Definitive answer on the IR filter.

[pat308]

Advertisement (gone after registration)

Has anybody seen a definitive statement that the IR filters solve the problem with IR sensitivity on the M8? I thought maybe Sean Reid, Erwin Putts, or somebody else with an M8, a range of IR Filters and an objective test setup might have made a comment by now.

[Guest guy_mancuso]

I have made hundreds . Look around there is a Cut filter thread or any thread i have started has shots with the filters on

[pat308]

I have read all the the threads. I have seen a lot of photographs made with various filters.

 

I was looking for something objective.

[Guest guy_mancuso]

I just put the green test thread and the cut filter thread back to the top but i don't know what you mean by objective . One with and one without is pretty objective to me

[waterlenz]

Has anybody seen a definitive statement that the IR filters solve the problem with IR sensitivity on the M8? I thought maybe Sean Reid, Erwin Putts, or somebody else with an M8, a range of IR Filters and an objective test setup might have made a comment by now.

 

I believe the Leica statement is about as a definitive statement as you are going to get.

Filters are it! What is unknown is if the B+W 486 filters are the same as the filters that Leica will be providing. This is important because they may require different profiles and cyan correction functions, if they are not the same.

Tom

[Guest guy_mancuso]

Tom i can't imagine there not to be honest. leica does not make filters that i know of and they vend this out the 486 seems to be the only one that works correctly that i know of.

[waterlenz]

Advertisement (gone after registration)

I have read all the the threads. I have seen a lot of photographs made with various filters.

 

I was looking for something objective.

 

What do you mean by objective? Yes, photo evaluations are fairly subjective. The Reid Review comparison pix have numerical values, is this what you want?

Tom

[waterlenz]

Tom i can't imagine there not to be honest. leica does not make filters that i know of and they vend this out the 486 seems to be the only one that works correctly that i know of.

 

 

Guy,

I guess we will find out. It would be nice if they told us it was OK to go ahead with the 486. The Heliopan cut filters may not be the same as the 486 for example.

Tom

PS. Good luck with NJ tomorrow!!

[Guest guy_mancuso]

Thanks Tom . I hope they are B+W also, I have every lens with a filter now

[rvaubel]

Thanks Tom . I hope they are B+W also, I have every lens with a filter now

 

I think what Pat meant by "objective" is quantitiative. We know from everything you and others have done that the B+W 486 filters work at least qualitatively. But I have the Heliopan filters (but no M8) as does Jorge of Rangefinder formum who claims that the Heliopans do a better job rendering black. I know that the Heliopans have a different cut frequency than the B+W so its possible that they would be a better choice. They are beatifully made with anodized brass threaded rings which prevents galling against aluminum lens threads. They also have some sort of cool coating over the interference layer that is supposed to resist soiling. But they are very expensive.

 

Somebody needs to do a comparision test of the various filters available. I certainly hope Leica does but I have a feeling that since the B+W work that they will probably contract with them to get some filters in a hurry.

 

I wish I had an M8. But then, so do you

 

Rex

[jaapv]

Leica does not make their own filters, but buys them afaik at B&W. There is a fair chance that the Leica filters will be identical to the 486. Anyway, who cares as long as they do the job, which they do.

[pat308]

Guy,

I guess we will find out. It would be nice if they told us it was OK to go ahead with the 486. The Heliopan cut filters may not be the same as the 486 for example.

Tom

PS. Good luck with NJ tomorrow!!

 

This was partially my point. The filters out there seem to cut the IR at different levels. What does the M8 need in order to cut the IR that is causing its sensor problems. Some of the pictures I have seen did not seem to clear out the magenta cast completely. I didn't say this in the original post because I didn't want people jumping all over me.

 

Using a filter is an OK solution. But using a filter and still have trouble is not OK.

 

Will Leica be providing specifications for an IR filter to its chosen manufacturer for those filters they are going to be shipping/giving away? If any filter will do OK, but I don't know that without some pretty rigorous testing and Leica does not seem to be saying anything beyond filters will be available in February. If any old IR filter worked I would expect Leica to make that point so the number of free filters it has to give away between now and when their own production is ramped up in February is reduced.

 

I am not negating the work of Guy here or Jorge over at Rangefinder. I was looking for something more formal.

