Jump to content

Speed - why go for fast lenses?


pgk

Recommended Posts

Advertisement (gone after registration)

A comment in another thread got me thinking. The poster seemed to suggest that the fastest aperture was not an indication of overall performance.

 

From my recollections I seem to remember that the old rule of thumb was that a lens performed 'best' 2~3 stops below maximum aperture, but that a fast lens was essential for low light use. Now things have changed, at least when I look at how I shoot I consider that they have.

 

Firstly I'd say that looking through the MTF graphs supplied by Leica it does seem that the optimum graphs coincide with 2~3 stops below maximum aperture - nothing's changed here then. But personally, I now tend to use fast apertures for their depth of field characteristics as opposed to their abilities to operate in low light (although to be fair, fast apertures do still extend camera capabilities in very low light).

 

I have not really considered the optimum performance at mid apertures when buying a lens, although I am aware of the stunning performance of the 80/1.4R at f/8 and I do use this lens at mid apertures more than anticipated (can anyone comment on the 75/1.4M's abilities at mid apertures?).

 

So I'd say that I'd buy a fast lens for its 'isolating' power - the ability to seperate subject and background - and 'bokeh' rather that the low ligh applications which might have been my motivation in the past. Is this the experience of others or are fast aperture lenses still bought and used because of their mid aperture perfomance?

Link to post
Share on other sites

Leica fast lenses not only allow shallow depth of field but also provide another dimension to control aberration at edge and corner area, thus allow you to make object at center more outstanding at will. The degree of sharpness at edge and corner can be controlled and varied by changing aperture value.

 

Whereas modern lenses has full performance at full aperture and sharpness over the whole field, this type of control no longer exists. Besides, the transition from in focus to out focus is small with modern lenses and will affect bokeh as well.

Edited by amoebahydra
Link to post
Share on other sites

Hmm. I have never thought it like that. That lenses that had bigger maximum apertures also had bigger optimum apertures, I mean. Is this really true? If it is, are the smaller apertures of the slower lenses therefore better than those of the fast ones, respectively?

Link to post
Share on other sites

Hmm. I have never thought it like that. That lenses that had bigger maximum apertures also had bigger optimum apertures, I mean. Is this really true? If it is, are the smaller apertures of the slower lenses therefore better than those of the fast ones, respectively?

 

Not necessarily, some modern lenses already have optimal performance at full aperture and can hardly improve by stopping down.

 

It is always easier to design and fabricate lenses with smaller aperture as it is easier to control the light flux. As such, if you don't need the extra stop then use slower lenses. I always use Summicrons intead of Summiluxes.

Link to post
Share on other sites

I think there isn't a general rule about... at smaller apertures the "worsening" is due to diffraction effect, which in turn, if I remember well, depends on the absolute diameter of the iris (so that, for instance, a 28mm suffers diffraction more than a 90 at f16).

Personally, I remeber only a test I made with 50s Summilux (pre asph) and Elmar (red scale) both at f16 : Summilux was slightly sharper, and the same for Summilux 35 and Summaron 35 3,5, always at f16.

Link to post
Share on other sites

Advertisement (gone after registration)

Luigi---

it is true that diffraction occurs when light passes close to an opaque edge, such as a line in a diffraction grating, or (more relevant here) a diaphragm blade. The smaller the aperture becomes. the lower is the ratio of area to circumference, so that a larger proportion of the light suffers diffraction. So, if we measure diffraction at a fixed, small distance behind the aperture, a 2mm opening will give rise to more diffraction than, say, a 20mm aperture.

 

By that reasoning, wide angle lenses should be more diffraction-prone than longer lenses, because e.g. a 1:5.6 aperture opening is smaller in a 21mm lens than in a 90mm lens. But ...

 

Diffraction means that light is deflected from its proper path. And the further it goes, the further does it stray. Light has further to go on its way between the iris and the sensor when you mount a long lens, than when you use a short one. So in practice, the 'longer path effect' balances the 'larger hole effect', and diffraction tends to be a function of relative aperture (f-stop) alone. Most current Leica lenses tend to lose some brilliance to diffraction from f:8 on. In older lenses, and in lesser brands, the effect may not strike us until later, because there are more residual aberrations, and the continued reduction of these, as you stop down to 11--16--22--32--45--f:64 ... masks the increase in diffraction. When I was a kid, most lenses were so badly corrected that you could forget about diffraction.

