The GetDPI Photography Forum

Great to see you here. Join our insightful photographic forum today and start tapping into a huge wealth of photographic knowledge. Completing our simple registration process will allow you to gain access to exclusive content, add your own topics and posts, share your work and connect with other members through your own private inbox! And don’t forget to say hi!

Question about c-mount vs M42 lens luminosity on 4/3 sensors.

J

jlduh

Guest
Hi, i don't post often here but read regurlaly what's going on in this section ;)

I've used many different manual lenses on my G1 and i'm facing a strange phenomenon with a Fujinon 50mm 1,4 TV lens (in c-mount): when i compare the test shots i get from it and the ones from my other Super Multicoated Takumar 50mm 1,4 in M42, i get speeds 25% higher with the fujinon c-mount lens (i precise that i've done this test in spot metering in a précise area of the image for light measurement where light was constant and uniform).

How can i interpret this? I was wondering if the fact that M42 lens were illuminating an image circle with a size almost double than the one produced by a c-mount lens could explain this "loss", a part of the image and of the light being "eaten" by things "around" the 4/3 sensor whereas the c-mount lens illuminates almost only the sensor?

An other way to interpret this would be to say that when we mount an M42 lens on a micro 4/3 mount, the adapter is much longer than the adapter for the c-mount which is very thin and that this extra-length is like an extension tube, eating light as we see it when a macro extension tube is used... What do you think?

At the end, on a micro 4/3, did other members of the forum saw c-mount lenses giving speeds 25% or so higher than with "big" lenses (lenses for full frame)?

Thanks for reading and maybe... answering! :thumbup:
 

RichA

New member
Hi, i don't post often here but read regurlaly what's going on in this section ;)

I've used many different manual lenses on my G1 and i'm facing a strange phenomenon with a Fujinon 50mm 1,4 TV lens (in c-mount): when i compare the test shots i get from it and the ones from my other Super Multicoated Takumar 50mm 1,4 in M42, i get speeds 25% higher with the fujinon c-mount lens (i precise that i've done this test in spot metering in a précise area of the image for light measurement where light was constant and uniform).

How can i interpret this? I was wondering if the fact that M42 lens were illuminating an image circle with a size almost double than the one produced by a c-mount lens could explain this "loss", a part of the image and of the light being "eaten" by things "around" the 4/3 sensor whereas the c-mount lens illuminates almost only the sensor?

An other way to interpret this would be to say that when we mount an M42 lens on a micro 4/3 mount, the adapter is much longer than the adapter for the c-mount which is very thin and that this extra-length is like an extension tube, eating light as we see it when a macro extension tube is used... What do you think?

At the end, on a micro 4/3, did other members of the forum saw c-mount lenses giving speeds 25% or so higher than with "big" lenses (lenses for full frame)?

Thanks for reading and maybe... answering! :thumbup:
Likely it is due to the throughput from the lenses, glass thickness and coating effectiveness. Takumars may have been (some of them) super multicoated, but multicoated in the 1960s and 1970's isn't the same as now, or likely as good as the Fuji coatings. Also, I don't think the inverse square law has any effect since it's the scene's brightness and distance that matters and not the distance from the back of the lens elements. I've found some Takumars give a yellowish image, which means there is some filtering going on.
 

kds315

Active member
I would have also mentioned lens design (how many individual lenses light has to pass with different glass transmission plus esp. air-glass surfaces) and the quality of the coating. Have a look at older cine lenses giving f-stop and T-values. T stands for transmission, so a f1.9 and T2.3 lens actually means 17% transmission losses. This is NOT given on ordinary photo lenses but for cinematography it is essential to know, when changing lenses to get correct exposure.
 
J

jlduh

Guest
thanks for the answers...

I want to add that it is a Super Multi Coated Tak (so a late one) and it's not too yellow in comparison with many Super TAks i have seen. But there is a small faint yellow cast, yes... Does it causes such a difference, that's the question.

Did other people tried already to make comparisons between C mount and 35mm image circle lenses on this matter?

In my understanding of optics, rays coming from the center or the lens have a better chance to go through the lens without reflexion/absorption than the ones from the outside (think about the different kinds of vignetting/fall off) so i could imagine that from this stand point, the 35mm lens would have a better transmission of "effective" light (the one hitting the 4/3 sensor) than the C-mount one on the same 3/4 sensor... But no, in this case that's the opposite!

On almost the same subject, i discovered by testing that some high end teleconverters (namely the Nikon TC-E15ED AND TC-E17ED, the best ones on the market in this category) had an interesting behaviour depending on the lenses on which they are mounted. When i put them on a Lumix lens (micro 4/3 45-200mm Lumix) i get a loss of speed of around 1/3 dipah for the x1,5 and around 2/3 diaph for the X0,7. That's pretty small when you know that TC's for reflex (put between lens ans camera and not in front of lens like the E15 and E17 téléconv) are known to create a 1 diaph loss for X1,4 and 2 diaph loss for 2X... When i replaced TCE17ED by TCON 17 on the 45-200mm Lumix, i got slightly higher losses than with the previous one -which is top class quality i have to say- but still it was around 1 diaph only. That's this observation that made me think that lengthening the back of a lens would cause light loss just like with macro rings (and i think it is hte case anyway right?). Some people can investigate if they get the same results by the way?

