T-stops better: then why f stops?

cowanw said:

If the camera is told that the amount of light transmitted at full aperture is (for example) 1.8 as an Fstop when it is actually 1.9 as a T stop, then the camera cannot take incorrect input and yield correct output.

Click to expand...

You mis-understand the system.

The "camera" is not told anything about the environment by the lens.

Metering is done with a hand-held meter. The camera operator sets the desired T/stop on the lens barrel.
The use of T/stops (which are actually measured for each stop on each individual lens), are as accurate as possible.

- Leigh

flavio81 said:

On a typical 35mm SLR, the camera's TTL photometer knows the f-stop and also measures the light that passes through the lens. So the meter value incorporates the light transmission loss caused by the lens.

Click to expand...

That's correct.

Any loss through the lens is the same as reducing the ambient illumination of the subject.
The exposure will be correct regardless of which interpretation you choose.

- Leigh

David Lyga said:

I think that Hollywood, whose exposures have to be correct, have T-stops incorporated into their lenses (taking into account all lens aberrations which compromise full accuracy). T- stops (transmission stops) are calibrated so as to give an actual doubling or having of light transmission with each step. F stops are supposed to also, but they are not perfectly matched to that precise requirement, although theoretically they are. Is there a reason why manufacturers opt for the somewhat less accurate f stop? - David Lyga

Click to expand...

For 35mm still cameras and all formats larger than that the T-stop is useless once you rack out to focus. With tiny format cameras, that are focused near infinity much of the time, there may be some utility in using T-stops.

The Jussi test shows the Olympus 50mm f/1.8 lens as having FL=51.4mm and max aperture = f/1.89
I just compiled this chart, using the Jussi data as well as Modern Photography data from three vintages of OM 50mm f/1.8 lens:


Using the Jussi data, although the lens measures f/1.89, its light transmission is f/2.05 due to inefficiencies in the design; but Modern Photography did no similar light transmission measurement.

f/stops work for film photography why change? Those with TTL meters in their slrs already have the transmission losses compensated. Each stop is a factor of two and well within the tolerance of the film.

wiltw said:

The Jussi test shows the Olympus 50mm f/1.8 lens as having FL=51.4mm and max aperture = f/1.89
I just compiled this chart, using the Jussi data as well as Modern Photography data from three vintages of OM 50mm f/1.8 lens:

Using the Jussi data, although the lens measures f/1.89, its light transmission is f/2.05 due to inefficiencies in the design; but Modern Photography did no similar light transmission measurement.

Click to expand...

I can't follow you.

The chart i posted does test for T-stop as well. I don't know which Modern Photography test are your quoting, or what "Jussi test" (?!)

(PS: Jussi = Nesster = fantastic flickr.com user that uploads lots of interesting pictures related to photo gear)

flavio81 said:

The chart i posted does test for T-stop as well.

Click to expand...

You cannot "test" for T/stops using a spreadsheet.

T/stops are a MEASURED value, and will differ, if only slightly, for each individual lens.
They are individually measured and engraved (the lines, not the labels).

That's the whole difference between T/stops and f/stops, the latter being calculated geometrically but not measured.

- Leigh

cowanw said:

I am not sure that TTL solves the problem. TTL is tied to the maximum aperture of the lens, which must be communicated to the camera. That aperture is still measured as an F stop and is inaccurate as any other F stop.

Click to expand...

Bill, there's a TTL meter for LF. Horseman made it, in two sizes. 6x9 (metering area 56 x 82 mm) and 4x5 (sorry, I don't know the metering area's size). It measures average illumination at the film plane at shooting aperture, doesn't know or need to know the lens' maximum aperture or maximum t/stop. It just measures how much light the lens passes. Works, too.

Yep, I have one, thanks
And I am getting the idea now. I set myself to thinking that meters think in EV's and then realized the loss of light within the lens is no different than the loss of light from an overhead cloud in exposure terms.

cowanw said:

Yep, I have one, thanks
And I am getting the idea now. I set myself to thinking that meters think in EV's and then realized the loss of light within the lens is no different than the loss of light from an overhead cloud in exposure terms.

Click to expand...

Interesting that you have a Horseman meter. I recently got one, after a bit of shopping hell, and calculated what averaging across the field means for w/a lenses with substantial falloff. In short, use a center filter. Do you agree?

I do except that I almost exclusively use the setup for portraiture and think the falloff is a bonus. This, as well other aspects of lenses (soft focus for example) make exposure for an individual lens, individual and best learned by experience.

I have also used my Horseman film plane exposure meter in the darkroom to help guide exposure when projection printing color transparencies to 4x5" panchromatic B&W photographic sheet film.

flavio81 said:

I can't follow you.

