Jump to content

A Sane Attitude to Rededge

lars_bergquist

By lars_bergquist
in Leica M9 / M-E

Recommended Posts

Maximus

Maximus

Posted
Posted

Advertisement (gone after registration)

http://uk.leica-camera.com/assets/file/download.php?filename=file_4776.pdf

 

Excerpt below :-

 

Improvements in the M9 Firmware Version 1.116

1. Camera Performance:

• Faster zooming into the last picture taken.

• Speed of SD-card formatting enhanced.

• Image quality of “thumbnail” pictures improved.

• SD- Card compatibility improved*

• Magenta colour shift correction for the following wide angle lenses improved:

o M 18mm/f3.8 ASPH (11649) o M21mm/f2.8ASPH.(11135) o M24mm/f3.8ASPH.(11648) o M 28mm/f2 ASPH. (11604) o M 28mm/f2.8 ASPH (11606)

 

Leaving aside the emotive nonsense, can anyone explain why Leica released information that simply isn't true?

Link to post
Share on other sites
  • Replies 230
  • Created
  • Last Reply
UliWer

UliWer

Posted
Posted
Assuming what we see is the result of overcorrecting the cyan-shift caused by the UV/IR cut filter, I wouldn’t expect this to be dependent on the f-stop. Did you find otherwise?

 

May be very, very slight differences. Here the 28 Summicron at F/5.6:

 

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!

 

With f/2 in #55 you have some more vignetting and you can see a slight red tone in the lower left and perhaps also the right edge (not to be seen in practical use I think). With f /5.6 this has - almost - vanished.

Link to post
Share on other sites
UliWer

UliWer

Posted
Posted
http://uk.leica-camera.com/assets/file/download.php?filename=file_4776.pdf

 

...

• Magenta colour shift correction for the following wide angle lenses improved:

o M 18mm/f3.8 ASPH (11649) o M21mm/f2.8ASPH.(11135) o M24mm/f3.8ASPH.(11648) o M 28mm/f2 ASPH. (11604) o M 28mm/f2.8 ASPH (11606)

 

Leaving aside the emotive nonsense, can anyone explain why Leica released information that simply isn't true?

 

If you would like to take the pains of comparing the results of one of these lenses before and after the firmware - and after a reset - we could discuss about wether it is true that the colour shift correction was improved or not. As long as you don't give us visual evidence your question is in vain.

Link to post
Share on other sites
Maximus

Maximus

Posted
Posted
If you would like to take the pains of comparing the results of one of these lenses before and after the firmware - and after a reset - we could discuss about wether it is true that the colour shift correction was improved or not. As long as you don't give us visual evidence your question is in vain.

There's already evidence in this thread of "after" shots showing that the firmware update claims are false.

Link to post
Share on other sites
mjh

mjh

Posted
Posted
Leaving aside the emotive nonsense, can anyone explain why Leica released information that simply isn't true?

“Improved” meaning “better but still not perfect” I haven’t seen evidence proving it wrong.

Link to post
Share on other sites
Maximus

Maximus

Posted
Posted
“Improved” meaning “better but still not perfect” I haven’t seen evidence proving it wrong.

"Improved" or "substantially the same"?

 

Improvement should be marked and clearly noticeable, as any reasonable person here already knows. Please provide evidence of such.

Link to post
Share on other sites
mjh

mjh

Posted
Posted
With f/2 in #55 you have some more vignetting and you can see a slight red tone in the lower left and perhaps also the right edge (not to be seen in practical use I think). With f /5.6 this has - almost - vanished.

I did download both of your images for comparison purposes and while I am not absolutely sure it seems to me that the main difference is in the normal vignetting. At f/2 there is visible vignetting in all the three color channels, i.e. including the red channel (just slightly less of it). At f/5.6 the red and green channel are almost free of vignetting, but there is is still some in the blue channel.

 

So in both cases there is some noticable red-shift towards the corners that isn’t reduced by stopping down (whereas normal vignetting is). But there is one thing I don’t understand: The UV/IR filter causes vignetting in the red channel, resulting in cyan corners. Now we would expect the correction to boost the corners in red channel to compensate for this effect, and if there is some overcorrection, the result should be negative vignetting – the corners in the red channel would be brighter than the center. But that’s not what we see; rather it is the corners in the blue channel that get darker.

 

Come to think of it this would make sense if we did assume that normal vignetting isn’t corrected for at all, so even with the added boost in the red channel the normal vignetting might still be stronger. In that case you would have to stop down even further (try f/11 or f/16) before negative vignetting in the red channel could be observed.

Link to post
Share on other sites
UliWer

UliWer

Posted
Posted

@maximus:

May be you find this example striking for red shift. It was taken with a pre-aspherical 21mm Elmarit - not mentioned in the list. Though I think it's useless to discuss about this. You already posted in the second enry of this thread, that you are not interested in it's topic.

