Stretching the M8

[rob_x2004]

Advertisement (gone after registration)

Manuel there are two things here. Camera operator error for the flat image that looks colour cast and might be might'nt be I dont know. I think it is more likely something else, raw processing error and seting error when the frame was made. Second issue is the regualr pattern of dots and those bands of contrasting density. Dunno. Seems pretty obvious to me though. I just highlighted them a bit. Frankly I think these images will sell lots of slide film, or canons, which is a shame. I am convinced the camera is capable.

[Advertisement]

The Leica Q3 is available for pre-order at B&H Photo Video and Adorama

[ArtZ]

I don't want discourage you but it seems to me a problem with the CCD sensor. I read something about the way that CCD sensos transfert the image to the image processor.

 

I do not want to make a polemic here but if some manufactures have choosen CMOS sensors it's for that reason.

 

I should send an email to Attila von Gyimes (info@leica-camera.com) and/or Ursula Brand-Gerheim (M8.Support@leica-camera.com) including your pictures in attachment and ask them what they think.

 

Saludos

[ho_co]

The vertical line is peculiar; it is not the same issue in the http://www.l-camera-forum.com/leica-forum/leica-m8-forum/23902-dng-processing-problem.html thread.

 

The M8 sensor is divided at the center across its 18mm height. (This is also the case with many or all medium-format backs.) The electronics interpreting the two parts can get out of synch. When that happens, the vertical line appears--but only in the center, as shown in the other thread.

 

This is something completely different, which I don't think I've seen.

 

Still, if one needs to mess with the image so heavily to uncover it, I'm not sure it counts as a problem.

 

Art, I'm not sure we can lay this to CMOS vs CCD. Maybe so. I know the problem in the other thread has to do with how the sensor is laid out, and I don't think all CCDs are divided in two like the M8's (at least until you get to medium format sizes). You obviously know more about it than I do, so tell me more about this difference between the two.

 

Rob, I'm not sure what your red arrows are pointing to besides the vertical line. Could it be spots on a dirty sensor?

 

Arthur--wonderful shot! Striking just to be able to say, "I've been there!"

 

--HC

[rob_x2004]

Howard the spots are in a regular rivet layout. I wasnt carefull to tag them, jsut went dot dot dot etc. The first mage is in portrait orientation so it is horizontal across the bottom of the frame, or the top? There is maybe some crud too. I thought I had seen it in another thread somewhere, but not the one referenced in Manuels response. Issues get OT sidetracked easily in here and I dont really follow the place so I dont know where to find it. Maybe it has been discussed at length. Is it a big problem? Someone said something about writing or converting to file. Software problem in camera or post processing? Sensopr definitely?

 

Hopefully the images will get sent as Manuel suggested and an answer can get posted?

[ArtZ]

If my memory is good, I think that CCDs move photogenerated charge from pixel to pixel and convert it to voltage at an output node. CMOS imagers convert charge to voltage inside each pixel. That's the reason why CMOS are faster and uses less energy than CDD.

 

The CMOS requires an additional chip to treat all this information before send it to the image processor. On CCD this is not necessary because the image processor can treat this information.

 

When I see these images it seems that there's a problem during the transfert at a particular line(s) of pixels.

 

Maybe the sensor didn't do the job correctly, maybe the image processor didn't treat it properly.

 

You can find useful information about CCD sensors and CMOS sensors at CCD vs. CMOS

 

Hope that helps

 

ArtZ

[ho_co]

Art, thanks for the pointers.

 

The lines of varying density bother me quite a bit. In one case the line runs along the long edge, in the other case along the short edge of the frame. Should be two different phenomena, n'est-ce pas? Strange.

 

Rob, I accept your judgment that the spots look like regular rivets. But tell me, do you attack all digital images this way? You've discovered elements that are all but invisible in the originals, so there is clearly result to your method; but it would never occur to me to do such things to an image. (I don't even know what half of them are--"Equalising, Balance tone, S curving...")

 

--HC

[rob_x2004]

Advertisement (gone after registration)

Art, thanks for the pointers.

 

The lines of varying density bother me quite a bit. In one case the line runs along the long edge, in the other case along the short edge of the frame. Should be two different phenomena, n'est-ce pas? Strange.

