Alright, i never said I would flip again! Here's another theory.
The thing that's still bugging me is that when I test the LEDs by them selves, they are fine, when I power all the colors I get white, but when I try to get "white" in the grid, by applying equal power to the 3 elements, it looks like a rainbow!
So I've just done a bunch of reading and it says that silicon diodes have a forward voltage drop of 0.7v, which makes sense, this is what could very well killing my blue diodes. I'm only running off about 3.4v, this is what my logic runs at and I thought it would be enough for the LEDs, but obviously there is too much of a voltage drop, even without the diode, I mean the LEDs are diodes anyway so they must have a voltage drop too!
RED: Typical: 2 V Max: 2.4V GREEN: Typical: 3.4 V Max: 3.6V BLUE: Typical: 3.4 V Max: 3.6V So possibly, all the LEDs create a considerable voltage drop which is causing the blues to be dim, come to think of it, not possibly but probably! Potentially a gross oversight on my part, owing to my newbness.
So maybe the answer is just to pump up the voltage! The only problem is, I can't just pump it up because my logic can't run at more then 3.4v. So I have to work out if I can run separate power to the LEDs, and if I can then I need 2 different power supplies. And having a look at this diagram it looks like I can do it, there is a separate Vcc and V(LED) !! So next step is to make a 2nd power supply for the LEDs, that shouldn't be too much of a problem.
So I'm not saying there isn't an inconsistency in the LEDs, or that this will 100% fix the problem, but I never noticed this kind of issue in my previous RGB LED projects when I was running on 5v like this http://sites.google.com/site/v3spin/Main/electronics/RGB_LED_3 , so there is hope yet.