Leica M9 with filter or au naturel ?

[Guest EarlBurrellPhoto]

Advertisement (gone after registration)

Hi Jono,

 

I would put the cut off at 75 mm rather than 35 mm. I think Leica's spec on this is too liberal.

 

Cheers,

 

Sean

 

Another place I think there's a bit of liberalism going on is the reports I've seen that say the M9's IR sensitivity is now on a par with other top-end digitals. I can't speak for the latest Nikons as I don't use them, but my Canon 1DS-III and 5D-II produce zero colour error attributable to IR (neither did the 1DS-II and 5D-I I used before them. ).

[Jamie Roberts]

{snipped} But now we've got a US$7000 camera which only partially solves the problem and at the same time complicates the use of IR filters when they are needed. For me that's a much more serious issue than the crop factor, which I'm quite used to by now.

 

From what I've seen so far this isn't true: the M9's levels of IR contamination are in line with similar offerings from Nikon and Canon. IOW, they all suffer from this to one extent or another, and some post processing is needed depending on how color-critical you need the results to be...

 

The M8, on the other hand, did have a real problem with IR / magenta contamination, especially under IR-heavy light sources. Many times careful post-processing can create acceptable results from an unfiltered M8, but the overall results are much better with the external filters.

[fotografr]

Does anybody have an idea what repair cost are to be expected for exchanging the front element?

 

Regards

Steve

 

Depends entirely on the lens. As you can imagine, the cost would be significantly more for the front element of the 0.95 noctilux than, for example, a pre-asph 35/1.4. Some front elements actually cannot be replaced at all because of the type of adhesive used to bond it in place (that according to Don Goldberg).

[Guest EarlBurrellPhoto]

From what I've seen so far this isn't true: the M9's levels of IR contamination are in line with similar offerings from Nikon and Canon. IOW, they all suffer from this to one extent or another, and some post processing is needed depending on how color-critical you need the results to be...

 

That just isn't true, at least of Canon. I can point you to at least a hundred busy pros besides myself shooting for glossy table mags and hyper-critical photo editors, who will attest they get jet-black blacks no matter what the fabric or light source, from Canon of at least the last two generations. With Nikon, I can't dispute you from personal knowledge.

 

 

Many times careful post-processing can create acceptable results from an unfiltered M8,

 

I'd agree to that as long as you add "tedious" to "careful". And the problem is, my M9-shooting colleagues are telling me that it's just as tedious to correct the less-contaminated M9 files as the more-contaminated M8 files. It isn't the degree of correction required, it's isolating the contaminated subject areas. Global tweaks mess up other colours...although granted, less with the M9 because the correction levels needed aren't as severe.

[Pointcolville]

 

In my book refinement has nothing to do with the gear you use or the filters on your lens and everything to do with the work you produce.

 

The "refined user" I was referring to always dons surgical gloves and checks the weather forecast before removing his M from his environmentally sealed alligator skin satchel.

[noah_addis]

The "refined user" I was referring to always dons surgical gloves and checks the weather forecast before removing his M from his environmentally sealed alligator skin satchel.

 

Understood... sorry I didn't get the context of your original post, it may have been cut off. Not sure how the story that ends with the front elements surviving began...

[carstenw]

Advertisement (gone after registration)

Another place I think there's a bit of liberalism going on is the reports I've seen that say the M9's IR sensitivity is now on a par with other top-end digitals. I can't speak for the latest Nikons as I don't use them, but my Canon 1DS-III and 5D-II produce zero colour error attributable to IR (neither did the 1DS-II and 5D-I I used before them. ).

 

Well, Leica claimed that the M9 is better than 2/3 of the competition, probably 35mm DSLRs. I suppose it is possible that both the 5D2 and 1Ds2 and 5D are above that. My own experience with the 5D tells me that I would sometimes see some leakage, but not that much.

[Guest EarlBurrellPhoto]

Well, Leica claimed that the M9 is better than 2/3 of the competition, probably 35mm DSLRs. .

 

In strict terms "the competition" for the M9, as a digital M-mount rangefinder, are the Epson R-D1, the M8, and M8.2 Therefore Leica's claim is not only unimpeachable, it's downright modest

[lars_bergquist]

Well, not *that* modest. They say -- I quote from the latest LFI issue, p. 27:

 

"Die M9 nutzt wie die M8 ein Absorptionsfilter direkt über dem Sensor, doch dieses besteht aus einem neuen, wesentich effizienteren Filterglas und ist zudem mit 0.8 gegen 0.5 Millimeter deutlich dicker" (now listen) "wodurch die M9 laut Leica in Sachen Infrarotfilterung im oberen Drittel der digitalen Systemkameras landet, während die M8 hier noch die rote Laterne trug."

 

Or for those that do not have Goethe's and Hölderlin's language: "The M9 like the M8 uses an absorption filter directly on top of the sensor, but it consists of a new, far more efficient filter glass, and is also at 0.8mm as against 0.5mm clearly thicker" (here it is) "so that the M8 in the matter of IR filtering lands in the upper third of digital system cameras, according to Leica, while the M8 in this case still flew the plague flag."

