purple RD1 shot with M8

[Guest stevenrk]

Advertisement (gone after registration)

I added a post to the warm and fuzzy earlier today, and am saddened to change my tune, but the first test shots in well lit tungsten light simply speak for themselves. This is a wonderfully crafted modified M7 body (as we know) that has a digital capture that isn't ready for release. The shot is Auto WB processed in C1 3.7.6 with no adjustments. The shot using tungsten WB is worse. We all know what an RD1 looks like. The strap is black and the foam blue in real life.

 

It seems as if the M8 is really only usable at the moment in daylight. Understand that I could correct in post. But that is unacceptable in a $400 camera made in 2001, let alone an M8. It's as if they built an M7 that only could use one type of film. And hard to believe that Leica was not aware that the M8 did not function correctly in tungsten light. This is simply different from banding -- which I also found -- and cannot be avoided in tungsten, and, as you can see, is quite obvious in its impact.

 

With respect to waiting for a firmware update, just ask the R9 owners who have chosen not to install the latest firmware update because of the problems it caused.

 

I'm very saddened to find this. But for those still waiting to purchase, thought worth posting.

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!

[Advertisement]

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

[dhoelscher]

This is probably the same issue seen with what Leica refers to as the purple tint on "black synthetic fabrics."

 

This issue arises because of the enhanced IR sensitivity of the sensor, which itself arises from the very thin IR-absorbing coverglass (made so thin because of the issues needed to make rangefinder lenses have less issues).

 

My understanding from a different thread on this matter is that Leica will sell an "IR barrier filter" along with "an updated firmware" at the "end of November" to deal with this particular issue.

 

DH

[simardr]

My understanding from a different thread on this matter is that Leica will sell an "IR barrier filter"

 

will sell ?

 

If this is a flaw, no one should have to pay to get it fixed...

 

At this point, it would also make sense for Leica to issue some formal communications related to these problems so future potential buyers get reassured as to the reliability of their product. This is a high-end product for which people have high expectations.

 

Roger

[ho_co]

Steven--

See also Leica's statement on the issue on their website, referenced at http://www.leica-camera-user.com/digital-forum/8867-leica-statement-auto-wb.html#post87048.

 

They're aware of the situation and have specific suggestions to use while awaiting the firmware update David mentioned.

 

--HC

[dhoelscher]

[quote name=

 

will sell ?

 

If this is a flaw, no one should have to pay to get it fixed...

 

Roger

 

My apologies ... the exact wording in the other thread said Leica "will offer" an IR barrier filter ... I don't know if that means for an extra charge or not.

 

DH

[sean_reid]

I added a post to the warm and fuzzy earlier today, and am saddened to change my tune, but the first test shots in well lit tungsten light simply speak for themselves. This is a wonderfully crafted modified M7 body (as we know) that has a digital capture that isn't ready for release. The shot is Auto WB processed in C1 3.7.6 with no adjustments. The shot using tungsten WB is worse. We all know what an RD1 looks like. The strap is black and the foam blue in real life.

 

It seems as if the M8 is really only usable at the moment in daylight. Understand that I could correct in post. But that is unacceptable in a $400 camera made in 2001, let alone an M8. It's as if they built an M7 that only could use one type of film. And hard to believe that Leica was not aware that the M8 did not function correctly in tungsten light. This is simply different from banding -- which I also found -- and cannot be avoided in tungsten, and, as you can see, is quite obvious in its impact.

 

With respect to waiting for a firmware update, just ask the R9 owners who have chosen not to install the latest firmware update because of the problems it caused.

 

I'm very saddened to find this. But for those still waiting to purchase, thought worth posting.

 

Hi Steven,

 

Before you draw too many conclusions, shoot another picture with a WhiBal or other known color-neutral card and then set your WB with the eyedropper in C1. Auto WB and presets are usually not accurate for most cameras in incandescent light, especially the R-D1 itself. More specifically, Auto-WB in household lighting is a disaster for most cameras and auto WB has never been a good method for critical color work in any lighting condition.

