Jump to content


Photo

f-stop to T-stop conversion?


  • Please log in to reply
17 replies to this topic

#1 Sal Caldecutt

Sal Caldecutt

    New

  • Basic Members
  • Pip
  • 5 posts
  • Other

Posted 30 October 2006 - 09:55 AM

Im new to posting on this forum, I wondered if anyone could tell me the formular for converting f-stops to T-stops? I carnt seem to find this out anywhere else on the website, Cheers
  • 0

#2 Mark Dunn

Mark Dunn
  • Basic Members
  • PipPipPipPip
  • 2422 posts
  • Other
  • London

Posted 30 October 2006 - 11:42 AM

T stands for transmission and takes into account the light absorbed (and scattered?) by the many elements of a zoom lens, so it depends on the lens. For the Angenieux 12-120, geometrical stop f2.5, it's T2.2 and so on. The loss of light also occurs in still camera lenses, of course, but the small differences can be corrected in printing, whereas in film, shot-to-shot consistency in density is much more important. Failing to allow for the T-stop would introduce an unnecessary variable to be corrected in printing or telecine.
  • 0

#3 Stuart Brereton

Stuart Brereton
  • Basic Members
  • PipPipPipPip
  • 3060 posts
  • Cinematographer
  • Los Angeles

Posted 30 October 2006 - 01:31 PM

T stands for transmission and takes into account the light absorbed (and scattered?) by the many elements of a zoom lens, so it depends on the lens. For the Angenieux 12-120, geometrical stop f2.5, it's T2.2 and so on. The loss of light also occurs in still camera lenses, of course, but the small differences can be corrected in printing, whereas in film, shot-to-shot consistency in density is much more important. Failing to allow for the T-stop would introduce an unnecessary variable to be corrected in printing or telecine.


It's not just zoom lenses that have to compensate for light loss. All lenses do to a greater or lesser degree. This is also the reason there is no 'formula' to be found, as different lens designs require different amounts of compensation.
  • 0

#4 K Borowski

K Borowski
  • Basic Members
  • PipPipPipPip
  • 3905 posts
  • Camera Operator
  • I.A.T.S.E. Local # 600 Eastern Region

Posted 30 October 2006 - 01:42 PM

T stands for transmission and takes into account the light absorbed (and scattered?) by the many elements of a zoom lens, so it depends on the lens. For the Angenieux 12-120, geometrical stop f2.5, it's T2.2 and so on. The loss of light also occurs in still camera lenses, of course, but the small differences can be corrected in printing, whereas in film, shot-to-shot consistency in density is much more important. Failing to allow for the T-stop would introduce an unnecessary variable to be corrected in printing or telecine.


Mark, please don't trivialize still photography as not needing to be as precise. Try matching color and density across a wedding with 500 different shots and you'll see that it can be just as tedious as a feature film. Also, my understanding is that the 12-120 Angenieux, which I own, is measured in "f-stops" not "t-stops". It says "f-stop" on my lens's barrel.

Just be aware that some cine lenses are measured in f-stops as well.

Regards,

~Karl Borowski
  • 0

#5 Michael Collier

Michael Collier
  • Basic Members
  • PipPipPipPip
  • 1262 posts
  • Gaffer
  • Los Angeles, CA

Posted 30 October 2006 - 02:14 PM

maybe he was refering to the 10-150 angie which is marked in T-stops. Or maybe he was trying to say that at the marked f2.5, the lens would be a T2.2 when you compensate.

I think the point he was making is matching is much more complicated for a film where a shot is directly backed against another, and any slight shift in density (even if corrected for) becomes very apparent. Still photography has its own set of challenges, but in my experience no two shots need to be matched to the degree a film does (they need to be matched, just not to such a fine degree).