[Guest guy_mancuso]

Pat no word on what filter they will be using. The only 2 that i know or seen work is the B+W and Heliopan. I would assume they will use the B+W since they are a preferred vendor. Now getting rid of the cast completely after the filter application probably at this point just comes down to better profiles which fingers crossed will be here very soon. That has been my biggest issue is the profiles. Seems we have everything else in order except the profiles. Not to say the fix not needs to be done but the profiles and firmware updates should be a priority at this point. At least in my mind you can work today before the camera gets the fix from Germany for the streaking if you have a cut filter,profile and the latest firmware update we really should be in excellent shape to work today. I believe leica recognizes that part now. I am hoping Dec. 6th besides further info and registering the camera for the recall that they do put up the latest firmware to download and C1 has better profiles to update as well. That is at least what I am hoping for on Wednesday. if they can at least get that in our hands we can work so much better. Than all that needs to be done for us is the repair

[cme4brain]

Has anybody seen a definitive statement that the IR filters solve the problem with IR sensitivity on the M8? I thought maybe Sean Reid, Erwin Putts, or somebody else with an M8, a range of IR Filters and an objective test setup might have made a comment by now.

 

It is obvious that IR filters are required for rendering true (black) colors in the M8 when photographing some objects as the Kodak chip is sensitive to IR light. I am a serious amateur photographer, not with the qualifications of some of the other users here, but I have researched the professional reluctance to use filters on lenses (causing addtional reflections and flare). It seems that the use of IR cut filters on the lens, that reflect wavelengths rather than absorb them, are much less resistant to this.

 

Apparently most digital camera maufacterers use a (relatively) thick glass in front of the image sensor to block IR light, something Leica could not do due to the exit pupil constraints. Thick glass alone in front of a sensor blocks IR light, with little additional intervention needed (from a non-engineer). I would wonder if Leica could coat the very thin sensor glass with the same coatings that B+W does to make their IR filters. That way, the IR is blocked at the chip, and there is no "wide angle cyan" effect as acute angle light rays in wide angle lenses cause red to be reflected as well. I would think the image sensor IR coating thickness surely cannot affect the vignetting problems of the chip (as coating are nanometers in thickness).

[stevme]

Pat no word on what filter they will be using. The only 2 that i know or seen work is the B+W and Heliopan.

 

If I may throw my 2/100 of a Euro into this discussion, its seems to me that both Guy and our interlocutor, Neuer Benutzer, Pat308, have a point. Guy has been saying that the problem is now profiles, since the B+W filter cuts the IR. Pat is looking for something measured. We do not know at this point that the IR reflectivity of the B+W 486 filter precisely removes IR frequencies to which the M8 sensor responds. It may remove too much and it may be too little. In either case, if there is a mismatch, we have incorrect visual information being reported across an entire image that a profile cannot correct, since profiles do not respond only to IR light, but fasten on any magenta. So we need mathematical precision to construct a filter that feeds a visual-only image to the profile.

 

The profile, in turn, is not quite "there" yet, even with the B+W 486 filter, hence Guy's comments on needing a profile. To my eye, using a recently calibrated monitor, the 486 filtered images look "hard" for lack of a better word, color changes appear too abrupt -- the changes in color gradation lack the subtlety of tone one can see in the 5D images (a girl's portrait) that was recently posted in response to a scanned film image, for example. It is as though contrast and saturation are too high. If this is not a correct description of "it", there is nevertheless something lacking in the filtered -- and profiled -- images that leads Guy and me to think more is needed. The images are "off".

 

So I think Pat and Guy are both right on this. We need to know precisely what IR frequencies the M8 sensor responds to and we need a profile that takes a visual light only image and "develops" the data into color information. It would help if an edict were issued from Solm's on what filter parameters are necessary and what color gamut, when filtered, will be delivered by the sensor so that the profile makers can do their work.

[rvaubel]

.......... We do not know at this point that the IR reflectivity of the B+W 486 filter precisely removes IR frequencies to which the M8 sensor responds. It may remove too much and it may be too little. In either case, if there is a mismatch, we have incorrect visual information being reported across an entire image that a profile cannot correct, since profiles do not respond only to IR light, but fasten on any magenta. So we need mathematical precision to construct a filter that feeds a visual-only image to the profile.

 

The profile, in turn, is not quite "there" yet, even with the B+W 486 filter, hence Guy's comments on needing a profile. .......... We need to know precisely what IR frequencies the M8 sensor responds to and we need a profile that takes a visual light only image and "develops" the data into color information. It would help if an edict were issued from Solm's on what filter parameters are necessary and what color gamut, when filtered, will be delivered by the sensor so that the profile makers can do their work.