 

Digital system cameras, NOT digital rangefinder cameras.

 

Silicon itself does always retain a basic sensitivity to IR, one of the reasons why the CCD chip has replaced film in astronomy. All quality digital cameras attempt to filter it away, but none seems to succeed totally. But some may just bias their colour rendering slightly against magenta ...

 

The old man from the Age of Film

[sean_reid]

Another place I think there's a bit of liberalism going on is the reports I've seen that say the M9's IR sensitivity is now on a par with other top-end digitals. I can't speak for the latest Nikons as I don't use them, but my Canon 1DS-III and 5D-II produce zero colour error attributable to IR (neither did the 1DS-II and 5D-I I used before them. ).

 

I agree actually. In my M9 review, I reported on, and demonstrated, the IR sensitivity in the M9 under various lighting conditions (which is minor but could be a problem for some under incandescent lighting). DPreview then followed shortly with their own tests of this as well. Leica included a statement about this in their FAQ. My results basically matched Leica's own tests (which I have seen but will not discuss in detail).

 

Yet I have read subsequent reviews that seem to state the M9 is not at all sensitive to IR. I don't know what to make of that. My testing (in my original M9 review) shows that it does exist at a weak level, varying under different lighting conditions, and the text explains why. It hasn't been a problem for me and I don't think it will be for many people but I *do* think photographers should know about it and be able to make their own choices.

 

Leica didn't drop the ball on this, IMO, they just had to choose a filter thickness which would filter as much IR as possible without adverse optical effects in the corners. They sweated the details in finding that balance.

 

You around Pascal?

 

Cheers,

[sean_reid]

From what I've seen so far this isn't true: the M9's levels of IR contamination are in line with similar offerings from Nikon and Canon. IOW, they all suffer from this to one extent or another, and some post processing is needed depending on how color-critical you need the results to be...

 

Hi Jamie,

 

Take a look again at my tests of this compared to the Canon 5D.

 

Cheers,

 

Sean

[Guest EarlBurrellPhoto]

 

Leica didn't drop the ball on this, IMO, they just had to choose a filter thickness which would filter as much IR as possible without adverse optical effects in the corners. They sweated the details in finding that balance.

 

 

I agree, and given that Leica themselves admit it, it seems ludicrous that people are saying the M9 shows no IR color effects. But where I think Leica did drop the ball is in not giving a setting for UV/IR "ON" and firmware correction for cyan drift with coded lenses just like the M8 has. That would let people use IR filters when needed, without having to resort to added postprocessing. Why I consider this a ball drop is because like the in-camera menu for lens selection, and (at first) the discrete mode, it appears more of a marketing decision than a technical issue. I suspect it isn't in the menu because marketing guys fear it might cause people to keep their M8s (or buy used ones) as backups/second cameras instead of buying a second or third M9. Personally I think that's unfounded because I believe most people (aside from those who have to buy the latest Leica no matter what) who are into the M9 is because they have a thing about full-frame, and for them the M8 was always a source of anxiety. Perhaps once Leica sells enough M9s that they can breathe easier, the marketing guys will back off and let the techies do the right thing and add the IF-filter setting in a future firmware update.

[mby]

Does anybody have an idea what repair cost are to be expected for exchanging the front element? (..)

 

About €125 for the first lens on an Elmarit 21mm (pre-ASPH) at Leica CS Solms.

 

Best,

Michael

[sean_reid]

I agree, and given that Leica themselves admit it, it seems ludicrous that people are saying the M9 shows no IR color effects. But where I think Leica did drop the ball is in not giving a setting for UV/IR "ON" and firmware correction for cyan drift with coded lenses just like the M8 has. That would let people use IR filters when needed, without having to resort to added postprocessing. Why I consider this a ball drop is because like the in-camera menu for lens selection, and (at first) the discrete mode, it appears more of a marketing decision than a technical issue. I suspect it isn't in the menu because marketing guys fear it might cause people to keep their M8s (or buy used ones) as backups/second cameras instead of buying a second or third M9. Personally I think that's unfounded because I believe most people (aside from those who have to buy the latest Leica no matter what) who are into the M9 is because they have a thing about full-frame, and for them the M8 was always a source of anxiety. Perhaps once Leica sells enough M9s that they can breathe easier, the marketing guys will back off and let the techies do the right thing and add the IF-filter setting in a future firmware update.

 

I can ask them about this but I think the answer is even simpler. They aren't expecting many M9 owners to use IR-cut filters. If a lot of buyers do want to use the filters Leica may need to re-visit the lens detection options. My guess is that most people won't bother.

 

Cheers,

 

Sean

[Guest EarlBurrellPhoto]

I can ask them about this but I think the answer is even simpler. They aren't expecting many M9 owners to use IR-cut filters. If a lot of buyers do want to use the filters Leica may need to re-visit the lens detection options. My guess is that most people won't bother.

