Well, research comes in when developing new types.

All weapons of the same type have the same affinity graph.

So if you and your opponent both use weapons of type "plasma" you both have the same affinities.

You could develop a new type thats closely related to plasma, "fusion plasma" or something else, but an established type cannot be changed anymore.

Well... a finite amount of something spread over infinite parts is zero per part.

The average over all possible affinities could and should be zero, that way no type is inherently better than another.

A type might be better in a limited spectrum over another, but over the whole it should even out.

A graph on the HUD would not be very useful i think... as you cant change it on the go...

It would just be a static graph.

Not much sense in cluttering your HUD with an image

And your talking about spectrum gives me an idea:

A weapons type is a spectral graph.

Every type is a more or less random graph in a form like the one showed in the sensor interface mk2.

And the affinity is in short how much those graphs match.

0% affinity is when the average deviation between the graphs is zero, which is the case between 2 completely random spectra or when the graphs perfectly match.

Positive affinity is when both graphs are the same but one of them is higher in its amplitude, same graph, higher amplitude -> positive affinity.

Negative affinity is when the average deviation is negative.

Same graph, lower amplitude -> negative affinity.

When we limit the total amount of spectrum we prevent the theorethic "all at max" type from appearing

This makes affinities asymmetric, though.

So a plasma weapon vs a laser shield would have +50% affinity, but a laser weapon vs a plasma shield would have -50% affinity.

It makes it a little bit more complicated, but very logical at the same time.

Plasma is good against laser, under all circumstances.

this also makes out of the infinitely connected graph of independent types a big coherent whole which follows strict, reproducable rules

the spectral approach also makes it more understandable for research, as you know what you have to do to build a counter to the enemies tech

@Hyperion:

Before you claim that this is the same as your RGB idea i say:

Yes, in a sense, but my system calculates the relative affinity differently and has a muuuuuuch higher variability.

With 24 bit rgb you have 256*256*256=256^3 possibilities.

With for example 256 spectral lines with 256 possible states you have 256^256 possibilities, which is sliightly more