what is a dot correction data ?


Joined Sep 7, 2009
I think I got confused on my last post when this was in a different thread:
I'm still not sure I understand it. Because there is an AND gate in the comparator, if you wanted to use it as an 8 bit brightness, the 6 bit section would have to be all set high. It looks like if you set all the 8 bit section high and put data on the 6 bit section it would work as a 6 bit brightness control.
I'm not really sure what happens under other circumstances.
If either section is all low then I think the output would be zero brightness.


Thread Starter

Embedded Lover

Joined Dec 22, 2008
The output flickers when I set the 6-bit to zeros , and full brightness when all the 6-bits and the 8-bits are ones.


Joined Sep 7, 2009
Anyone else fancy having a look at this one? - I'm not quite understanding it.


Joined Nov 6, 2005
I belive the 'dot correction' idea is to calibrate the individual LED outputs for 100% brightness.

If you get a batch of LEDs, the brightness for a given current will vary from LED to LED.
Using that chip, each LED can be measured after the whole unit is assembled and a correction table created, which can be saved in the MCUs flash or eeprom.

With that loaded into eg. the 6 bit register bank, the 8 bit bank can the be used for greyscale image data with the knowledge that any given value will be the same brightness for every pixel.

Without that facility, either the LEDs have to be pre-graded for exactly the same brightness (which adds to the cost) or your display program would have to do the maths for each pixel update to apply the brightness compensation the the PWM value for each LED.