Jump to content
UltravioletPhotography

[Filter Test] Hrommagicus Magic


Andrea B.

Recommended Posts

What is the raw color of the Hrommagicus filter?

 

I like to determine this for my filters by photographing an UV/Vis/IR white standard which has broad uniform, diffuse reflectivity such as the 5x5 inch Spectralon square shown in the photo. The Hrommagicus filter is a cyan filter having a bit more blue than green as you can see from the raw histograms. Cyan is at 180° on the color wheel. The Hromaggicus is at approximately 189°. Just call it 10° toward the blue away from cyan.

 

I made a white balance on the Spectralon in Photo Ninja. The screen shot shows that the Temperature = 15000 and the Tint = 59. I saved this white balance in Photo Ninja for use on a later test scene.

d610_uvNikkor_hrommagicus_sun_spectralon_20200126sf_18343rawCompLabel.jpg

 

 

When photographing the Spectralon, the blue channel was the first to blow out. The shot is actually somewhat underexposed to prevent hitting the wall in the blue channel because Photo Ninja will not white balance on a blown out area. This is probably a correct approach. Interpolation is good for dealing with highlights, but not so useful for a correct white balance which should not be based on guessing. The raw histogram in log format shows that I got fairly close to the right hand wall in the blue channel. That is not obvious in the linear version.

The raw histogram was created from only the cyan area. The left-hand side of the photo was excluded.

d610_uvNikkor_hrommagicus_sun_spectralon_20200126sf_18343rawHistoCenterLin.jpgd610_uvNikkor_hrommagicus_sun_spectralon_20200126sf_18343rawHistoCenterLog.jpg

 

d610_uvNikkor_hrommagicus_sun_spectralon_20200126sf_18343makeWhiteBal01.jpg

 

 


 

Hrommagicus Test Scene

 

I photographed some sky and a dark green juniper. There are a bunch of dried grasses in the foreground. OK, maybe not the greatest shot ever, but it was cold out there!

 

I made an in-camera white balance against the scene just as seen in the photo. When attempting an in-camera UV or IR WB with a Nikon, remember Bob Friedman's advice to first set a proper exposure in manual mode. Under a colored filter the metering will not always show the correct exposure. For this Hrommagicus a good exposure was about 2 stops longer than what the meter was showing. At the metered exposure, the D610 could not make the WB measurement.

 

This in-camera WB turned out to be the one having the foliage colour closest to red (as either 0° or 359° on the color wheel.) That has been important for some members here who are working on achieving the Aerochrome "look". But the outcome here was a pure surprise to me. And I'm not sure you would necessarily get pure red foliage for a different scene.

 

The version you are seeing here is the extracted JPG. No formal conversion was done. No edits were made. Resizing and labeling was done in Photo Mechanic. The in-camera sharpening was preserved as well as the Vivid picture control. (Resizing may have dulled the sharpening, of course.)

d610_uvNikkor_hrommagicus_sun_juniper_shotInCamWB_vividPC_20200126sf_18395asShotLabel.jpg

 

 

Here is the test scene in all its raw cyan glory, as shot.

The red, green and blue channels are also shown. (The two green channels were averaged by Raw Digger.) The blue channel is the brightest. The raw histograms are as expected - blue leads green just a bit.

d610_uvNikkor_hrommagicus_sun_juniper_shotInCamWB_vividPC_20200126sf_18395rawCompLabel.jpg

 

d610_uvNikkor_hrommagicus_sun_juniper_shotInCamWB_vividPC_20200126sf_18395rawRedChan01.jpgd610_uvNikkor_hrommagicus_sun_juniper_shotInCamWB_vividPC_20200126sf_18395rawGreenChan01.jpgd610_uvNikkor_hrommagicus_sun_juniper_shotInCamWB_vividPC_20200126sf_18395rawBlueChan01.jpg

 

d610_uvNikkor_hrommagicus_sun_juniper_shotInCamWB_vividPC_20200126sf_18395rawHistoLin.jpg

d610_uvNikkor_hrommagicus_sun_juniper_shotInCamWB_vividPC_20200126sf_18395rawHistoLog.jpg

 

 


 

Two Test Scene Variations

 

This version of the test scene was converted in Photo Ninja. The white balance made from the Spectralon was applied to the file. No camera color profile was added. The color of the foliage has shifted about 10 degrees away from pure red (0°/359°) back towards blue. This produces more of a cerise (cherry red) colour.

 

Is this a more valid colour version of the test scene? Who can say? We are dealing with false colour here. Anything goes! Using a Spectralon white balance made just prior to shooting any scene is a good way to standardize false colour outcomes when needed. But feel free to push your false colour in whatever direction pleases you best.

 

[side Note: It is good practice to photograph a white standard for every scene. Our cameras do not always measure white balance correctly under UV, IR or color filters even when you are measuring WB against the white standard itself. If you make documentary photographs, this step is mandatory indoors or out. Outdoors maybe more so because sunlight is changing composition and colour every second. ]

 

This conversion does not have the same amount of saturation in the sky colour as does the extracted JPG above because I was only making simple conversions. But you can see from the sampled sky colour (pushed to 100% brightness and 100% saturation) that a richer sky colour can be easily attained.

d610_uvNikkor_hrommagicus_sun_juniper_shotInCamWB_vividPC_20200126sf_18395noProfLabel.jpg

 

 

This version was also converted in Photo Ninja. The preceding Spectralon WB was again applied to the file. Then I added a Nikon D610 visible color profile made in sunlight for my converted D610. The red foliage shifts a few degrees more. And the sky colour moves also - away from cyan towards blue.