Link to post
Share on other sites
Maximus

Maximus

Posted
Posted
@maximus:

May be you find this example striking for red shift. It was taken with a pre-aspherical 21mm Elmarit - not mentioned in the list. Though I think it's useless to discuss about this. You already posted in the second enry of this thread, that you are not interested in it's topic.

On the contrary, I am VERY interested in this topic. Read ALL of my posts and you'll see why.

 

It is simply unacceptable that a high-end camera manufacturer cannot guarantee colour-cast free results with their own expensive lenses without at least telling its customers about the technical limitations it cannot overcome. Then at least we could make an informed decision.

 

http://www.l-camera-forum.com/leica-forum/1266042-post118.html

 

There by the grace of God go all of us. I would have been very disappointed if I had just bought this lens and achieved similar results. To expect others not to be is brutally naive at best.

Link to post
Share on other sites
UliWer

UliWer

Posted
Posted
...

Come to think of it this would make sense if we did assume that normal vignetting isn’t corrected for at all, so even with the added boost in the red channel the normal vignetting might still be stronger. In that case you would have to stop down even further (try f/11 or f/16) before negative vignetting in the red channel could be observed.

 

I'll try this with small apertures the other day. Funny we are discussing this on examples from the M8 where the problem was never noticed before the times of the M9.

Link to post
Share on other sites
xtoph

xtoph

Posted
Posted
If you think it is that easy to correct, I advise you to offer your services to Leica. I'm sure you will be welcomed with open arms- and at a substantial salary.

 

it's possible you weren't replying to my post, but it looked like you were--maybe i don't understand the threading here.

 

but regardless, i certainly never claimed it was 'easy' to correct the color casts.

 

that, however, is irrelevant. even if we assume that it is difficult--and i did assume so--the relevant fact is that it was *known to be difficult* ahead of time. this means that testers should have known--did know--exactly what to look for.

 

your defense of leica over the red shift has shifted from 'i am certain they did not know about the problem before release' to 'you're expecting too much--it's really hard to correct for this problem'.

 

the two defenses are contradictory. and inappropriate--they offered a ff camera, implying that it worked the way a ff camera is supposed to. i didn't buy an m9 because it was *almost* a solution to a difficult engineering problem. i bought it to take pictures that bear a reasonably photographic resemblance to reality.

Link to post
Share on other sites
ho_co

ho_co

Posted
Posted

chris and Max--

At the top, Lars presented a couple suggestions for seeing a bit of sanity in the red-edge matter.

 

You've made clear how you feel. Personally, I don't find it necessary for you to keep repeating yourselves.

 

If you just want to complain about the operation of the new firmware, why not start a new thread on the topic?

Link to post
Share on other sites
UliWer

UliWer

Posted
Posted

In the meantime until we see some examples from the M9 I take my freedom to present some more with the M8. This time photos taken under indirect daylight, WB "auto", EV+1, nor corectio in C1 besides processing for JPG. All with UV/Ir-Filter, lens detecion on with UV/IR.

 

First 3,8/18

 

@f3,8:

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!

 

f/5.6

 

 

 

f/16:

Link to post
Share on other sites
UliWer

UliWer

Posted
Posted

Now 28 Summicron:

 

F/2

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!

 

 

F/5.6

 

 

F/16

Link to post
Share on other sites
pgk

pgk

Posted
Posted

So far the only poster who has shown us that he has had genuine and specific problems has been Adan. UliWer's images do illustrate variations but whilst interesting still don't show whether the red edge is a problem in the real world. As an M8 user and potentially an M9 purchaser I am interested in real world information not emotive accusations. So far I see nothing that convinces me that this problem will impinge on final images any more than a host of other 'problems' on my various cameras, all of which I have to live with because there is no perfect solution to most of them.

 

What I still find irritating is the continual moaning about 'serious problems' without any illustrative evidence to back it up (accepting that on the web this may not be as easy as desirable). So come on, show us real world examples, please.

Link to post
Share on other sites
jaapv

jaapv

Posted
Posted
it's possible you weren't replying to my post, but it looked like you were--maybe i don't understand the threading here.

 

but regardless, i certainly never claimed it was 'easy' to correct the color casts.

 

that, however, is irrelevant. even if we assume that it is difficult--and i did assume so--the relevant fact is that it was *known to be difficult* ahead of time. this means that testers should have known--did know--exactly what to look for.

 

your defense of leica over the red shift has shifted from 'i am certain they did not know about the problem before release' to 'you're expecting too much--it's really hard to correct for this problem'.

 

the two defenses are contradictory. and inappropriate--they offered a ff camera, implying that it worked the way a ff camera is supposed to. i didn't buy an m9 because it was *almost* a solution to a difficult engineering problem. i bought it to take pictures that bear a reasonably photographic resemblance to reality.

You are misquoting and putting words in my mouth. Quite unlike your normal posts. - I never said they didn't know, and I said, given the problem, would you have preferred waiting for it whilst Leica attempted solving it. And yes, given the parameters, it may be a very difficult solution causing along wait.

Link to post
Share on other sites

Archived

This topic is now archived and is closed to further replies.

Go to topic listing

  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...