The idea is to extend on the sharpen function, which is essentially a highpass filter mixed with the dry (original) signal to emphasize the edges of the original source.
My idea is to go wild with the edge width (highpass filter) and gain & polarity.
So the edge will be on both sides (instead of the left side) , adjustable in gain and mix
The original signal can be adjusted in gain as well, for additional effect.
The Sharp module will be something like this: (changes will be made, but this is current at 19-02-22)
pot controls:
Edge mix pos
Edge mix neg
Edge gain
Original signal mix
switch: / or pot
edge width
jacks:
in
out
cv? for HighPass cutoff filter
progress:
The proto works and looks great.
I’m trying to go to the max so you can too.
So I’m still working out the features.
Neg DRY + edge looks cool too (in an experimental way), so maybe I’ll add that as extra.
CV on gain is possible but I don’t think it has great effect (but I have to test this with video rate input). so maybe if the edge width can be controlled in the extreme, it might be a cool feature.
I’m not sure yet if this is going to be 4 or 8HP
We’ll see
Cool, Martijn! CV control on the filter cutoff is pretty essential, I think. For my purposes, it would only need to support LFO rates.
The Sandin Differentiator never impressed me due to its one-sided effect. Edge detection only on the left side? Not worth the effort, in my opinion. You’re solving that problem, so thank you in advance!
Voltage control of cutoff would be very cool but ideally all I/O in a video synthesizer (digital memory modules excluded because we can’t time travel) should operate on wideband frequencies. Designing the main parameter’s modulation range only up to LFO rate would be unfortunate. You probably need a LT125X to implement this, I’d bet.
Sandin Differentiator can be a little underwhelming if you just patch one signal into one of its 6 inputs and monitor one of its two outputs. The trick is to feed its outputs through another module and back into its inputs. Since it’s a HPF, we’re extracting high frequency content so it’s important to use detailed images to get anything to show up. Feedback typically generates higher order harmonics so it’s a great way to get Differentiator chewing on the signal range it wants. You can also change the caps for different cutoff ranges.
OK, I misspoke… it’s not that the Differentiator only has edge detection on one side. Rather, I meant to say that the effect is one-directional. Looks like an emboss filter. With extra smearing. Not really my thing.
In analogue video, the effect can only really be one way (to the right) because things to the left occur earlier in time than things to the left.
You can’t have an analogue effect that goes to the left because the event/change that it is responding to hasn’t yet occurred. That would be a bit like an echo occuring before the initial sound is made.
Yeah, I know how a raster scan works. But there’s no reason why an analog video circuit can’t distinguish between a rising voltage and a falling voltage, and process them differently.
