Voightlander Lenses, LoCa and Purple Fringing

[BrianS]

Advertisement (gone after registration)

If you have an M9 or M Monochrom, use a UV filter. At least Test the lens and camera combination before making general assumptions.

 

Leica M9, Voigtlander Nokton 50mm F1.5 V2, Multi-coated version, wide-open. Same lamp as used with the M8.

No Filter, B&W UV/IR Cut Filter, and B&W UV/IR Cut Filter AND a second UV multicoated filter.

I have a number of lenses used with the M9 and M Monochrom with UV/IR cut filters on them. I've done tests of purple fringing before.

On the plus side: I can use the M9 and M Monochrom for UV photography and they cost a lot less than other cameras for this intended purpose.

 

The color rendering is nothing like the eye picks up. The M8- far less UV sensitivity. The M9- "Gee, Leica should have given away two free UV filters with each one sold"... M8 owners will get the reference. 

 

Cool- I get a free UV capability that's better than I hoped for.

 

Welcome, dear visitor! As registered member you'd see an image here…

Simply register for free here – We are always happy to welcome new members!

Edited March 26, 2021 by BrianS

[Advertisement]

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

[BrianS]

Since I have the Spectral Response of the sensor with the S8612- and I suspect the new cover glass and Dye was made to match this closely,

You can see the uptick in Red sensitivity in the UV region. Red+Blue- going to produce Purple response in the UV region. The plot cuts off, does not show the full range of sensitivity in UV. 

Welcome, dear visitor! As registered member you'd see an image here…

Simply register for free here – We are always happy to welcome new members!

Edited March 26, 2021 by BrianS

[Guest]

2 hours ago, hdmesa said:

I'm not hearing that message here. If anything, I feel like I read about equal affection for both CV and Leica lenses on this forum and elsewhere. There probably is less sharing of CV lens images here – most of that is on other forums where they are being adapted to non-M bodies.

CV 35 and 50 APO put performance right up there with the Leica 35/50 APO. And maybe Voigtlander will more quickly fill out their APO line over the next year or two.

Some fantastic images from the voigtlander 21mm f1.4 on the fred miranda forum albeit the sony mount version.

[BrianS]

Nikon DF (CMOS Sensor), AF-S Nikkor 50mm F1.8, wide-open. Filter placed in front of the UV Lamp.

 

And with this- I will continue to use UV Filters on my lenses.

Welcome, dear visitor! As registered member you'd see an image here…

Simply register for free here – We are always happy to welcome new members!

[LarsHP]

2 hours ago, hdmesa said:

The click-bait article title says they're useless, but they come right back and say a good one may be useful. From the article (Zeiss T* UV is 410nm cutoff):

"Some small combinations that could do something" is sloppy writing – saying, "There are, however, a few situations where UV filtration can be beneficial" would have been more clear.

The title is true - except if you have a quite old digital camera. Even on a Sony, the aggressively cutting Zeiss UV filter won't do much, if anything, since it cuts the same as the sensor filter.

2 hours ago, BrianS said:

I used the modern Jupiter-3+, which is multi-coated.

Nano-Coated optics tend to greatly diminish longer wavelength IR, but more into the band that is past sensitivity of a Silicon based sensor. I gave my Meade 1000/11 to a group that discovered the $4000  nano-coated field-scope killed IR that they were trying to collect. The Meade saved their experiment.

The Canon 50/0.95 is a fine performer on Sensors Unlimited Sensors past 1550nm. I used mine on a $25K imager.

 

You need to know the spectral characteristics of the camera in use to decide on what cut filters to select. For the M8, M9, and M Monochrom: if you use a classic lens and do not want UV light in the image, use a filter. The CCD and cover glass used is by-design sensitive to UV. These are the finest 35mm format  CCD based cameras that money can buy.

What kind of sensor is the "Sensors Unlimited Sensors" that goes past 1550nm? (From what I have read so far, regular silicon based sensors only go as far as about 1100nm.)

[BrianS]

The Sensors Unlimited array is InGaAS, went to around 2um. I spent the 1980s working with Long Wave (9um-12um) and Medium Wave IR (3um to 5um).

[LarsHP]

Advertisement (gone after registration)

13 minutes ago, BrianS said:

The Sensors Unlimited array is InGaAS, went to around 2um. I spent the 1980s working with Long Wave (9um-12um) and Medium Wave IR (3um to 5um).

