum posts - it might only take a few minutes with the schematic for someone out there to spot the problem.
All schematics/modules I post are free for all to use - but a credit is always polite!
Don't stagnate, mutate to create!
User avatar
trogluddite
 
Posts: 1727
Joined: Fri Oct 22, 2010 12:46 am
Location: Yorkshire, UK

Re: Lin to freq

Postby Drnkhobo » Thu Jan 10, 2013 9:14 pm

Yeah I get you Trog, the math is too much either way.

Heres my filter.
Filter Madness.fsm
(129.87 KiB) Downloaded 1290 times
....
Drnkhobo
 
Posts: 312
Joined: Sun Aug 19, 2012 7:13 pm
Location: ZA

Re: Lin to freq

Postby trogluddite » Thu Jan 10, 2013 9:22 pm

Got it!
I haven't use a multimode filter for a long time - take a peek inside; the cutoff input doesn't go directly to the filter, it's going via a pitch to frequency convertor.
So what's happening is that the knob value is being pitch->Freq converted twice. So either take the cutoff value directly from the knob, or keep your knob as it is, and get rid of the P-F inside the filter, unless you want to keep it for extra modulation inputs

What a difference the schematic makes, eh! - took about 30secs.
All schematics/modules I post are free for all to use - but a credit is always polite!
Don't stagnate, mutate to create!
User avatar
trogluddite
 
Posts: 1727
Joined: Fri Oct 22, 2010 12:46 am
Location: Yorkshire, UK

Re: Lin to freq

Postby Drnkhobo » Thu Jan 10, 2013 9:37 pm

:oops:

aww, I submit, please go easy on me master :ugeek:

so why does it have a pitch 2 freq converter inside for modulation? And why the *144???
Drnkhobo
 
Posts: 312
Joined: Sun Aug 19, 2012 7:13 pm
Location: ZA

Re: Lin to freq

Postby Drnkhobo » Thu Jan 10, 2013 10:18 pm

Thanks again Trog, these results im getting are FAR more accurate than before!
Drnkhobo
 
Posts: 312
Joined: Sun Aug 19, 2012 7:13 pm
Location: ZA

PreviousNext

Return to DSP

Who is online

Users browsing this forum: No registered users and 15 guests