Aperture blades

[&Denekamp]

Alright, since i've read this article on how the shape of the aperture affects the recorded image, I've been having this question.

The article was mostly on how out of focus areas were rendered by a lens. and how the number of blades affects its bokeh (boke). From the article I got that a three blade aperture isn't really any good, a four blade aperture isn't good either. Five is pretty acceptable, six is getting better. eight even better, and so on, and so on.

This is because the shape of the aperture is getting closer to the circle shape as it goes up trought the number of aperture blades (right?)

So, ofcourse I had to have a look at my own lenses. My 50mm has 6 blades. but, when I set it wide open @ 1.7, you no longer see the blades, making the aperture the shape of a circle.

Would that mean (and here comes the actual question) that full open, the lense has the nicest bokeh because then its shape is closest to that of a circle? or am I talking rubbish now?

oh, and while I'm at it; one more question, if you dont mind.

Why dont the manifactures dont just go the extra mile and put in a few more aperture blades? I mean, are they really that expesive that you have to cut corners on the number of aperture blades? or are there other problems involved with putting in a few more?

Thanks in advance,


Niels

 

[Big Mike]

I think that's the idea...more blades makes more of a circle, which make the out of focus areas look better...although better is pretty subjective. I don't know if that means that wide open gives 'better' bokeh...but I do know that wide open give a shorter DOF, which makes it easier to render things out of focus.

There are probably other factors to how the bokeh looks, but I'm not an expert in lens design.

I'm guessing that more blades are more expensive to design and produce. Weight & size might also be an issue.

 

[voodoocat]

The benifit of more apeture blades is only evident when you start closing those blades.

 

[DocFrankenstein]

Would that mean (and here comes the actual question) that full open, the lense has the nicest bokeh because then its shape is closest to that of a circle? or am I talking rubbish now?s

No, not necessarily - depends on the lens
No

Usually the problem with bokeh wide open, is that you get the cat's eye effect.

Read here:
http://www.vanwalree.com/optics/vignetting.html

And also here:
http://www.vanwalree.com/optics/bokeh.html

And them might as well read everything:
http://www.vanwalree.com/optics.html

From my experience... the most pleasing bokeh often comes from slow german lenses... maybe that's why they're so expensive.

 

[Hertz van Rental]

I'm guessing that more blades are more expensive to design and produce. Weight & size might also be an issue.

You get a problem when the aperture is wide open because all those blades would have to go somewhere - and they layer over each other. A lot of blades could give rise to a situation where you have 6 or more blades on top of each other which would require quite a lot of space - and if one was even slightly out the risk of jamming would be much greater.
You also have the simple mechanical problem of moving all those blades in synch, and the more blades the more force is required to move them.
Lens designers stick to an optimum number that strikes the best balance between all the problems.

 

[DocFrankenstein]

Lens designers stick to an optimum number that strikes the best balance between all the problems.

main problem being the manufacting cost it seems. :er:

If you look at any superb professional primes, then all have 8+ number of blades.

 

[Unimaxium]

main problem being the manufacting cost it seems.

That would be my guess too. I can't imagine it's cheap to make the mechanism that moves each of those little blades. And to do it in synchronocity, and to accurately achieve the correct diameters for each f-stop. I'd assume it can't be easy to add a lot of blades. And there's all the stuff Hertz mentioned, too.

 

[&Denekamp]

Thanks guys, for all the info!

So, not a 'more pleasant' (as I should say, instead of nicer) bokeh at full open. Instead a whole lot of additional side affects that should be taken in account.. hmm, too bad.. :meh:

I think you guys are dead right about the cost thing (and all the problem solving involved) with more aperture blades. Just look at those 10+ blades in Leica lensen and similar. They're not going for $1000 and up for no reason :er:

Doc, I've read the whole site. really interesting! thank you.

 

[Rob]

This is quite interesting if you happen to like Ken Rockwell:

http://www.kenrockwell.com/tech/bokeh.htm

 

[DocFrankenstein]

It doesn't cost much money to make those 3-5 additional blades... But it's purposeful division of products into "consumer" and "pro" grades. Just the same as it doesn't cost more than 5 cents to make the metal mount for 50/1.8 canon lens, but yet canon molds out of fn plastic.

 

[Hertz van Rental]

It doesn't cost much money to make those 3-5 additional blades... But it's purposeful division of products into "consumer" and "pro" grades. Just the same as it doesn't cost more than 5 cents to make the metal mount for 50/1.8 canon lens, but yet canon molds out of fn plastic.

The blades? No. But the design of the mechanics, putting it all together, the alteration in lens body design to accomodate it all? The costs add up and you would probably make a lens that no one would buy because it the advantages (if there were any) would not justify the expense.

As for mounts. Metal mounts can be cast quite cheaply - but you always have to finish off on a lathe or milling machine. This is a skilled job and not cheap. It is also quite easy to go off tolerance with a lot of waste.
Plastic can be moulded these days to very high tolerances and plastics are also custom designed for optimum qualities. It's a lot cheaper in many ways - and the savings are split giving extra profits to investors and savings to customers.
These considerations far outweigh the disadvantages.

 

[DocFrankenstein]

I may be wrong. My source of the costs and designs were two engineering professors who specialize in manufacuring.

Milling/lathes are computer controlled and automated.

But the design of the mechanics, putting it all together, the alteration in lens body design to accomodate it all?

Size is going to be increased by the thickness of the 3 blades. Lens design is not affected... Too many variables to be having an internet discussion over it.

The professor approximated the manufacturing costs of canon 50/1.8 with metal mount to be from 4 to 6 dollars. Not too bad, eh?