Forgot Password
Pentax Camera Forums Home
 

Reply
Show Printable Version Search this Thread
02-20-2016, 10:22 PM   #61
osv
Pentaxian




Join Date: Mar 2007
Location: So Cal
Photos: Gallery
Posts: 2,080
maybe image circle that the lens was designed for would be the most accurate way to put it.

wrt field curvature and such for legacy glass, 24mm is by far the worst, 28mm is a huge step up, 35mm is still showing weak corners for many wide lenses, etc... when you hit 135mm, they all look similar on the a7r, all of the f/2.8 legacy glass has ugly ca.

afaik:
1)film doesn't show ca as much as digital does: https://www.flickr.com/photos/genotypewriter/6147351879
2)some types of ca often get worse towards the outer regions of the ff image circle, it's another reason why ff lenses on crop don't represent how the lens really performs
3)higher rez ff sensors will show more problems than lower rez ff sensors, and color film in particular is the lowest rez there is on ff

for example... supertak 135/3.5, wide open i think, bad ca on the irrigation pipes, and definitely hosed up on the right side: https://www.dropbox.com/s/w4vtk7fiufpa04y/135mmSuperTak3point5prime2861.JPG

crop sensor would knock out the worst of the right side, and i suspect that if you could see the ca on color film, it's not going to look that bad... it certainly won't be as sharp on color film... no ca on b&w film? just another tonal graduation that's lowering the resolution?

---------- Post added 02-20-16 at 09:35 PM ----------

QuoteOriginally posted by stevebrot Quote
Of the lenses on my shelf, only the Helios 44M is particularly soft in the corners relative to the center.
that's because you are looking at it on low rez film... this old 35/2.8 minolta glass is really nice, mechanically it's on a par with some of the best taks ever made, but it fails in the deep corners, and it has a very slight wavy focus field that you'd never see on film: 35mm lens comparison, second round, Minolta 35/2.8 & Pentax FA35/2, on the Sony a7R.

my m35/2.8 is better in the deep corners, but not perfect, and it's not as good as the fa35 is in the corners.

all of the ugly warts start showing up on 36mp

02-21-2016, 12:46 AM   #62
Site Supporter
stevebrot's Avatar

Join Date: Mar 2007
Location: Vancouver (USA)
Photos: Gallery | Albums
Posts: 26,231
QuoteOriginally posted by osv Quote
that's because you are looking at it on low rez film
Ummmm, no. It is because it is a Biotar-derived lens meaning that it is characteristically sharp in the center at all apertures, but sucks in the corners and not just a little bit. You don't need imatest or even a big print. Those characteristics were known to the Zeiss designers and were intentional.

Like you, I also own both the MC W.Rokker-HG 35/2.8 (ca 1970) and the FA 35/2 referenced in your linked test, but I shoot both on film. The Rokkor is lower contrast and softer overall than both the FA 35/2 and my LZOS Jupiter-12 35/2.8 on that medium. Again, the performance difference is obvious. It is a pleasant enough lens to shoot with on the SRT 101, but not my first choice at that focal length. Translation? I don't need 36 Mpx for the warts to show. Both the lower contrast and corner softness are visible on a 1100 dpi scan. The contrast is easily dealt with in post and the softness is only apparent on close examination and not a problem up to the maximum size I can print (13"x19" from a 4000 dpi scan) using my equipment. Still though, I get decent results from the lens on its native format and based on the full resolution corner crop on your test shot, so do you on your A7R.

Your comments regarding CA on film vs digital are curious. CA is an optical quality of the lens and is not related to the capture medium. It may be measured on an optical bench with no camera present. I have no problem capturing CA on film. When it is there it is obvious and is often visible even on B&W negative. It was an issue of optical design long before the advent of color emulsions since it severely degrades performance and has been a component of lens testing since at least the late 1960s when I started shooting. Back to your comments, you are correct that LaCA is more pronounced in the corners. It is also more pronounced at narrower apertures. Fortunately, it is easily corrected in PP with a resulting improvement in both sharpness and local contrast. LoCA (so-called bokeh CA) does not show this pattern and is harder to correct.

