What is the problem with IR filters?

[bebert]

Advertisement (gone after registration)

Hello everyone,

 

Having ordered an M8 in Paris and looking now for the miraculous IR cut filter, I have a question : what is the problem of putting an IR filter on Leica lenses with the M8 (I am not talking about the additional price to pay for this filter and do not want to enter into a discussion about Leica's responsibility)?

 

Is it really detrimental to the quality of the pics (in particular with respect to focus and sharpness)?

 

I have read a lot of negative opinions on this forum concerning this solution as being something provisory and actually not 100% satisfactory.Why?

 

IMHO what is key is the quality of the final picture with or without filter.

 

Jean-Marc Humbert

[Vivek Iyer]

IR cut filters will ONLY improve color fidelity and sharpness.

 

No idea about what 'detrimental' effect you are referring to!

 

There is a bunch of folks who complain that they have to put a filter on their lenses

(most leica lenses that I see getting traded always boast of having used them with

UV "protection" filters. Now all the non filter-puritans are coming out of the wood works!).

 

Another bunch is complaining that they have to buy extra filters.

 

Since you are concerned with neither, you should be just fine. Have fun!

[plasticman]

and another bunch is complaining that there's a detrimental effect on color fidelity in the corners of wide-angle lenses.

 

But that is just whining, of course.

[Vivek Iyer]

Mani, Is there any demontration (that filters affect the color fidelity on wide angle lenses)

to that effect or just speculative whining? How is the color fidelity of the images from wides without the IR cut filters? Better or worse?

[scott kirkpatrick]

and another bunch is complaining that there's a detrimental effect on color fidelity in the corners of wide-angle lenses.

 

But that is just whining, of course.

 

No, it's real. See the "cut filter" thread for examples at 21 mm. This is the reason that Leica asked Kodak to put the weaker (but less angle-dependent) IR filter into the KAF 10500 imager that is used in the M8. They hoped that an absorbtive filter, which is less angle sensitive, would be easier to deal with for all kinds of lenses (not just coded ones), than an interference filter as is used in the similar chip that is in the DMR.

 

The dichroic IR filter seems to work just fine in cleaning up the IR false color problem for lenses of 28mm and longer focal length. For wider angles you need some post processing action to reduce the cyan shift around the edges. The effect is magnified for the ultra-wides, compared to what we would have seen if the dichroic filter had been on the sensor, because these are retrofocus designs. The range of angles at which light strikes the sensor is less than the range of angles at which light enters the front of the lens.

 

scott

[john_r_smith]

What I want to know is why the hell Leica don't sack their existing R&D team and put Scott in charge at Solms. And this is not meant to be a sarcastic comment, either. Every time Scott comes up with something, it's sensible.

 

J

[rick_dykstra]

Advertisement (gone after registration)

What I want to know is why Leica haven't made a positive out of this 'feature'. It all about the imaging performance. So why not be up front about it, explain why it is the way it is and offer a range of well priced IR filters to get the most out of the system?

 

Unless of course some clever marketing types worked out that the hoo-haa that would break out all over the web as users discovered the 'feature' would generate a great deal of passionate, animated discussion? Which in a couple of weeks would end up with users talking about the positives that result from the design philosophy. Is it not better to allow the market to discover how good a new product really is despite initial hiccups, than to try to force feed the masses with the benefits of a design peculiarity?

[plasticman]

But that is just whining, of course.

 

this was ironic, of course.

 

or just speculative whining?

 

I actually take personal offense at your phrasing.

 

Your response is PRECISELY the sort of thing that i think is wrong with this 'gung-ho', 'let's all be positive', there's-no-problem attitude which we're all being bullied into adopting on this forum.

 

Because of light-incidence issues, corners of wide-angle lenses are seriously detrimentally affected by IR filters. Simple fact.

But don't let the 'facts' get in the way of your cheerleading.

[gravastar]

One problem I've experienced (with the Epson R-D1) is that the filters when viewed off axis look like a highly reflective magenta colored mirror. Several times people I've photographed have commented on this. It detracts from the unobtrusiveness of a rangefinder. It's not what people expect to see.

 

Bob.

[Vivek Iyer]

Mani, Why take personal offense when your original comment itself is attributed to a group rather than yourself.

 

Let me clarify a couple of things:

 

1. I am not cheerleading a $5,000 10mp camera.

 

2. What happens to color fidelity with the wides when an IR-cut filter is not used (topic of the thread)? The fringing still exists, doesn't it?

 

On the DSLRs which use retro focus design wide angle lenses, color fringing is a well known problem. It is also known that the color fringing varies from lens to lens. There are wide angle lenses with minimal (or almost no) chromatic aberrations (CA) and others with increased CA.

 

All the discussion about the weak IR cut filter, IR leak, color casts, etc have already been played out on the Epson R-D1. No one bothered to notice it to be this big of a problem then (It wasn't exactly a cheap camera when in debuted!).

 

No one has, so far reported on the new tri-elmar and how it fares with the M8.

There is also no comparative evaluation of wide angles on the M8.

 

So, generalisations about wide angle lenses on M8 should be dealt with caution.

[peterv]

The dichroic IR filter seems to work just fine in cleaning up the IR false color problem for lenses of 28mm and longer focal length. For wider angles you need some post processing action to reduce the cyan shift around the edges. The effect is magnified for the ultra-wides, compared to what we would have seen if the dichroic filter had been on the sensor, because these are retrofocus designs. The range of angles at which light strikes the sensor is less than the range of angles at which light enters the front of the lens.

