Jump to content

Freedom from Filters (here lies heresy)


Recommended Posts

  • Replies 83
  • Created
  • Last Reply

Top Posters In This Topic

Top Posters In This Topic

Hi everyone.

I knew that either I was going to be ignored . . . . or not!

Still, I've made my point.

Nothing is simple about colour with bayer filter sensors, and putting an IR filter in front isn't simple either.

Leica don't appear to think that you need an IR filter in all circumstances . . . and neither do I.

I am now off to do some serious damage to a couple of bottles of Fleurie, and then I have a long drive to do in the morning, so I'll thank everyone who has contributed and press the escape button.

Hopefully I've made at least one person question the meaning of life :p

 

Have a great evening / weekend

Link to post
Share on other sites

My two cents on the optional/required UV/IR filter issue:

 

CV15

I've convinced myself that the cv15 doesn't need a UVIR filter and works fine enough with John's special, codeable (sp?) adapter. I did some test shooting both ways and seem to get better color separation using lens rec:ON but no filter (only a standard Leica UVA for some protection). Like one other poster commented, color preference is an emotional preference (as well as a choice based on some visual expertise for those in this forum) so there is some latitute over what is 'correct'. Remember we used to choose films based on their different qualities in recording color. Biggest difference here came from getting the special codeable lens mount.

 

28 2.8 Elmarit (new)

Have the Leica one on here and colors are downright perfect.

 

Zeiss 50 f/1.5 Sonnar

Have the proper filter but no special coding. Result is just about perfect, no corner issues and can slide contrast/saturation preference in desired direction in post process.

 

75 cron (new)

Have Leica filter (my two free ones (for the 28 and 75) came this week) on and color is great.

 

I'm using C1 for all processing because I haven't found a better quality output. Work process is a bit of a pain and I keep trying to like Lightroom (mostly cause I paid for it) but it does seem to overdo it a bit with saturation - the C1 files just look better -- finer range in color and more detail somehow.

 

Thanks, Jono, for starting this thread and daring to say that filters could be a photographers preference and, for the testing. As someone with many, many extra MB of versions 1,2,3...10 etc files on my system, and reams of test prints, I can appreciate what it takes to finally make an informed decision about some of these issues.

 

Mark

 

A quick Lightroom tip Mark, regardless of whether filters were used: open the file, reduce saturation by 5-10, increase blacks by 2-5, then try a slight bump to vibrance. Works for me!

 

Tim

Link to post
Share on other sites

Just out of interest, when you don't use filters to correct the IR magenta problem but use Photoshop or using C1 profiles instead, what happens to things that are genuinely magenta ;-)

 

yeah..... like my magenta suit

Link to post
Share on other sites

wasn't not established that there is a focus shift without an IR cutoff filter ?

 

No--the M8 is tack sharp with or without filter. Honestly, when I go back and forth between the 5d files and the M8 files, it's really a night and day difference (not that the 5d can't be sharpened, but you know what I mean).

 

And FWIW your magenta suit, with the right profile, will still be magenta ;) It will just be less saturated.

 

If you want to have fully saturated magenta, like certain flowers, then you need to use filters with the M8 or do some PS LAB corrections in post.

 

Of course, if you're printing to CMYK afterwards, you probably need to de-saturate your magenta suit anyway ;)

Link to post
Share on other sites

Advertisement (gone after registration)

Hi Alan - I'm not sure quite what you're trying to say?

The point of my thread is that I haven't seen ANYONE saying that it's okay to shoot the M8 without filters for landscape work, but I've seen a LOT of people saying it isn't. I am simply saying that it's possible to get 'good' (not necessarily 'right') colours without using the filters - nothing else

 

Hi Jono,

 

Of course, none of us needs anyone's OK to photograph as we like. Heck, if someone wants to shoot textiles with an M8 but no filter, who are we to object? It's useful to know that mounting an IR-cut filter on an M8 lens will (generally and assuming good color profiles) provide more accurate color (color that has fidelity to the subject) than working without it.The photographer can then decide whether or not that accuracy (to the degree that such can exist) is important or not.

 

As far as I'm concerned, its OK to paint the whole camera orange and shoot with lenses made from bottles. If a client is paying for something, then they need to get what they asked for. Otherwise, photographers should do whatever they want.

 

Cheers,

 

Sean

Link to post
Share on other sites

Just out of interest, when you don't use filters to correct the IR magenta problem but use Photoshop or using C1 profiles instead, what happens to things that are genuinely magenta ;-)

 

You open Pandora's box. The magenta shift of black is only part of the changes IR makes, for better or worse.

 

Cheers,

 

Sean

Link to post
Share on other sites

wasn't not established that there is a focus shift without an IR cutoff filter ?

 

Not a focus shift but rather, sometimes, a slight softening of detail in some subjects that reflect a lot of IR light. The visible light is in-focus but the IR is out of focus. That said, it has never caused a problem for me and I often shoot bareback.

