Samuli Vahonen wrote:
Snapsy, I don't think the issue is CoC 0.010mm, that is pretty accurate (the actual area of DOF, which is perceived as sharp). Based on graph Barnack outputs, the rear bokeh should have CoC somewhere around 0.035mm (=6000px on 36mm==>166.666px/mm =====> 0.035mm=5.83px), but the blur is more than this, much more.
BTW. Most mathematical DOF calculations, are based on theoretical single lens element model, and doesn't apply very well for example to retrofocus lenses, lenses with aberrations etc. - I did the background study in 2005/2006 when I wrote my own DOF calculator, so I may not have most up to date information. However thanks for link to Barnack, I never bothered to calculate the theoretical stuff outside DOF in my tools I programmed for photography.
Martin, I'm not complaining - I just want to understand. This makes G45 just more interesting for me, as you know front to back all sharp isn't my (or your) style. But as any lenses, I really want to understand inside out how they work to get best results. Naturally I can just shoot thousands of photos and analyze some tens of hours, but understanding theory/reasons why this happens would make the process of understanding MUCH faster and efficient (at least typically it does...).
As you mention, there is something same in ZE/ZF 2/35. I have also some "weird" shots, where DOF is really long for being shot f/2.0-2.8, but rear boke is gigantic (compared to my expectations). As far as I remember this mostly happened on quite far focus distances (>5 meters).
i have on occasion noticed similar phenomenon with my g45, but i always chalked it up to misremembering my shooting aperture. i'll have to test mine out in similar shots. no idea what the cause would be optically – the g45 is a pretty basic double gauss design. i've mentioned before that it's drawing style does remind a lot of the ZE 35/2 (but with courser bokeh).
Frank,
I have difficulty understanding how the CY28 2.8 can be so sharp, from corner to corner. It is a lovely, small lens on the A7R and the colour rendition is wonderful
That and the CY 35-70 are my two fav lenses to take bushwalking (landscape and forest photos), and on the A7R are a great match
dovey wrote:
Frank,
I have difficulty understanding how the CY28 2.8 can be so sharp, from corner to corner. It is a lovely, small lens on the A7R and the colour rendition is wonderful
That and the CY 35-70 are my two fav lenses to take bushwalking (landscape and forest photos), and on the A7R are a great match
Right it is sharp. Here at FM many rightfully praise that lens. And it is still (fairly) cheap. And I'd add that it is a very fine lens for the "renowned" CY object isolation. F/2.8 aint bad at 28mm.
I have not tried the CY 35-70 but I do have the CY 28-85. Both are exceptionally good zoom lenses om par and sometimes ahead of the primes.
Samuli Vahonen wrote:
Snapsy, I don't think the issue is CoC 0.010mm, that is pretty accurate (the actual area of DOF, which is perceived as sharp). Based on graph Barnack outputs, the rear bokeh should have CoC somewhere around 0.035mm (=6000px on 36mm==>166.666px/mm =====> 0.035mm=5.83px), but the blur is more than this, much more.
If memory serves the blur circle calculations are appropriate up to mid-distance objects but not for far-distance, where blur quantity is instead determined by the physical aperture size.
wfrank, that b&w portrait of the young lady is stunning.
A couple months ago I was fortunate to secure an almost BNIB CY 28/2.8 and 35-70 lenses. They've been sitting in the closet and I was going to sell them. After seeing your shots with the 28 I really want to try the lens on my A7r. thanks for posting and rekindling my interest in that lens.
Picture This! wrote:
wfrank, that b&w portrait of the young lady is stunning.
A couple months ago I was fortunate to secure an almost BNIB CY 28/2.8 and 35-70 lenses. They've been sitting in the closet and I was going to sell them. After seeing your shots with the 28 I really want to try the lens on my A7r. thanks for posting and rekindling my interest in that lens.
shelt wrote:
^^^ How do you like your Ultron so far?
Shelt, I like it a lot but it is not without problems. At night with long distance scenes in the city(lights) I did notice a lot of flare and strange color reflections. I think they are caused by the chrome parts in the adapter and lens mount.
I just bought a black rubber ring which fits around the lens and cover the chrome parts.
I don't know yet if it helps because I did not yet have the time to test it.
Scenes closer by are allright at night. (see first photo, cropped to the right for composition)
It is sharp in the corners but does have some colour cast on the sides/corners which is correctable. Mostly only clearly visible in skys and clouds. The Canon FD 20/2.8 doesn't have these problems, but has less punch/sharpness I think.
The second photo is uncorrected for color cast, in this case it is acceptable for me.
In the third photo for example,I don't know what that ray of light is, I don't remember it was there.
But is is a wonderfull made lens, feels good and I could not let it go for the price I payd.
Picture This! wrote:
wfrank, that b&w portrait of the young lady is stunning.
A couple months ago I was fortunate to secure an almost BNIB CY 28/2.8 and 35-70 lenses. They've been sitting in the closet and I was going to sell them. After seeing your shots with the 28 I really want to try the lens on my A7r. thanks for posting and rekindling my interest in that lens.
Thanks so much - though my long haired son would rage on being called a young lady
With regards to CY I collected a bunch of them for my late 5D2 over the years. With the entrance of the A7's I expected I would get rid of some of them for smaller (despite usually slower) RF lenses. But I dont think so anymore. Theyre too good.
I have to correct on that last picture. In this case it is a spotlight on the back of the church but I will show you what I mean in amother picture. Just a moment.
Samuli Vahonen wrote:
Snapsy, I don't think the issue is CoC 0.010mm, that is pretty accurate (the actual area of DOF, which is perceived as sharp). Based on graph Barnack outputs, the rear bokeh should have CoC somewhere around 0.035mm (=6000px on 36mm==>166.666px/mm =====> 0.035mm=5.83px), but the blur is more than this, much more.
BTW. Most mathematical DOF calculations, are based on theoretical single lens element model, and doesn't apply very well for example to retrofocus lenses, lenses with aberrations etc. - I did the background study in 2005/2006 when I wrote my own DOF calculator, so I may not have most up to date information. However thanks for link to Barnack, I never bothered to calculate the theoretical stuff outside DOF in my tools I programmed for photography. ...Show more →
Aberrations are not included in simple DOF calculators, but retrofocus lenses are, so long as they are well corrected. You get a departure in the macro regime, but that is not relevant here. For what it is worth, I find an infinity blur diameter of 0.051 mm for an object distance of 4 m, and 0.041 mm for an object distance of 5 m (@ F/11).
What is your estimate of the blur disk diameter?
It is not uncommon for lens apertures to differ from their nominal value by a few tenths of a stop, so 30% more blur than mentioned above could be accounted for without resorting to aberration theory.
Great portrait. 
