Raspberry Pi 4

Aiptek doesn't get me any touch screen hits, just cameras. I've an ex checkout display that has a touch, but its relatively slow 1200 baud, and I assume boxed responses, so I've never considered it for machine control. Its displaying a scan doubled output from a legacy computer, a trs80 color computer-3 in the basement. Way off topic for this list. lol

Just found out that the Touch is non functional on the LCD I got from Ebay.

So ... here we go again to find a replacement LCD.

non-functional or needs a command to make it poll itself? I've looked at the output on the one I have, but haven't tried to write a mouse driver like for it. Or possibly change the serial mouse driver I did write to receive it. It outputs a hit with co-ords but only when touched. ISTR its 1200 baud, slower than a seriel mouse like the old DeXxa I use on the coco. logitek mice and ms mice have only one diff in the protocol, both running at 2400 baud, so I wrote the driver to start in logitek mode, and autoswitch the first time it see's the ms version for the 3rd button. Only works in os9, the aftermarket, unix-like os we've re-written extensively over the last 30 years. Finding several bugs and expanding the file system to handle up to 4GB had drives, we call it Nitros9 now. Discovering new commands hidden in the hitachi 6309 clone of the moto 6809, but its cmos, we've roughly doubled the speed of the os. it even has some 32 bit mulls and divides. Heady times those. Hitachi had perms to clone it, but they started from scratch and filled up the empty spots in the opcode array, but were enjoined from ever admitting they had improved it. Among other things by taking one clock out of the machine cycle. And 25 years later they're still mum.

I wish I still had the brains I had then. But a pulmonary embolism about 6 years back cost me me some IQ. Survival rates on those run about 2%. I sure as hell don't recommend it as a way to die though.

Take care now. Oh, that shrunken kernel, was that for a pi-4?

Touch = No Go on Windows10 / Win8 = dead. Refund on its way.

Yes the shrunken is for Pi4 and it works OK so far.

Good, if and when I get video out of it, I'll take it for a longer walk, but I'm not using gpio for direct control, I using the hm2_rpspi.ko driver for a 7i90 card, buffered by a trio of 7i42A's which greatly simplifies the wiring and protects the 3.3 volt 7i90 from higher voltage gate destroying noise. That was written by Prof Bertho Stultains, but I already found it doesn't like an arm64 install, it was originally written for armhf. Bertho has sent me a version that should run on either, but without video I've not been able to test it on buster 1.1 yet. I found the ship message from banggood in my spam folder, that massive heat sink lid and a pair of the micro to std hdmi adapters s/b here around 10/03/19 if the schedule holds. And about a $20 bill cheaper than amazon. But thats another 2 weeks down the log. Just about the time they are working on me again. I'll turn the rpi4 upside down so the old video card fan can blow on the new sink and the 1" jumper should work just like it does on the rpi3. Its upside down too.

Thats the plan anyway. But Murphy (of Murphy's law fame) is about, that guy even drinks my last near beer occasionally.

Take care now.

Actually the 7I90 inputs are not 3.3V with normal jumpering but are 5V tolerant and
will take +7V maximum inputs. What usually causes damage is negative voltages < -0.5V
with unlimited current destroying the bus switch or FPGA input protection diodes.

This is often caused by things like shorting bare FPGA card inputs to
frame grounds on equipment that has motor drives (step drives, servo drives, VFDs, etc).
The frame grounds may have as much as a couple of volts of HF noise with amps of available
current relative to the FPGA card ground. This will instantly destroy the input protection diode.

If you don't want to use a fancy input protector, a 100 Ohm series resistor will do 99% of the negative
input protection.