... but those aren't used in normal digital cameras.

Edited March 26, 2021 by LarsHP
Checked out.

[LarsHP]

1 hour ago, BrianS said:

Since I have the Spectral Response of the sensor with the S8612- and I suspect the new cover glass and Dye was made to match this closely,

You can see the uptick in Red sensitivity in the UV region. Red+Blue- going to produce Purple response in the UV region. The plot cuts off, does not show the full range of sensitivity in UV. 

Welcome, dear visitor! As registered member you'd see an image here…

Simply register for free here – We are always happy to welcome new members!

Thanks for the graph. This is quite informative. I find it interesting that the 390nm wavelength will translate into violet and 400nm to indigo colors while 410nm will be blue - if I am not mistaken.

[BrianS]

1 hour ago, LarsHP said:

... but those aren't used in normal digital cameras.

That's what amazed me about the old Canon 50mm F0.95 lens: it was quite sharp on the Sensors Unlimited IR camera. This sensor was $25,000 almost  twenty years ago. We were a bit disappointed with the lens that came with it- I had the Canon with the C-Mount adapter. "Like night and day" difference.

 

My first "Commercially Available"  Digital camera was the Kodak DCS200ir, 1993. The Blue sensitivity of the first-generation CCD's was very low, Infrared- was very high. So I asked for clear coverglass. They made it for me- full-spectrum camera, the first sold by Kodak. Before that- all custom made sensors. 40 years working at a research lab- have seen some neat stuff. 

I have the 50/2 APO-Lanthar on my M Monochrom, and the 50/1.5 Nokton V2 on the M9. It's a great setup.

 

Comparing UV leakage on the M8, M9, and Nikon Df- the M8 with the UV/IR filter gives the colors closest to what I observe.

Edited March 26, 2021 by BrianS

[BrianS]

Found the M8 Spectral Response Curve. You can see why it does so well in IR.

Note the RED Dye uptick for short wavelength is smaller than that of the M9. 

 

Most lenses are corrected for the visible portion of the spectrum. Digital Sensors are sensitive to a greater portion of the spectrum, attempts have been made to attenuate the portions of the spectrum outside of the visible range by integrating filters into the sensor stack. These filters are not 100% effective. Filters used on Leica cameras must be thin, due to the geometry of the lens mount. Additional filtration is required to further attenuate light outside of the visible region than what is incorporated into the sensor stack. UV and IR leakage will show up as unnatural colors due to the response curves of the dye used in the color mosaic filter, usually as "Purple". I use UV/IR filters with several of my lenses on the M8 and M9 to reduce fringing caused by wavelengths outside of the visible range.

Welcome, dear visitor! As registered member you'd see an image here…

Simply register for free here – We are always happy to welcome new members!

Edited March 26, 2021 by BrianS

[LarsHP]

14 hours ago, BrianS said:

Found the M8 Spectral Response Curve. You can see why it does so well in IR.

Note the RED Dye uptick for short wavelength is smaller than that of the M9.

Most lenses are corrected for the visible portion of the spectrum. Digital Sensors are sensitive to a greater portion of the spectrum, attempts have been made to attenuate the portions of the spectrum outside of the visible range by integrating filters into the sensor stack. These filters are not 100% effective. Filters used on Leica cameras must be thin, due to the geometry of the lens mount. Additional filtration is required to further attenuate light outside of the visible region than what is incorporated into the sensor stack. UV and IR leakage will show up as unnatural colors due to the response curves of the dye used in the color mosaic filter, usually as "Purple". I use UV/IR filters with several of my lenses on the M8 and M9 to reduce fringing caused by wavelengths outside of the visible range.

To be honest, no I don't see why the M8 sensor stack should do so well in IR based on that graph. Above 750nm there's practically no transmission. Compared to a sensor without hot mirror filter this is practically nothing. Given the M8 doesn't have a dichroic UV/IR cut filter in addition to the absorptive ... well it might help, but this still looks very, very far from what an IR modded camera will let through.

16 hours ago, BrianS said:

Nikon DF (CMOS Sensor), AF-S Nikkor 50mm F1.8, wide-open. Filter placed in front of the UV Lamp.

And with this- I will continue to use UV Filters on my lenses.