 

Rob, I accept your judgment that the spots look like regular rivets. But tell me, do you attack all digital images this way? You've discovered elements that are all but invisible in the originals, so there is clearly result to your method; but it would never occur to me to do such things to an image. (I don't even know what half of them are--"Equalising, Balance tone, S curving...")

 

--HC

 

I dont follow?

 

I pointed out the areas between the "various densities" which you seem to be concerned about. There are at least two bands in the second image, I didnt look hard. Regular rivet pattern is on another thread somewhere. I'd like to find it.

 

I remarked that the apparent coulour cast might also be attributed to contrast in the mid tones and I gave my reasons rather than just mouthing off.

 

Yes I look at a lot of m8 images, because I am interested in where the m8 is at.

 

People level criticisms ... or pointed observations about the m8 without putting them into perspective, which is why I occasionally ask about the conditions or full frame so we can see the extent of the crops or whatever.

 

Regards

[RSL]

I posted a previous thread about the amazing dynamic range (DR) of the M8 and most of you agree with it. While I am still at the Canadian Rockies, I wanted to see how far this DR will stretch ... this was shot in late morning. The color of the lake is true due to the silt from the glacier. I have question about the tones and colors on the slopes of the mountains. What are your thots?

 

 

Zeiss 21mm/2.8 Biogon ZM

 

Arthur,

 

It's a fine shot. Bring it up in CS3 Camera Raw, pick up the white balance eyedropper and click on the gray underside of one of those clouds. I just downloaded the picture and tried that, and almost immediately you get the kind of color balance you're looking for. I can't judge the results I got here because even though I work with a well calibrated screen, I'm not sure how far off the original the copy is, but if you're working in .DNG, which you should be, you should be able to get what you're looking for almost immediately.

[Bob Ross]

I dont follow?

 

I pointed out the areas between the "various densities" which you seem to be concerned about. There are at least two bands in the second image, I didnt look hard. Regular rivet pattern is on another thread somewhere. I'd like to find it.

 

I remarked that the apparent coulour cast might also be attributed to contrast in the mid tones and I gave my reasons rather than just mouthing off.

 

Yes I look at a lot of m8 images, because I am interested in where the m8 is at.

 

People level criticisms ... or pointed observations about the m8 without putting them into perspective, which is why I occasionally ask about the conditions or full frame so we can see the extent of the crops or whatever.

 

Regards

Hi Rob,

I think perspective can be used is a useful tool, but when you take web size images and push them to extremes, it is more of a warping exercise. I would say that any digital image can be pushed to see how it will break up and they all do. Perspective on image quality and processing flexibility would be better achieved by spindling an original file and, indeed, I do that when considering any new camera.

Bob

[Bob Ross]

Other than the yellow cast tonally it's quite flat an easy way of checking is to convert the image to greyscale turn it upside down.

While the apparent yellow color cast can different due to our monitors, your tip about using gray scale upside down is an excellent way to check tonality. Something that I've notice in the digital image medium is that we can meter well, but seldom adjust the contrast to the scene. Fortunately it can be done in PP. I fooled around with Arthur's image using your technique to address the flatness and came up with this, which might cure some comments, but destroy Arthur's knowledge of what he intended to capture. The picture brings back memories of suffering from "Canadian Rocky Green-out", after being up there a few days....LOL

Bob

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!

[nryn]

A good way to enhance haze in these long-distance images is to enhance local contrast. You can do this by applying unsharp mask settings which have a very large radius (250 or close to 250) and a relatively small amount. This would help define the texture of the mountains a bit more, as Bob's adjustments have done. You can also do this on a duplicated layer in PS then adjust the opacity of that layer to get very finely tuned adjustments.

 

I think the latest version of Lightroom has a setting called Clarity which does something to this effect.

[Philinflash]

Second image is full of it too...theres a lot more than I have highlighted or marked in crop. What gives? Brains trust?

 

[ATTACH]45283[/ATTACH]

 

Crud on the sensor?

[ho_co]

... an easy way of checking is to convert the image to greyscale turn it upside down.

... your tip about using gray scale upside down is an excellent way to check tonality.

Bob--

The approach certainly shows results. Nice job.

 

Bob & Imants--

Could you be more specific about the upside-down gray scale technique?