 

Cheers,

 

Sean

Link to post
Share on other sites

You open Pandora's box. The magenta shift of black is only part of the changes IR makes, for better or worse.

 

Cheers,

 

Sean

 

Sean, with the greatest of respect, Pandora's box is already open, and has been ever since they made colored pigments ;)

 

In all seriousness, I don't shoot anything from anyone's digital camera that doesn't need colour correction in post.

 

That's a fact of life for a working pro: plain and simple. From the moment I apply an input profile, to the moment I send it to a lab (or print it myself) it's all about process control and predictability, not just color management.

 

Every camera does some things well and other things not so well.

 

Now, do I need to do the same corrections with a 5d that I would with an M8?

 

No, but believe it or not they're similar, once you've worked with the M8's own characteristics.

 

IOW, the client pays for what I give them, and I make darned sure what they get will print properly (eg by the numbers). But what "properly" is depends on the output method, the viewing conditions, and the print technology; the camera is only the first point in the chain.

 

And it's quite weak in terms of human vision, actually. So I don't know any camera that produces "print ready" art with anything like regularity. Even in the film world, choice of film changed colour balance. And while it's fine to "print the negative" there are of course a thousand choices between the latent image on undeveloped film and a final print on photo paper or in a book.

 

That the prepress folks, or the Lab, or whoever else made those decisions in the past aren't always making them anymore (or at least aren't making them to your advantage) just means you have more control than ever. Including the control and ability to screw up in new and better ways :)

 

@ Jono--I think I've actually been pretty consistent in saying you can shoot the M8 without filters--landscapes or anything else, actually ;) Filters--as they always were--are just a means to an end.

Link to post
Share on other sites

Guest guy_mancuso

My last stab at this than the drinking starts. LOL

 

I still go back to what i orginal said and believe okay maybe the word normal and neutral are not as accurate as words but there has to be a value that all OEMs try to find or get to to cut the IR value coming in I think that is 750 something. So what i was trying to say was with the IR filter the M 8 gets back to that OEM value that manufactuers need to hit to block IR. Yes every camera has different firmware and different color palette . What i am saying has nothing really to do with color yet, what i am saying is getting the camera in the correct IR blocking value. From there it's up to profiles and firmware to determine what looks correct or not. My theory is simply use the filters at all times because leica built the firmware with ON/IR ON around the use of leica filters and there basing there color with that switch IR on. From this point on it is up to me when using C1 or lightroom to have the right profile or calibration. it's all variable but my thought is the amount of blocking IR needs to hit a certain value that takes the unwanted IR off the sensor to start with. Basically kick it out before it ever comes near the sensor like other camera'sand use the filter at all times. That is my stance and sticking to it and BTW been real successful at it. If folks want to work without filters that is there call but remember this IR does not only effect magenta but the whole spectrum of visable light in every lovely color just more with some and less with others

Link to post
Share on other sites

1. This has been fun to read - hope it continues

 

2. Back when I took "Basic Color Photography" in about 1974, the instructor told us that there were many kinds of "correct" color.

 

Among others: Colorimetrically correct color, in which the wavelengths of light in the final image exactly replicate the original wavelengths the subject reflected (which is impossible via primary tripack photography (i.e. CMY film or RGB sensors - but may have been achieved, once, in the 19th century**); visually correct color, which "looks" right; and emotionally correct color, which "feels" right (i.e. conveys to others the feelings that drove one to shoot something).

 

3. Certainly I took 5000 or so pictures with the M8 before acquiring any filters - and a lot of those ultimately looked beautiful. The filters are only REQUIRED to solve specific IR issues. And those issues may occur in 95% of any particular photographer's pictures - or only in 5%. Depends on what one likes to shoot under what lighting.

 

4. Different lenses definitely do handle IR differently. My older 90 f/2 makes blacks more magenta than some of my other lenses (while on film it is far more resistant to UV). Optical glass covers a wide spectrum of characterisitics (pun intended), and absorption of certain wavelengths is among those.

 

**One photographer (Du Hauron?), in the mid-1800s, managed to occasionally produce color pictures by shooting with mirrored plates coated with silver emulsion. He couldn't reproduce the effect predictably, nor could others. But the theory says what he achieved was possible, and that it did, in fact, reproduce the exact color wavelengths that had made the exposure.

 

The idea was that if, say, green light of a specific wavelength hit his plate, some of the light would go all the way through the emulsion, hit the mirror, and bounce back, the lightwaves "interfering" with the incoming green light to form interference bands layerwise in the emulsion. I.E. Where the waves interfered constructively, extra silver was exposed, and where they interfered destructively, no silver was exposed.

