Corner sharpness: Four Thirds vs 35mm (bandwidth warning)

Started Oct 27, 2007 | Discussions
Ominous Veteran Member • Posts: 4,304
since when was the 5D pro level?

Its a 20D with a bigger sensor.
--
http://www.pbase.com/ewhalen

Jim Attfield Veteran Member • Posts: 3,930
Yes...

Amin Sabet wrote:

I suppose it was inevitable. This is why I almost didn't start this
thread. Bleh.

... that would have been FAR better. This whole 'equivalence' thing has become 'equivalent' to a tiresome presentation at a tiresome conference given by Professor T. Iresome in the University of Tiresomeness, Tiresome, NZ.

Jim

Amin Sabet
OP Amin Sabet Veteran Member • Posts: 6,763
Re: Ah, Chez to the rescue.

joe mama wrote:

Nice to see you two teaming up again. What's it been? A month?

We'll I've said what I had to say and don't wish to further contribute to the negative aspects of this thread. For anyone, Joe included, who got anything of value out of his thread, I'm glad. Most seem to find the test entirely worthless because of my choice of lenses, RAW conversions, and lack of tripod =).

 Amin Sabet's gear list:Amin Sabet's gear list
Sony a9 Panasonic Lumix DMC-GX85 Nikon D500 Panasonic Lumix G 14mm F2.5 ASPH Panasonic Lumix G Macro 30mm F2.8 +11 more
Amin Sabet
OP Amin Sabet Veteran Member • Posts: 6,763
Re: reiterate the corner sharpness case

Thanks Riley. I'd be interested in seeing an ultrawide shootout as well.

Rriley wrote:

first, many thanks to Amin for a qualified test experience. I just
knew your experience would be more than handy around here.

while i am reading, with agreeably just a tad of a smirk about how
corner sharpness isnt of 'personal' concern, i feel i need to press
the case where it is. Not that Im into landscape, but i just cant see
why anything less than sharp corners would be acceptable in the
majority of cases. However YMMV.

Yet more to my own field, corner sharpness, along with edges (which
is what disappears next) is a critical value. It might/might not be
so obvious in many views of nature, but within architecture it
becomes very obvious. Where as you might expect with interiors in
particular, all 4 corners may present this issue. The patterns and
textures become awash, necessary detail is lost and to make matters
worse; the more transient nature of mixed light sources of electrical
lighting and daylight perhaps flash, vastly complicate a repair
process.

Within the escape of architecture the congruence of heavy use of UWA
and the requirement for consistent sharpness over the entire frame I
strongly suspect that true UWA tests would be even more revealing
than those witnessed here.

 Amin Sabet's gear list:Amin Sabet's gear list
Sony a9 Panasonic Lumix DMC-GX85 Nikon D500 Panasonic Lumix G 14mm F2.5 ASPH Panasonic Lumix G Macro 30mm F2.8 +11 more
Amin Sabet
OP Amin Sabet Veteran Member • Posts: 6,763
Re: Then I messed up.

Xargo wrote:

I was just a bit concerned at first as I have had bad
experience in converting E-410 files to .dng (I had to use
photorescue software to bring back the original .orfs as the .dng
quality was so horrible). I use .dng all the time with E-330/L1
though. To me it seems like the NR-control just isn't implemented in
.dng (yet?) and that causes problems with very finely detailed photos.

That's interesting to know. I wonder if its a problem with the way ACR and LR render Oly DNGs. I find that the Adobe products add excessive NR and are inadequate at removing high ISO coarse color noise. Using them to convert to DNG and then viewing the DNG in C1 beta, I haven't noticed a problem. In fact, this has been my main workflow. I'm looking forward to true E-410 support in C1!

 Amin Sabet's gear list:Amin Sabet's gear list
Sony a9 Panasonic Lumix DMC-GX85 Nikon D500 Panasonic Lumix G 14mm F2.5 ASPH Panasonic Lumix G Macro 30mm F2.8 +11 more
Amin Sabet
OP Amin Sabet Veteran Member • Posts: 6,763
Re: Corner sharpness: Four Thirds vs 35mm (bandwidth warning)

Stu 5 wrote:

That is why you use a lens testing chart....... they are flat so
there are no D.O.F problems.

Life has DOF. Why create a scenario for testing that doesn't apply to real life?

 Amin Sabet's gear list:Amin Sabet's gear list
Sony a9 Panasonic Lumix DMC-GX85 Nikon D500 Panasonic Lumix G 14mm F2.5 ASPH Panasonic Lumix G Macro 30mm F2.8 +11 more
Amin Sabet
OP Amin Sabet Veteran Member • Posts: 6,763
Re: Corner sharpness: Four Thirds vs 35mm (bandwidth warning)

Andy Hewitt wrote:

Hi Amin,

While I wouldn't want to criticise your efforts, I have to say that
this test was rather useless, being perfectly honest.

