Pages: [1]   Go Down

Author Topic: Fixing the M8 problems with profiles-likely not  (Read 5032 times)

Stephen Scharf

  • Full Member
  • ***
  • Offline Offline
  • Posts: 168
Fixing the M8 problems with profiles-likely not
« on: November 24, 2006, 09:47:32 pm »

There's been a lot of discussion on the site about the use of profiles to fix the M8's magenta cast problem. While I know that this is the magic bullet that many have been hoping for, according to Thomas Knoll, profiles are NOT the solution to this problem.

I think it might be of interest to readers to read what Thomas Knoll has to say on the subject-these quotes were originally posted at Digital Outback Photo:

"Profiles cannot fix IR sensitivity issues. The problem is the camera sees the world differently than humans do. It is possible to have objects in the real world that the camera sees as exactly the same color, but humans see as different colors. That is impossible to fix with a profile.

There is are also cases where the camera sees two objects as different colors, but humans see them as the same color. A profile could "fix" this, but not with out messing up other colors.

For example, the Leica M8 sometimes renders some black objects as purple. A profile could just map all purples to blacks, but that would mess up all photographs of of objects that really are purple.

The only way to really fix this is with a better IR filter."


While I was having lunch the other, a I saw someone wearing a dark magenta sweater that looked almost exactly like the magenta casts that the M8 renders of black textiles. In this scenario, how could a profile possibly know tha the sweater was really magenta and not black?

According to the points by Thomas Knoll above, it could not.
« Last Edit: November 24, 2006, 09:48:21 pm by Stephen Scharf »
Logged

michael

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 5084
Fixing the M8 problems with profiles-likely not
« Reply #1 on: November 24, 2006, 10:10:10 pm »

No one is suggesting that profiles can fix the problem. But, the profiles that Jamie has produced remove the meganta cast from most images where it is evident without significant adverse effect on other colours.

I have now tested the profiles on a number of different images that display the magenta problem. If the use of filters wasn't available, or desired, I could live with the results 90+% of the time. In fact it produces quite a pleasant palette, with some unique characteristics of its own. Reminds me of Agfachrome.

So while strict science says it isn't the solution, pragmatic evidence says that it can be of help.

Michael
« Last Edit: November 24, 2006, 10:10:50 pm by michael »
Logged

Stephen Scharf

  • Full Member
  • ***
  • Offline Offline
  • Posts: 168
Fixing the M8 problems with profiles-likely not
« Reply #2 on: November 24, 2006, 10:34:40 pm »

Quote
No one is suggesting that profiles can fix the problem. But, the profiles that Jamie has produced remove the meganta cast from most images where it is evident without significant adverse effect on other colours.

I have now tested the profiles on a number of different images that display the magenta problem. If the use of filters wasn't available, or desired, I could live with the results 90+% of the time. In fact it produces quite a pleasant palette, with some unique characteristics of its own. Reminds me of Agfachrome.

So while strict science says it isn't the solution, pragmatic evidence says that it can be of help.

Michael
[a href=\"index.php?act=findpost&pid=86931\"][{POST_SNAPBACK}][/a]

The use of profiles for solving this problem reminds me of putting a steering damper on a motorcycle to prevent tankslappers.....it's a band-aid to fix intrinsic flaws in chassis or suspension design.

Leica would have done well to have used a Design for Six Sigma approach in designing the M8; it they had true Voice of the Customer (VOC) and used a Concept Engineering approach along the lines suggested by Creveling et. al., they would have known from their VOC that accurate blacks were a Critical Functional Response that was highly valued by customers, and that the magenta cast problem would not have been acceptable design trade-off to their customers in the first place, and could have shaped their design space to incorporate the fixes that they only now have apparently implemented. This would have saved them considerable damage to their reputation, not to mention incurring a considerable Cost of Poor Quality.

Ref: Creviling, Slutsky, D'Antin: Design for Six Sigma, 2003, Prentice Hall
Logged

madmanchan

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 2115
    • Web
Fixing the M8 problems with profiles-likely not
« Reply #3 on: November 25, 2006, 08:29:51 am »

Stephen, Leica's mistakes have been discussed thoroughly on this board already and on several other boards. Yes, they goofed with some aspects of the M8 and now they are taking measures to fix/address the issues. Continuing to criticize Leica for poor craftsmanship, design errors, or whatever isn't constructive.

Eric
Logged
Eric Chan

michael

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 5084
Fixing the M8 problems with profiles-likely not
« Reply #4 on: November 25, 2006, 08:42:18 am »

I'll add an additional observation, that was first identified publicly by Sean Reid, but which I have been seeing as well.

Because of its high IR sensativity the M8 is an incredable B&W camera. Whereas a lot of dark subject matter becomes of a single luminance with digital cameras that have IR cut filters, with the M8 there are beautiful subtlties of tonality to be seen. Quite remarkable.

This was seen in years past with film cameras when using B&W film with a bit of extra IR sensativity.

So, though everyone would wish that Leica hadn't screwed up the way they did (which in my view had more to do with business practice than technology or design failure), the fact is that photographers now have a valuable instrument available. Unique B&W capability without a lens filter, and good colour with.

And for those that prefer to shoot without filters, Jamie's profiles are a viable solution much of the time.

