[Digitalis]

what are you using to adapt e-mount glass to your bench?

I haven't tested any E-mount glass on my bench yet.

I have to get an electronic adapter board that allows for control of lens AF and aperture from Zeiss.The lens mount boards are expensive, especially the electronic ones.

---

Last edited by Digitalis; 02-16-2016 at 01:44 AM.

[clackers]

Roger Cicala also tests the lens instead of the image file - the RAW is written modified.

[Digitalis]

Which is why I trust Roger Cicala's results more than Imatest - I have been able to replicate Rogers results with my own testing bench with Leica,Canon and Nikon lenses he has tested.

[wildman]

shuffle data around to make things look better than they really are

That sounds fair enough to me.
In photography human visual perception is reality.

[Digitalis]

sounds more like nostalgia: making things in the past sound better than the crap you are going through at the moment.

[wildman]

sounds more like nostalgia: making things in the past sound better than the crap you are going through at the moment.

................?

[osv]

Which is why I trust Roger Cicala's results more than Imatest - I have been able to replicate Rogers results with my own testing bench with Leica,Canon and Nikon lenses he has tested.

roger uses imatest all the time, it provides real-world numbers from the entire imaging system.

sony isn't the only manufacturer that "cooks" the raw, leica and some of the m4/3 bodies do it as well.

the problem with optical benches is that they only test at infinity; afaik, it's difficult if not impossible to optimize a lens for both infinity and mfd, so if you are shooting at less than infinity, the bench probably doesn't represent what the lens is actually doing.

roger kludged e-mount lenses into working on his bench, but he had to add a layer of optical glass to the rig: https://www.lensrentals.com/blog/2015/10/sony-e-mount-lens-sharpness-bench-tests

be sure and read the comments, a lot of assumptions in the article were called into question in the comments section.

[Digitalis]

roger kludged e-mount lenses into working on his bench, but he had to add a layer of optical glass to the rig

yeah, I would never do that, but Rodger had good reason for it: I saw the MTF results without the glass, and they were atrocious. For some reason SONY use a very thick sensor stack, so there was a need to emulate it through the addition of a planar glass surface near the exit pupil.

the problem with optical benches is that they only test at infinity; afaik, it's difficult if not impossible to optimize a lens for both infinity and mfd, so if you are shooting at less than infinity, the bench probably doesn't represent what the lens is actually doing.

At lensrentals they use an OLAF bench, which is vertical. My Zeiss bench is horizontal, this gives it greater flexibility with which lenses that can be placed on the lens mount platform as it can be moved to allow for testing the lenses at different focus distances and it also allows for unusual lenses such a superwides such as the Canon 11-14mm f/4L and Sigma 8-16mm to be tested and it can also handle lenses as long as 1200mm*. But like everyone elses bench, mine freaks out when you put a fisheye lens on the platform.

roger uses imatest all the time,

But that isn't all he uses.


*Which are relatively common in the world of large format.

[GeneV]

They just vary a lot. The FE 35mm f2.8 has 2.6 stops of vignetting wide open. Which is a lot, considering that wide open is only f2.8. Border sharpness really isn't great. The Zeiss FE 24-70 f4 has fiercesome distortion throughout and pretty bad vignetting at 24mm and f4. My perception from the reviews is that a lot of these lenses rely pretty heavily on software manipulation to make them usable.

People have told me that this is the wave of the future and that they would rather have a smaller, cheaper lens and use software to correct them, but I have always felt that if you could have a decently corrected lens, then you don't have to manipulate it as much in post.

Is this distortion something you have experienced or just heard about. The only Sony lens I have used which has "fearsome distortion" is the 16-50. I have not used the FF 24-70, but my Zeiss 16-70/4 on a crop format is the best lens of that range I have ever used, even raw.

---------- Post added 02-17-16 at 02:01 PM ----------

That sounds fair enough to me.
In photography human visual perception is reality.

That was my reaction. I am not really that concerned with whether I have a Ferrari which theoretically goes over 217 mph if I never have a reason or a place to go that fast.

If the end result looks good, then I'm happy, but YMMV,

[clackers]

Is this distortion something you have experienced or just heard about.

Roger Cicala points out that looking at the Sony RAW file gives about 1% distortion on IIRC the 35mm f1.4, Gene, but his testing of the lens itself shows about 4%!


ie the RAW files are 'cooked' with JPEG techniques. There may be vignetting and sharpening adjustments applied too to cope with the corners and edges.

[Digitalis]

35mm f1.4, Gene, but his testing of the lens itself shows about 4%

4% distortion on a 35mm f/1.4 prime?...um, WTF?

[petrakka]

There are two conversations going on here which seems a little strange: there is the one about numbers that have zero to miniscule impact in most every day applications for like, 99% of photographers - and the one that says, well the lenses work well and look great and the proof is in the files returned from the camera - algorithms, resolving power etc be damned. They work. I mean, every time I drive into downtown Chicago I see iPhone images on billboards that look good.

[clackers]

4% distortion on a 35mm f/1.4 prime?...um, WTF?

Uh huh.

Lot of m43 style trickery going on, with expensive lenses.

https://www.lensrentals.com/blog/2015/10/sony-e-mount-lens-sharpness-bench-tests

I don't know if he'll ever get around to the rest of the line up.

[Digitalis]

I just looked up the Sony 30mm macro lens - which has noticeably more distortion than the DA35mm f/2.8. And when a Macro lens has a problem with distortion, things aren't going well.

[Sliver-Surfer]

I just looked up the Sony 30mm macro lens - which has noticeably more distortion than the DA35mm f/2.8. And when a Macro lens has a problem with distortion, things aren't going well.

I looked this up as well one has better distortion less sharp the other is more sharp worse distortion