I don't think he was trying to trivialize stills, just point out a difference. In still photography the instant you snap a photo is way more important than when you call action on set, would that comment trivialize motion picture making?
  • 0

#6 Nathan Milford

Nathan Milford
  • Sustaining Members
  • 692 posts
  • Director
  • New York, NY

Posted 30 October 2006 - 02:17 PM

Your Angx. 12/120 is measured in both, unless you have some odd one-off version.

The white numbers are callibrated for F/stops (F/2.2 wide open) the red numbers are calibrated for T/stops (T/2.5 wide open).

Generally only ancient cine lenses come with only F/stop markings.

Also, I believe Mark wasn't trivializing the needs and practices of stills shooters. What he was trying to illustrate is that, in still photography, you have a lot more room for error as you can fix much more in processing and printing among other niceties. Matching is matching, and for stills it can take hours in the dark room and is in many cases more tedious than with cinematography because you have more options in tweaking developer times or even developers, enlarger lenses, etc...

In traditional cinematography the labs are strictly controled to give consistant results and printing is a bit more tiresome to get precisely what you want.

Photochemically speaking, matching a tenth of a stop exposure differential is a lot easier for a still being printed to paper then for a 10 second motion picture shot being printed to print stock. That's not to say with DI's you don't have a ginormus new set of possibilities and options (like for still photographers with Photoshop), but again... I'm just talking photochemically.
  • 0

#7 Dominic Case

Dominic Case
  • Basic Members
  • PipPipPipPip
  • 1357 posts
  • Other
  • Sydney Australia

Posted 31 October 2006 - 02:04 AM

Getting back to the original question . . .

One stop is one stop. F or T, if you change by one stop you are doubling or halving the amount of light that reaches the film.

F stops are calculated by the manufacturer as the ratio of lens diameter to focal length. The result tells you about the theoretical brightness of the image that is formed, and also about the depth of field.

In practice, the image brightness is also affected by light loss due to internal reflections etc. No two lenses are the same so there is no constant conversion from F to T stops.

The T stop of a lens is the same as the F stop would be if the lens transmitted 100% of the light it was meant to. Usually it's a higher number (ie less light gets through). So it's a practical measure that tells you about exposure.

If your exposure meter says 3.5, then t 3.5 on any lens will give the correct exposure. F 3.5 will vary a bit from lens to lens.

If your depth of field table (for a given focal length) says 8, then f 8 will be right - for any lens of that focal length. T 8 will be a bit less depth of field, not the same for every lens.
  • 0

#8 Mark Dunn

Mark Dunn
  • Basic Members
  • PipPipPipPip
  • 2422 posts
  • Other
  • London

Posted 31 October 2006 - 04:20 AM

That's what I meant to say.
  • 0

#9 Michael Collier

Michael Collier
  • Basic Members
  • PipPipPipPip
  • 1262 posts
  • Gaffer
  • Los Angeles, CA

Posted 31 October 2006 - 04:56 AM

Getting back to the original question . . .

F stops are calculated by the manufacturer as the ratio of lens diameter to focal length. The result tells you about the theoretical brightness of the image that is formed, and also about the depth of field.


I believe that was a typeo (dominic as everyone knows is very smart, smarter than I), but I am quite sure that F-stops are calculated as a ratio of the lens diameter to the apeture size. Otherwise, your F-stop would change as you zoom, not as you iris down.
  • 0

#10 Mark Dunn

Mark Dunn
  • Basic Members
  • PipPipPipPip
  • 2422 posts
  • Other
  • London

Posted 31 October 2006 - 07:21 AM

It's the ratio of focal length to aperture diameter. So, at its simplest, a 100mm. lens with the iris closed down to 40mm. is at f2.5 . If its maximum aperture is 50mm., it will have f2 on the barrel.
Some cheaper zoom lenses for stills do change in focal ratio- it increases at the long end. Presumably this is the design constraint on cine zoom lenses, where this wouldn't be acceptable.
  • 0

#11 Charles MacDonald

Charles MacDonald
  • Sustaining Members
  • 1157 posts
  • Other
  • Stittsville Ontario Canada

Posted 31 October 2006 - 10:17 PM

It's the ratio of focal length to aperture diameter. So, at its simplest, a 100mm. lens with the iris closed down to 40mm. is at f2.5 . If its maximum aperture is 50mm., it will have f2 on the barrel.
Some cheaper zoom lenses for stills do change in focal ratio- it increases at the long end. Presumably this is the design constraint on cine zoom lenses, where this wouldn't be acceptable.