Regarding your Nikon Df, if you check the graph I linked to twice (at Petapixel, but data are from Kolari Vision) it has a hard cut at about 415nm, so whatever you see is within the visual range as long as you don't have a proper UV photography filter stack in front of the lens.

Which filter did you put in front of the UV lamp? If it's one of the dichroic UV/IR cut filters, then the angle of the filter may explain the color change.

Since apparently nobody scrutinized it, here it is. We can't see exactly which of the green Nikon graphs is the Df though. Nikon Df is listed in the first row.

Welcome, dear visitor! As registered member you'd see an image here…

Simply register for free here – We are always happy to welcome new members!

 

Trying to go back to something related to the original topic (chromatic aberration) of this thread, putting an aggressively UV cutting filter like the Zeiss T* may help if the sensor stack doesn't cut much itself. Modern DSLR and mirrorless cameras won't benefit, but digital Leica M cameras may benefit if they don't have a dichroic UV/IR cutting filter. This seems to be the case with the M8 and M9. Regarding the newer models, I don't know. (Using dichroic UV/IR cut filters on wide angle lenses will interfere with visual light, so that is another story.)

[LarsHP]

18 hours ago, BrianS said:

If you have an M9 or M Monochrom, use a UV filter. At least Test the lens and camera combination before making general assumptions.

Leica M9, Voigtlander Nokton 50mm F1.5 V2, Multi-coated version, wide-open. Same lamp as used with the M8.

No Filter, B&W UV/IR Cut Filter, and B&W UV/IR Cut Filter AND a second UV multicoated filter.

I have a number of lenses used with the M9 and M Monochrom with UV/IR cut filters on them. I've done tests of purple fringing before.

On the plus side: I can use the M9 and M Monochrom for UV photography and they cost a lot less than other cameras for this intended purpose.

The color rendering is nothing like the eye picks up. The M8- far less UV sensitivity. The M9- "Gee, Leica should have given away two free UV filters with each one sold"... M8 owners will get the reference. 

Cool- I get a free UV capability that's better than I hoped for.

If you don't have a UV band pass filter like the Baader "Venus" or a stack of - say - a Hoya 2mm U-360 and 2mm Schott S8612, then you don't know what you are seeing. It may be squarely in the visual range or at its outer range. (BTW: Some consider 390nm as the threshold to UV.)

[BrianS]

Cut filters are gradual. The IR leakage on the M8 is about 5%- that caused major problems. Under extreme circumstances such as sunlight reflecting off out-of-focus metal objects, UV contamination still shows up.

As stated- and  by my own measurements, I will continue to use the UV and UV/IR cut filters. Other people should use combinations that they find satisfactory. The unnatural purple fringing that people complain about might be reduced using filters- you will not know until you try. I fiund this to be true with the 50/1.1 Nokton used on my M9 when I bought it 10 years ago.

[BrianS]

Made me drag out the Olympus EP2 (CMOS)  Full-Spectrum camera and my UV-APO Lens. This lens is designed for UV Photography.

First image: Nikkor L37c, lets a lot of UV out. I have an L39 somewhere- need to find it.

Second Image, Tiffen Haze1 on the left and Leica UV/IR on the right. These cut into the UV much more than the L37c. The Haze1 and UV/IR filter produce about the same amount of filtration, I attribute that to the UV Lamp emitting mostly light in the UV region, little IR. The camera was bought for IR photography, specified to be full-spectrum. 

Note that the "purple" color of the UV appears to be close to the color that people complain about. It is my belief that it is caused by near-UV contamination that the lens is not corrected for. 

I do not use UV or UV/IR filters on my UV-APO and Pentax Ultra-Achromat. Would defeat the purpose.

 

Welcome, dear visitor! As registered member you'd see an image here…

Simply register for free here – We are always happy to welcome new members!

Edited March 27, 2021 by BrianS

[LarsHP]

3 hours ago, BrianS said:

Cut filters are gradual. The IR leakage on the M8 is about 5%- that caused major problems. Under extreme circumstances such as sunlight reflecting off out-of-focus metal objects, UV contamination still shows up.

As stated- and  by my own measurements, I will continue to use the UV and UV/IR cut filters. Other people should use combinations that they find satisfactory. The unnatural purple fringing that people complain about might be reduced using filters- you will not know until you try. I fiund this to be true with the 50/1.1 Nokton used on my M9 when I bought it 10 years ago.

I think it's most likely that the M8 is without a UV/IR cutting dichroic filter, a filter type that is very sharp cutting (not gradual).