 

Without setting WB from a known color-neutral card, one can't draw accurate conclusions about the M8's color rendering. It would be worth doing this for you to get a better reference point as to what the camera delivers. I stress, however, that the reference card must be absolutely color-neutral. One doesn't need to do post-processing color correction if the WB setting process for C1 is accurate.

 

I made my test pictures for the review using WhiBal sampling in various kinds of light and the color rendition is quite good. That's the same method I currently use when testing all cameras. It certainly is not true that the M8 is a camera that can only render color well in daylight. Example to follow.

 

Cheers,

 

Sean

[Guest stevenrk]

Advertisement (gone after registration)

The problem here is that you are seeing purple in fabric and metalic surfaces -- with different parts of the RD1 coming up black and purple. I'm asuming based on the way light is reflecting off of the surface. This is in additon to the surface areas of the strap -- not an esoteric fabric, but rather quite basic -- and the foam padding.

 

The idea from Leica that the solution is to tell us to use an IR filter on our M lenses is frankly slightly absurd for them to even try to suggest. Especially since we have the R9. Could you imagine what would happen if Phase One or Leaf offered up an IR filter to put on lenses? And the M8 is not meant to keep up with them on resolution, but in terms of color and offering up a professional level camera, it should be held to the very same standard.

 

And as mentioned, the problem with waiting for a firmware update is that the R9 owners have chosen not to install the most recent firmware update because it caused as many problems as it fixed.

 

Really unfortunate they didn't wait -- as other camera makers have recently -- until they were able to get it right.

[sdai]

They're aware of the situation and have specific suggestions to use while awaiting the firmware update David mentioned.

 

I found it extremely hilarious to read this in Uwe's quote from Leica's note to him:

 

For best results, it is further recommended that the specific lighting situation within the WB setting is selected.

 

While Steven reported in his post:

 

The shot using tungsten WB is worse.

 

Sigh ............... a would be beautiful camera otherwise.

[sdai]

Could you imagine what would happen if Phase One or Leaf offered up an IR filter to put on lenses?

 

Exactly ... perhaps they should throw a WhiBal card into the box as well.

[sean_reid]

Case in point. The picture below was made with the M8 in very ordinary household tungsten lighting and the only color correction it has received was my setting WB via sampling from a WhiBal card. That is the same way I set color for any critical application. I use fruit and vegetables for these pictures because most people can relate the colors seen in the picture to the colors of these things in life. Yes, I know, the bananas are not quite ripe yet and the stock C1 profile is a touch high in saturation (for my taste, some may love it).

 

Cheers,

 

Sean

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!

[sean_reid]

Exactly ... perhaps they should throw a WhiBal card into the box as well.

 

That would be a good idea for all of the manufacturers. I've yet to see any digital camera that gives dead-on color with auto WB in incandescent light. Using a WB reference card is an excellent idea for all color work from any camera where critical color reproduction is needed.

 

There's a separate issue which is that the M8 doesn't filter IR very much and that's giving a magenta tint in some circumstances. That's mostly a matter of getting a tweaked C1 color profile. This kind of color fine-tuning (after a camera has been released) has been common for many new cameras - Nikon, Canon, etc.

 

Sean

[sean_reid]

The problem here is that you are seeing purple in fabric and metalic surfaces -- with different parts of the RD1 coming up black and purple. I'm asuming based on the way light is reflecting off of the surface. This is in additon to the surface areas of the strap -- not an esoteric fabric, but rather quite basic -- and the foam padding.

 

Hi again Steven,

 

That purple/magenta cast issue (in blacks in particular) definitely does seem to be related to the IR sensitivity. Again, I think it can be nailed with revised color profiling.

 

Cheers,

 

Sean

[grober]

Another bonus of going digital with the M8 (or R-D1): freedom from all the color corrective filters we carried around in various diameters to fit all our lenses when used on film bodies.

 

There may be times when a certain filter is still a good idea, but I'll enjoy the added color control I'll gain with the M8.

