FLARE! fly in the ointment of that beautifu 35/1.4!

noimmunity

scratch my niche
Local time
7:09 PM
Joined
Jul 1, 2007
Messages
3,102
The 35/1.4 has by now been recognized as a superlative lens that compares favorably with lenses even far above its modest price.

My copy is, however, MUCH more susceptible to flare than any of my M mount lenses from Zeiss, Leica, and Cosina Voigtlander used on film.

Is this a problem common to all digital capture? Have other people using the 35/1.4 and other XF lenses noticed a problem with flare?

Here are some examples, all taken with the stock hood mounted:


flare example 1 by areality4all, on Flickr


flare example 2 by areality4all, on Flickr


flare example 3 by areality4all, on Flickr


flare example 4 by areality4all, on Flickr
 
Hmm.. The only one I'm seeing true reflective flare in is the 3rd one. The others just look like the lights are blown out and hazing a bit because they're so bright. Doesn't seem much different to other digital cameras/lenses.
 
Wow, that's a great look, very on-style!

Trying to remember what movie I saw over Christmas that was all shot with that 70s flare, the g/f and I were wondering how they achieved it so consistently....
 
You've got back lit subjects - my guess is you're not using a flash and you may be just upping the exposure compensation - I get flare too - with all my lenses, every single one of them, when I point them at light sources. . . .

Cheers,
Dave
 
1. If you shot these at f/1.4, it's not surprising. Use the special lens hood (if you aren't) and stop it down to f/2.

2. It looks like you have some pretty extreme sensor overload in here, too.

3. If you are using a cheapo skylight/uv filter, get a better one. With some lenses, you can get back-forth flare between the filter and the lens. Sometimes it's unavoidable with point light sources (it also seems to happen a lot on lenses with negative front elements).

4. The 35/1.4 flares the same amount a 50/1.4D Nikkor does.

5. Count your blessings. This lens likely very likely outperforms your other lenses, and the bigger (flare-attracting) elements are in part a safety margin to assure even illumination. And you would never get shots like this with color film, so comparing it to film lenses is something of an extraneous worry.

Dante
 
I don't have one of these cameras, but when the pictures first started appearing on the internet, I thought that lots of them had large amounts of flare. Not so much the ghosting type of flare (image forming flare), but the kind you are showing. I think it's referred to as veiling flare. Sure, the conditions you are shooting in is going to exacerbate the problem, but I noticed it any time there was a bright light source in the frame, like a window or a bit of sky.

I've shot plenty in the conditions similar to what you show, and other lenses can handle that situation MUCH better.
 
1. If you shot these at f/1.4, it's not surprising. Use the special lens hood (if you aren't) and stop it down to f/2.

Not shot wide open. With the excellent ISO performance of the XP1, it isn't necessary, and shallow DOF would not have added anything to the image. As I mentioned in the OP, the hood was mounted for all images.

3. If you are using a cheapo skylight/uv filter, get a better one.

Since I discovered that using a filter exacerbates flare, I stopped using one after a couple of days.

5. And you would never get shots like this with color film, so comparing it to film lenses is something of an extraneous worry.

I have several years's experience shooting color and black and white film on-stage in very similar lighting situations. My M lenses, especially the ZMs, did not exhibit this kind of veiling flare.

But yes, with this camera, I do count myself very fortunate, and as I mentioned in my previous thread "very pleased after one month" (http://www.rangefinderforum.com/forums/showthread.php?t=118700), I LOVE this camera, find it very reliable, and it fits a much needed place in my kit.
 
Here is an example photo taken with a lens famous for being prone to flare (35mm Summilux-M pre-ASPH) on film ... :eek:. Not that close as your photos but also with some bright light spots in and around the frame ...

2185177104_8c50a23420_z.jpg
 
Here is an example photo taken with a lens famous for being prone to flare (35mm Summilux-M pre-ASPH) on film ... :eek:. Not that close as your photos but also with some bright light spots in and around the frame ...

2185177104_8c50a23420_z.jpg

Personally, I think that is a lovely photo, Gabor. With the XP1, it could be a challenge to get that kind of result with the placement of those lights.
 
Gross overexposure leads to odd artifacts in diigital photography. This is one advantage of film. These problems have nothing to do with the lens.

At the same time the APS-C Xes have a short distance between the rear lens elements and the sensor. This makes the lenses more susceptable to reflection artifacts.

So there are three things going on: gross overexposure, reflection artifacts and lens flare.

I have no problem telling people to avoid the Fujis if they need to severely overexposed bright lights and shoot into the lights at the same time.
 
