LED matrix delay (brightness issues)

I am creating a 32x32 led matrix using these matrices:

http://www.flickr.com/photos/mowcius/sets/72157620898562245/

and the CCShield from Andrew Stone (which I haven't yet worked out how to use properly):

http://www.arduino.cc/cgi-bin/yabb2/YaBB.pl?num=1235795499/0

and some transistors for the v+ lines...

This is hopefully going to be like the 'peggy' LED board...

I have tested one display just using the arduino, and by turning gnd and v+ lines on and off in a row I can create the appearance that the LEDs are lit simultaneously...

The only issue with this is that at the moment it is very long code because I am just turning LEDs on to show an image (very basic at the moment) and because I am just turning the LEDs on and off with no delays, they are on for the same amount of time that they are off, resulting in 50% brightness.

If I could put in a delay smaller than 1ms then I would be sorted but I'm not sure how to do it. I would like the LEDs to be as bright as possible but not visible flashing (which I get with 1ms delay x8 for the 8 rows).

Thanks in advance,

Mowcius

Have you seen the code in my project:- http://www.thebox.myzen.co.uk/Hardware/Econo_Monome.html

That turns each row on at a time and in sequence with the data in an array. Yes for a row with no LEDs on it is a waste of time but it keeps things simple.

The problem you will have is the refresh rate, an individual LED needs to be on a minimum of 32 times a second to stop it from appearing to flash. As the number of LEDs go up this is increasingly difficult and the on / off ratio reduces the brightness.

To reduce the dimming you can increase the current through the LEDs, look at the data sheet and you will see the pulse current rating is very much higher than the continuous current rating. However if the multiplexing ever stalls then you will burn something out.

If you implement your refresh routine as an ISR (interrupt service routine) called from a timer you can set it at any interval you like until the time to execute it exceeds the calling rate.

I can't find a datasheet for my matrices... The forward voltage is 2v (I put 3.3 through it before I read that and it seemed fine - if a little warm)

What would you have thought that the max pulse current rating will be? 5v? That would be handy!

I am also looking at your monome project. I had seen it before (your projects get around!) but I didn't have chance to read through it fully...

Thanks,

Mowcius