But i also discovered that repeating the same experience with E15 and E17 teleconverters on some 35mm manual focus 200mm/300mm focal length lenses was giving me a significantly higher losses of light than the ones i got on the micro 4/3 lenses (up to twice as big depending on the lenses)! Just like if the light captured by the teleconverters was "converged" towards the sensor on micro 4/3 lens and dispersed on a bigger surface around the sensor on the 35mm lens...

Did other people already tested or observed this?

That's why with the fujinon/takumar comparison i was thinking of "something" specific to this difference in images circles...
 
Last edited:

Pat Donnelly

New member
Welcome JLDuh!

Very interesting point.

I believe you may have a point about the distance from the sensor!

Also, when using a 50mm 1.4 OM lens on my 4/3 (E-410)I noticed that closing down a stop did not always mean a stop less in speed, particularly at the wide end. I may repeat this with the m4/3 (EP-1).
 
C

ChrisJ

Guest
I thought f8 (for instance) is f8 and, regardless of the manufacturer or focal length would give the same level of light on every square mm of sensor/film.

Obviously the yellowing in the coating of Takuma lenses will make the lens 'less transparent' so the careful calculations of Pentax when they originally designed the lens will be off, so f8 might not be f8 anymore in that instance.

The reason why the light level diminishes when you fit extension rings is you are looking at a smaller area of subject so there is less area reflecting light into the lens, same for the teleconverters, the area of subject on a 2X converter is only a quarter (half the height X half the width) of the area hence the 2 stop drop.

Chris
 
Last edited:

Godfrey

Well-known member
I thought f8 (for instance) is f8 and, regardless of the manufacturer or focal length would give the same level of light on every square mm of sensor/film.
An f/number is a single index normalization to indicate light transmission based on geometry which is calculated by focal length divided by physical lens opening. Given all else being equal, f/X allows the lens to transmit the same amount of light regardless of focal length.

BUT, all else is rarely exactly equal. Focal length often varies with focus distance, depending on lens design. Transmissive qualities of the optical glasses differ depending on glasses/materials, coatings and lens design. So f/numbers (f-stops) are really just an approximation, close enough for most purposes in still photography.

T-stop values take into account the additional effects of these modifiers to light transmission. Cine lenses were calibrated in T-stops because variations in exposure with turret mounted lenses, used often in cine cameras, made exposure more critical than for still work.

A more extensive discussion of f/numbers and T-Stops is on Wikipedia. It's reasonably accurate:
http://en.wikipedia.org/wiki/F-number
 
C

ChrisJ

Guest
An f/number is a single index normalization to indicate light transmission based on geometry which is calculated by focal length divided by physical lens opening. Given all else being equal, f/X allows the lens to transmit the same amount of light regardless of focal length.

BUT, all else is rarely exactly equal. Focal length often varies with focus distance, depending on lens design. Transmissive qualities of the optical glasses differ depending on glasses/materials, coatings and lens design. So f/numbers (f-stops) are really just an approximation, close enough for most purposes in still photography.
[/url]
Following your comment I've just checked some of my lens collection on my G1, they all gave the same shutter speed when the lenses were set at the same aperture, using the artificial light in my lounge, not exactly scientific I know. May be I am just lucky.

All the varables you list are known by the lens designer, except in the case of the Takumar lens where the optical properties of the glass alter over time I have never had any problem with any lens I've fitted to my G1 which is a pretty wide range, even on zoom lenses where the dimensions of the aperture change with the varying focal length are usually accurate to 1/3rd of a stop or better and that is probably accounted for by mechanical wear, new zoom lenses perform better than that.

T stops was an older system which only applied to fixed focal length turret mounted lenses as far as I can remember, zoom lenses were always calibrated in f stops and would keep their setting over their focal range. Mind you the memory is not what it used to be. LOL

Chris
 

Godfrey

Well-known member
Following your comment I've just checked some of my lens collection on my G1, they all gave the same shutter speed when the lenses were set at the same aperture, using the artificial light in my lounge, not exactly scientific I know. May be I am just lucky.

All the varables you list are known by the lens designer, except in the case of the Takumar lens where the optical properties of the glass alter over time I have never had any problem with any lens I've fitted to my G1 which is a pretty wide range, even on zoom lenses where the dimensions of the aperture change with the varying focal length are usually accurate to 1/3rd of a stop or better and that is probably accounted for by mechanical wear, new zoom lenses perform better than that.

T stops was an older system which only applied to fixed focal length turret mounted lenses as far as I can remember, zoom lenses were always calibrated in f stops and would keep their setting over their focal range. Mind you the memory is not what it used to be. LOL

Chris
You misunderstand. f/stops may be only approximately equivalent lens to lens, but that "approximately" has, for most purposes, very small practical variation.

T-stops apply to any lens system where critical, matched exposure is essential between a set of lenses, it isn't some "older system". When zoom lenses and TTL metering became the norm, such precision became less important, so few if any manufacturers continue to supply T-stop settings.

In my experience with the G1 and Micro-FourThirds or FourThirds lenses I find that frame to frame variations in exposure across lenses and even with the same lens at a given f/stop setting with controlled lighting run about 1/6 stop. This is perfectly acceptable for still work, and comparable to what I used to see with Nikon, Canon, Olympus, Pentax and Leica film cameras. It would waste thousands of dollars worth of film on a cine camera with three prime lenses on a turret.
 
Top