Combining a standard WB with a colour profile is another legitimate approach for standardization of false colour results, when desired.

d610_uvNikkor_hrommagicus_sun_juniper_shotInCamWB_vividPC_20200126sf_18395pnD610sunProfLabel.jpg

 

 


 

Just Playing Around

 

I wondered what would happen if I photographed a Color Checker Card under the Hrommagicus Filter and used it to create a Hrommagicus false-colour profile. Mind you now, this makes absolutely no sense. I just thought it would be fun! The result produced a very rich, deep colours for the test scene.

 

The CC Card is shown in its raw colours first.

d610_uvNikkor_hrommagicus_sun_ccChecker_shotMono_20200126sf_18316rawCompCol.jpg

 

Here is the profiling step in Photo Ninja.

d610_uvNikkor_hrommagicus_sun_ccChecker_shotMono_20200126sf_18316makeProf01.jpg

 

This is the Hrommagicus false-colour profile applied to the CC Card photo.

[side Note: This photo also illustrates why we do NOT use a visible colour card for white balance under our UV or IR or other filters. The Spectralon white balance was applied to this photo, but you can see that the top row of visible neutral patches are not at all neutral because the CC Card is designed for visible reflectivity only.]

d610_uvNikkor_hrommagicus_sun_ccChecker_shotMono_20200126sf_18316pnLabel.jpg

 

Now the Hrommagicus false-colour profile is applied to the test scene. This is not the Aerochrome look, but it is quite beautiful.

d610_uvNikkor_hrommagicus_sun_juniper_shotInCamWB_vividPC_20200126sf_18395pnLabel.jpg

Link to comment
I am really bad at editing myself. So I am always happy to be informed of any errors, typos, mis-interpretations or other goofiness. Thank you kindly!!
Link to comment

Interesting analysis.

My first thoughts reading through, was you need more blue. The last shot looks like you dialed in 6000K to the CWB, using the filter.

But its good to know where its color lands and with this we can push the blues into blue, not cyan and try to hold the reds. Many other cheap alternatives just fail to keep the reds red, they move into orange when you get the blues looking nice.

Link to comment

Hi David !! Hope this analysis is useful in some little way. As you said, knowing where the color lands can be helpful - especially if you want to try to produce an Aerochrome look without too much processing. :smile:

 

There is a question at the end of this next paragraph.

To make these color shifts, you have to use a color tool which can deal separately with each hue in order to attain blue without affecting red (and v.v.). For Photo Ninja that would be the Color Enhancement color patches. Choose the cyan patch and push it to blue. In NX2 it is easy to pop a color point onto the cyan and shift it to blue. But what does one use in Big Photoshop to accomplish this shift? I have only PS Elements which has a Hue/Saturation tool that I find less easy to use for such a shift.

 

What I found interesting was that the on-scene WB produced the best red. How well does that hold up for other scenes? I gotta try a few more. Also, like your comment, I want to try a range of K settings to see what we get with this filter.

 


 

I just realized that I didn't do any color samples on the raw composite. Going to do that now and then update the topic.

 


 

Hi Col !! How are you doing with the wildfire situation where you live? I hope all is OK for you.

 


Link to comment

Thanks Andrea, All OK where I am, just got a bit smokey, that was all thank goodness, but a very bad experience for too many people.

Some areas burnt were larger then Great Britain.

Link to comment

Just out of curiosity, would anyone happen to have a link to a transmission profile for a "Hrommagicus filter"?

 

I am completely wrong from what I posted here originally.

Cadmium has informed me that the filter is:

"2mm KG3 + 0.08mm 729 Polyester + 1mm clear" All glued together.

That would be a good combination for protection.

 

I have damaged at least two similar combinations that I have made myself. Do not put Isopropyl alcohol on any Lee filter. It will draw the color dye off and smudge the polyester.

The transmission spectrum is:

post-188-0-20480900-1580215655.jpg

Link to comment

That chart is considerably more green than what I got.

So the IR must be passing thru the blue dyes and shifting that peak? Or what?

 

The peak 500 nm color looks approximately like this at full brightness & full sat.

 

green500nm.jpg

 

Caution: Please don't over-interpret the wavelength to RGB approximation. It is only an approximation.

https://academo.org/...r-relationship/

 

Working backwards, my above cyan 189° ~ (0 218, 255) ~ 481 nm.

 

This all reminds me to go look up the actual tristimulus values for some of these filters or filter layers.

Link to comment

Andrea,

This review for Capture one is interesting. Seems the color can be easily manipulated with mouse click.

 

https://m.dpreview.com/reviews/phase-one-capture-one-20

 

I only own Capture one version 6. So a little out of date. So is my DXO versions 9pro and basic 11. With my new computer,, I will have to see if I buy either new. I also have Afinity photo to play with.

 

But Bernard has found some useful edits in Gimp. So I might just try Gimp, Glimpse and the new Raw therapee first.

 

My Adobe Photoshop is CS2, when I was a student. So haven't ever updated it. Also my lightroom is version 3, which I never liked the catalog.

Link to comment
In NX2 it is easy to pop a color point onto the cyan and shift it to blue. But what does one use in Big Photoshop to accomplish this shift? I have only PS Elements which has a Hue/Saturation tool that I find less easy to use for such a shift.

In Photoshop you can use the Selective Color settings.

Choose cyan and use the sliders to push it to whatever color balance you want.

Link to comment

Please sign in to comment

You will be able to leave a comment after signing in



Sign In Now
×
×
  • Create New...