a quarter of a stop faster.... difficult or gimmick?

K

kross

sonnarism
Local time
12:47 AM
Joined
Jan 12, 2007
Messages
168
Location
Atlantis Island
View My Gallery
i would like to understand the limitation, if any, that is behind all these little
"increments" in achieving faster light transmission for rf lenses.

leica "increase" the noct speed from f1.1 to f1.0 then f0.95. why can't they
just go for f0.7 or something faster.... i would think this is more of a business
strategy than a physical limitation so that they have room to sell f0.9, f0.85,
f0.8 and so on at a later time. 😛

as for cv, why f1.1 and not f1.0? they will most probably release a f1.0 say in
5yrs time or thereabout...

F0.X, is there a real difference? it would be interesting to put the nok f1.1 and
noct f1.0 up for a light transmission test.

just some thoughts to spice things up....😀
 
Well, since the size increase it's logaritmic (I hope it's correct 😀) the lens become really bigger every step.
And, having so little DOF at f1.0, I can't find the use of a f0.7 lens...
 
The practical limit is the size of the hole in the camera body -- which is why the Noctilux is f/0.95 instead of f/0.9. If they could have gone faster, they would. This also explains, I suspect, why the new lens is 'only' f/1.1 and comes under the Voigtländer label instead of Zeiss. Zeiss tends to make 'landmark' lenses in Germany, and I suspect that (like Leica) they can't make a faster lens than f/0.95 that meets their standards, so a fast ZI woukd have been seen as 'me-too'.

Also, the faster the lens, the harder it is to control aberrations, illumination, vignetting, focus shift, etc. Although there have been VERY fast lenses -- I seem to recall f/0.7 as a limit -- they have normally been corrected for one wavelength at one distance, e.g. X-ray screen relay lenses.

No, f/1.1 is not really significantly faster than f/1.2, but again it's a slippery slope: the extra half-stop from f/1.4 to f/1.2 isn't that much either...

Basically, if you want/need the speed, there's very little choice (and always has been) and the new 50/1.1 is astonishingly affordable. I'd be very surprised, too, it it were not a vast improvement over the old Canon f/1.2.

Tashi delek,

R.
 
Last edited:
The new CV Nokton might be also limited to f/1.1 due to patents hold by Leica on the Noctilux. Otherwise, every small step towards larger apertures means dealing with a lot of optical and engineering problems.

Also, when Dr. Mandler designed the Noctilux 50/1.0 without aspherical elements 40 years ago, optical calculations took a lot more time, optical coatings were more difficult to make, and CNC milling didn't exist.
 
If the throat diameter of a Leica M is 44mm (http://www.photoethnography.com/ClassicCameras/index-frameset.html?Lens-LSM.html~mainFrame) then the fastest possible lens would be 0.88 (assuming none of the 44mm is needed for rf cams etc (and that I am doing my maths right...)

The new CV Nokton might be also limited to f/1.1 due to patents hold by Leica on the Noctilux. Otherwise, every small step towards larger apertures means dealing with a lot of optical and engineering problems.
Click to expand...

I'm not sure if patents would have anything to do with it, as they would have expired by now. There were 35mm camera lenses @ 1.1 long before the Noctilux, so I can't imagine Leica had any fundamental patents necessary for making this kind of lens.

But I can well imagine that it becomes a lot more expensive to make even small improvements in lens speed at this level.
 
I dont think anyone really needs the 1.1 compared to 1.2 in terms of shooting at dark. But there might be some difference when comparing the dof. Especially when comparing 1.4 or 1.5 to 1.1 there is an obvious change with the depth of field you get when shooting wide open...

And this is what a tripod or stable hands cant help at all.
 
historicist said:
If the throat diameter of a Leica M is 44mm (http://www.photoethnography.com/ClassicCameras/index-frameset.html?Lens-LSM.html~mainFrame) then the fastest possible lens would be 0.88 (assuming none of the 44mm is needed for rf cams etc (and that I am doing my maths right...)



I'm not sure if patents would have anything to do with it, as they would have expired by now. There were 35mm camera lenses @ 1.1 long before the Noctilux, so I can't imagine Leica had any fundamental patents necessary for making this kind of lens.

But I can well imagine that it becomes a lot more expensive to make even small improvements in lens speed at this level.
Click to expand...

I think you may be doing the maths backward i.e. 0.88:1 as 1:0.88 and I’m unsure the throat size has anything at the maximum aperture, I had a OM 180 f2 at one time and it didn't need a 90mm lans mount, do we have a expert?

They couldn’t patent an aspect ratio, if one could some enterprising chap would have registered Pi years ago

🙂
 
Last edited:
F1.1 vs F1.2 is about the difference between 1/125 and 1/100. Whatever is moving moves less and produces less blur in the Image. And between 1/15 and 1/12th. You are at the limits of holding a camera steady, shake will be less.

I've handles Kiu's 5cm F1.1 Nikkor and my Canon 50/0.95 at the same time. My lens was about twice as heavy as the Nikkor. It had a LOT more glass in it. Wider, and thicker. It's not easy suck in all that light and get all of it to fall on one flat surface. I've watched over the shoulders of optical engineers running the models and having lenses designed. When they had the metal of the lens barrel formulated to account for change in temperature and the resulting change in the optical path, you knew it was not easy.
 