 

The old man from the Age of Chromatic Aberration

Link to post
Share on other sites

Thanks for the clear explanation Lars... I wasn't indeed sure of the diffraction-to-focal length relation; I wonder... think of the long Telyts (400/560)... their iris is positioned a lot recessed in respect to the 2 elements lens unit (much nearer to the focus plane) : does this have a positive effect on diffraction ?

Link to post
Share on other sites

Interesting explanation of diffraction, Lars.

 

Most Leica lenses are excellent performers wide open, and that extra speed can make a real difference. Bokeh is far more apparent with standard to longer lenses, 50 and up -- another reason, apart from perspective, that they are better for portraits.

 

But small apertures can help with DOF, for example when you want take a picture of a building with something in the foreground, a flower, for instance, and you want both sharp.

 

I used to find it surprising that some photogs recommend stopping down as far as possible for DOF when it isn't really necessary -- landscape at infinity. Picture would be far sharper with a wider aperture and faster shutter speed.

 

A big advantage of Leica Ms is you can have BOTH a slow shutter speed and shoot wide open with a medium ISO and still get sharp pictures.

Welcome, dear visitor! As registered member you'd see an image here…

Simply register for free here – We are always happy to welcome new members!

Link to post
Share on other sites

I used to find it surprising that some photogs recommend stopping down as far as possible for DOF when it isn't really necessary -- landscape at infinity. Picture would be far sharper with a wider aperture and faster shutter speed.

 

Yeah... I've taken some landscapes with the 35 Lux stopped down too far, and the results weren't nearly as sharp as I'd hoped because of the shutter speed.

Link to post
Share on other sites

There are of course times when only a very fast lens cuts the mustard. And pictures that are equally sharp all over are as a rule quite boring. Selective sharpness and unsharpness is one of the main ways we have to direct the attention of onlookers to the things we want to point out. That does not mean that everything else has to be blurred out of recognition. Attention can be directed by quite subtle means at times. And a boring picture does not become less boring just by being very sharp. Conversely, an arresting image does not always have to be totally sharp!

 

This said, you do often need a bit of depth of field--however you define it. The important thing is to use it deliberately, in a conscious and purposeful fashion.

 

Then, a bit of nitpicking. 'Bokeh' does NOT mean simply 'out of focus'. It means 'the quality of that part of the image that is out of focus'. And 'quality' does not simply mean 'good' or 'bad'. It means simply that out of focus objects can look different. What happens does to a very great extent depend on spherical aberration, corrected, undercorrected or overcorrected, because this determines how point highlights look. And if spherical is not perfectly corrected, they will look different in front of and behind the plane of focus, respectively.

 

Modern Leica lenses are usually very well corrected for spherical, so that they exhibit a smoother bokeh than older designs. There, you have it! When people say that modern aspherical lenses have 'harsh bokeh', then they are not speaking of bokeh at all. They simply mean that they find sharply focused pictures vulgar.

 

The old man from the Age of Pictorialism

Link to post
Share on other sites

Thanks for the clear explanation Lars... I wasn't indeed sure of the diffraction-to-focal length relation; I wonder... think of the long Telyts (400/560)... their iris is positioned a lot recessed in respect to the 2 elements lens unit (much nearer to the focus plane) : does this have a positive effect on diffraction ?

 

I'm fairly the position of the iris (as well as the f/stop) affects certain aberrations, but I can't remember which, can't find anything on Google and don't have an optics textbook here to check.:confused:

Link to post
Share on other sites

Thanks for the clear explanation Lars... I wasn't indeed sure of the diffraction-to-focal length relation; I wonder... think of the long Telyts (400/560)... their iris is positioned a lot recessed in respect to the 2 elements lens unit (much nearer to the focus plane) : does this have a positive effect on diffraction ?

The level of diffraction is basically the size of the Airy disk in the image plane and in the ideal case is only a function of the aperture, not of the focal length. The diffraction tells you how 'sharp' the optimally focussed part of the image is. The maximum achieveable sharpness increases as the difraghm gets larger, but this is assuming the abberations etc. are properly corrected. In practice the largest apertures will suffer from lens imperfections more, hence the "2-stops" rule of thumb for optimal sharpness of the image plane. In practice the modern Leica lenses are pretty well corrected when full open so the 2-stops rule is somewhat of a relic from the past.

