next topic »

[Reply #45 on:]

Could be  you're just going too fast, got too much flashing too quick, the eye can't take it in.
Slow it down some, and clearly turn off a column before turning on the next one.

No, I've played with the delay - from 0 to 5 and it works as expected. So there's nothing wrong with that. My mistake, the multiplexing does work fine. It explains why a single row is quite bright even though a single column is not. It's because rows are generated by switching on one anode at a time while keeping the cathodes constant. But the columns are generated by several anodes at once and a single cathode turned on. Which causes the brightness to dim, possibly because of the low sink value of the 4094.

[Reply #46 on:]

Looks like I'm gonna keep this thread alive even if it kills me! 

My friendly neighbourhood parts guy didn't have any of the high current TPIC6B595s. But thankfully, he finally got the 74HC595s and it has made a world of a difference over the HCF4094Bs.

One curious thing to note. Initially I had the same problem as earlier where the rows were bright and columns weren't. But when I changed the orientation by 90 degrees (in the code I mean), it suddenly looks great. All LEDs are of more or less equal brightness.

Couple of things to note. I switched from using shiftOut to SPI and wow! The speed has increased by more than 10 times. Secondly, I got some pretty 8x8 bit fonts from Richard Prinz's Pixel ROM Font Editor. It has a whole bunch of pre-built fonts from the C64, Macintosh era.

[Reply #47 on:]

Cool.