Andy, I respect your opinions on this. Regards, Amin

 Amin Sabet's gear list:Amin Sabet's gear list
Sony a9 Panasonic Lumix DMC-GX85 Nikon D500 Panasonic Lumix G 14mm F2.5 ASPH Panasonic Lumix G Macro 30mm F2.8 +11 more
Amin Sabet
OP Amin Sabet Veteran Member • Posts: 6,763
Re: FF corner test using my primes

Chez Wimpy wrote:

For the most part I think the point is moot because serious landscape
work (for serious output sizes) begs for MF+. Every year my favorite
office room calanders are filled with landscapes of Hokkaido, and
they are indeed sharp to the edges. Truly breathtaking, but obviously
from large format film. I would hate to be a landscape photographer
dedicated to Canon FF - unless I were a genius at stitching. Of
course if Canon intends to challenge that market with 40MP FF
sensors (and why not? they are slowly going that way) they will have
to do something about the problem. I saw another calander with
aerial photography done in africa, and with that one the left halves
of two of the images were horribly soft, clearly from a decentered
lens. I guess publishers all have different standards

CW, good point. I've personally seen some amazing (to me) landscape work with 35mm digital, but it just comes down to standards as you say. If I were a pro scaper, I'd definitely have some digital medium format in my kit.

 Amin Sabet's gear list:Amin Sabet's gear list
Sony a9 Panasonic Lumix DMC-GX85 Nikon D500 Panasonic Lumix G 14mm F2.5 ASPH Panasonic Lumix G Macro 30mm F2.8 +11 more
Ominous Veteran Member • Posts: 4,304
Re: Corner sharpness: Four Thirds vs 35mm (bandwidth warning)

Amin Sabet wrote:

Stu 5 wrote:

That is why you use a lens testing chart....... they are flat so
there are no D.O.F problems.

Life has DOF. Why create a scenario for testing that doesn't apply
to real life?

Are you testing DoF, or the lens corners?

You can't say a lens has soft corners if the shots you are taking the corners are simply not in focus.

-- hide signature --
Amin Sabet
OP Amin Sabet Veteran Member • Posts: 6,763
Re: Corner sharpness: Four Thirds vs 35mm (bandwidth warning)

Ominous wrote:

You can't say a lens has soft corners if the shots you are taking the
corners are simply not in focus.

If two images in a comparison have the same point of focus and the same DOF, you cannot say the corners are out of focus in one and in focus in the other. Furthermore, if you shoot two test chart pics at different DOFs and then try to generalize that information to real life situations where you want the same DOF, you will often fail. Finally, it is known that some lenses, or even copies of the same lens, are optimized for short subject distance versus long distance. I find charts to be generally disappointing in predicting real outcomes.

 Amin Sabet's gear list:Amin Sabet's gear list
Sony a9 Panasonic Lumix DMC-GX85 Nikon D500 Panasonic Lumix G 14mm F2.5 ASPH Panasonic Lumix G Macro 30mm F2.8 +11 more
Stu 5 Veteran Member • Posts: 3,277
Re: Corner sharpness: Four Thirds vs 35mm (bandwidth warning)

Ominous wrote:

Amin Sabet wrote:

Stu 5 wrote:

That is why you use a lens testing chart....... they are flat so
there are no D.O.F problems.

Life has DOF. Why create a scenario for testing that doesn't apply
to real life?

Are you testing DoF, or the lens corners?

You can't say a lens has soft corners if the shots you are taking the
corners are simply not in focus.

Exactly! The way these lenses have been tested you can't tell if the lens has soft focus on the edges or if it's out of focus. It's all right going on about 'life has DOF' but when it comes to testing lenses your testing needs to be controlled or the results are meaningless but worse than that, misleading. You need to use a lens chart on a wall and a camera on a tripod that is parallel to the wall, with the lens framed on the centre of the chart. You then take photos with both lenses through the whole aperture range on the lowest 'iso' setting on the camera. You then can look at the resulting photos and use the chart to measure the 'lines' per 'mm' reading in the centre and on the edges. You will also be able to tell if the lens suffers from pin cushion or barrel distortion as the chart has readings for that as well. Rule of thumb is that a lens performs best one f stop from fully closed BUT this is not always the case everytime. As I said before the only reason I know this is because I have studied lens design and have done lens testing.

Ominous Veteran Member • Posts: 4,304
Re: Corner sharpness: Four Thirds vs 35mm (bandwidth warning)

Amin Sabet wrote:

Ominous wrote:

You can't say a lens has soft corners if the shots you are taking the
corners are simply not in focus.

If two images in a comparison have the same point of focus and the
same DOF, you cannot say the corners are out of focus in one and in
focus in the other. Furthermore, if you shoot two test chart pics at
different DOFs and then try to generalize that information to real
life situations where you want the same DOF, you will often fail.
Finally, it is known that some lenses, or even copies of the same
lens, are optimized for short subject distance versus long distance.
I find charts to be generally disappointing in predicting real
outcomes.