Their are tricks that Zoom designers use, some put the stops in a place in the optical path where the ratio is constant, others use cams to change the appature. Some zooms for still cameras ignore the change and let the behind the lens meter take care of it.

There are often 12 or more elements in a zoom, and their are points that the appature can be placed that the zooming does not affect the light transmission.

Cine Zooms have to take care of all that and Not "breathe" as you zoom. That is why they are expensive, and because the compesation is hard to work right, they are often seen as not as good as a prime.
  • 0

#12 Mike Rizos

Mike Rizos
  • Basic Members
  • PipPipPipPip
  • 330 posts
  • Other

Posted 31 October 2006 - 11:14 PM

It's the ratio of focal length to aperture diameter. So, at its simplest, a 100mm. lens with the iris closed down to 40mm. is at f2.5 . If its maximum aperture is 50mm., it will have f2 on the barrel.



This would only be true if the iris was located in front of the lens.
To find the f-stop the lens focal length is divided by the effective aperture. The effective aperture is usualy not the same as the physical size of the iris opening, because a lens will bend light rays as they enter it.
Effective aperture is defined as the diameter of the entering beam of light that will just fill the iris or diaphragm. Relative aperture is another name for f-stop.
To caclulate depth of field use f-stops, exposure T-stops.

Question: So T is for trasmission surely. What does f in f-stop mean? I've heard focal length, but that doesn't make much sense. I've also heard it may refer to (f)unction.
  • 0

#13 Dominic Case

Dominic Case
  • Basic Members
  • PipPipPipPip
  • 1357 posts
  • Other
  • Sydney Australia

Posted 01 November 2006 - 02:28 AM

I believe that was a typeo

Not really a typo (typeo?) more of an oversimplification. I knew what I meant, I just didn't write it right :rolleyes: My error :o

Yes, of course you are right, it's the effective aperture diameter that determines the f stop. But that's really quite a difficult thing to measure except in a simple single-element lens. It's not necessarily the actual diameter of the aperture diaphragm - as Mike has explained.

What does f in f-stop mean? I've heard focal length but that doesn't make much sense.


Traditionally it's written not as f-stop but f/stop. That's the clue. Obscure historical attempt at explanation coming up . . .

The way I think of it is that if f is the focal length and d is the effective aperture diameter, then the "aperture (a)" is given by a=f/d. That's the mathematical form of the expression we've quoted. Flip it over and you get d=f/a.

So for any lens to get a certain exposure, you need to feed the focal length into the equation and you can deduce the diameter you need to set the stop to. (Not at all practical for a modern compound lens with an internal iris, but very useful in the nineteenth century when you inserted a Waterhouse Stop to control the exposure.

The point of all this? The particular size Waterhouse stop you need is given by f/a. If the required exposure is a setting of 3.5, then you need a stop of diameter f/3.5 (focal length of lens divided by 3.5).

The precise usage (and the link between the formula and the exact words used) has become sloppy. But clearly the "f" derives from "focal length" which is invaribly used as "f/some number".
  • 0

#14 Sal Caldecutt

Sal Caldecutt

    New

  • Basic Members
  • Pip
  • 5 posts
  • Other

Posted 01 November 2006 - 08:52 AM

Cheers for all the replies, for the original question what I meant to say was - If my light meter reads say f/4 what would my T-stop be? As light is lost reaching the film plane due to the glass elements in the lens, surely the aperture would need to opened wider than T-4? If so how much?
  • 0

#15 Nathan Milford

Nathan Milford
  • Sustaining Members
  • 692 posts
  • Director
  • New York, NY

Posted 01 November 2006 - 09:26 AM

Cheers for all the replies, for the original question what I meant to say was - If my light meter reads say f/4 what would my T-stop be? As light is lost reaching the film plane due to the glass elements in the lens, surely the aperture would need to opened wider than T-4? If so how much?


