Orange filter leads to sharpness reduction ?

[valtof]

Advertisement (gone after registration)

I experienced using an orange filter (39mm Heliopan) on a CV 2.5/35 Type II.

I noticed that using such a filter leads to a noticeable sharpness reduction.

The two images below were taken with TriX and scanned with Vuescan

(100% crop of a 4800 dpi scan, a bit of accentuation in PS for presentation).

The first one is with the orange filter and the second without.

I think the difference is obvious.

I even made several tests with the M8 (JPEG forced in B&W) and it's even more visible.

 

Does anyone has an experience, explanation of this unexpected phenomenon ?

 

Cheers

Christophe

.

[Advertisement]

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

[giordano]

I experienced using an orange filter (39mm Heliopan) on a CV 2.5/35 Type II.

I noticed that using such a filter leads to a noticeable sharpness reduction.

The two images below were taken with TriX and scanned with Vuescan

(100% crop of a 4800 dpi scan, a bit of accentuation in PS for presentation).

The first one is with the orange filter and the second without.

I think the difference is obvious.

I even made several tests with the M8 (JPEG forced in B&W) and it's even more visible.

 

Does anyone has an experience, explanation of this unexpected phenomenon ?

 

The orange filter will be cutting out a great deal of the blue and a good bit of the green, so most of the image will be formed by the longer wavelengths towards the red end of the spectrum. Most general purpose lenses are not perfectly colour corrected and tend to bring these longer wavelengths to a focus slightly further back than the green and blue.

 

Normally this has little if any visible effect but the orange filter increases the importance of the longer wavelengths in forming the image: hence the very slightly diminished sharpness you see in these highly magnified images.

 

To check that this is the cause, try focusing a fraction closer than the rangefinder shows: focus normally, then turn the focusing ring about half the distance between the index mark and the f/4 depth of field mark. With luck this will give you a better result with the orange filter and a worse one without.

 

This effect is likely to be more prominent on the M8 than on film, where the thickness of the emulsion can mask it to some extent. Also, on B&W film the entire film is sensitive to all visible wavelengths, while on the M8 only 25% of the photosites on the sensor are sensitive to red light (plus 50% green and 25% blue).

[Michael Hiles]

The orange filter will be cutting out a great deal of the blue and a good bit of the green, so most of the image will be formed by the longer wavelengths towards the red end of the spectrum. Most general purpose lenses are not perfectly colour corrected and tend to bring these longer wavelengths to a focus slightly further back than the green and blue.

 

Normally this has little if any visible effect but the orange filter increases the importance of the longer wavelengths in forming the image: hence the very slightly diminished sharpness you see in these highly magnified images.

 

To check that this is the cause, try focusing a fraction closer than the rangefinder shows: focus normally, then turn the focusing ring about half the distance between the index mark and the f/4 depth of field mark. With luck this will give you a better result with the orange filter and a worse one without.

 

This effect is likely to be more prominent on the M8 than on film, where the thickness of the emulsion can mask it to some extent. Also, on B&W film the entire film is sensitive to all visible wavelengths, while on the M8 only 25% of the photosites on the sensor are sensitive to red light (plus 50% green and 25% blue).

 

I think this is a good explanation - but not of this phenomenon. If we were looking at focus shift due to wave length differences, i.e. slight shifting of the plane of focus, there should be a sharp plane in the filtered photo - but not the same plane of sharpness as in the non-filtered photo. This is not the case. In the filtered photo, the whole picture has degraded sharpness. It also looks as though the aperature was small enough to ensure quite good depth of field in the entire picture, making the precise plane of focus less an important issue. The impression is of a film with less accutance, or a lens of lesser quality overall, i.e. some softness.

 

I don't have a good explanation that I can stand behind, but it would surprise me if a competent filter from a trusted manufacturer would cause such an obvious drop in quality, but I am prepared to be wrong.

 

Michael

[Kent10D]

I assume that different apertures were used due to the filter factor(?).

 

Is it possible that the wider aperture used for the less-sharp image taken with the orange filter is near the maximum aperture of the lens?

 

Or if you used a slower shutter speed rather than wider aperture to compensate for filter factor might that be a contributing factor?

[valtof]

Thanks John but I'm afraid I agree with Michael since these shots were taken at least by something like f:5.6 or 8 at 1/250s.

Therefore no chance of shaking or misfocussing difference between them.

And since Heliopan is a serious brand I wouldn't incriminate the filter itself.

Below is a test with the M8, same lens and filter, 100% crop ;

First with filter second without.

The flower jar in the image is orange in the reality hence logically lighter in the filtered shot.

 

Cheers

Christophe

.

[Michael Hiles]

Thanks John but I'm afraid I agree with Michael since these shots were taken at least by something like f:5.6 or 8 at 1/250s.

Therefore no chance of shaking or misfocussing difference between them.

And since Heliopan is a serious brand I wouldn't incriminate the filter itself.

Below is a test with the M8, same lens and filter, 100% crop ;

First with filter second without.

The flower jar in the image is orange in the reality hence logically lighter in the filtered shot.

 

Cheers

Christophe

.

 

Christophe,

 

The difference in the M8 pictures is startling. It seems that the only variable is the filter, so suspicion is unavoidable. Everything else is constant. The lens evidently makes sharp pictures, as does the M8 and Tri-X without the filter.

 

If we believe that, then the next questions is "Is the variable Orange or Helioplan or filter in general?" It would be interesting to try again with i) another orange filter from Helioplan to see if it is THAT filter, ii) another orange filter from another manufacturer , iii) a yellow filter from Helioplan and another manufacturer, iv) same for a red and green filter. Essentially, change most likely variables and keep testing (with the M8 - save film and time, and it shows the effect most clearly)

 

All very curious.

 

Cheers,

 

Michael

[Kossi008]

Advertisement (gone after registration)

Shutting out shorter wavelengths MUST have a negative impact on achievable sharpness in theory, because the airy disc (diffraction-limited image of a point source) would be larger for orange light than, say, for blue light.

 

I am only unsure whether this effect is large enough so that it should show up here or not.

 

Another guess would be the filter in general. Even if it is ideally plane-parallel, it would introduce slight spherical aberrations at fairly open diaphragms, again, I am not sure it would be enough to show.

 

Or how about all those ions in the filter glass matrix contributing some stray light?

 

Hey, I'm only putting out some wild guesses here, so please don't bash me if the orders of magnitude don't work out. I guess what I'm trying to say is there's plenty of possible causes that would not necessarily point to a "bad" filter...

[usefeet]

Another something for the cooking pot is that about a year ago we did a test with a professional (laser) collimator and I when shown the faults, I was surprised.

 

You need the VERY best glass, and the glass needs to be loose, not tight.

 

(I use Leica or B+W only)