Spectrum Block

Post Reply
wde
Posts: 257
Joined: Sun Jan 08, 2012 9:28 am

Spectrum Block

Post by wde »

Spectrum Block uses FftSharp to visualize three frequency zones among other things. Unzip to generators folder. Modern Sequence Editor is needed to see the output.

Create a long pattern, right click to choose connection, change parameters to adjust things.
Attachments
SpectrumBlock.zip
(17.89 KiB) Downloaded 8 times
SpectrumBlock_src.zip
(24.13 KiB) Downloaded 6 times
Last edited by wde on Mon Apr 05, 2021 3:47 pm, edited 3 times in total.

User avatar
IXix
Posts: 964
Joined: Wed Nov 23, 2011 3:24 pm

Re: Spectrum Block

Post by IXix »

So cool! You're a legend. 8-)

User avatar
Candle
Posts: 285
Joined: Sun Apr 23, 2017 1:03 am
Location: Canada
Contact:

Re: Spectrum Block

Post by Candle »

Worth the wait :!:

:dance: :D 8-)

See You In The Shadows…

User avatar
HerrFornit
Posts: 344
Joined: Sat Feb 25, 2017 12:27 pm
Location: Steinberg UR12, AKG 702, Old Hi-Fi, A-49, PC: Xeon 1230v3 8GB, Win10x64pro, Buzz 1503x86
Contact:

Re: Spectrum Block

Post by HerrFornit »

Yeah. :!:
Just a few observations after short testing:
-Maybe SB does not clear the result when changing to any "dummy" input connection (eg MSE). How to clean then?
-After playing with parameters, the resulting relief changes with every loop (?)
-A cross or hook for marking the chosen connection would be nice.
-Perhaps a line for a phase correlation :idea: between input signals (usually L/R or any machines(!)) would be a useful extension. ;)

Thanx a lot!

wde
Posts: 257
Joined: Sun Jan 08, 2012 9:28 am

Re: Spectrum Block

Post by wde »

Fixes and cleanups.

Now it resets view when connection is changed. Tooltip shows connection. Should be also accurate (tested with static sample). Phase is out of scope :)

User avatar
HerrFornit
Posts: 344
Joined: Sat Feb 25, 2017 12:27 pm
Location: Steinberg UR12, AKG 702, Old Hi-Fi, A-49, PC: Xeon 1230v3 8GB, Win10x64pro, Buzz 1503x86
Contact:

Re: Spectrum Block

Post by HerrFornit »

thanx for fixes. There is still some "random" factor for results. Even after off commands before looping. Probably no bug of SB rather VST variance! With matilde sample its stable results, too.
wde wrote:
Sun Apr 04, 2021 9:38 am
Phase is out of scope :)
I see. Is phase not a result of the fft anyway? Phase is for me the most useful spectral information when composing, to find interferences. But not essential. Perhaps in another Block machine. :)

wde
Posts: 257
Joined: Sun Jan 08, 2012 9:28 am

Re: Spectrum Block

Post by wde »

Added stereo balance and correlation coefficient calculation which can be used to understand stereo signal width. I cleaned up the code further so it should be pretty straightforward to add new visualizations if someone wants to do that.

User avatar
HerrFornit
Posts: 344
Joined: Sat Feb 25, 2017 12:27 pm
Location: Steinberg UR12, AKG 702, Old Hi-Fi, A-49, PC: Xeon 1230v3 8GB, Win10x64pro, Buzz 1503x86
Contact:

Re: Spectrum Block

Post by HerrFornit »

Cool 8-)
Usually Correlation meters are from -1 to +1. /s. pic. Span and TT Dynamic Range Meter
Is yours (red line 0 to +1 ?)
EDIT: Have seen that you can show the values in tooltips. its also negative so -1 to +1 prob. But values (at least the bars) differ from the red line. In my example SPAN and TT's didnt go below 0 while the red line peaks below 0 to -0.4 or so.
I definitely need same dashed help line fully right/middle/fully left ;)
Correlation Meters.jpg
Correlation Meters.jpg (40.04 KiB) Viewed 120 times

Post Reply