I was doing a non-scientific test of my 35mm f1.4 and 105mm f2.5 Nikon lenses and found something odd. The 105 @ f2.5 was much dimmer than the 35mm @ f2.8. The difference was probably around an f-stop as I had to go from 400 iso to about 800iso to get similar exposure (in dim lighting). I seem to remember discussions about how a wide-angle converter does not add exposure but does a prime wide lens gather more photons per pixel at the same f-stop?

For your future reference, perhaps this will help:
F-number - Wikipedia, the free encyclopedia (http://en.wikipedia.org/wiki/T_stop#T-stops)
Essentially, the f/stop system is an idealistic approach the tries to provide a reference number(f/stop) wherein the light transmission thru a lens is a constant. In reality, there is light loss in a lens element system that is not accounted for in the f/stop method because, usually, the loss is insignificant. In lenses with many elements, such as a cinematic zoom lens, this can be more than 1 f/stop.

I was doing a non-scientific test of my 35mm f1.4 and 105mm f2.5 Nikon lenses and found something odd. The 105 @ f2.5 was much dimmer than the 35mm @ f2.8. The difference was probably around an f-stop

You're right Marcus, your tests weren't scientific enough. If you photographed a grey card (or a colour photo) and filled the frame with this card or photo by moving forwards or backwards, both primes would give the same exposure at the same designated aperture.

I'm guessing both the 35mm lens and the 105 mm lens have roughly the same number of elements so the light loss will be very slight between them, but note that the 105 (especially) will only be f/2.5 when focused on infinity - it can easily lose a stop when racked out for close-up work.

A wide-angle converter will absorb a small amount of light, as will anything you place in front of your front element.

tom.

"it can easily lose a stop when racked out for close-up work."

Thanks! That is the piece of information I was missing and have never seen/heard discussed in all the years of doing video/photography. Focus distance and light transmission loss is something that should be mentioned since non-flash photography needs every stop it can get. I went to every page I could find on the web last night regarding f-stops and the only information I found was light transmission loss due to lens elements and coating, none of which should be so dramatically different between the two lenses.

I was doing a low-light test indoors and was focusing near infinity for the 35mm but nowhere near that with the 105mm. I'm glad I went ahead and ordered a faster 85mm as 105mm is a bit too long indoors and now I know it loses more light due to being focused well short of its infinity. I'll be using the new, f1.8 85mm at much closer to its infinity.

"it can easily lose a stop when racked out for close-up work."

Thanks! That is the piece of information I was missing and have never seen/heard discussed in all the years of doing video/photography.

That's probably because modern zooms focus by moving internal elements rather than by moving the large front element or moving the entire lens away from the focal plane. When internal element focusing is employed there's far less light loss but the disadvantage is that they can't focus very close (unless they resort to wide-angle or tele macro which brings with it its own problems and sacrifices).

All lenses are calibrated when focused at infinity, even macro lenses. And f stops are mathematically derived, so a lens that is full of dust and dead insects is still classed as an f/2,8 (say), but its T stop (T for transmission) will be a lot less.

tom.

That's probably because modern zooms focus by moving internal elements rather than by moving the large front element or moving the entire lens away from the focal plane. When internal element focusing is employed there's far less light loss but the disadvantage is that they can't focus very close (unless they resort to wide-angle or tele macro which brings with it its own problems and sacrifices).

All lenses are calibrated when focused at infinity, even macro lenses. And f stops are mathematically derived, so a lens that is full of dust and dead insects is still classed as an f/2,8 (say), but its T stop (T for transmission) will be a lot less.

tom.

And that's why cinema lenses are rated by T-stops and not F-stops - you simply can not rely on the "f" when dealing with light and grip and all the effort and money that goes into a scene.

For example you can take a modern MF f1.4 85mm Samyang lens and it exhibits all the focal/bokeh properties of a 1.4 but it "T"ransmits as f2-2.8 (and sure it is sharper and better contolled at that)

Actually all those old sub f1.0 lenses are rated at really odd circumstances one of which is internal flare that will "buid-up" some more light.

T

Actually all those old sub f1.0 lenses are rated at really odd circumstances one of which is internal flare that will "buid-up" some more light.

Can't go along with that. A lens is rated as f/1,0 when it's focal length and diaphragm diameter are the same numerically. It can be full of dead rats and flare like 60's loons, but it'll still be f/1,0.

tom.