IR Sensitivity Advantages

[edlbell]

Advertisement (gone after registration)

I hope this topic has not been covered, but since we have the feature I would like to learn more about IR.

 

What are the advantages to having a camera that is IR Sensitive?

 

I shot some stars with the M8 and noticed that some of the stars, or maybe they are planets, are red and some are yellow. I assume this is because the camera has an IR sensitivty feature!!

 

Ed

 

Please do not use this thread to re-hash "old" issues!!

[jaapv]

I'm waiting for my IR filter to do IR photography....

[Jean LeBlanc]

Since infrared wavelengths refract differently than visible wavelengths, IR can only degrade images with visible light. If you look at most high quality lenses, you'll see a little red dot on the focusing scale. That's for infrared photography. You first focus the lens as normal. Then, let's say you're focused at exactly ten feet. You move the lens from the ten foot position to the red mark on the scale. Now the lens is properly focused at ten feet for infrared light. Infrared film is shot with very dark red filters that eliminate visible light (shorter wavelengths, red through violet) and of course UV so only infrared light reaches the film. If visible light could reach the film, it would be out of focus at ten feet, and your subject would be soft, the infrared light being in focus, but visible wavelenths out of focus.

 

So it is a bit of a puzzlement how Leica thought that by allowing more IR to reach the sensor, that would improve image quality. Clearly the amount of infrared light getting through isn't enough to make the M8's images soft, but exactly how it makes things better isn't at all clear.

[marknorton]

I have this feeling that IR photography is a bit like buying a fish-eye lens. It's great for a couple of days, then you've been there, done that, and it's just tedious. The fact that the M8 can unexpectedly be used for IR photography is a pretty marginal benefit in my book.

[jaapv]

I have this feeling that IR photography is a bit like buying a fish-eye lens. It's great for a couple of days, then you've been there, done that, and it's just tedious. The fact that the M8 can unexpectedly be used for IR photography is a pretty marginal benefit in my book.

So it is. But it is but one filter...

[marknorton]

So it is a bit of a puzzlement how Leica thought that by allowing more IR to reach the sensor, that would improve image quality. Clearly the amount of infrared light getting through isn't enough to make the M8's images soft, but exactly how it makes things better isn't at all clear.

 

I don't think that's the correct way to look at it. Leica wanted to filter out IR but other constraints prevented them using an IR filter which was fully effective and their judgment was to let the IR stay to avoid compromising the image quality.

 

Everyone focusses on the black fabrics but it also seems to be messing up white balance and colour generally including foliage. I'm also concerned that without the filters, there's an out of focus IR induced haze in the images.

[drjon]

Advertisement (gone after registration)

I don't think that's the correct way to look at it. Leica wanted to filter out IR but other constraints prevented them using an IR filter which was fully effective and their judgment was to let the IR stay to avoid compromising the image quality.

 

Everyone focusses on the black fabrics but it also seems to be messing up white balance and colour generally including foliage. I'm also concerned that without the filters, there's an out of focus IR induced haze in the images.

 

Based on the pictures with the M8 posted so far on the web, your concern about an "IR induced haze" appears groundless. The M8 AWB needs refining but this is unlikely to be related to IR. As you indicate, Leica chose relatively more IR contamination as the trade off to ensure as little inteference between the M lenses and the sensor as possible - a sound choice in my view.

[marknorton]

Errr.... take a look at the pictures which show what the camera is recording in infra-red by using a visible light blocking filter. What you see is a purple out of focus image. With no filters, that image will be combined with the visible light image. It may be several stops down but it will be there. So I respectfully disagree with you.

[scott kirkpatrick]

Errr.... take a look at the pictures which show what the camera is recording in infra-red by using a visible light blocking filter. What you see is a purple out of focus image. With no filters, that image will be combined with the visible light image. It may be several stops down but it will be there. So I respectfully disagree with you.

 

Also note the increased apparent sharpness of the first few high-detail street shots that appeared once Guy Mancuso got his filters. That's the result of removing the "IR haze" a few stops down in the Arizona sunlight.