 

-g

[ho_co]

Steven & Simon--Grungewumpus!

 

Have you read Leica's own announcement on the issue? If not, don't jump in with opinions of what they should and shouldn't do.

 

How do other companies handle the issue? Well, have you noticed that although you can manually set white balance to common home tungsten color temperature on the D200, Nikon's auto white balance specifically doesn't go that low. It can't disappoint because it just doesn't work.

 

Leica is aware of and working on the problem and has posted that fact on their web site. As Sean pointed out, one can't judge results without using a specific white balance procedure. And as Leica points out, one shouldn't even be trying to comment on white balance unless using Raw capture.

 

Obviously, Epson uses different materials for different parts of their cameras. Under certain light our eyes may mistake them for the same colors, under other light they may appear quite different. Remember, the RGB system is only supposed to be an approximation to what our eyes see. Remember, there were some colors that we could see but certain films (the Kodachromes, as I recall) couldn't capture.

 

So get off your high horse and consider the problems before you start complaining.

 

As you can see, I'm unhappy because I don't have my M8 yet.

 

--HC

[sean_reid]

My understanding from a different thread on this matter is that Leica will sell an "IR barrier filter"

 

will sell ?

 

If this is a flaw, no one should have to pay to get it fixed...

 

At this point, it would also make sense for Leica to issue some formal communications related to these problems so future potential buyers get reassured as to the reliability of their product. This is a high-end product for which people have high expectations.

 

Roger

 

They have and I quoted them specifically on this issue in the last revision of my M8 article (Part 3).

 

I don't know what level of color-management experience various members of this forum have but based on some of the comments I've been reading in various threads, I think that some people are going to be very surprised by how much of this IR/magenta cast can be dealt with simply using revised color profiles for RAW conversion. This is not as big a deal as some seem to fear.

 

For those who are new to RAW conversion color profiles, they don't relate at all to a firmware change. The firmware change can help to improve the accuracy of auto-WB and the WB presets (in-camera for JPEGs). The RAW conversion color profiles, however, are references that C1 and other RAW conversion programs use to interpret the RAW data when making color decisions. New profiles can come from C1 itself, commercial profile makers or even knowledgeable members of this forum who have access to C1 Pro or the like. In fact, I have a friend who is a color management expert and I'll be sending him some M8 RAW files soon so that he can look at tweaking profiles.

 

Cheers,

 

Sean

[innerimager]

Hi again Steven,

 

That purple/magenta cast issue (in blacks in particular) definitely does seem to be related to the IR sensitivity. Again, I think it can be nailed with revised color profiling.

 

Cheers,

 

Sean

I'm sorry Sean but this makes no sense to me. Over sensitivity to IR is not a WB issue, and Leica wouldn't be considering using a filter (hot mirror I assume) for the problem if it were. Incorrect WB could make it worse if the red channel is too saturated but the inherent problem would be there. This is reminiscent of the problem the D2H had, I certainly wouldn't want to have to use an external filter to deal with it. Your lovely fruit image is not an appropriate comparison for IR sensitivity which renders black fabric, in particular, purple. We wedding shooters had many a correctly white balanced image have purple tuxes when the groom bought black with the D2H. best...Peter

[jrgeoffrion]

I added a post to the warm and fuzzy earlier today, and am saddened to change my tune, but the first test shots in well lit tungsten light simply speak for themselves. This is a wonderfully crafted modified M7 body (as we know) that has a digital capture that isn't ready for release. The shot is Auto WB processed in C1 3.7.6 with no adjustments. The shot using tungsten WB is worse. We all know what an RD1 looks like. The strap is black and the foam blue in real life.

 

It seems as if the M8 is really only usable at the moment in daylight. Understand that I could correct in post. But that is unacceptable in a $400 camera made in 2001, let alone an M8. It's as if they built an M7 that only could use one type of film. And hard to believe that Leica was not aware that the M8 did not function correctly in tungsten light. This is simply different from banding -- which I also found -- and cannot be avoided in tungsten, and, as you can see, is quite obvious in its impact.

