Theoretical Background Blur in Fujinon lenses (incl. 90mm F2)

[Maurice]

Found this place where you can make nice graphs:

 

http://howmuchblur.com/#compare-1.5x-56mm-f1.2-and-1.5x-90mm-f2-and-1.5x-35mm-f1.4-and-1.5x-23mm-f1.4-and-1.5x-16mm-f1.4-and-1.5x-14mm-f2.8-and-1.5x-10mm-f4-on-a-0.9m-wide-subject

 

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!

[Gaffman]

Nice.  It's interesting to see how closely the 90mm follows the 56mm's curve but never exceeds the amount of blur offered by the f/1.2.  I suspected this might be the case but there it is plotted on a graph, how handy.

 

And to me this isn't just a "pissing contest" either because I work in photographic retail and get asked this question a lot when comparing lenses across brands and formats.  Thanks again.

[Maurice]

And the 35mm 'beats' the 90mm below <1m for some reason. Interesting indeed.

 

I have a Minolta Rokkor 85mm F1.7, and it sits right in between the 90 and 56mm.

Even beats the 56mm at above >15 meters eventually. Of course its fairly soft wide open, but that's character.

[Iko]

 

And the 35mm 'beats' the 90mm below <1m for some reason. Interesting indeed.

 

16, 23 and 35 indeed...

 

At first I was skeptical about the results, but probably the difference is due to the stop advantage of the

three over the 90mm, which at lower distances overcome the focal lenght advantage.

 

I am curious of hearing an explanation of the phenomenon by someone more practical on the argument

[Maurice]

I think yours is good enough.

 

Here's one with the 35mm F2 and 27mm F2.8

 

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!

[mjh]

Keep in mind that the distance is the distance between subject and background, not the distance between camera and background. The graph will look different for subjects of a different size – the width of the subject can be specified. With a 35 mm lens you need to get closer to capture the same subject than with a 90 mm lens which gives a fast 35 mm lens a headstart, but as the background distance increases, the longer focal length wins out eventually. The break-even point depends on the subject size and thus the distance of the subject.

[Maurice]

Indeed indeed, nothing's ever simple with optics.

[Iko]

I hadn't noticed the words under x-axis, now it's more clear

[aceflibble]

That sort of works, but it's based on very clinical/theoretical/lab subjects and conditions, which makes it biased in favour of medium input perimeters. (Until the very extremes, where nothing can deny the blurry of a 200mm f/2 or 400mm f/4.) It also doesn't seem to quite have crop factor figured out correctly.

Not bad, and great if you like DxO and all that kind of thing, but I think the point at the bottom of the page about interpreting the info should be the thing you pay most attention to. 


edit: I'm an idiot (and it's half past 2 in the morning), I left a decimal point in the crop factor box. That explains why its maths seemed off. Nevermind, seems to check out okay!