 

scott

[drjon]

Errr.... take a look at the pictures which show what the camera is recording in infra-red by using a visible light blocking filter. What you see is a purple out of focus image. With no filters, that image will be combined with the visible light image. It may be several stops down but it will be there. So I respectfully disagree with you.

 

It will be there in principle in those areas of the particular image containing the particular fabrics with the particular IR-relective properties. And several stops down in those areas. Perhaps that is why I cannot see this "IR-haze" on the pictures I have seen posted on the web. As you say, we can respectfully agree to differ on this point as it would need some sophisticated measurement of so-called "sharpness" to really identify this extra sharpness with the filter - and/or some pretty large prints done with some pretty good printer so we could actually perceive it clearly. Or maybe my eyes just can't hack it anymore.

[plasticman]

Can someone possibly clear up a genuine mystery for me in all this: the reason why the IR filter was left off the sensor was apparently in order to NOT sacrifice lens detail and sharpness, but (as Mark says) i would agree that apparent sharpness is actually increased by the addition of the IR-filtering. And the areas where this is true are definitely (imho) not restricted to patches of synthetic material - Guy's and other's shots have shown that there are significant differences in color, contrast and sharpness across the frame.

 

Wouldn't it therefore be at least theoretically possible to contrive a compromise solution where a (relatively weak) IR filtering layer was added to the sensor, without sacrificing detail-rendition?

[jaapv]

IR filtering would have been by interference filter. Which means it is sensitive to angle of incidence. Which means the undesirable effects would have been quite apparent in the corners and edges of the sensor on a rangefinder camera, with the lens close to the sensor, which produces acute angles.

[blakley]

Mani, the sensor has a "relatively weak" IR absorptive filter on it already.

[marknorton]

Mani, I think it's a difficult balancing act. On the one hand, they want to filter out IR because of the colour shift and also because the IR image formed by the lens is out of focus compared to the visible light image. If the sensor is sensitive to both, it will render both and the IR element will show up as fuzziness in the image, a haze of overall softness.

 

On the other hand, if they apply an aggressive IR filter, the IR will go but there will be image problems with the visible light image, colour fringing and astigmatism, made worse by the particular geometry of M lenses.

 

I now believe then that Leica have done the right thing to keep the cover glass as thin as possible - to preserve the visible light image - and that an external filter taking out the IR will lead to an overall improvement in sharpness because it will take out the IR fuzz.

[plasticman]

I was a little unclear i think - what i meant was that it must be possible to find an optimum compromise solution, with moderately increased IR-filtering increasing sharpness, as against the possible degradation of the lens detail - thereby achieving the status quo in sharpness, with slightly less sensitivity to IR.

 

But the angle of incidence is obviously the biggest obstacle to this... and thereby maybe the microlens solution to vignetting needs an IR counterpart, in the next release of the camera...?

 

Mark: we were posting at the same time.

I see your point - my only reservation here is that my first (and so far only) lens bought in anticipation of the M8 was a Noctilux - and i'm very, very reluctant to stick any filter on the front of that.

At the moment i'm back on the fence with regard to the M8.

[lars_bergquist]

--------------------

 

But the angle of incidence is obviously the biggest obstacle to this... and thereby maybe the microlens solution to vignetting needs an IR counterpart, in the next release of the camera...?

 

I really do think that if a ready solution to that problem had been available, then Leica would have implemented it. Now we have to live with the quirks, or use film -- but colour film does have some quirks too! It's just that we have come to an accommodation with these quirks, one that we have internalised to the degree that we simply don't notice them anymore!

 

I do however have a strong suspicion that the M8 was designed to be 'upgradeable', the way the old screw-mount cameras were. I was in fact more or less told that this is the case. So I believe that as improved hardware (sensors, electronics) becomes available,

we can have tit installed. This would in fact be a very sensible strategy for Leica, as it would ensure customer loyalty.

 

The old man who couldn't afford a IIIf when in his teens ...