Center filter for CV15 on M9 ?

[C_R]

Advertisement (gone after registration)

Has anyone tried a Schneider center filter to compensate for vignetting with the CV15 ? Sometimes I dont like the light falloff, and have to correct with postprocessing. I use Cornerfix very often for this (many thanks to Sandy !), but for the same reasons I use a grad ND I would prefer to start postprocessing with a properly exposed file, sometimes. The red corner issue will be still there, but can be fixed very well with Cornerfix. A 2 stop underexposed corner, compensated by software in postprocessing, leads to poor tonal quality with higher ISO.

Schneider Kreuznach offer three different 52 mm center filters for the Apo Digitar lenses. Their light falloff is tailored for the special wide angle, but might fit the CV15. Any LF freaks out there ?

[Advertisement]

The Leica Q3 is available for pre-order at B&H Photo Video and Adorama

[jklotz]

I use center filters on my Schnider 35xl LF lens sometimes. Loosing a stop and a half seems like it might be a disadvantage on a hand held camera unless you were shooting a static image tripod mounted.

[jaapv]

Isnt the problem that a centre filter, which is grey, caters for optical vignetting only? It cannot compensate for IR (filter) induced colour shifts.

[C_R]

Isnt the problem that a centre filter, which is grey, caters for optical vignetting only? It cannot compensate for IR (filter) induced colour shifts.

 

As I said, the red corner issue will be still there, but can be fixed perfectly with Cornerfix.

 

But sometimes it would be nice to have a uniform exposition from center to corners. With exposure bracketing and tripod you can achieve this with masking in PS - however, you get misaligned parts when there is movement during the bracket series. Foliage, for example.

[jaapv]

You can do that extremely simply in photoshop or C1 and any number of postprocessing programs. Just click lens correction and slide the vignetting slider. This is one of the filters that has lost its function in digital.

[ho_co]

C_R, my initial reaction is that since CornerFix is designed to correct both vignetting and color shift, why add a filter when you're planning to use CornerFix anyway?

 

 

But overall your question has me very interested.

 

I know that some companies make filters for specific lenses, but I wasn't aware of general-purpose graduated filters. Does Schneider make these filters for specific lenses, for example also for the 12 mm Heliar?

 

Do you have a link that might point me in the right direction?

 

Thanks!

[C_R]

Advertisement (gone after registration)

You can do that extremely simply in photoshop or C1 and any number of postprocessing programs. Just click lens correction and slide the vignetting slider...

 

I know about the software solutions, but it is not the same result when you dodge the corners, as compared to a correctly exposed corner. If you "raise exposure" in the darker periphery 2 stops, you see a tonal quality loss, especially when you start with a higher ISO exposure.

 

...This is one of the filters that has lost its function in digital.

 

Respectfully disagree. Grad filters, radial or grad ND, have still their value in digital, if your goal is a well-exposed digital file.

[C_R]

... Does Schneider make these filters for specific lenses, for example also for the 12 mm Heliar?

 

Do you have a link that might point me in the right direction?

 

Thanks!

 

Howard, the Schneider filters are specifically designed for their LF lenses. Thats exactly the problem, I am not sure that they will work for the CV15 or CV12. I am afraid there is only try-and-error...

[ho_co]

Thanks, C_R! I'm glad you reminded me that there are ways other than software to get the results we want!

 

I think maybe Puts is onto something when he says that with digital there may be too great a tendency to use our post-processing tools simply to even the playing field, so that everyone's Gretag Macbeth chart will look the same.

[C_R]

Thanks, C_R! I'm glad you reminded me that there are ways other than software to get the results we want!

 

I think maybe Puts is onto something when he says that with digital there may be too great a tendency to use our post-processing tools simply to even the playing field, so that everyone's Gretag Macbeth chart will look the same.

 

Howard, after a learning curve with PS I enjoyed postprocessing, sometimes as makeshift for lazy in-camera work :-) To a certain extent it is ok, but a makeup should not be overdone..

 

You mention color management, another good example for potential in-camera work. Because with my Sigma DP1 and DP2 it is a bit tricky to get the color right, I switched over to manual WB with a grey card. It is sometimes nearly impossible to get the right WB from a Auto-WB RAW file in postprocessing. The Leica Auto WB is definitively better, but manual WB could be useful in high ISO and difficult light to avoid clipped channels from postprocessing.

[Mauribix]

Grad filters, radial or grad ND, have still their value in digital, if your goal is a well-exposed digital file.

Actually I respect what you say, but I can't share it.

Why wouldn't you get a well-exposed digital file without one of those filters?

I guess that using a center filter, or expose properly and then correct vignetting in PP should give almost the same result, with, in my experience and personal opinion, a larger latitude in which to operate when using DPP instead of the "mechanical" (filter) one.

