F stops or T stops?

[Solarize]

This is probably just me not thinking straight but.....

If a wide angle lens is inherently quicker than a say a telephoto lens, when we use an external light meter will the resulting exposure not be different depending on the lens we have used? Or is the difference between them considered negligable; hence using f.stops instead of calculating t.stops?

Is 2.8 on a wide angle the same as 2.8 on telephoto, or what about readings for say a large format camera vs a 35mm?

What happens if we use say a wide angle lens adaptor on a standard lens, how does this affect exposure?

I really should know this but hey.......

Thanks

 

[Jonathan Brewer]

An F-stop is a fraction/ratio between the diameter of the iris diaphram at the rear of the lens, and the effective focal length of the lens. If the iris diaphragm opens up to 50mm and the effective focal length of the lens is 100mm, you have F1/2, and since nobody engraves a fraction on the lens barrel, the value engraved is '2', for F2.

The 50mm width at the widest opening of the diaphragm isn't engraved on a lens barrel because it doesn't mean anything, the ratio does,..........the above example of a 100mm lens w/an iris diaphragm at the rear opened up to 50mm giving you F2, would be the same thing as a 50mm lens where the iris diaphragm at the rear is opened up to 25mm, the two lenses are different focal lengths and different sizes, but the ratio is the same.

Whether it's a prime lens or zoom, the principle is the same.

After constructing a lens, and after some tests, the folks who've made the above 100mm lens, discover that the light bounces around in the lens
and there's light loss by the time it gets to the iris diaphragm, so that when the iris diaphragm is set to 50mm they not actually getting F2.........................................so they'll calibrate the lens and engrave on the lens barrel "T2' where F2 is actually being transmitted, this is a 'T' stop.

You meter and calculate in Fstops and you find you need F2 for an exposure, if you have a lens calibrated in Tstops, and you set it to T2 you know you're getting the exposure you calculated for, just think 'T' for trasmission.

 

[keithwms]

Is 2.8 on a wide angle the same as 2.8 on telephoto, or what about readings for say a large format camera vs a 35mm?

Yes it is the same. Because an f/# is a ratio of focal length and pupil diameter, it defines the open solid angle through which light enters, which of course is what you really need to know to determine how much light makes it to your film. The f/# is quite a universal concept, you can apply it to any type of camera lens or format size.

 

[DrPablo]

Light intensity decreases exponentially with increasing distance from the light source (the "inverse square law"). So a long focal length lens will require a wider aperture to illuminate the film to the same degree as a short lens.

So the f/numbers in general are statements of illumination; i.e. the physical aperture is presented in a way that controls for focal length.

If instead of f/stops we used aperture pupil dimensions in metering, you'd be right that we'd have to take focal length into account. As is a given f/stop means the same from lens to lens with respect to exposure.

 

[Helen B]

An f-stop is a relative aperture, and the 'relative' part takes the focal length of the lens into account - so f/2.8 on a 28 mm lens results in effectively the same illumination at the centre of the film plane as f/2.8 on a 280 mm lens, apart from any difference caused by transmission losses in the lens.

The transmission aperture (T-stop or T-number) takes the losses in the lens into account: reflections at air-glass surfaces mostly. The T-number is the f-number of a lens with 100% transmittance that would give the same on-axis illumination for an object at infinity. The difference between the f-number and T-number is nothing to do with the focal length of the lens, it is more to do with the number of elements in the lens and the efficiency of the coating etc. Most modern lenses have something between a quarter and a half a stop difference between the f-number and the T-number, which is pretty insignificant. Older lenses, especially zoom lenses with many elements, may have up to a stop difference.

As an aside, f-numbers are used for focus calculations, whether or not T-numbers are used for exposure, because f-numbers are geometrical.

The effect of an adapter (converter) depends on the adapter, and where it is. A wide angle adapter is almost always in front of the lens (Tele adapters can go behind or in front of the lens). The f-number of the lens is the focal length divided by the diameter of the entrance pupil (the image of the lens iris when seen from in front of the lens). When you put a converter in front of the lens both the focal length and the diameter of the entrance pupil change, and they pretty much change in proportion (this is a simplified version of the whole story, maybe this thread will evolve into something more rigorous). This means that the f-number doesn't change significantly (this is an advantage of teleconverters that go in front of the lens - the downside is that they have to be physically large enough to maintain the new size of the entrance pupil). The important concept is that the f-number is not calculated from the physical diameter of the iris, but the apparent diameter*, and the apparent diameter can change.