BTW – the factory fix has nothing to do with IR sensatibity, ( rather, green ghosts and streaking) that was a design decision made long ago. Leica's mistake was not being upfront about what its consequences would be. As I wrote. A bad business decision.

I have a full article on this subject in preparation, pending Leica's more comprehensive public announcement next week.

Michael
« Last Edit: November 25, 2006, 08:53:15 am by michael »
Logged

image66

  • Full Member
  • ***
  • Offline Offline
  • Posts: 136
Fixing the M8 problems with profiles-likely not
« Reply #5 on: November 25, 2006, 11:33:41 am »

Quote
Because of its high IR sensativity the M8 is an incredable B&W camera. Whereas a lot of dark subject matter becomes of a single luminance with digital cameras that have IR cut filters, with the M8 there are beautiful subtlties of tonality to be seen. Quite remarkable.

This was seen in years past with film cameras when using B&W film with a bit of extra IR sensativity.

[a href=\"index.php?act=findpost&pid=86971\"][{POST_SNAPBACK}][/a]

Michael, thank you for pointing this out.  I wrote about this a few days ago and mentioned how Ilford Delta 400, for example, is one film that reacts to near-IR in a very similar manner. Delta 400 is the best portrait B&W film I've ever used.

The fact is, B&W from digital cameras really is a step backwards.  Sure, we've been able to "get by", but we've also had to change our "look" so much too.  What makes for a good color photograph isn't necessarily what makes for a good B&W photograph.  Just look at the color response curves of various B&W films to see what I'm talking about.  They are NOT linear in either color response or tonal response.  We can play games with the images in Photoshop to get a "look" that may be pleasing, but the characteristics of a good B&W film just aren't there.

It could be, that the M8, with it's IR flaws, is finally THE digital camera that breaks down the barrier of digital B&W and we might now start to see some quality digital B&W images that don't scream "Photoshop".

Something else I'm remembering here--last year I was running a video camera for a Christmas production--the entire choir was dressed in black.  Yup, you guessed it, under the stage lights, about half the clothes were red or purple.  One SONY camera picked up the red and purple, another had a switchable IR filter and didn't pick up the colors while the Panasonic cameras didn't have any problem at all.  I also recall different color print films doing the same thing too.  This isn't a new problem.
Logged

John Camp

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 2171
Fixing the M8 problems with profiles-likely not
« Reply #6 on: November 25, 2006, 01:46:18 pm »

Leica has said that one of its reasons for opting for a weak IR filter was to enhance the camera's qualities for B&W. I'm finding that it also does interesting things in color in what are essentially natually low-color situations (ambient city light at night, etc.) You get what at first glance seems to be a B&W photo, with just some really subtle color tones.

JC
Logged

mtomalty

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 541
    • http://www.marktomalty.com
Fixing the M8 problems with profiles-likely not
« Reply #7 on: November 25, 2006, 03:45:06 pm »

Looking at the IR 'shortcomings' from a different perspective I wonder if the increased
IR sensitivity,combined with the appropriate filter (I don't recall the #, but they are almost black),
can the M8 be used to create those dreamy color images where foliage goes pink,etc

Mark
Logged

Stephen Scharf

  • Full Member
  • ***
  • Offline Offline
  • Posts: 168
Fixing the M8 problems with profiles-likely not
« Reply #8 on: November 26, 2006, 01:07:59 am »

Quote
Stephen, Leica's mistakes have been discussed thoroughly on this board already and on several other boards. Yes, they goofed with some aspects of the M8 and now they are taking measures to fix/address the issues. Continuing to criticize Leica for poor craftsmanship, design errors, or whatever isn't constructive.

Eric
[a href=\"index.php?act=findpost&pid=86964\"][{POST_SNAPBACK}][/a]


The point of this thread was not to rehash the issue about the M8's image quality problems, it was to point readers to Thos. Knolls views on whether profiles would be effective for the IQ problems.
Logged

Brian Gilkes

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 443
    • http://www.briangilkes.com.au
Fixing the M8 problems with profiles-likely not
« Reply #9 on: November 26, 2006, 06:04:46 am »

Sure Stephen, this thread has drifted, but that is what makes conversations so interesting and sermons so boring. I'm happy to accept Thomas Knolls analysis of the ineffectiveness of profiles as a perfect fix. Thoms is usually right.Done.
Now this biz re B&W is really interesting. Leica photography has always been largely about monochrome. Henri, Sebastaio, the PJ wedding folks around the corner. Sure Leitz lenses have done wonderful stuff for colour but we stuck the Kodachrome or Agfachrome in the R8. The Tri X went into the M3>/7.
Then we all bought digital, loved it, but for B&W the channel mixer just didn't do it. Maybe, just maybe, Leica has done it again. Off track we often discover something wonderful. Let's hope this has happened this time . If that is so Leica must not panic and make unmodified M8s into rare classics.
Let's see some results in B&W without any of the fixes.
Cheers,
Brian,
www.pharoseditions.com.au
Logged

jani

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1624
    • Øyet
Fixing the M8 problems with profiles-likely not
« Reply #10 on: November 26, 2006, 05:27:23 pm »

Quote
Leica would have done well to have used a Design for Six Sigma approach in designing the M8 [...]
Wow, I think Scott Adams is reading these forums ...  
Logged
Jan
Pages: [1]   Go Up