 

When viewed after development, the layers of silver in the emulsion would only allow wavelengths of the exact same frequency as the original exposure to weave their way through, filtering out all others. Not unlike the interference layers that filter IR in the IR-cut filters.

Link to post
Share on other sites

The M8 is interesting in comparison with many other cameras because, with the option of UV/IR filters, it gives a wider range of color rendition options than can be achieved with the narrower sensor gamut most other cameras provide. With filters you get one color palette; without, you get another - which is not available on most other cameras. With an IR-pass filter (transmits IR & blocks some or all visible light) you get yet another option which isn't available on many other digicams.

 

There's room for artistic experimentation here.

Link to post
Share on other sites

What I haven't heard said in the past 20 or so posts is that working without filters on the extreme wide angles (the CV15 and CV12 being the popular examples) allows you to use the older "lens detection ON" routines (if you are coded as something), which have much more powerful overall vignetting corrections than do the "ON+UV/IR" routines. The CV15 in particular vignettes strongly so in OFF its pictures give attention to the center. For landscape and active street work, you might want to see the full frame, so use "ON."

 

scott

Link to post
Share on other sites

It would sure beiinteresting to know why Leica made the vignetting correction weaker with IR on....we know they have a reason for everything, although not always one we agree with...

 

GUY- It would be worth asking them about this, maybe we are missing something here and there is a reason why we wouldn't want full correction with IR on?

Link to post
Share on other sites

Guest guy_mancuso

Hmmm not tested this but you saying the vignetting correction is different with ON / ON IR.

 

Now that seems strange . Anyone else notice this difference

 

The only thing i could think of and really speculating here is the cyan drift and it's effects.

 

We should be careful here also and really need to talk about Leica only lenses because the firmware is squarely aimed at coded lenses and there respective focal length. The CV lenses maybe there is some differnce not sure but we really have to remember this very important point . The firmware is color corrected and cyan drift correction for Leica glass and the coding of that glass, also the use of Leica filters, there is no question in my mind on this one becuase they told me for one and i do see a difference between B+W and Leica filters in color. Everything else is a guess, although they work good it is just not optimized for Zeiss and CV lenses. I can't say this point enough because outside of the Leica lenses it's a crap shoot on what you are getting. Luckily the results so far have been good for Zeiss and CV lenses and slipping in the leica codes and also for B+W filters for most lenses we been able to cheat pretty good

Link to post
Share on other sites

Guest tummydoc

Obviously someone had to write a software routine for the 1.102 firmware to correct the cyan drift, and obviously there's separate algorithms for each lens code, so the routine has to have adjustable parameters. Therefore is there any technical reason why the same routine, complete with lens-specific presets and manual control, can't be offered as a plugin to the RAW conversion? It would allow advanced users full control: full auto (in-camera), semi-auto (in-camera + manual tweak), manual (cyan done in post based on EXIF detection of lens). Also it would give those with un-coded lenses an easier workflow while respecting Leica's "reasons" for not putting a lens menu in the firmware.

Link to post
Share on other sites

No--the M8 is tack sharp with or without filter. Honestly, when I go back and forth between the 5d files and the M8 files, it's really a night and day difference (not that the 5d can't be sharpened, but you know what I mean).

 

I think you need a better 5D - and my profiles for 5D, maybe. When I profiled the 5D I bought a $100 lens, and the sharpness and color tone blew me away (50/1.8). And as you know I own a 1Ds, a 1DsII and some of the best Canon glass. In fact, it was so good that I'm thinking of using a 5D for travel.

 

Edmund

Link to post
Share on other sites

Guest guy_mancuso

That is something you may see with C1 and version 4 BUT from the rolling rumors of C1 and it's release is we have a wait on our hands. I wish i knew more on the C1 front and i was a beta tester for them awhile back but have not heard a word from them on Version 4. So everything could be internal

Link to post
Share on other sites

Obviously someone had to write a software routine for the 1.102 firmware to correct the cyan drift, and obviously there's separate algorithms for each lens code, so the routine has to have adjustable parameters. Therefore is there any technical reason why the same routine, complete with lens-specific presets and manual control, can't be offered as a plugin to the RAW conversion? It would allow advanced users full control: full auto (in-camera), semi-auto (in-camera + manual tweak), manual (cyan done in post based on EXIF detection of lens). Also it would give those with un-coded lenses an easier workflow while respecting Leica's "reasons" for not putting a lens menu in the firmware.

 

No 'technical' reason why not at all. I'm surprised that someone like DxO haven't managed to build the corrections into their DxO Optics Pro product by supporting the M8. Similarly, if you were prepared to measure each lens to build an appropriate table of correction coefficients you could definitately build a plug-in to do this. The Panotools Radial Luminance tool almost does this but requires manual trial & error setting each colour but it is doable. Knoiwing Leica's algorithms and weighting factors (which are locked up in the firmware) could make this process even more accurate and applicable to any lens.

Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...