When you put a converter behind the lens things are different. The focal length changes, but the diameter of the entrance pupil doesn't. Therefore the apertures marked on the lens are no longer correct, so they need to be offset. Once again, this isn't the full story because the exit pupil diameter has changed, and that has some bearing on the issue.

I'll just reiterate that this isn't a rigorous explanation, but I hope that it is OK to begin with, especially in conjunction with the other good replies you will get here.

Best,
Helen

*There is a special case where the iris is in front of the lens elements. In this case the entrance pupil and the iris are the same.

 

[jstraw]

Right.

F-4 (for example) is a larger diameter hole on a long lens than it is on a short lens. F-4 on a 50mm lens is 12.5mm in diameter. Focal length divided by the diameter of the aperture = f-stop. So, F-4 on a 100mm lens is an aperture 25mm in diameter.

 

[Jonathan Brewer]

This isn't rocket science, you'll pick it up fairly quickly. In terms of calculating for exposure I believe there's a broader consideration,......cumulative error, your meter may be off/the way you're pointing your meter is off for what you want, your film(from batch to batch), what the lens is supposed to be delivering at the rear, the performance of the shutter, and so forth.

A bunch of little errors can add up to one big error, which is why I bracket in both directions exposure wise, regardless of format if it's an important shot.

Now this isn't technically part of your original question, but I think it's an important issue related to calculating for an exposure.

 

[Jonathan Brewer]

OMT..............The entrance pupil is the image of the aperture stop in the optics that come before it. The entrance pupil depending on the construction of the lens can be fairly close to what you've got back at the iris diaphragm, I know, I've checked some of my lenses regarding this.

In terms of calculating for an f number using the entrance pupil, that doens't hold true for a lens that doesn't have a front cell and some lenses don't, and in that case, it's using the iris diaphragm which is in front the rear cell.

 

[Solarize]

Thanks for the help all, I think I'm starting to understand this a bit more.

Just to clarify something... When someone talks about using a wider lens for the speed it offers them, is this only because the lens they are using happens to stop down further than say their standard lens. So a wide angle lens with a minimum aperture of 2.8 is effectively no quicker than a standard lens with 2.8 as the minimum aperture?

Cheers

 

[Helen B]

***

In terms of calculating for an f number using the entrance pupil, that doens't hold true for a lens that doesn't have a front cell and some lenses don't, and in that case, it's using the iris diaphragm which is in front the rear cell.

Jonathan,

If there is no front cell the entrance pupil is the physical iris, so it does hold true. I've added a note to my original post to clarify that, in case it isn't obvious.

Solarize,

Could you give us an example of this reference to using a wider lens for its "speed"? In general, wide lenses have a smaller maximum relative aperture than standard lenses. For example for 35 mm still cameras the fastest production 50 mm lens is an f/1, the fastest production 35 mm is an f/1.2 and the fastest production 24 mm lens is an f/1.4 (I think - I'm sure that someone will correct me if I'm wrong).

I wonder if it has something to do with usable shutter speed, or the idea that you can use a wide lens at wider apertures for the same depth of field (not entirely true, but it all depends on how you think of depth of field).

Oddly enough an f/2.8 lens is said to have a maximum aperture of f/2.8, not a minimum aperture - the minimum and maximum refer to the relative aperture, not the f-Number, which is a reciprocal. f/2.8 is larger than f/4. Similarly 'stopping down' usually refers to going from f/2.8 to f/4, for example - ie making the relative aperture smaller. Making the relative aperture larger can be called 'opening up', even though you may be going from f/4 to f/2.8.

Best,
Helen

 

[jstraw]

Thanks for the help all, I think I'm starting to understand this a bit more.

Just to clarify something... When someone talks about using a wider lens for the speed it offers them, is this only because the lens they are using happens to stop down further than say their standard lens. So a wide angle lens with a minimum aperture of 2.8 is effectively no quicker than a standard lens with 2.8 as the minimum aperture?

Cheers

That is correct.

 

[bjorke]

...f-numbers are used for focus calculations, whether or not T-numbers are used for exposure, because f-numbers are geometrical.

It's rare that a still photographer needs to worry about T-stops anyway.

T-stops are most important for movie cameras, because changes in exposure from shot-to-shot break editorial continuity. But still, in general, have no such concept.