 

Let me say, if I expose in "A" mode, the center filter is a good help, you're right, to get a better exposed,balanced all across the frame, file. But if you work manually, the center filter is becoming superflous IMHO.

Avoiding the use of a center filter can let you pull up the vignetted corners via any software you like, and still get a better resolution because of less glass between your sensor and your subject.

[C_R]

Actually I respect what you say, but I can't share it.

Why wouldn't you get a well-exposed digital file without one of those filters?

I guess that using a center filter, or expose properly and then correct vignetting in PP should give almost the same result, with, in my experience and personal opinion, a larger latitude in which to operate when using DPP instead of the "mechanical" (filter) one.

 

Example: without center filter, you get a brightly exposed central part, and, say, -2 EV underexposed periphery. To keep those highlights, you cannot expose longer, they would be clipped. So, an underexposed corner is always the consequence with heavy vignetting lenses like the CV15.

 

Let me say, if I expose in "A" mode, the center filter is a good help, you're right, to get a better exposed,balanced all across the frame, file. But if you work manually, the center filter is becoming superflous IMHO.

 

Also when you work manually, the corners will be underexposed, you have to expose for the highlights in the center.

 

Avoiding the use of a center filter can let you pull up the vignetted corners via any software you like, and still get a better resolution because of less glass between your sensor and your subject.

 

This is indeed the question. In your experience, to pull up the corners with DPP gives almost the same result as a really spot-on exposure.

 

I give you right for low ISO, the M9 files have a very good shadow detail quality and can be pushed to a great extent. But if you are picky, there is a visible difference in tonal quality even in low ISO. To "mask" the center with a grad filter allows for brighter, correct exposure of the corners. You dont have to push exposure with DPP then, and get the best possible tonal quality for that given ISO. This will be crucial when you shoot higher ISO, just try out how ISO 1000 looks when pushed 2 stops in DPP...

 

I give you right that you can fix vignetting also without filters, but you have to use a tripod and exposure bracketing, and blend two exposures with DPP. Sometimes tricky when the scene isnt still enough, the in-camera worklflow with grads lets you grab the whole scene in one shot

 

For my Elmarit 28, I used a B+W soft grad ND (screw-on, with step-up ring) and got nice results. A screw-on is not as flexible as the filter holder system, but ok for most of my landscape work, and doesnt block the finder. I just ordered a 52 mm Tiffen grad ND for my CV 50/1.5 (and, if usable, also the CV15), this grad is very soft over the entire diameter. Same ultra-soft grad ND material worked perfectly on my 6x9 Fujica GSW, a Formatt HiTech grad ND is always screwed-on for my landscapes to balance the exposure

[Mauribix]

Example: without center filter, you get a brightly exposed central part, and, say, -2 EV underexposed periphery. To keep those highlights, you cannot expose longer, they would be clipped. So, an underexposed corner is always the consequence with heavy vignetting lenses like the CV15.

....

Also when you work manually, the corners will be underexposed, you have to expose for the highlights in the center.

 

That's right, I'm not sayin' that I would expose on the supposed vignetting... i.e. I expose the picture balancing and protecting the highlights, unless I'm looking for a wanted effect (burn highlights with some portraits for example).

 

 

This is indeed the question. In your experience, to pull up the corners with DPP gives almost the same result as a really spot-on exposure.

 

I give you right for low ISO, the M9 files have a very good shadow detail quality and can be pushed to a great extent. But if you are picky, there is a visible difference in tonal quality even in low ISO. To "mask" the center with a grad filter allows for brighter, correct exposure of the corners. You dont have to push exposure with DPP then, and get the best possible tonal quality for that given ISO. This will be crucial when you shoot higher ISO, just try out how ISO 1000 looks when pushed 2 stops in DPP...

 

Yes, I confirm, in my experience pulling up the corners with DPP gave me the same results. Quite often even better.

 

Anyway, I see a contradiction here in what you state.

We know that the higher is the ISO used, the smaller is the margin to PP until we get noise.

So how could be useful to use a gradient filter and "diminish the actual speed of the whole lens"? (please pass me the statement, english's not my mother language)

 

I mean, for the same shot you'd have to use longer exposure or higher iso, and get more noise all across the frame.

 

That's not an issue probably if you shoot landscapes with a tripod, but that's to be specified. Anyway, even in that case longer exposures generates higher noise (with every ISO setting), that's a fact.

 

If we're shooting, let's say @ f 5.6, and we suppose to have 2 stops of vignetting on the corners, we should used a center filter to compensate those 2 stops lost in the corners.

That's ok, but now we have to shoot with a de-facto 2 stop slower lens all across the frame.