 

Scott, you apperantly you know quite a lot about this;

 

Does this mean that in your opinion a filter in front of the sensor would be the best option?

 

Can this be done?

 

What would happen to present image quality?

[lct]

One problem I've experienced (with the Epson R-D1) is that the filters when viewed off axis look like a highly reflective magenta colored mirror...

Hi Bob, would you have one pix or two showing that look on your R-D1 body?

Just curious.

[jlancasterd]

IR cut filters will ONLY improve color fidelity and sharpness.

 

No idea about what 'detrimental' effect you are referring to!

 

There is a bunch of folks who complain that they have to put a filter on their lenses

(most leica lenses that I see getting traded always boast of having used them with

UV "protection" filters. Now all the non filter-puritans are coming out of the wood works!).

 

Another bunch is complaining that they have to buy extra filters.

 

Since you are concerned with neither, you should be just fine. Have fun!

 

Does the IR filter also filter out UV - Will it cut UV haze at high altitudes? If not will we get a 'UV-blue' effect?

[Vivek Iyer]

John,

 

The B+W UV-IR cut filter will filter UV. I have no idea about the UV sensitivity of the Kodak sensor (despite my attempts to find out more, see : http://filmlives.net/community//netscape6/viewtopic.php?t=261).

 

Most modern Leica lenses will not transmit much UV. If you are using old Elmars it could be a problem provided the CCD in M8 is sensitive to UV.

[alexr]

and another bunch is complaining that there's a detrimental effect on color fidelity in the corners of wide-angle lenses.

 

But that is just whining, of course.

 

It is not whining. I refer you to Guy-s comments on WA use with IR filter. he reports a cyan cast in the corners. A direct effect of the IR filters.

[plasticman]

It is not whining. I refer you to Guy-s comments on WA use with IR filter. he reports a cyan cast in the corners. A direct effect of the IR filters.

 

hehe - that is the last time i attempt irony when discussing the M8. ;-)

 

But back on topic - i'd especially like to see what the effect would be when using the CV 15 or even the 12 - which i'd intended to experiment with, when the Noctilux novelty had begun to wear off.

Anyone have experience with IR filters on lenses as wide as these?

[stunsworth]

Mani, a bit difficult as the 15mm at least doesn't have a filter thread - I don't think the 12mm has either, but I only own the 15mm. You could hold a filter in front of the lens, but that would be far from ideal.

[plasticman]

oh - one more reason why the filter solution won't work then... :-(

 

i've read some very good things about the 15 - what do you think? And do you have some samples of it in action on the M8 (or R-D1?)

 

edit: sorry, i should of course written the filter 'solution'... ;-)

[lars_bergquist]

Mani/plasticman,

 

the Cosina/Voigtländer 15 mm lens is like most other lenses of that and similar focal length: there is no filter thread. The bulging front lens is protected by a built-in shade of the now traditional 'petal shape'. In any case, tests with that lens that I have seen have been very good. I am beginning to suspect that the IR false colour problem too will be fairly adequately dealt with by firmware, i.e. profiles. It seems that already, JPEG profiles do a better job than those used with DNG. The remaining problem will probably be with incandescent lighting only, and I'm already using correction filters there!

--A 15 mm 1:4.5 lens is not exactly what you use with incandescent, and neither really are a 21 or 24 mm 1:2.8. One single 46 mm filter will take care of my 28 mm Summicron and 35 and 50 mm Summiluxes.

--On the other hand, when I get my M8 I may well buy a deep-red IR filter, set the camera to black-and white and go out run some experiments ...

 

The hopeful old man from the Age of Flashpowder

[Advertisement]

Hallo,

Look here: Interesting Leica gear

[gumshoecamus]

Scott, you apperantly you know quite a lot about this;

 

Does this mean that in your opinion a filter in front of the sensor would be the best option?

 

Can this be done?

 

What would happen to present image quality?

 

to add to what scott has written...

as I understand it, the unavoidable problem with m rangefinders is the short distance between rear lens element and film/sensor plane. it means that in the corners light strikes the film/sensor at acute angles and causes vignetting -- especially with wide angle lenses, where the light is travelling at an acute angle when it enters the lens anyway.

modern lens design makes this much less of a problem with film. however sensors are more susceptible to vignetting, and so far (even with microlenses) have to occupy less of the imaging circle than film to produce adequate results. this is probbaly the main reason why the M8 is not full frame -- I hardly think cost is a primary concen at leica.

the M8 does use a larger chip than the R-D1, but vignettes much less (see andy piper's examples with the CV 15). as I understand it, an unusually thin and weak IR filter was one of the deliberate design decisions to achieve this (and to reduce CA in the corners with wide angle lenes). the microlenses also contribute (and like the phase P30, probably improve sensitivity).

the R-D1 is sensitive to IR too, though not as much as the M8. it probably also uses a weaker-than-normal IR filter for the same reason.

the downside (which we all know about) is IR interference in images. the solution seems to be a special cut filter and some firmware/profile adjustments. I imagine the coding will allow the camera to handle cyan casts in the corners before the raw file is written to the card.

I too would rather not put a filter on the front of my lens, but I would prefer that to vignetting and CA with wide-angle lenses and/or a smaller sensor. leica really should have explained this early on -- but they didn't.

I will be interested to see if canon -- with all its R+D budget -- can overcome this inherent M rangefinder problem if it lauches its own DRF, as is rumoured.

dave