IR not just black -- faces take on red blotchy quality?

[Guest stevenrk]

Advertisement (gone after registration)

The photograph at the top of this recent thread seems to confirm something I've read:

 

http://www.leica-camera-user.com/digital-forum/9149-m8-red-lining-maybe-new-problem.html

 

The facial image clearly presents an accentuated red blotchy quality. Apparently another consequence of the IR problem --- the IR accentuates blood and blood vessels near the surface. In many ways even more significant a problem than the other results of the IR sensitivity, at least if you are planning to use the M8 to photograph people in color. Maybe impacting BW shots also?

 

And it causes human skin to appear "blotchy". Areas with less subsurface blood appear unusually light, areas with increased subsurface blood appear as darker blotches. (Such irregularities in blood concentration can be due to circulation problems, or to quite normal physiological reactions that constrict or dialate capillaries). There's also a tendency for surface veins to be unusually prominent, especially in fair haired people.[/Quote]

 

Here's the link for the quote above and a full discussion of the effect of IR sensitivity on various colors, including the now infamous black:

 

Comments on the Leica statement...: Leica Talk Forum: Digital Photography Review

 

Anyone else run in to this?

 

Related question: for anyone using an IR cut filter, does it give off a red mirror like reflection to someone who is looking into the lens or just generaly in your direction?

 

Steven

[stunsworth]

Perhaps he just had a blotchy red face. Then again you may be correct, since I wasn't there, don't know the man and don't know how the conversion parameters were set I can't say.

[chrism]

Yet the face shown in part in the photo above the one you're referencing doesn't have any blotchiness or even undue redness, and it should at least show something if the problem below is all camera generated.

 

Chris

[Jamie Roberts]

Ok let's put a stake in this one, ok?

 

I'm pale, and my son is paler. His skin is red, and if anyone was going to blotchy due to IR, it would be me--or him. In fact, I have rosecea, and am blotchy a bit.

 

Now please look carefully at this M8 shot:

 

 

 

Now look at the 100% crop. Apart from the very apparent "razor sharpness" and lack of noise from the 50 1.4 summilux--at 1.4, mind--do you see any artifacts? If you do, you need a new monitor, guys.

 

 

 

How about magenta? No?

 

Well, that's because flourescent light doesn't have much. But this is also with a good c1 profile, made for the--wait for it--the ETC 5d profile.

 

The C1 profile for the M8 is terrible. Just terrible, as is usual with C1's first profile efforts--believe me. I love C1, but they're far from great getting it right the first time with profiles.

Ask anyone with a Canon 1 Series.

[Guest stevenrk]

Ok let's put a stake in this one, ok?

 

I'm pale, and my son is paler. His skin is red, and if anyone was going to blotchy due to IR, it would be me--or him. In fact, I have rosecea, and am blotchy a bit.

 

Now please look carefully at this M8 shot:

 

Now look at the 100% crop. Apart from the very apparent "razor sharpness" and lack of noise from the 50 1.4 summilux--at 1.4, mind--do you see any artifacts? If you do, you need a new monitor, guys.

 

How about magenta? No?

 

Well, that's because flourescent light doesn't have much. But this is also with a good c1 profile, made for the--wait for it--the ETC 5d profile.

 

The C1 profile for the M8 is terrible. Just terrible, as is usual with C1's first profile efforts--believe me. I love C1, but they're far from great getting it right the first time with profiles.

Ask anyone with a Canon 1 Series.

 

Thanks Jamie, that's helpful. If you find yourself in a spot where the lighting is conducive to IR (such as with room lighting) and feel like taking another shot of your son, that would be helpful to. The shot I saw on the other thread and the thread describing the effect of IR seemed to connect, but maybe not. Or maybe not correct.

 

I also was thinking back to the D1X and then the 1DsII, and the 1X had a real problem with enhancing the blotchiness in skin, while the MkII (and MKI) do not. Interesting that we now find out the D1X was much more sensitive to IR than the Canons. Never knew what caused it, just knew it was and it went away with the 1D.

[Jamie Roberts]

{snipped}I also was thinking back to the D1X and then the 1DsII, and the 1X had a real problem with enhancing the blotchiness in skin, while the MkII (and MKI) do not. Interesting that we now find out the D1X was much more sensitive to IR than the Canons. Never knew what caused it, just knew it was and it went away with the 1D.

 

Steven--I think it's all a trade off...

 

Ok, not going to let Leica off the hook here. There is stuff that needs fixing.

 

But the 1d2 and 1ds2 were quite sensitive to IR in the Caucasian skin range. Much virtual ink was spilt on this over on the RG forum. Of course, you quickly learn to work around it (though a lot of people didn't, and downgraded their 1ds2s to 1ds).

 

Interesting, eh?

 

The funny thing is, skin on Canons isn't blotchy because, well, it's often "smoothed" due to high DR and less-than-optimal color workflow Oh, and the AA filter too, I guess... but I was never convinced of that, honestly.