There is no set F/stop to T/stop differential. There is no set compensation.

To recap: An f-stop is mathematical (focal length \ iris diameter, i.e. 50mm lens with a
25mm iris is F/2.0) and a T/stop is an actual measurement of the light transmission of the lens.

Different lenses have different designs using a different amount of elements, air-to-glass surfaces, differnet coatings etc... Thier efficiencies differ. The T/stop takes in to account, were there no elements to disrupt light transmission (read: pinhole), what the proper amount of light an F-number should allow.

When your light meter say's F/4.0, set the lens to T/4.0. It means the same thing for all intents in purposes. By using the T/stop ring on the lens you're choosing to let the actual amount of light through that should be transmitted by a mathmatical F/stop in a perfect lens. You're compensating for the lens design, just a smidge, and that all there is to it.
  • 0

#16 John Adderley

John Adderley
  • Guests

Posted 01 November 2006 - 12:28 PM

>[quote name='Nathan Milford' date='Nov 1 2006, 02:26 PM' post='135734']
>There is no set F/stop to T/stop differential. There is no set compensation.

I have a 9 page official explanation of T stop calibration by Peter Merigold, Lens Designer TTH Ltd. (signed with a personal dedicaction no less) It explains everything with bells and whistles. I interviewed Peter about two years ago about his long association with TTH and how he worked through the whole period from ray tracing with human 'computers' to modern day electronic computers. TTH Cooke were the first commercial people in UK to embrace the new computer technology for industrial design.

I can scan the document and put it on my web site if anyone would care to see it.

John Adderley DOP UK
john@adderley.net
www.adderley.net
  • 0

#17 Sal Caldecutt

Sal Caldecutt

    New

  • Basic Members
  • Pip
  • 5 posts
  • Other

Posted 01 November 2006 - 06:10 PM

Thanks for the explanation, I understand it now. I?m interested to see the official Peter Merigold explanation of T-stop calibration too, cheers for all the info.
  • 0

#18 John Adderley

John Adderley
  • Guests

Posted 06 November 2006 - 02:57 PM

Im new to posting on this forum, I wondered if anyone could tell me the formular for converting f-stops to T-stops? I carnt seem to find this out anywhere else on the website, Cheers


I have now posted the treatise on T-Stops by Peter Merigold. This is the definitive paper from Taylor Hobson.

Please look on the CONTACTS AND ADDRESSES page of my web site for the link. http://www.adderley.net Or if you would rather skip my web site then go to http://www.adderleykine.net/

I hope you find this helpful and maybe I should put this posting on the General section, any thoughts on that?

Don't forget in UK 2030 BBC4 Friday Night, Nation on Film... excellent series!

Edited by John Adderley, 06 November 2006 - 02:59 PM.

  • 0


Media Blackout - Custom Cables and AKS

Paralinx LLC

Aerial Filmworks

Ritter Battery

Tai Audio

The Slider

Metropolis Post

rebotnix Technologies

CineTape

Glidecam

Abel Cine

Wooden Camera

Broadcast Solutions Inc

CineLab

Willys Widgets

FJS International, LLC

Gamma Ray Digital Inc

Technodolly

Opal

Visual Products

Rig Wheels Passport

Willys Widgets

Technodolly

Abel Cine

Ritter Battery

Tai Audio

Media Blackout - Custom Cables and AKS

The Slider

CineLab

Metropolis Post

CineTape

FJS International, LLC

Aerial Filmworks

Broadcast Solutions Inc

Glidecam

Visual Products

Opal

rebotnix Technologies

Paralinx LLC

Rig Wheels Passport

Gamma Ray Digital Inc

Wooden Camera