Purple fringing (PF) is another matter. While it is colorful, it is not CA per se. It is a digital sensor phenomenon* and does not occur on film nor can it be measured on an optical bench. A year or so ago, I did a test using parallel setups with my FA 77/1.8, using both Ektar 100 and the K-3. I can reliably generate LoCA on both film and digital with that lens, but not PF. I could generate PF on the K-3 only. The K-3 is more resistant to PF than my K10D was with the same lens. I presume that is due to differences in the sensors.

I appreciate the effort you go through testing the various lenses on your A7R. But lacking an actual comparison for detail capture of the same subject from the same tripod under the same conditions, I don't believe you are in a position to say whether your camera "stresses" a particular lens more at 36 Mpx than if it were shot with an A7 at 24 Mpx or with a high resolution film scanned to the same nominal pixel count. Don't get me wrong, I am not denying that there some incredible modern lenses out there, only that I am not convinced that the older designs (I will restrict this to primes) are categorically less capable or that the latest lenses have less problems with corner sharpness, vignette, or flat field.

So much for my part of this thread hijack.


Steve

* PF was once a matter of hot discussion and still remains a bit of a mystery. It is easy enough to tease out of most lens/sensor combinations, but some lenses are worse than others regardless of sensors. The most compelling theory I have heard is that it is LoCA at the near UV end of the spectrum that interacts with the sensor micro-lens array resulting in a violet bloom on the sensor. Studies were done with tunable narrow band LED lights and when the illumination was UV sparse, PF was essentially absent.

Last edited by stevebrot; 02-21-2016 at 01:22 AM.
02-21-2016, 01:09 AM   #63
Pentaxian
Digitalis's Avatar

Join Date: Mar 2009
Location: Adelaide.
Posts: 8,535
QuoteOriginally posted by stevebrot Quote
lacking an actual comparison for detail capture of the same subject from the same tripod under the same conditions, I don't believe you are in a position to say whether your camera "stresses" a particular lens more than if it were shot with an A7 at 24 Mpx or with a high resolution film scanned to the same nominal pixel count.
Which is why optics testing bench scores are so important. As they reduce the number of variables that invariably crop up during lens/camera testing. When you are testing lenses on cameras there are so many things that can go wrong to invalidate your test data. There are things that a bench can do that a camera film/sensor can not. And in your case with the 135mm f/3.5 - I'd suspect that your lens adapter mount didn't have the mount on the camera or lens side perfectly parallel to the sensor plane. I have never seen a 135mm f/3.5 with that degree of astigmatism in my life, and I have tested a few.
02-21-2016, 01:39 AM   #64
Site Supporter
stevebrot's Avatar

Join Date: Mar 2007
Location: Vancouver (USA)
Photos: Gallery | Albums
Posts: 26,231
QuoteOriginally posted by Digitalis Quote
I have never seen a 135mm f/3.5 with that degree of astigmatism in my life, and I have tested a few.
I don't have the long history of the rigorous lens testing, but agree that something is simply not right with either the lens or the adapter. Perhaps there is something growing inside?


Steve

02-21-2016, 01:45 AM   #65
Pentaxian
Digitalis's Avatar

Join Date: Mar 2009
Location: Adelaide.
Posts: 8,535
QuoteOriginally posted by stevebrot Quote
Perhaps there is something growing inside?
That would reduce contrast, which isn't what we are seeing here: and the chances of encountering a de-centered lens from that era is very low....but not impossible. There may have been some curious owner of the lens decided to clean it out, and inadvertently upset the optical alignment. Testing a lens for de-centering is rather difficult on camera, on a test bench it only takes seconds. You just project a test pattern through the lens, spin the lens on its mount: and if the test pattern wobbles..the lens is clearly out of alignment.