 

The problem is that there is no one answer. There is a lot of subjectivity when it comes to color information, some people like it one way, some like it another. But let's face it, some kind of filter is required and the only two candidates that seem to be in the running are the B+W and the Heliopan. Leica is going to have to pick one or the other for expediantcy because they don't have time to develop a whole new IR filter.

 

My question is, of the two filters, is one doing a better job? This is a question that has to be answered quick because Leica is going to go down one road or the other if they haven't already. I hope the answer is the B+W 486 as it is cheaper and Leica has had a relationship with B+W in the past. But wouldn't it be a hell of a note if the Heliopan was clearly superior and Leica chose B+W because nobody raised their hand? I can just imagine the whining and bitching about the incompatentcy of Leica and the evils of their PR machine,etc,etc,etc.

 

Does anyone out their have both filters to do, at least, some informal tests? If so, don't forget to include some of those tricky greens in the test...those seem to be the killer. The Greta color chart seems pretty useless, no matter what you do it seems to come out fine if you white balance it at least.

 

Rex

[Jamie Roberts]

{snipped}the changes in color gradation lack the subtlety of tone one can see in the 5D images (a girl's portrait) that was recently posted in response to a scanned film image, for example. It is as though contrast and saturation are too high.{snipped}

 

In those images the difference is lighting, not profiles. Besides, you should never judge colour and tonal response just from the monitor, unless you're doing it numerically.

 

{snipped} It would help if an edict were issued from Solm's on what filter parameters are necessary and what color gamut, when filtered, will be delivered by the sensor so that the profile makers can do their work.

 

We don't need an edict from Solms, and sensors will also vary individually. What they need to do (and I hope for JPEG shooters they make this a switch in firmware) is update their JPEG profile to take whatever affect the filters have on visible light into account and 2) get Phase to do the same thing for RAW profiles.

 

BTW--except in the case of a workaround profile like the ones I've been doing (compensating for magenta blacks), I wouldn't expect the difference to be that great. I *do* find the base M8 profile overly yellow and green, but some people like it that way

[pat308]

We don't need an edict from Solms, and sensors will also vary individually. What they need to do (and I hope for JPEG shooters they make this a switch in firmware) is update their JPEG profile to take whatever affect the filters have on visible light into account and 2) get Phase to do the same thing for RAW profiles.

 

I was trying to have a discussion about IR filters and what the sensor needs to have blocked. I know you have worked very hard on profiles and I am sure they help, but I would like to see what frequencies need to be blocked to get a good raw capture. What does the filter need to do so what I see - through the view finder - is what appears on "film" [generally speaking].

 

Profiles to me feel like a coercion of the data, a compromise. Changes to the profile seem to effect the whole image and my understanding is that only certain parts of the image are actually generating the IR that makes it to the sensor.

[ptomsu]

In those images the difference is lighting, not profiles. Besides, you should never judge colour and tonal response just from the monitor, unless you're doing it numerically.

 

 

 

We don't need an edict from Solms, and sensors will also vary individually. What they need to do (and I hope for JPEG shooters they make this a switch in firmware) is update their JPEG profile to take whatever affect the filters have on visible light into account and 2) get Phase to do the same thing for RAW profiles.

 

BTW--except in the case of a workaround profile like the ones I've been doing (compensating for magenta blacks), I wouldn't expect the difference to be that great. I *do* find the base M8 profile overly yellow and green, but some people like it that way

 

Exactly, I like that profile, and I like it much more - sorry to say - than the profiles you created.

 

Peter

[Advertisement]

Hallo,

Look here: Interesting Leica gear

[jaapv]

BTW--except in the case of a workaround profile like the ones I've been doing (compensating for magenta blacks), I wouldn't expect the difference to be that great. I *do* find the base M8 profile overly yellow and green, but some people like it that way

It may even be a cultural thing - personally I've always preferred the more greenish-brownish Agfa style over the red and blue balanced films of Kodak and bluish Fuji films - Don't stone me - I know Kodak and Fuji make differently cast films - it is a matter of taste -but I mean as a general and historical trend. Anyway with this background I find the choice Leica made rather pleasing- desaturate your chrome profile a bit, - just fine. Or the generic M8 profile with IR filter and slightly more saturation, a sniff of cyan maybe -not bad at all.