1) The spectral power distribution curve of the light source is certainly
the place to begin the comparison. But remember that it is the *area* under
that curve, between appropriate wavelength limits, that counts, not just
the *height* of the peaks. (Technically: the spectral power per unit
bandwidth must be integrated over the wavelength range).
2) But, of course, the intrinsic response of the sensitizer to light also
varies with wavelength, and has to be scaled into the calculation. And it
should be remembered that the sensitivity of iron-based coatings (e.g.
cyanotypes or Pt/Pd) is different intrinsically from that for
photohardening of dichromated colloids (e.g. gumbi and Carbon). So it is
essential to compare like with like, and we should not be surprised at the
apparent 'differences' in experience which have been recently decribed (You
can leave the rest of your crow for the cat, Judy).
3) And then there is the question of how much of your original light (at
each wavelength) is actually arriving at the surface of the print: you need
to take into account the transmittance of the glass, as Klaus has pointed
out, and any other apparently clear material in the path; e.g. plastic
masks, and even the negative film base itself, which do have a cutoff in
the UV, and are absorbing an important part of the light you can't see.
If this were all there was to it, we might still have a chance of doing
some useful scientific calculations and predictions. But if you are making
a real print (a strongly recommended activity, as opposed to just amusing
oneself with laboratory photochemistry) there is also (horribile dictu):-
4) *The Internal Filter Effect* (A gasp of horror is appropriate here).
Most 'real-life' alt-proc coatings also contain coloured substances that do
not contribute significantly to the light-sensitivity (i.e., the pigment in
Gumbi and Carbon; the potassium ferricyanide in cyanotype; the noble metal
salts in Pt/Pd). On the contrary, these substances are absorbing your
printing light as it passes through the coating, and giving nothing in
return. E.g. the presence of these coloured (and UV-absorbing) substances
is like putting a coloured filter over your light source - except that the
filter is 'built-in', and inescapable. Once you add this into the equation,
it is obvious that it is not strictly valid to compare the efficiency of
your light sources unless the coatings are *identical in all respects*.
Two differently pigmented carbon tissues, for instance, will not be of the
same 'speed' even if the dichromate concentration is the same.
At this point, the scientist bows gracefully, admits that the real world is
far too complicated for him to quantify, and retires to make some prints
instead.
Mike