 

(I'm more familiar with PSCS2, so if there are buttons to push there, I'd appreciate help on doing this in PS--but any info would be appreciated.)

 

Thanks!

 

--HC

[Guest stnami]

Howard. it's pretty easy........ You can lets say have a blue statue next to a green box and and a orange background. Sure the colours will distinguish one from another but they may have the same tonal value and the image looks flat and lifeless. By changing to greyscale i.e. desaturate you can see how flat the image is or how contratsty it is,

The upside down caper.............. basically to trick the mind and lose the figure ground aspect of the image. You see the photo as a group of shapes and your bias towards the subject changes it becomes a purely structural exercise in changing the tones.

I use it quite regularly...........works for me

[ho_co]

Thanks, Imants! That's brilliant.

 

Seems I might have heard of turning a print upside down at some time, but I didn't think of that here. Somehow, my mind got short-circuited into converting to grayscale but then reversing the grayscale--i.e. making it a negative. And from there I wallowed off into self-pity.

 

Thanks for the explanation and for the idea!!!

 

--HC

[geoffreyg]

I fooled around with Arthur's image using your technique to address the flatness and came up with this, which might cure some comments, but destroy Arthur's knowledge of what he intended to capture. The picture brings back memories of suffering from "Canadian Rocky Green-out", after being up there a few days....LOL

Bob

[ATTACH]45339[/ATTACH]

 

Very nice reworking. Looks like all the positives in the original shot simply got better. What did you do?

[Bob Ross]

Very nice reworking. Looks like all the positives in the original shot simply got better. What did you do?

Hi Geoffrey,

I think that I might have over-cooked it a bit. Things look different a day after:rolleyes:

I use Picture Window Pro, so you might have to translate to the editor that you use. I started out making tone masks for the highlights, midtones and decided not to do a shadow mask. This alows me to adjust the tones and colors in each area independently. I used the masks to recover some detail by increasing contrast using the curves, while the image was still in a low contrast state. Then I used Imants' monochrome technique, by making a plain monochrome conversion and an upside down version, too. I adjusted the global tone, which turned out to be just pulling the left end of the curve to the right a smidge (after my previous work with the masks). The tones looked good, so I just applied them to the color version. To reduce the haze on the mountains, I used a bit of local area contrast enhansement (described by Ethrfarm) and then made a side by side composite of before and after. Now, I see some flaws, caused by going to fast with the masks and working on a small file, but I thought it showed off Imants' technique and answered the "it looks flat" comments.

Bob

[Bob Ross]

Thanks, Imants! That's brilliant.

 

Seems I might have heard of turning a print upside down at some time, but I didn't think of that here. Somehow, my mind got short-circuited into converting to grayscale but then reversing the grayscale--i.e. making it a negative. And from there I wallowed off into self-pity.

 

Thanks for the explanation and for the idea!!!

 

--HC

Hi Howard,

Imants gave an excellent explaination. Another trick that I use when I have a composition problem is to make a high contrast (lithograph) version to see how the light areas compose and sometimes I will use that to crop. Our eyes are drawn to the brighter parts of an image first and the better they are composed, the more likely the viewer will stay with the image to really see it, even when the subject might be disturbing. This also helps me to know when to say screw it and trash the image;)

Converting to a B&W negative is also a good tool, for those who have stared at negatives projected on the base board of their enlargers for too long. It works with scanning negatives to help retain the highlight detail, too. Adjust the negative to what you know is the "perfect negative" and reinvert to a positive. Some old habits die slow lingering deaths:eek:

Bob

[Advertisement]

Hallo,

Look here: Leica M

[ho_co]

For me, this thread is an eyeopener.

 

I thought Arthur's image was superb, and now I see how so many can improve it!

 

Great job, Arthur, for posting a great shot and opening it for discussion.

 

And it's wonderful to see all the ideas that can be realized just to tweak the image to perfection!

 

It sure ain't a point-n-shoot world round these here parts, stranger!

 

--HC

[arthury]

ArtZ, Rob A & ho_co,

 

I am quite amazed you found a way to reveal the vertical lines and the pattern of spots. I opened the images in PS/CS3, zoom in to 100% and did all of kinds of things to it but I cannot see the vertical lines. The only spots I saw are several large black spots , which are clearly dust particles.

 

Tell us what you did to help us follow your steps.

 

Thanks!