The fact that you hand held them says they don't have the same focus point, or would we even know if the blurring is a result of motion blur.
--
http://www.pbase.com/ewhalen

Ominous Veteran Member • Posts: 4,304
I'd also like to point out...

That the Oly camera has blow the detail in the house (the white thing) even though it is shot at ISO 200, while the detail is still there in the 800 ISO shot from the 5D.

-- hide signature --
Xargo Contributing Member • Posts: 850
Re: Corner sharpness: Four Thirds vs 35mm (bandwidth warning)

Stu 5 wrote:

As I said before the only reason I know this is
because I have studied lens design and have done lens testing.

Ok, I don't see why that means other people are not allowed to post tests taken outside the lab. I think OP managed to show the point very well with this kind of test. If the 5D corner smearing was because of the motion blur, it would obviously be apparent in other places around the shot as well. DOF was the same for both cameras and the possible focus difference doesn't explain the smearing.

I do a lot of scientific lens testing at our university as well but that doesn't make tests like this useless. Scientific? Not. Useless? Not.

-- hide signature --

DMC-L1, E-330, 11-22mm, 14-50mm Vario-Elmarit-D, 25mm Summilux-D, 180mm Apo-Telyt-R, 250mm Telyt-R, 60mm Macro Elmarit-R, 135mm Elmarit-R, 90mm Elmarit-R, 35mm Summicron-R, FL36, EC-14, 2x Leica-R extender

Rriley
Rriley Forum Pro • Posts: 21,846
are you guys drinking again ?

how do you get the corner OOF and not the whole bottom edge which is the same distance ?
--
Riley

real men couldn't care less

 Rriley's gear list:Rriley's gear list
Sigma DP2 Merrill Canon EOS 5D Olympus E-3 Olympus E-5 Panasonic Lumix DMC-GH2 +1 more
Xargo Contributing Member • Posts: 850
Re: I'd also like to point out...

Ominous wrote:

That the Oly camera has blow the detail in the house (the white
thing) even though it is shot at ISO 200, while the detail is still
there in the 800 ISO shot from the 5D.

Actually the detail is there in the raw-file. OP gave a link where to download them.

5D simply has a different tone curve. Here's a nice read about this:

http://www.jayandwanda.com/photography/DRtesting/drtest.html

Also I don't see how DR has anything to do with corner smearing which was the point of the whole test.

-- hide signature --

DMC-L1, E-330, 11-22mm, 14-50mm Vario-Elmarit-D, 25mm Summilux-D, 180mm Apo-Telyt-R, 250mm Telyt-R, 60mm Macro Elmarit-R, 135mm Elmarit-R, 90mm Elmarit-R, 35mm Summicron-R, FL36, EC-14, 2x Leica-R extender

Tal C Contributing Member • Posts: 547
oly is sharper?

The way I see it, the dry leaf on the upper left of the crop is much sharper on the oly... unless you've said it and I missed it somehow....
--
Tal C

Ominous Veteran Member • Posts: 4,304
Re: are you guys drinking again ?

The corner and the "bottom" are not the same distance from the lens.
--
http://www.pbase.com/ewhalen

Ominous Veteran Member • Posts: 4,304
Re: I'd also like to point out...

Xargo wrote:

Ominous wrote:

That the Oly camera has blow the detail in the house (the white
thing) even though it is shot at ISO 200, while the detail is still
there in the 800 ISO shot from the 5D.

Actually the detail is there in the raw-file. OP gave a link where to
download them.

I missed it...can you point it out to me?

5D simply has a different tone curve. Here's a nice read about this:

http://www.jayandwanda.com/photography/DRtesting/drtest.html

Also I don't see how DR has anything to do with corner smearing which
was the point of the whole test.

-- hide signature --

DMC-L1, E-330, 11-22mm, 14-50mm Vario-Elmarit-D, 25mm Summilux-D,
180mm Apo-Telyt-R, 250mm Telyt-R, 60mm Macro Elmarit-R, 135mm
Elmarit-R, 90mm Elmarit-R, 35mm Summicron-R, FL36, EC-14, 2x Leica-R
extender

-- hide signature --
Rriley
Rriley Forum Pro • Posts: 21,846
Re: are you guys drinking again ?

Ominous wrote:

The corner and the "bottom" are not the same distance from the lens.

in this orientation
the centre of the bottom of the frame is the closest point
therefore IF the corner of the frame is OOF, the bottom edge should be too
if you look at the full res images, you can see that its not

indeed even in the crops, it is obvious it is just the corner that appears soft, devoid of detail. And that points equal in range from the lens to the corner. maintain focus

-- hide signature --

Riley

real men couldn't care less

 Rriley's gear list:Rriley's gear list
Sigma DP2 Merrill Canon EOS 5D Olympus E-3 Olympus E-5 Panasonic Lumix DMC-GH2 +1 more
Keyboard shortcuts:
FForum MMy threads