[Guest stevenrk]

Advertisement (gone after registration)

Steven--I think it's all a trade off...

 

Ok, not going to let Leica off the hook here. There is stuff that needs fixing.

 

But the 1d2 and 1ds2 were quite sensitive to IR in the Caucasian skin range. Much virtual ink was spilt on this over on the RG forum. Of course, you quickly learn to work around it (though a lot of people didn't, and downgraded their 1ds2s to 1ds).

 

Interesting, eh?

 

The funny thing is, skin on Canons isn't blotchy because, well, it's often "smoothed" due to high DR and less-than-optimal color workflow Oh, and the AA filter too, I guess... but I was never convinced of that, honestly.

 

Thanks Jamie. Absolutely, none of it is ever free:)

 

On the MkII, I remember some of that ink, and agree that was never clear, and mostly takes just getting to know and appreciate the beast -- or buy a DMR. I stuck with the beast.

[Joseph S. Wisniewski]

Related question: for anyone using an IR cut filter, does it give off a red mirror like reflection to someone who is looking into the lens or just generaly in your direction?

 

It's less visible when the lens is pointed directly at a centered subject, more visible when the subject is off center, and most visible to people out of frame at 30-45 degrees off axis.

 

What I found got the most attention from street subjects and event (wedding and corporate) subjects was bringing a camera to bear, either up from a relaxed holding position, or a panning motion. As you hit that 30 degree point, and they saw the big red "flash", they got quite alarmed.

 

In the studio, it was when changing lenses...

[Joseph S. Wisniewski]

Ok let's put a stake in this one, ok?

Delighted. That would involve actual evidence, in short, a comparison between an IR and non IR situation. Can you do the same shot under problem light (sunlight, strobe, or incandescent) both with and without an IR blocking filter with minimum subject motion or composition change between the shots? Or maybe the same shot under both problem and no problem (fluorescent) light?

I'm pale, and my son is paler. His skin is red, and if anyone was going to blotchy due to IR, it would be me--or him. In fact, I have rosecea, and am blotchy a bit.

 

Now please look carefully at this M8 shot:

 

Now look at the 100% crop. Apart from the very apparent "razor sharpness" and lack of noise from the 50 1.4 summilux--at 1.4, mind--do you see any artifacts? If you do, you need a new monitor, guys.

Or just a more open attitude. I see three artifacts: veins left of eye, unusual dark circles (for a child) right of eye, and red cheeks that look line windburn.

 

Again, you didn't try a comparison with and without filter. My own experience with IR contamination is that the picture would have noticeably improved (less veins, dark circles under eyes, and "wind burn") with the filter.

 

How about magenta? No?

 

Well, that's because flourescent light doesn't have much.

Anyone here working exclusively under fluorescent light?

[Jamie Roberts]

Delighted. That would involve actual evidence, in short, a comparison between an IR and non IR situation. Can you do the same shot under problem light (sunlight, strobe, or incandescent) both with and without an IR blocking filter with minimum subject motion or composition change between the shots? Or maybe the same shot under both problem and no problem (fluorescent) light?

 

Joseph--delighted you may be, and right about actually making a worst case situation, which I did not do, but in this case, I believe, you're probably quite wrong.

 

We'll see though. So far, what I've seen is that the IR shift is is not in skin tones. I will post under problem light later (travelling next week). If I can get my hands on one of the right filters (they're a little hard to come by right now) I'll do an AB comparison.

 

My personal comparison is with film plus the Canon 1 series stuff--but they read IR too, even if they're not as sensitive in the same places.

 

Or just a more open attitude. I see three artifacts: veins left of eye, unusual dark circles (for a child) right of eye, and red cheeks that look line windburn.

 

Um no--you actually need your attitude opened here, if anyone needs it I know what my son looks like, and wouldn't have posted if all of those things weren't actually quite visible.

 

The veins left of eye are visible to anyone. Like mine, they're very close to the surfavce of his skin. They are not IR artifacts in the slightest. The unusual dark circles are, in this case, simply shadows. And finally, yes, the 5d profile I used for the M8 is a little red. But his cheeks are actually quite red, as kids his age and skin tone actually can be. He's been sick with a cold, and is a little flushed to boot.

 

Not the camera. But, again, it's flourescent, and you're right, you don't see that much problem there.

 

Again, you didn't try a comparison with and without filter. My own experience with IR contamination is that the picture would have noticeably improved (less veins, dark circles under eyes, and "wind burn") with the filter.

 

Joseph--you're probably right about the filters being an even better proposition, but I'm honestly not sure how much IR bleeding there was in this scene. I just haven't seen any blotchy faces due to IR with the M8.

 

Much magenta in neutrals. But skin tones appear so much better than the Canons that I'm prepared to live with it.

 

On the other hand--once Leica announces its fix--I have a feeling I'll be using IR cut filters anyway

 

Anyone here working exclusively under fluorescent light?

 

Well, of course not. That's why I mentioned it in my post.