 

With respect to waiting for a firmware update, just ask the R9 owners who have chosen not to install the latest firmware update because of the problems it caused.

 

I'm very saddened to find this. But for those still waiting to purchase, thought worth posting.

 

This looks like IR to me. Basically, the sensor sees light that our eyes don't see and calculates 'false colors'. It may appear in other situations too and, in portraits, will typically give a magenta bag under the eyes. Changing the profile won't eliminate the issue if it's IR.

 

I don't know how many lenses you have, but if you have 1 or 2, the simple fix is to stick a B+H IR cut filer (#486) and it will solve this issue if it is IR.

[sean_reid]

I'm sorry Sean but this makes no sense to me. Over sensitivity to IR is not a WB issue, and Leica wouldn't be considering using a filter (hot mirror I assume) for the problem if it were. Incorrect WB could make it worse if the red channel is too saturated but the inherent problem would be there. This is reminiscent of the problem the D2H had, I certainly wouldn't want to have to use an external filter to deal with it. best...Peter

 

Hi Peter,

 

I'm sorry that it doesn't make sense to you but check back in with me after I've gotten some good color profiles to try on these files. Then we'll see how they do. Again, I think some people may be surprised.

 

Look over my posts above. There are two separate issues. The first is getting a proper WB so as to have a baseline. The second is dealing with the IR sensitivity which, again, I suspect can be dealt with through color profiling. I'll know more when my color-management wizard friend can take a crack at some M8 files.

 

Cheers,

 

Sean

[Advertisement]

Hallo,

Look here: Interesting Leica gear

[jrgeoffrion]

Sean: OT: Did you have all these fruits and vegetables on hands? I'm lucky if I have an apple and a banana around.

[Guest stevenrk]

Steven & Simon--Grungewumpus!

 

Have you read Leica's own announcement on the issue? If not, don't jump in with opinions of what they should and shouldn't do.

 

How do other companies handle the issue? Well, have you noticed that although you can manually set white balance to common home tungsten color temperature on the D200, Nikon's auto white balance specifically doesn't go that low. It can't disappoint because it just doesn't work.

 

Leica is aware of and working on the problem and has posted that fact on their web site. As Sean pointed out, one can't judge results without using a specific white balance procedure. And as Leica points out, one shouldn't even be trying to comment on white balance unless using Raw capture.

 

Obviously, Epson uses different materials for different parts of their cameras. Under certain light our eyes may mistake them for the same colors, under other light they may appear quite different. Remember, the RGB system is only supposed to be an approximation to what our eyes see. Remember, there were some colors that we could see but certain films (the Kodachromes, as I recall) couldn't capture.

 

So get off your high horse and consider the problems before you start complaining.

 

As you can see, I'm unhappy because I don't have my M8 yet.

 

--HC

 

HC, with all due respect, I appreciate WB. I nearly always use selective manual WB (including the use of color WB meter in critical applications). But that's not the point here.

 

I would not dream of comparing the M8 to a prosumer cam like a D200. Don't mean to insult Nikon users, but I would compare it to the R9, or the 1Ds (I and II) and Phase One and Leaf. (Unless the M8 is breaking the M tradition, which it clearly is not.) And compared to these, what we are seeing here is not acceptable. Certainly many of us can work aroung this. Sean's shots (including the lovely daylight shots) and others posted on the forum speak to this. But a professional level camera in late 2006 that can't produce even a simple black is not ready for release. period.

 

And at the same time, I wish Leica all good fortune in pulling this off. What they are trying to do -- bring us a professional level digital RF with Leica quality and Leica lenses -- is admirable and important. But that simply does not explain why they didn't wait until they had this sorted out before releasing the camera. Better -- and more professional -- to have delayed the introduction.

 

And Sean what I don't understand is that we were told that C1 was producing great results. Leica apparently knew all the time though that these problems existed. How can we now trust that some future fix will solve this, and not create additional problems as the R9 firmware update has?