Last edited by Digitalis; 02-21-2016 at 01:53 AM.
02-21-2016, 11:10 AM   #66
osv
Pentaxian




Join Date: Mar 2007
Location: So Cal
Photos: Gallery
Posts: 2,080
QuoteOriginally posted by stevebrot Quote
Ummmm, no. It is because it is a Biotar-derived lens meaning that it is characteristically sharp in the center at all apertures, but sucks in the corners and not just a little bit.
no, you claimed that "only the Helios 44M is particularly soft in the corners relative to the center." which is not true, my reply was that you just can't see softness in other lenses because you've been shooting with weak low rez film.

we've already been through the problems with film many times out here. Digital cameras vs. film, part 1

QuoteOriginally posted by stevebrot Quote
Like you, I also own both the MC W.Rokker-HG 35/2.8 (ca 1970) and the FA 35/2 referenced in your linked test, but I shoot both on film. ...Both the lower contrast and corner softness are visible on a 1100 dpi scan.
oh? i'd be interested in seeing a film demonstration of the soft corner differences that i illustrated in that hg35/2.8 link...

QuoteOriginally posted by stevebrot Quote
Your comments regarding CA on film vs digital are curious. CA is an optical quality of the lens and is not related to the capture medium. It may be measured on an optical bench with no camera present.
duh...

QuoteOriginally posted by stevebrot Quote
I have no problem capturing CA on film. When it is there it is obvious and is often visible even on B&W negative.
i gave a specific link with photos, it's pretty clear: "film doesn't show ca as much as digital does: https://www.flickr.com/photos/genotypewriter/6147351879 "

one of the points i made was how do you fix it on b&w film, which you didn't address.

QuoteOriginally posted by stevebrot Quote
Purple fringing (PF) is another matter. While it is colorful, it is not CA per se.
no, that's not true: "Examples of axial color (longitudinal chromatic aberration)" Chromatic aberrations

QuoteOriginally posted by stevebrot Quote
Purple fringing (PF) is another matter. While it is colorful, it is not CA per se. It is a digital sensor phenomenon* and does not occur on film
no, here is a specific example showing pf on film, bottom of the page, notice how one example is clearly purple, and the other is green: "Figure 7. Chromatic aberrations recorded on film. The church cross is photographed off a slide projection screen; the vane, taken at the "best" focus and after slight defocusing, is a 2900 dpi scan obtained with a slide scanner. Notice the similarities with Figs. 3 and 4." Chromatic aberrations

whatever you saw in your testing might have been due to a misfocused lens.

QuoteOriginally posted by stevebrot Quote
I appreciate the effort you go through testing the various lenses on your A7R. But lacking an actual comparison for detail capture of the same subject from the same tripod under the same conditions
i've provided all that with numerous lenses, so i have no idea what you are talking about... i shoot the same scenes back-to-back with different lenses, i edit and post the comparisons into one pic so that it's easy to see, and i shoot it with a level of focusing precision that you can't duplicate with any of the gear you own.

---------- Post added 02-21-16 at 10:51 AM ----------

QuoteOriginally posted by Digitalis Quote
And in your case with the 135mm f/3.5 - I'd suspect that your lens adapter mount didn't have the mount on the camera or lens side perfectly parallel to the sensor plane. I have never seen a 135mm f/3.5 with that degree of astigmatism in my life, and I have tested a few.
if it was astigmatism, the left side would be blurry, and it's not... the right side is a blurry mess because the lens is decentered, and no, you don't need an optical bench to see that

here is a canon 135/3.5, different adapter of course, notice how both sides are clean, not just the left side... similar blurry foreground, tho, because it's shot wide open just like the supertak135/3.5 pic was:

https://www.dropbox.com/s/5rkipvss1snzaji/135mmCanon3point5prime2867.JPG

this is astigmatism/field curvature, notice how it's blurry on both sides, and the top and bottom: Astigmatism and field curvature

this is what it looks like in the real world, when the edge is in focus the center is slightly out, blow the pic up to 100%: http://www.codectest.com/2014/24mm-camera-lens-shootout-on-the-sony-a7r-third-round-konica-24mm
02-21-2016, 06:25 PM   #67
Site Supporter
stevebrot's Avatar