BTW: Do you know if the Leica UV/IR cut filter is the same as the B+W or Hoya? Or is it a different one? (Baader, for instance, makes an excellent UV/IR cut filter which could be what Leica uses too.)

2 hours ago, BrianS said:

Made me drag out the Olympus EP2 (CMOS)  Full-Spectrum camera and my UV-APO Lens. This lens is designed for UV Photography.

First image: Nikkor L37c, lets a lot of UV out. I have an L39 somewhere- need to find it.

Second Image, Tiffen Haze1 on the left and Leica UV/IR on the right. These cut into the UV much more than the L37c. The Haze1 and UV/IR filter produce about the same amount of filtration, I attribute that to the UV Lamp emitting mostly light in the UV region, little IR. The camera was bought for IR photography, specified to be full-spectrum. 

Note that the "purple" color of the UV appears to be close to the color that people complain about. It is my belief that it is caused by near-UV contamination that the lens is not corrected for. 

I do not use UV or UV/IR filters on my UV-APO and Pentax Ultra-Achromat. Would defeat the purpose.

Thanks for posting further test shots. However, these shots just show how much the various filters pass UV and what color the UV takes. It does support your theory, but violet / purple / magenta color "fringing" in an image could just as well be from a combination of red and blue wavelengths. We see that in tests for LoCA and LaCA from cameras that have effectively cut away both UV and IR too.

I will do a test with my full spectrum converted Sony a7 + Nokton 50mm f/1.2 Asph VM with and without Zeiss T* UV filter and post my findings.

[LarsHP]

Okay, I did a quick test with my full spectrum converted Sony a7 + Nokton 50mm f/1.2 Asph VM with and without Zeiss T* UV filter. Below are the two shots @ f/1.2 in 100% from the center of the image. Purposely unfocused to show purple fringing.

I can see that there is more purple fringing in the shot without the UV filter.

In other words, those (BrianS and others) who say that a UV filter will reduce purple fringing seems to be right, but this requires that the camera is without a built-in UV filter which most modern digital cameras have. Leica M cameras may be a notable exception.

Welcome, dear visitor! As registered member you'd see an image here…

Simply register for free here – We are always happy to welcome new members!

Edited March 27, 2021 by LarsHP
update / correction

[Cronilux]

Some people in this forum need to start working on their skills instead of pixel peeping the last tiny fringes. It’s unbelievable how many brick walls, test charts and fruit bowls I’ve seen here. 

[lct]

19 minutes ago, Cronilux said:

Some people in this forum need to start working on their skills instead of pixel peeping the last tiny fringes. It’s unbelievable how many brick walls, test charts and fruit bowls I’ve seen here. 

Unbelievable how many people like working nowadays .

[Advertisement]

Hallo,

Look here: Leica M

[Alan Friedman]

Here is a look at the level of apochromatic correction of my copy of the VM Apo-Lanthar 35 at f2. This image is a cropped jpeg at 100%, using a macro M-L adapter on the Sigma fp to bring close focus to within 10" of the ruler. Click to open the full size file. The camera is hand held at 1/60 to avoid banding from the LED lighting in my office. Looking at the words, one can see a faint green hue beyond focus and purple in front of focus... but close enough for government work, in my opinion.

 

Welcome, dear visitor! As registered member you'd see an image here…

Simply register for free here – We are always happy to welcome new members!

Edited March 29, 2021 by Alan Friedman

[paulcurtis]

23 minutes ago, Alan Friedman said:

Here is a look at the level of apochromatic correction of my copy of the VM Apo-Lanthar 35 at f2. This image is a cropped jpeg at 100%, using a macro M-L adapter on the Sigma fp to bring close focus to within 10" of the ruler. Click to open the full size file. The camera is hand held at 1/60 to avoid banding from the LED lighting in my office. Looking at the words, one can see a faint green hue beyond focus and purple in front of focus... but close enough for government work, in my opinion.

 

Welcome, dear visitor! As registered member you'd see an image here…

Simply register for free here – We are always happy to welcome new members!

Yes, that's pretty good. As you say i can see both the green and purple. Shame you don't have a Leica to compare with! And shame the MFD isn't the same as the E mount version (35cm) i think the VM is 50cm?

I look forward to seeing how the lens performs and if it matches the 50 well!

cheers
Paul