Comment by boznz

3 days ago

Nothing exciting if you look at the code near the end.

for(i=0;i<8;i++) LEDRegister[i]=LEDRegister[i]+TMR0+addr;

Note this is also technically a bug, the for(i=0;i<8;i++) loop executes so fast that TMR0 (an independent PIC timer that increments every 740uS) is usually the same value added to each LEDregister. I actually added +i to the end of this line in my version, but I'm guessing someone making this will tweak the code to do whatever they want or store some patterns as consts. The code really is quite crappy now that I look at it sober, the watchdog timer is turned off and should be on, there is no checksum on the serial data, procedures called from interrupts, defines with double-slash comments, all stand out like a sore thumb. Lucky it just blinks lights and doesn't run a dialysis machine!

3 comments

boznz

Reply
mulmen  3 days ago

Pretty sure that’s the random junk it makes up if there’s no serial data coming in. The comment says:

> // no data coming in so do something interesting

My question is what they send over serial.

  • boznz  2 days ago

    Apologies for not reading the question.

    I currently have a small program on the RPi NAS which polls my AWS server showing a few metrics, uptime, load, traffic and latest sessionID, but it was off at the time of the video as it is pretty static. Only the top panel is actually used for this the others are just random for now work in progress I guess.

    For my retro computer, it is still a bit of a mess, but should be easy to clock out the 16 CPU registers into a serial stream which I will hook a small micro into to send it over a UART. I have actually gotten more excited about completing the retro computer project since having the lights to look at.