 

Consequence: for diffraction the location of the diaphragm makes no difference.

 

The location of the diaphragm does influence the out-of-focus rendering of the lens (AKA the bokeh), where you can have "sharp" or "doughnut" or "fuzzy" or "whirly" & a whole pile of other flavors. This is not related to diffraction but to the fact that a point light source will render a (fuzzy) image of the diphragm in the the OOF part of the image. Easiest example is the bokeh of a mirror objective which looks horrible due to all the images of the small-mirror blocking the primary light path.

 

See below an example from Douglass Herr:

badbokeh.jpg

 

Or how about this one (source):

090321-1.jpg

Edited by SJP
Link to post
Share on other sites

Interesting, so how is diffraction handled in a lens which provides a very small aperture, such as a macro lens where you need to stop down heavily to get any depth of field? Maybe it isn't...

 

One of my Nikon macro lenses goes down to f57 and the other f45.

Link to post
Share on other sites

The answer is that you get tons of diffraction. You also get paper-thin depth of field. It is simply (well ...) a matter of balancing one horror against another in the hope of obtaining a useful image.

 

The ultimate in diffraction is the pinhole camera. It is essentially afocal: nothing is out of focus, but nothing is really in focus either. Field of view is determined 'trigonometrically' by the physical distance of the pinhole from the film.

 

The old man from the Age of Box Cameras

Link to post
Share on other sites

If you look at Leica's info sheets on their new lenses, the slower ones have MTF graphs wide open at f/5.6 and at f/8 whilst the fast (f/1.4) lenses have MTF graphs wide open, at f/2.8 and at f/5.6 which makes me think that these lenses do optimise at wide apertures. Its also clear from the graphs that the smaller aperture lenses perform better at say f/5.6 (and probablt f/8 since no comparison is shown) which makes me think that they are better performers for work which does not require use at very wide apertures.

 

My personal experience of diffraction leads me to consider that its effect varies and is dependant on the interaction between lens and subject matter and as such is not as simple as being an obvious degredation at small apertures.

 

There is a lot of interesting information in "EF Lens Work III" from Canon, which is surprisingly general once you've got past the Canon Lens section!

Link to post
Share on other sites

Diffraction lowers contrast (edge sharpness) so it is less visible with high contrast subjects. All optical defects of lenses are seen in their interactions with the subject, which is pretty obvious. Curvature of field e.g. is more obvious with flat copy matter than with three-dimensional subjects, and similarly linear distortion is most easily seen in architectural photography.

 

The old man from the Age of Box Cameras

Link to post
Share on other sites

All optical defects of lenses are seen in their interactions with the subject, which is pretty obvious.

Actually, this forum is pretty good at understanding/appreciating such matters however there is a wealth of disinformation on the web and the obvious is all too often overlooked.

For example diffraction limitation kicks in earlier with poorly designed lenses (as any further degradation quickly adds to existing aberrations) which might account for at least one of the factors which reduces the performance of high MPixel sensors.

Link to post
Share on other sites

Actually, this forum is pretty good at understanding/appreciating such matters however there is a wealth of disinformation on the web and the obvious is all too often overlooked.

For example diffraction limitation kicks in earlier with poorly designed lenses (as any further degradation quickly adds to existing aberrations) which might account for at least one of the factors which reduces the performance of high MPixel sensors.

 

Now I'm totally confused. I always thought that diffraction effects became noticeable earlier (that is, at a wider aperture) with HIGH quality lenses, because they were not masked by other abberations!

Link to post
Share on other sites

Lens diffraction is not the same as diffraction limitation, although the two are related.

Aperture diffraction is the loss of resolution of the lens on stopping down, i.e. the size of the Airy disk the lens produces. Other lens faults are of influence on this value as well, obviously.

Diffraction limitation is the interaction between the size of that Airy disk and the pixel size of the sensor, in other words the projected resolution of the lens and the resolution of the sensor.

 

 

Diffraction Limited Photography: Pixel Size, Aperture and Airy Disks

 

 

And then, to confuse matters, a lens can be diffraction limited as well, which means so well corrected that the Airy disk has the size of the theoretical disk produced by the diffraction of the aperture, in other words, the resolution loss caused by aberrations in the lens system are smaller than the resolution loss caused by the diaphragm. Very few lenses meet this criterium, a prime example being the Apo-Telyt 280/4.0 R

Edited by jaapv
Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...