What we could shoot, let's suppose without a 2stop center filter at 1/120 F/5.6 ISO 320, then would become one of the following:

 

1/30 f/5.6 ISO 320 (i.e. eventual moving subjects, would be probably blurred)

 

1/60 f/5.6 ISO 640 (probably still blurred moving subjects, with higher iso and noise as well)

 

1/120 f/5.6 ISO 1250 (the same scene, but with very higher noise than what we could have had at 320ISO)

 

and so on....

 

If we kept the first solution (320 1/120 f/5.6 without center filter) I'm sure that even a 2 stop digital correction for vignetting would have generated better results with the M9 (and M8 too), I cannot say the same for 2500iso, I know, but then with a center filter we should have used theoretically a 10.000 ISO sensitivity:eek:

 

 

So, still I can't see the advantage of using a center filter.

Most of these lenses then, when stopped down by 1 or 2 stops, they sensibly reduce vignetting, so your PP would be anyway less extreme and aggressive, but then you'd got the advantage of a calibrated correction. Spot on for what that concerns quality.

 

The fact that you seem to ignore IMHO, is that correcting the Vignetting in DPP, is much much more precise than any center filter you could ever have.

 

When I used the Zeiss 15 f2.8 for example, provided with its center filter, that filter, especially with the M8 was good (and not that much indeed) only @ full aperture.

Shooting a white wall with smaller diaphragms for example and keeping the filter on was generating weird vignetting effects.

What I mean, you're looking for a better image all across the frame, but then you can't have a filter dedicated to a specific lens such as the CV15, or 12 FWIW... So, I guess, you're accepting a compromise.

Giving up to DPP in favor of all these issues, just to use a center filter, well... that's not what I can call a smart solution.

 

 

 

 

I give you right that you can fix vignetting also without filters, but you have to use a tripod and exposure bracketing, and blend two exposures with DPP. Sometimes tricky when the scene isnt still enough, the in-camera worklflow with grads lets you grab the whole scene in one shot

 

That's not always like that in my experience, I don't use tripods usually (cause when I need such shots I use different cameras, but then that's just me).

 

For my Elmarit 28, I used a B+W soft grad ND (screw-on, with step-up ring) and got nice results. A screw-on is not as flexible as the filter holder system, but ok for most of my landscape work, and doesnt block the finder. I just ordered a 52 mm Tiffen grad ND for my CV 50/1.5 (and, if usable, also the CV15), this grad is very soft over the entire diameter. Same ultra-soft grad ND material worked perfectly on my 6x9 Fujica GSW, a Formatt HiTech grad ND is always screwed-on for my landscapes to balance the exposure

I keep on sayin' that it seems like a compromise to me, using a standard center filter to fix a vignetting problem of a specific lens, doesn't sound that good to my ears.

My 2cents

[C_R]

Maurizio, it boils down to the question, how much can the underexposed DNG be pushed.

 

A picture is worth a 1000 words....

 

This is one from a bracket series, -2EV underexposed and pushed 2 stops (left) against 0 EV, ISO 320, M9 with CV15 @ f:8, tripod; 1/30 sec left, 1/8 sec right.

 

Conversion with standard settings, untouched (except the +2EV for the underexposed shot), from uncompressed DNG with CaptureOne V.

 

I think there is a visible difference. It is a personal choice to accept the minor quality loss, or not. If I have the time and a tripod, I make a series of bracket shots for possible composite work later on.

 

BTW, very interesting to hear your experiences with the Zeiss and CF, it is doubtful that a Schneider CF will work with the CV15 then. Thanks a lot for your observations ! I think, I will work with grad ND when handheld, or exposure bracketing when on tripod.

 

[Mauribix]

C_R,

that's obvious that a "pushed" picture is always worse than a properly exposed one.

What I mean, you're comparing two shots with the same iso setting (made with a tripod?)

The first is properly exposed @ 1/8 f8 iso 320

The second one is underxposed and then pulled from 1/30 f8 iso 320

 

What I mean, I'd like you to make the first shot @ 1/30 f8 iso 320, and the second one @ 1/30 f8 iso 1250 (with a 2stop ND filter), I guess that would change a bit.

 

1/8 vs 1/30, as I wrote before, is the difference between a blurred picture and a static one.

It's good till we want to shoot landscapes with a tripod, you're right; but we get a totally different scene if we have to catch moving subjects indeed.

 

BTW, I see slight color differences between these shots, what is due that to?

 

anyway you're right, it may worth 1000 words, I think we're here to compare opinions and experiences, that's the invaluable thing, isn't it!?

[C_R]

Maurizio,

 

the color differences can be due to underexposure, and/or red color shift since this crop was from the lower left part (the dreaded red-corner problem of the CV....) I didnt make any corrections, as I said, except +2EV for the underexposed shot. Here on the forum I read a comment, that the red corner issue is less evident in brightly exposed shots, mor ein underexposed. This could be the case here