Hi,
f-stop and light loss aren't the same thing. The f-stop is the relationship between the effective diameter and focal length of the lens. The problem is as one aproaches 1:1 distortion increases in a non-linear fashion.
So, you can have a f-stop less than one, but the design is going to have to do some fancy tricks to minimize the distortion. With modern computer controlled, laser measured, glass shaping systems, this sort of thing is easier to do than it used to be. I think there was a commercially available Canon 50mm f0.9 at one time, but I can't recall for sure. I know there's a 50mm f1.0, though, available right now, for a sizable chunk of change.
There is always light loss in a lens, mostly from reflection of light off the air/glass interface of each element. This is why coatings were developed, to reduce this loss. There are also designs that use a very small number of elements (such as a Tessar design).
This is the reason why prime lenses tend to outperform zooms - less elements. Of course, the elements are necessary to correct various distortions. The best thing would be to use
one element, like the Mark-I Eyeball, but that's a bit too hard to do.
Lens design is, in reality, a large set of trade-offs. Here's a few practical examples:
I have a special 58mm f1.2 lens that has a specially shaped objective (front) element in it. It's designed to minimize coma (the star points seen from lights in night shots), and works very well for this. It's not all that great in any other application, though, being intended to shoot wide open, set at infinity, at night. It's great for cityscapes and such. Use a different lens for other shooting, though. Oh, and it cost over $2000 fifteen years ago. Quite a bit of money for a lens you can't use all the time....
I have another special lens, a 45mm f2.8, which is a Tessar 4-element design. It's a very flat field lens and great for reproduction (copy stand) work. It does tend to make everything look flatter than other lenses do, though. Sometimes that just what one wants, though. At least this wasn't a costly lens like the 58mm ($300).
I also have a standard 50mm f1.4 lens, which is best as a general purpose unit, and we all probably know how this lens performs (a 50 f1.4 or f1.8 being the mainstay of 35mm photography seemingly forever).
So, I have
three prime lenses in the same basic focal length range, since each does something a bit different than the others. Would I need to add a 50mm f1.0 to the mix? Maybe someday - if I needed a lens with a very, very shallow DOF (like shooting the tips of eyelashes or something).
And, after admitting to having all these primes, I shoot most of the time with a zoom lens! Why? To minimize the foot work, of course. The key here is that I know what levels of performance I'm giving up by using the zoom. Should I need the extra performance, it's easy enough to swap the glass for the correct one to give me the specific performace needed.
However, I don't think that there will ever be any single lens that will work for everything all the time. This is why I dislike non-interchangeable lens cameras. I know the value of being able to pop those suckers on and off.
Stan
Is it not also that light never perfectly passes through glass?
Canon have a habit of making f1.0 lenses (they did one for EOS, and
I have several Super8 cine cameras with such a claim on the lens,
but unless you physically put MORE light in at the start you are
going to lose some in the glass. I think this is why the f1.0 lens
is such a wierd shape.
Correct me if I am wrong, but isn't there a mystery NASA lens tale
about Stanley Kubrick having this lens with an f-stop of less than
1? I believe this is also a 'trick' because there is no such thing
as 'light amplifying glass', despite that fact, companies manage to
make lenses that have EFFECTIVELY zero light loss, or in the NASA
case a lens that actually stops so little light its effective light
loss is negative (totally impossible right??)
Man its all too much to think about, and I think I remember
something about motion picture shutter angles affecting 'effective'
light levels, where stills cameras have no such boon, so it seems.
--
Amateur Photographer
Professional Electronics Development Engineer
