[sdiy] MIDI velocity

rsdio at audiobanshee.com rsdio at audiobanshee.com
Thu Apr 7 01:01:37 CEST 2016


On Apr 6, 2016, at 3:46 PM, "Richie Burnett" <rburnett at richieburnett.co.uk> wrote:
>> The worst way I could imagine would be for one set of columns to contain the first switch, and then an entirely separate set of columns would contain the second switch. Then, the firmware would have to scan all the first switches in one pass, then the second switches on a second pass. Of course, even that wiring couldn't prevent the firmware from scanning the columns in a different order to keep the reads localized to one part of the overall key bed at a time.
> 
>> I suppose I could just look for schematics of any velocity keyboard, and it would show a typical wiring...
> 
> 
> http://www.doepfer.de/DIY/Matrix_61.gif

Thanks.

Looks like Doepfer have specifically separated the Break and Make switches onto separate columns. That means you cannot simultaneously read both. But an interrupt could feasibly read two columns in succession for nearly simultaneous reading.

The left keyboard would have had 4 columns and 8 rows without velocity, but has 8 columns and 8 rows with the separate columns for doubled switches. The right keyboard has the same arrangement, but only uses 5 rows of each of the last 2 columns.

I suppose this wiring makes the firmware a little simpler since the bits positions are the same for both switches on the same key. I have no measurements to say that reading both switches simultaneously would actually provide a timing benefit, but it seems possible.

Brian

p.s. Looks like they're also using dual diode packages with common anode wiring. I've found those packages to be cheaper than buying twice as many single diodes. There are packages with more than two diodes, but for some reason those tend to cost more per diode, despite the reduction in packaging. Probably supply and demand.



More information about the Synth-diy mailing list