Join Date: Mar 2007
Location: Vancouver (USA)
Photos: Gallery | Albums
Posts: 26,231
QuoteOriginally posted by osv Quote
no, you claimed that "only the Helios 44M is particularly soft in the corners relative to the center." which is not true, my reply was that you just can't see softness in other lenses because you've been shooting with weak low rez film.
Glad you have knowledge of both the films I use and the maximum performance attributes of the lenses on my shelf. Is it safe to assume that your film experience is similarly limited?

QuoteOriginally posted by osv Quote
oh? i'd be interested in seeing a film demonstration of the soft corner differences that i illustrated in that hg35/2.8 link...
Ummmm...your example shows little or no corner degradation. That is easily matched. My copy is a little softer, though I am unwilling to waste the film to demonstrate in the interest of this discussion.

QuoteOriginally posted by osv Quote
no, that's not true: "Examples of axial color (longitudinal chromatic aberration)"
I guess if you are willing to call LoCA purple fringing, then apparently you have limited experience with PF. End of discussion, you win. You also seem pretty convinced that film is incapable of displaying CA, field curvature, or any other aberration that can be demonstrated by a photo of a strip mall or measured by imatest on a pure digital capture. Again, end of discussion, you win. The world begins and ends with what can be captured by a 24x36 36 Mpx sensor. See preceeding...




Steve

(...by the way...am very aware of how to test for field curvature. The first requirement is that the target is planar across the full field and parallel to the focal plane...I am also aware that astigmatism is not the same thing...)

Last edited by stevebrot; 02-21-2016 at 06:52 PM.
02-21-2016, 06:47 PM   #68
Site Supporter
stevebrot's Avatar

Join Date: Mar 2007
Location: Vancouver (USA)
Photos: Gallery | Albums
Posts: 26,231
QuoteOriginally posted by Topsy Quote
I haven't really informed myself much about it, but I'd assume that if the lens performed well on film and still is performing well on digital cameras the corners shouldn't be much different than on film either?
Sounds reasonable to me.


Steve

02-21-2016, 07:24 PM   #69
Site Supporter




Join Date: Oct 2009
Location: North
Photos: Gallery
Posts: 3,778
QuoteOriginally posted by osv Quote
you missed the point... of course a slow lens that's decades old isn't going to be optimal, everyone knows that, so why did they make an ignorant comment about its pq in the review? it was a cheap shot at the k-1.

by comparison, did they put an ancient nikon 50/2 lens in the d800 first look? and mention how much it sucked on 36mp? i'm gonna guess no

dpr can be such a fail, for example, the forums are overrun with sock puppets trolling multiple fake ids, that seldom get banned: Please put a stop to this character: Site Feedback and Help Forum: Digital Photography Review

i thought that adam wrote an overall better intro to the k-1 than dpr did.

has any article mentioned what the efcs capability of the k-1 is?
+1 for me.

Overall, I get a hint that some guys over at DPR like it from how the article is overall uplifting/enthusiastic.
However, in typical DPR fashion regarding Pentax stuff, some folks there will have to throw in some low blows like that 50/2 comment and the AF tracking with a UWA.
02-21-2016, 08:24 PM   #70
osv
Pentaxian




Join Date: Mar 2007
Location: So Cal
Photos: Gallery
Posts: 2,080
QuoteOriginally posted by stevebrot Quote
Ummmm...your example shows little or no corner degradation.
it's significantly worse than either of my pentax 35mm primes, maybe you didn't examine the corner crop at 100%?? because the differences between sharpness and blurring is obvious.