Sparrow said:
I think you may be doing the maths backward i.e. 0.88:1 as 1:0.88 and I’m unsure the throat size has anything at the maximum aperture, I had a OM 180 f2 at one time and it didn't need a 90mm lans mount, do we have a expert?

They couldn’t patent an aspect ratio, if one could some enterprising chap would have registered Pi years ago

🙂
Click to expand...

Dear Stewart,

Throat size, flange to film (or sensor) distance, focal length and maximum aperture are all intimately related. The logic behind f/1 or close to it as the maximum aperture was related to me independently by Zeiss and Leica at photokina, and one of the people who explained it made a very convincing demonstration. Take a fast lens (f/1.5 or larger) and look at it from the front and the back. If there's a big difference in the apparent size of the hole, the lens is a long way from symmetry and harder to correct.

Long, fast lenses are easier to design and build, hence the popularity of 58mm (f/2 Biotar, f/1.4 Nikkor, etc,) and 55mm. But they're also bigger and heavier and not well suited to rangefinders with fixed 50mm frames.

Cheers,

Roger
 
Seen decades ago, over the entrance to a butcher shop in Germany:

a small video camera with a f/0.75 lens, to trigger the automatic sliding door and give a stamp size image on a stamp size screen inside ...

Improve sensor/film sensitivity, without creating more grain/noise I would say is the answer to razor thin DOF at f/0.9, f/0.7 or below.
 
The f1.1 number might be due to Cosina's engineering philosophy:
http://keppler.popphoto.com/blog/2006/05/the_contrary_mi.html#more
"Usually we first design the very best possible lens, regardless of glass price," explains Kobayashi. "Then we try to substitute less expensive elements wherever possible without noticeably affecting quality. We stop when we have lowered production costs sufficiently, but have retained quality, and where the difference from our original lens will be negligible to the user."
Click to expand...
This is just a guess, but maybe Cosina was aiming for an f1.0 lens, but after the "cheapening" phase, it turned out to be an f1.1.
 
Throat size, flange to film (or sensor) distance, focal length and maximum aperture are all intimately related. The logic behind f/1 or close to it as the maximum aperture was related to me independently by Zeiss and Leica at photokina, and one of the people who explained it made a very convincing demonstration. Take a fast lens (f/1.5 or larger) and look at it from the front and the back. If there's a big difference in the apparent size of the hole, the lens is a long way from symmetry and harder to correct.
Click to expand...

OK, so the pupil can be bigger than the throat size itself but it just gets harder to do when it is less symetrical? I'm way out of my depth here but it seems to make sense, I guess a long lens like a 200 f2 would be more symetrical than a 50 f1.
 
Roger Hicks said:
Dear Stewart,

Throat size, flange to film (or sensor) distance, focal length and maximum aperture are all intimately related. The logic behind f/1 or close to it as the maximum aperture was related to me independently by Zeiss and Leica at photokina, and one of the people who explained it made a very convincing demonstration. Take a fast lens (f/1.5 or larger) and look at it from the front and the back. If there's a big difference in the apparent size of the hole, the lens is a long way from symmetry and harder to correct.

Long, fast lenses are easier to design and build, hence the popularity of 58mm (f/2 Biotar, f/1.4 Nikkor, etc,) and 55mm. But they're also bigger and heavier and not well suited to rangefinders with fixed 50mm frames.

Cheers,

Roger
Click to expand...

And you were convinced?

How did they explain the Zeiss Planar 50mm f/0.7?


http://www.luciolepri.it/lc2/marcocavina/articoli_fotografici/Planar_50_0,7_ancetres/26.gif
 
It's like in racing, every incremental increase in speed requires a magnitude increase in costs. It's not just making a bigger hole, it's also dealing with all the abberations that come with it.
 
Most of these super-fast F0.7 (and such) lenses have extremely small backfocus, and are special purpose lenses. CRT and X-ray film applications. I have a Rodenstock 42mm F0.75 and a couple of lenses similar in design.

Now- If Cosina had wanted to do an F1 lens, they would have. If they wanted the Nokton to be an F1.4 lens, they would have. I suspect that Cosina made the Nokton an F1.5 lens and the Super-Nokton an F1.1 lens because no other company has brought these max-apertures out in over 40 years. Nostalgia.
 
The theoretical max. aperture is f0.5, two full stops faster than f1.0
The fastest lens for photographic work is the mentioned f0.7 Planar but it only covers 18x24mm.
The problem is the performance with faster lenses, there have been many fast designs but most were used for their distinctive (I don't even mean thin DoF) look, not real performance.
Some optical problems enhance by the factor four when you use one stop faster aperture, some even more! So going from f1.4 to f1.2 to f1.1 to f1.0 and even f0.95 might not be as easy as it sounds, if you really want a usable performance...
I would love the possibilites of the new Noct, but my Summiux Asph was already expensive enough... ;-)
Fast lenses are great, especially with SLRs and their "look" and shallow DoF cannot be imitated, even when someday sensors will be fast enough already...

f-stop is just simple math, the ratio between the diameter of the entrance pupil and focal length, by the way, the front lens diameter has to be at least as big as the entrance pupil = 2/200mm = >100mm front lens diameter.
T-stop is the "real" speed of a lens, used in motion picture lenses and a bit slower than f-stop (T1.3 instead f1.2) because it takes the actual amount of light into account (including reflections).
 
You must log in or register to reply here.
Back
Top Bottom