Can someone explain me the relation between the out of focus and number of blades of the diaphragm. Some companies add more blades to have a better out of focus (bokeh) in their pictures... All technical comments are welcome....Thanks. JPAuger.

-- Jean Auger (paphoto@videotron.ca), May 06, 2002

Answers

The bokeh of the lens is only partly controlled by the number of blades. Many blades, well designed, can eliminate the "hexagons" of out of focus highlights, and make them into more circular patterns.

The nature of the lens design (optical fomula) has more to do with the bokeh than anything else. Some designs just have lousy OOF areas (mirror lenses for example), some have great OOF areas, that are soft and have gentle gradients from light to shadow.

-- Charles (cbarcellona@telocity.com), May 06, 2002.

A lens with only a few aperture blades usually produces out of focus highlights with noticeably geometric shapes at the mid apertures. Many people think this is unnatural looking, and that a smoother, softer, rounder out of focus "blob" looks nicer. I have lenses with both types of apertures, and have to say I prefer the look of the highlights created by the many bladed, near round aperture styles.

-- Andrew Schank (aschank@flash.net), May 06, 2002.

Keep in mind that shape of the aperture blades does matter too. I.e. say 8-blade diaphragm with curved aperture blades can givr more pleasant highlight rendition then 10-blade with straight blades. Placement of the diaphragm relatively to the film plane also matters.

-- Alexander Grekhov (grekhov@wgukraine.com), May 06, 2002.

>Some companies add more blades to have a better out of focus (bokeh) in their pictures.

Now THAT sounds nutty to me. IMO they concentrate on making them SHARPER. To hell with bokeh.

-- Alec (alecj@bellsouth.net), May 06, 2002.

>Some companies add more blades to have a better out of focus (bokeh) in their pictures.

Now THAT sounds nutty to me. IMO they concentrate on making them SHARPER. To hell with bokeh. --------------------------------------------------

Thus another Nikon photographer is born. . . .

-- Keith Davis (leica4ever@yahoo.com), May 06, 2002.

Contax even went through the trouble to get a near perfect round aperture in their little T3 point and shoot camera. You can have both sharp lenses with nice bokeh, or you can have very sharp lenses and lousy bokeh--like most AF, internal focusing macro lenses.

-- Andrew Schank (aschank@flash.net), May 06, 2002.

FWIW, I recently picked up a Tamron 90mm Macro lens for my Canon. Which BTW is a surprisingly sharp lens -- I think even sharper than my Nikon 105 macro was. Anyway, it has 9 curved baldes that make an almost perfectly circular aperture -- and the Bokeh is quite excellent. While certainly not definitive proof, this at least seems to support the hypothesis that rounder apertures give better Bokeh ;;;-)

-- Jack Flesher (jbflesher@msn.com), May 06, 2002.

An attempt to be slightly more technical...

First let assume we have a POINT light source out of focus. If our lens is an ideal lens (gaussian thin lens without any aberration) then the shape of that point light source on the film is exact copy of the shape of the figure formed by aperture blades; the brightness of the shape on film is constant across all the shape. If our lens is a more real thing (with all the aberrations) then A) the shape gets distorted and B) the brightness of the shape on film is no more constant. The last issue is well known and there is even a sort of classification ("ring", "core" etc).

Second, lets consider an extensive, lengthy light source out of focus. Since in principle such a source can be described as a continuum of point light sources, the shape on film from such a blob can be derived as soon as we know the shape from point light source. If our lens is ideal it is an easy elementary geometry problem. Interesting consequence is that in the latter case the brightness of the shape on film is not constant even if the light blob has constant brightness (e.g. a light box); the brigtness across the edges does gradually decrease, and the character of this decreasing depends exclusively on the figure shape formed by aperture blades. If the lens is not ideal, the things get more complicated of course.

-- Andrey Vorobyov (AndreyVorobyov@hotbox.ru), May 07, 2002.

Better than any explanation, how about a visual comparison? I posted this before, but it works well with your question. I use the 105mm Nikkor and the 90mm Elmarit M, both fine lenses as far as sharpness and contrast. The one mechanical difference that is readily noticeable is the number of aperture blades. The Nikkor has only six simple blades, and any lens opening beyond the maximum aperture is distinctly hexagonal. The Leica Elmarit M has two more blades, and the effect is a more round aperture opening... not a perfect circle, but round enough to look less geometric than the hexagon of the Nikkor.

When there are out of focus highlights in the frame, the Nikon shots have that hexagonal shape in the highlights. The Elmarit M on the other hand reproduces those highlights as a more clean circular effect. Some might find the hexagonal shapes to be a distraction, (on another forum, I had a Nikon portrait critiqued negatively because the geometric highlights took attention from the face), and other people might not notice or care.

Take a look at two side by side examples here, both shot at f/4.0: Aperture blade effect

-- Al Smith (smith58@msn.com), May 07, 2002.

It is also true that current Leica lenses have less blades than the older equivalents? eg im sure older pre ASPH cron 35 had more blades than my now 35 ASPH cron?

-- Karl Yik (karl.yik@dk.com), May 07, 2002.

Keith: Alec ain't no Nikon photographer...at least not one who knows the AF-DC Nikkors. ;-)

Karl: yes you're right. The bean-counters are not exclusive to Japanese makes.

-- Andrew (mazurka@rocketmail.com), May 07, 2002.

Let's use "polygonal" instead of "geometric" - a circle/disc is no less "geometric" than a hexagon, heptagon or an octagon, etc. :)

-- Andrew (mazurka@rocketmail.com), May 07, 2002.

My 180 Elmarit (E67 version) has eight aperture blades - the Nikon 180 ED has either eight or nine.

The Elmarit shows very smooth blur circles at any aperture with no hint of 'polygonal' edges.

The Nikon projects very hard-edged OOF circles that definitely show polygonal forms when stopped down even a little.

(OTOH the Nikon is sharper in the center at 2.8 - no joke)

Another comparison - my 90 TE has 10 blades - my 90 'cron has 11. The TE tends to give a slightly 'busier' background than the 'cron at the same aperture (but is sharper down to at least f/8). Same manufacturer, same focal length, same era of design (mid/late-70's).

So the number of blades may play a roll, but it is clearly secondary to the optical design, the choice of aberration corrections, the location of the aperture within the light path, and other factors.

A perfectly circular aperture in a lens with 'hard' corrections will give perfectly circular (except for vignetting) hard-edges circles - while in a lens with 'softer' corrections it will give softer circles.

-- Andy Piper (apidens@denver.infi.net), May 07, 2002.