 

Cheers,

 

Sean

 

Maybe you're right, and it would be a shame if the M9 was not allowed to reach its full potential as a camera for the most discerning pros (and amateurs). An IR-filter setting in the firmware would make the M9 more of a pro machine without complicating things for the S-mode crowd one bit.

[sean_reid]

Maybe you're right, and it would be a shame if the M9 was not allowed to reach its full potential as a camera for the most discerning pros (and amateurs). An IR-filter setting in the firmware would make the M9 more of a pro machine without complicating things for the S-mode crowd one bit.

 

I suspect that it might come down to how many M9 owners feel that this feature is important. There would need to be a lot of R&D time put in to develop the corrections for each lens and to then add them to the firmware. This might be a question to revisit in a couple of months when more people have the camera and we have a better sense of how many are wanting to use the IR-cut filters.

 

Cheers,

 

Sean

[Jamie Roberts]

That just isn't true, at least of Canon. I can point you to at least a hundred busy pros besides myself shooting for glossy table mags and hyper-critical photo editors, who will attest they get jet-black blacks no matter what the fabric or light source, from Canon of at least the last two generations. With Nikon, I can't dispute you from personal knowledge.

 

Earl, at the risk of offending hundreds of busy pros, we don't agree. All digital cams shift colour from the way we see things or like to print them, and many of them suffer from IR contamination, but not on the scale of the M8. Yes, Canon is better (a lot better, IMO) than Nikon. In fact, I'd even grant that it's the best Japanese dSLR out there for IR rejection and overall colour. But it's not wholly immune either, and color from a Canon depends as much on processing as anything else.

YMMV.

 

I'd agree to that as long as you add "tedious" to "careful". And the problem is, my M9-shooting colleagues are telling me that it's just as tedious to correct the less-contaminated M9 files as the more-contaminated M8 files. It isn't the degree of correction required, it's isolating the contaminated subject areas. Global tweaks mess up other colours...although granted, less with the M9 because the correction levels needed aren't as severe.

We don't agree at all on this, sorry. I didn't find working with an unfiltered M8 "tedious" at all once I'd created a profile for it--of course, I should never have had to do that... and I'll repeat again that the color of a filtered M8 is much, much better (and very easy to work with).

 

What I find, actually, is that for my work (where it's skin that I care about, though of course neutrals are very important too), I get the easiest results in post--so the least amount of work--for good colour first from the DMR (which I suppose is like saying a Phase back equivalent), next the M8 with filters (I find skin tones better than Canon almost every time under almost all lighting conditions), next the Canon (I've shot tens of thousands of RAW shots on everything from a d60, 1ds, 1d2, 1ds2, and 5d), next the D3 at low ISOs (by a hair), next the M8 unfiltered, and finally the D3 at high ISOs (even in Nikon Capture NX).

 

As I intimated before, I don't have an M9 yet so I can't say. But from the few DNGs I've seen so far the only way you'd find them tedious to process is if you're used to processing Canon files That's ok, but let's not blame the tool, here, ok? Just to be clear, if you use Lightroom or ACR and are used to Canon files, the M8 / M9 is going to be a struggle.

 

So, again, YMMV.

[Jamie Roberts]

Hi Jamie,

 

Take a look again at my tests of this compared to the Canon 5D.

 

Cheers,

 

Sean

 

Sorry Sean--done in Lightroom, so I can't / won't comment.

[sean_reid]

Sorry Sean--done in Lightroom, so I can't / won't comment.

 

You're concerned about the lack of color profiles in LR then? If it helps, Leica engineers have seen my tests (of IR sensitivity) and I've seen theirs and we come up with basically the same results. They don't dispute that the camera is still slightly sensitive to IR, they've just filtered as much as they can (given optical performance goals in the corners).

 

At some point, I could re-run the RAW files in C1 but I'm quite sure the IR will be the same.

 

Did you read the section that explains why they couldn't filter IR 100%?

 

Cheers,

 

Sean

[Advertisement]

Hallo,

Look here: Leica M

[doug_m]

I brought this up in another thread but never saw a response:

 

An expert at Leica told me that perhaps there may not be noticeable cyan drift in the corners when using the 24 lux on the M9 because of how the front element is designed (no recess I think) as compared to other 24mm lenses. In fact, I have used the 24 lux on my M9 and so far do not see any cyan in the corners but my experience is very preliminary.

 

Can any of the experts on this forum comment on this issue? I'd really appreciate an opinion from others as I use an M8.2 and M9 and do not want to remove the UV/IR filter from the 24 lux unless I have to. Also, the B + W series VII UV filter I bought does a lot more rattling around in the lens hood than the Leica UV/IR. Does anyone also have a suggestion on fixing this rattle? The problem is that the B+ W filter ring seems to be slightly too small a fit within the 24 lux lens hood.