 

[DrPablo]

Just to clarify something... When someone talks about using a wider lens for the speed it offers them, is this only because the lens they are using happens to stop down further than say their standard lens. So a wide angle lens with a minimum aperture of 2.8 is effectively no quicker than a standard lens with 2.8 as the minimum aperture?

I think this idea (of faster speed with a wide lens) is a misconception that results from the fact that short lenses can be handheld for longer exposures. It feels faster, but in reality it's not.

 

[Jonathan Brewer]

Regardless of what you want to call it, in terms of a lens w/no front lens cell, you'll caluculate by using the iris diaphragm/aperture, period.

 

[Christopher Walrath]

God bless. How many times do I need a smack on the back of the head? I knew f-stop numbers were the ratio of focal length to aperture diameter, but I did not figure transmitting the same alues from lens to lens. I am so glad I am here and not at some digital hack group like BP.com.

 

[Jonathan Brewer]

I don't know what you mean by 'special', you might call a Kodak portrait lens 'special' in terms of no glass in front the aperture, certainly a convertible lens is more common in LF photography.

Also, I learned on the way up, and this is probably moreso with a LF lens, the idea of 'slop', that is, I always go wide open and then close down to the taking aperture to eliminate as much as possible any 'slop' in the aperture blades.

 

[Helen B]

It's rare that a still photographer needs to worry about T-stops anyway.

T-stops are most important for movie cameras, because changes in exposure from shot-to-shot break editorial continuity. But still, in general, have no such concept.

I agree entirely. Thanks for adding the correct perspective.

Best,
Helen

 

[Helen B]

I don't know what you mean by 'special', you might call a Kodak portrait lens 'special' in terms of no glass in front the aperture, certainly a convertible lens is more common in LF photography.

By 'special case' I just mean a particular case in which the entrance pupil is the physical iris. I have no idea why you are making such a big thing out of this.

Also, I learned on the way up, and this is probably moreso with a LF lens, the idea of 'slop', that is, I always go wide open and then close down to the taking aperture to eliminate as much as possible any 'slop' in the aperture blades.

It's also standard practice in cinematography, which we both work in, I believe.

Best,
Helen

 

[Jonathan Brewer]

'I have no idea why you are making such a big thing out of this.'..........

.............THAT'S RIGHT, you don't have any idea of why I say something, so don't try to guess. I'd appreciate no more off topic comments that get personal. Thanks.

 

[Solarize]

I wonder if it has something to do with usable shutter speed, or the idea that you can use a wide lens at wider apertures for the same depth of field (not entirely true, but it all depends on how you think of depth of field).

Best,
Helen

No need for the reference as this answers my question. I clearly wasn't thinking straight last night.

Thanks for the clarification on 'stopping down' Helen and thanks to everyone else.

 

[Solarize]

It's rare that a still photographer needs to worry about T-stops anyway.

T-stops are most important for movie cameras, because changes in exposure from shot-to-shot break editorial continuity. But still, in general, have no such concept.

Well I might as well learn it all now; I hope eventually to be a cinematographer.

 

[Woolliscroft]

With a lot of cameras TTL metering will sort out any variability anyway. I have often wondered whether zooms might trasmit less light at a given aperture, simply because they have so many elements and so more, potentially reflective, surfaces?

David.

 

[Helen B]

Hi David,

Based on the marked f-stops and T-stops on the ciné lenses I’ve used I’d estimate that normal primes lost about a quarter to a third of a stop; and zooms and retrofocus primes with many elements lost a third to a half a stop – but it is difficult to be anywhere near precise when the stops are only given to two digits (eg a zoom of f/1.1 and T1.3 – which don’t mean f/1.10 and T1.30 !). Older lenses with poorer coatings could have more of a difference.

Best,
Helen

 

[Roc Chan]

Macro lens such as Luminar and other technical lens also use T stops .

 

[Nicholas Lindan]

I have found the f-stop/t-stop difference is only significant for a wide open lens, the manufacturers get pretty optimistic - I have measured up to 1/2 a stop on a prime lens, a zoom - I don't want to think about it. If there is appreciable error in the remaining stops it tends to be a constant error - all the stops are 1/4 off. This behaviour applies to still camera lenses - where getting the the maximum apperature out of lens is de riguer marketing. Cine lenses are different beasts.

The f-stop shown on the lens should be used when calculating depth of field: an f1.4 lens may in reality be an f1.8 but it will have an f1.4's depth of field - or should have.