Light a spot on the wall with one LED at night. Then add a second LED operating at the same current and shining on the same spot from the same distance. The spot has twice the luminosity.
Does the spot look twice as bright? No.
Can you notice any difference? Hardly.
The relationship is logarithmic, not linear.
relative luminous flux is roughly linear with forward current. perceived brightness is roughly logarithimic with luminous flux.
My analysis was only for luminous flux.
I leave it up to the reader to calculate the perceived brightness.
EDIT: I've changed "brightness" in my analysis to "radiation." Hope that helps to aviod confusion.
I think that you didnt understand the problem I presented.
First, I wasnt speaking about the other color, only about the blue color.
Second, the problem which I presented was that the blue color's VF ranges from 3V to 3.8V therefore I dont know what current limiting resistor to pick.
As i said in the topic: "On the one hand, I dont want all 3V blue LEDs to get burned, but on the other hand, I dont want the 3.8V ones to have a weak lighting intensity".
Talking about a dual LED I assume that both colours are intended to be used.
If the Vf of the blue LED varies in such large dimensions (0.8V) just use a constant current source to supply the LEDs. I hope the forward current (If) is pretty much a constant value for all blue LEDs you've got.
BTW, have the LEDs been grown in a backyard in China?