QuoteOriginally posted by stevebrot Quote
I guess if you are willing to call LoCA purple fringing, then apparently you have limited experience with PF.
i quoted toothwalker, who stated that pf is a form of ca, and he's not the only one: "Axial (longitudinal) chromatic aberration (purple and green fringes)," New Color Fringe Correction Controls

you film guys who have never shot legacy glass on ff digital are in for a bit of a surprise
02-21-2016, 08:28 PM - 1 Like   #71
Lens Buying Addict
monochrome's Avatar

Join Date: Mar 2008
Location: Kirkwood (St. Louis) MO
Photos: Gallery | Albums
Posts: 19,494
Oh man.
02-21-2016, 08:38 PM   #72
osv
Pentaxian




Join Date: Mar 2007
Location: So Cal
Photos: Gallery
Posts: 2,080
QuoteOriginally posted by Digitalis Quote
Testing a lens for de-centering is rather difficult on camera, on a test bench it only takes seconds. You just project a test pattern through the lens, spin the lens on its mount: and if the test pattern wobbles..the lens is clearly out of alignment.
sometimes i'll spin the lens on the adapter mount, nearly 180 degrees, to see if the aberration travels with the lens, which proves that it's not an adapter problem.

i've never seen a case where the soft side failed to move with the lens rotation; it's never an adapter problem.

if the problem follows the camera body being flipped 180 degrees, the sensor is parallel to the target... that's not a concern at the near-infinity distances that i test at, tho, it's only something to worry about for people who shoot brick walls or targets up close.
02-21-2016, 10:53 PM   #73
Pentaxian
Digitalis's Avatar

Join Date: Mar 2009
Location: Adelaide.
Posts: 8,535
QuoteOriginally posted by osv Quote
sometimes i'll spin the lens on the adapter mount, nearly 180 degrees, to see if the aberration travels with the lens, which proves that it's not an adapter problem.
If the adapter mount is moving with the lens it doesn't prove much, the adapter mount can still be misaligned.
02-22-2016, 08:28 AM   #74
osv
Pentaxian




Join Date: Mar 2007
Location: So Cal
Photos: Gallery
Posts: 2,080
i've always kept the adapter fixed and locked on the body.

most of the time i'm lucky to get a 90 degree spin out of the lens, it's really not enough.
Reply

Bookmarks
  • Submit Thread to Facebook Facebook
  • Submit Thread to Twitter Twitter
  • Submit Thread to Digg Digg
Tags - Make this thread easier to find by adding keywords to it!
color, corners, dslr, film, focus, full frame, full-frame, glass, image circle, impressions, k-1, k1, lens, lenses, pentax k-1, pf, pic, review, slide
Thread Tools Search this Thread
Search this Thread:

Advanced Search


Similar Threads
Thread Thread Starter Forum Replies Last Post
Very first impressions of the K-S2 paulvzo Pentax K-S1 & K-S2 6 01-16-2016 10:10 AM
First impressions of the K-01 Scorpio71GR Pentax K-01 12 04-23-2015 03:58 PM
My first impressions of my new K-50 Dewman Pentax DSLR Discussion 7 11-01-2014 05:22 AM
K-30 first impressions of a K-x user mee Pentax K-30 & K-50 10 03-02-2013 05:00 PM
First impressions of K-r from a K-x owner devorama Pentax K-r 28 03-30-2011 08:55 AM



All times are GMT -7. The time now is 12:23 AM. | See also: NikonForums.com, part of our network of photo forums!
  • Red (Default)
  • Green
  • Gray
  • Dark
  • Dark Yellow
  • Dark Blue
  • Old Red
  • Old Green
  • Old Gray
  • Dial-Up Style
Hello! It's great to see you back on the forum! Have you considered joining the community?
register
Creating a FREE ACCOUNT takes under a minute, removes ads, and lets you post! [Dismiss]
Top