Gross overexposure leads to odd artifacts in diigital photography. This is one advantage of film. These problems have nothing to do with the lens.

At the same time the APS-C Xes have a short distance between the rear lens elements and the sensor. This makes the lenses more susceptable to reflection artifacts.

So there are three things going on: gross overexposure, reflection artifacts and lens flare.

I have no problem telling people to avoid the Fujis if they need to severely overexposed bright lights and shoot into the lights at the same time.


It's funny. I guess you would think it is gross overexposure, yet I bracketed exposure, tried dialing various different levels of exposure compensation, etc. yet none of it worked. I had plenty of time to try it. Yet in all cases, either the image was completely dark, or else the image was washed out by veiling flare. It was impossible in such situations to get a middle ground. Hence, I do not think it was a problem of overexposure.
 
"Overexposure" maybe isn't the most useful phrase in this context because it implies that there's a right that was missed. The dynamic range in this scene is vastly more than a sensor can cope with. Exposing so that the musicians are usefully illuminated means that the lights will be grossly overexposed -- those are some bright lights, even by stage light standards and close in. They are overwhelming the sensor & causing (as others have noted) a sensor related, not a lens related glow. Which is not to say that the XP1 is useless in this situation -- it clearly isn't. It may or may not have been possible to manage exposure better than you did given the light situation but certainly in a wide range of stage lighting situations the XP1 seems to work just fine. In the pictures thread both furcafe and copperB have posted nice looking things with stage lights in the frame and those shots seem to use both the 35 and the 18.
 
I have noticed some veiling flare w/my 18/2 (the only X-Pro1 lens I have) @ f/2, but I think your results seem to show some overexposure too.

Per JHutchins's post, if your bracketed shots were either completely dark, i.e., extremely underexposed, or washed out, then they must have been changing the intensity of the stage lighting by a huge amount (some bands/venues for a strobe-like effect), & it looks to me you still overexposed when the lighting was bright. In my experience, when shooting these kinds of situations, you "shoot to the left" (or more like slide film) because the highlights blow out rapidly.

As a point of comparison, I perhaps had something similar happen to me when I shot Social Distortion a couple years ago. The lighting for their set was unexpectedly much, much brighter (@ least 3 stops) than the opening acts. I normally shoot shows in manual (because stage lighting throws off most in-camera metering), & I was caught off-guard. Because they were strictly enforcing the 3 song rule & Social D songs aren't long, I failed to readjust in time & most of my shots were overexposed. You can see the results here: http://www.flickr.com/photos/furcafe/sets/72157627487370509/ Now that was some serious sensor overload; not too much veiling flare from the 35/1.4 Summilux ASPH, more from the old 50/1.4 Nikkor-S. I think some of the shots still worked as a kind of high-key thing, but another lesson learned.

It's funny. I guess you would think it is gross overexposure, yet I bracketed exposure, tried dialing various different levels of exposure compensation, etc. yet none of it worked. I had plenty of time to try it. Yet in all cases, either the image was completely dark, or else the image was washed out by veiling flare. It was impossible in such situations to get a middle ground. Hence, I do not think it was a problem of overexposure.
 
I think the first three images have suffered from bad white balance settings, or simply from too unnatural light spectrum at the stage at that moment. They look perfectly normal photographs to me if I run them thru Photoshop > Channel mixer > B&W with green filter preset.
The fourth image really seems just overexposed and gets at least betetr when you run it thru Curves > Auto.
There is some veiling lens flare, but I really couldn't even see it until on the treated images.
 
Looked at this again in CS5. Overexposure primarily. Assuming the meter was not fooled, I'd check to see if the aperture unit in the lens is actually stopping down.

Dante
 
I have to disagree. Three of the four photos are severely overexposed. The dynamic range of the sensor was way below the dynamic range of the scene. What happened here was you were forced to exceed the sensor's capabilities. It happens.

The situation was very difficult. Of course the lens flare didn't help. I think the close lens element-sensor distance didn't help either. But no sensor could handle that dynamic range. The fact that you tried a wide range of exposures and nothing worked is consistent with extreme dynamic range. A modern Zess lens on a D800 would have been better. But even then the severe overexposure would generate significant sensor artifacts. As mentioned above, ago AWB failed under these conditions (or the WB was manually misset).

I think your post's title is misleading because I think the lens flare was a minor problem compared to the operating the sensor way beyond the conditions it was engineered to handle. You had to do the best you could in impossible circumstances and that's what you did.
 
Back
Top Bottom