The chart i posted does test for T-stop as well. I don't know which Modern Photography test are your quoting, or what "Jussi test" (?!)

(PS: Jussi = Nesster = fantastic flickr.com user that uploads lots of interesting pictures related to photo gear)

Click to expand...

The LINK which YOU posted says FL was 51.4mm, the max aperture was determined to be f/1.89, and the T-stop value was f/2.04 measured transmission.
...whose values I inserted in Column C of my posted spreadsheet

The Modern Photography test results which I added into columns D thru F were taken from
Apr 1973, Sep 1979, and May 1983 publications

Perhaps the difference between the F stop and the T stop is more pronounced with zoom lenses (some of them 10 -1 ratio) which I assume have more elements and therefore more transmission losses, and are often used in cinematography. Does anyone know if this is true?

Yes, the losses increase with the number of elements.

Each air/glass surface has roughly the same loss, so the aggregate increases with the number of surfaces.

The same should be true of internal surfaces, as with a cemented doublet, though I'm not sure of the characteristics.

- Leigh

The basic answer is that when Kodak, and possibly others, tried them the public rejected them

jgoody said:

Perhaps the difference between the F stop and the T stop is more pronounced with zoom lenses (some of them 10 -1 ratio) which I assume have more elements and therefore more transmission losses, and are often used in cinematography. Does anyone know if this is true?

Click to expand...

Well, the 8-64/1.9 Angenieux 8x8B that came with my Beaulieu 4008ZM t/stopped around t/3.3. And the 7-70/1.4 Nikkor on the R-10 I tested and returned to the seller was roughly as dim. On the other hand, the 6-66/1.8 and 6-70/1.4 Schneiders on my ZM-2 and 5008S-MS t/stopped at t/1.8 and t/1.4 respectively. Yes, I was surprised.

David Lyga said:

I think that Hollywood, whose exposures have to be correct, have T-stops incorporated into their lenses (taking into account all lens aberrations which compromise full accuracy). T- stops (transmission stops) are calibrated so as to give an actual doubling or having of light transmission with each step. F stops are supposed to also, but they are not perfectly matched to that precise requirement, although theoretically they are. Is there a reason why manufacturers opt for the somewhat less accurate f stop? - David Lyga

Click to expand...

I own several cine lenses that are T-stop rated. T-Stop lenses have the aperture control on a ring on the lens barrel and it's "de-clicked", giving the glass a smooth step-less variable aperture (if you ever actually shoot with one, it's pretty awesome). From there, it's a simple calibration step at manufacturing time to measure the light transmission and put the markings on the lens barrel in the right places. The reason this is done is most cine lenses are primes (zooms exist, but are super expensive), and on set, once you have your lighting and exposure set up, changing lenses becomes super simple: make a note of the t-stop setting on the lens, put the new lens on and set it to the same t-stop setting. The exposure in the camera is exactly the same and you don't have to spend another block of time changing your lights to get the same exposure. Lots of lights used in cine land also come in standardized light outputs (e.g. a 1K light, etc.) it's simpler and easier to shoot with a matte box, take a light reading on the subject, then calculate how much ND you need to drop in your matte box to shoot at the T-Stop you want to shoot at. Again, when it comes time to change lenses, you don't have to go and re-figure the lighting back out.

It just makes it easier to set the lighting up where your ratios and dynamic range are what you want for the scene, and the overall brightness is a little on the bright side, and doesn't change from shot to shot and between lens changes, and you simply control your in-camera exposure and DOF with matte box ND and what you set your lens T-Stop to. Like lenses, good cine ND filters are calibrated, so 1 stop is 1 stop.

The reason for F-Stop ratings on most still cameras comes down to something very simple: electronic aperture control. It's reasonably accurate, and for a given lens and f-stop the DOF is accurate for that focal length. The downside, is the same f-stop on different focal length lenses don't transmit the same amount of light, though you can very accurately figure out what the DOF will be. On t-stop lenses they do transmit the same amount of light, but at the expense of not being able to calculate exactly what your DOF will be. For stills it's a big deal to have a razor thin DOF, and that has it's uses in motion picture land too, but the reality of the matter is, more often than not, the DOF is a lot larger than what you'd think it'd be simply because it's extremely difficult keep moving things in focus when your DOF is 1 inch. For motion picture work, you typically want to at least have all of a person's face in focus if not most of their head, but still shoot with a small enough DOF that you can do focus pulls between two people having a conversation, or pull focus to an object to emphasis it. It's not unusual to shoot a lot of stuff at T-4 to T-8 if you have enough light, and maybe get down to T-2.8 or T-2 for closeups and such (if the lens opens up that much).