Andrea B. Posted September 28, 2016 Share Posted September 28, 2016 See Also: [Filter Test 340/10 #1] Gotta work out a problem or two, but here is a 340/10 result anyway [Filter Test 340/10 #2] Cured the flare but it's ring around the posy here [Filter Test 340/10 #3] here [Filter Test 340/10 #4] Edmund Hard-coated 340/10 Rear-mount Experiment [Filter Test 340/10 #5] Edmund Hard-coated 340/10, Sunflower, Sun [Filter Test] 340/10 #6 Progress was made, finally. UPDATE: Progress was then lost. Given that the mirror-like Edmund Otpics 340/10 induces reflection rings in a photograph, I decided to try to work around that by placing the subject of the photo in the center of the ring patterns. This sacrifices a few thousand pixels or so, but at least I got some workable -- sort of -- photos this time. I also managed to keep the reflection problem somewhat under control by not photographing in bright sunlight. I shaded the subject or shot in the shade and used my UV-flash for the primary illumination. Of course that created another interesting little problem with false colours. The appearance of a subject in sunlight with the 340/40 is quite different from its appearance under the UV-flash. And so it goes. This is a Gaura flower. It has a UV-absorbing signature with a bit of bright reflection at the base of the petals. Equipment [Nikon D600-broadband + SB-14mod UV-Flash] Visible Light [f/11 for 1/200" @ ISO-400 in Sunlight with Baader UV/IR-Cut Filter] You are seeing the flower from the side here. Typically the two rounder petals are at the top with the stamens and style arcing gracefully downward between the two narrower side petals. Ultraviolet Light [f/11 for 1/160" @ ISO-400 in Sunlight with Baader UV-Pass Filter] The slightly darker petal veining seen here under the BaderU is less noticeable under the 340/10. Ultraviolet Light [f/11 for 15" @ ISO-1600 in Sunlight with Edmund Optics 340/10 Hard-Coated UV-Pass Filter] I started with the raw colours and tried various edits to see what I would get. Then I made a typical NX2 conversion to see what that yielded. I expected to get false greens and yellows. Instead I got false blues and yellows. There remains a certain amount of "veiling flare" in spite of my placement of the flower. That can be dealt with somewhat by application of more of a dehaze tool to increase local contrasts. AS SHOT in Monochrome Picture Control Raw Composite from Raw Digger with No Edits Think Orange! The Red and Green channels caught most of the light. Raw Composite with Contrast and Sharpness Edits in Photo Ninja Sharpness should be applied locally for better results, i.e., a less grainy background. Raw Composite with One Levels Edit in Capture NX2 Raw Composite with Two Levels Edits in Capture NX2 After the second levels edit the reflection rings started to show up more. Conversion in NX2 with White Balance Made on Background I really think I like the preceding version derived from the Raw Composite better than this standard conversion which did not work so well for this particuar filter. Link to comment
Andrea B. Posted September 28, 2016 Author Share Posted September 28, 2016 Here is a Rudbeckia experiment with the 340/10. VisibleMy late season Rudbeckias always have little quirks like dark streaks, curled petals or quilling. BaaderUWhat a Rudbeckia is supposed to look like in UV????? :D The next 4 were made with the Edmund Optics 340/10 using SB-14 UV-flash illumination. This time I was less successful at evading the reflection rings. And I had no idea where to go with the false colours. 340/10 As shot in monochrome 340/10 Variation 1Blue and yellow seems to pop up automatically with click-white tool. 340/10 Variation 2With some effort I cranked things over to green and orange-yellow. But, meh! It's rather heavy looking. 340/10 Whizbang Variation 3Sometimes Photo Ninja goes nutso and coughs up something strange and wonderful.OK, maaaaybe I was a little heavy-handed with the Detail slider. But I have no idea where this pale blue and pretty lime-yellow came from. Link to comment
Andrea B. Posted September 28, 2016 Author Share Posted September 28, 2016 Conclusion: I do not like this 340/10.In addition to the reflection ring mess, I am not seeing the lovely detail from it like we typically get in our UV photos.And shooting narrowband around 340 does not seem to show us anything we don't already know. Oh well.We live. We learn.And then we just throw our hands up in the air and give up on hard-coated narrowband filters. Link to comment
Andrea B. Posted September 28, 2016 Author Share Posted September 28, 2016 Melampodium with BaaderU and SB-14 UV-flash Melampodium with 340/10 and SB-14 UV-flash Finally, for the record, here is a Sunflower shot with the 340/10 in Sunlight only. This gives the "expected" false colours of green and yellow. :D :D :D Photo Ninja ConversionBetter detail, eh?? As ShotTry to aim for the middle of the target. Raw Composite Link to comment
Andy Perrin Posted September 28, 2016 Share Posted September 28, 2016 For what it's worth, I love the colors in the 340/10 Variation 1! The third one is neat in a gee-whiz kind of way, but I like the first variation best. Complementary colors, as my 9th grade art teacher would say. Link to comment
Cadmium Posted September 28, 2016 Share Posted September 28, 2016 Very nice and interesting!I agree with Andy, 340/10 Variation 1 would be my favorite also.I see the 'dichroic ring' in some more than others. It is the same lens/focal length for all of these? Link to comment
JCDowdy Posted September 28, 2016 Share Posted September 28, 2016 Andrea,I seem to recall you were going to rear mount this filter. Are these rear mounted? Link to comment
Andrea B. Posted September 28, 2016 Author Share Posted September 28, 2016 Not yet! I'll get there eventually. Too much going on these days. :D Link to comment
BruceG Posted October 1, 2016 Share Posted October 1, 2016 Nice review! I was just curious to know how would a narrow band filter perform. The ring of light could be due to different transmission at different incident angle, therefore I am guess might a lens hood help to reduce the effect. Link to comment
Andrea B. Posted October 1, 2016 Author Share Posted October 1, 2016 I was using a lens hood, but perhaps not the right one. :D The first trial with the 340/10 showed a lot of flare without lens hood.[Filter Test 340/10] Gotta work out a problem or two, but here is a 340/10 result anyway The 2nd trial with a lens hood cured the flare.[Filter Test 340/10] Cured the flare but it's ring around the posy here I'm not entirely sure this filter is dichroic. Hard-coating is a single-coat plasma deposition. But that is not to say that angle of incidence might not play a role here. More experiments will tell the tale. Link to comment
BruceG Posted October 4, 2016 Share Posted October 4, 2016 I see, then rear mounting would very likely solve the problem. It would be easy for me, because I can just put whatever filter in front of the sensor for my mirrorless nikon J5. I am surprised the filter does not leak any IR, some bandpass filters only come with a portion of the spectrum curve, how did you know it would be IR proof when you bought it? Link to comment
Andrea B. Posted October 4, 2016 Author Share Posted October 4, 2016 The chart for this hard-coated 340FWHM10 filter showed approximately OD 4 blockage or better outside the 330-350 nm range until past 1100nm. (We can't record past 1100nm.) There are some 3.5 OD spikes in a few places, but that is still strong enough blocking that it would take a lot of forcing to push the IR through those spikes. I have not yet done an IR forcing test with this filter because of that blasted reflection problem. Link: http://www.edmundoptics.com/optics/optical-filters/bandpass-filters/hard-coated-od4-10nm-bandpass-filters/65190/On this page click on the Curve icon to see the transmission chart. Link to comment
OlDoinyo Posted October 5, 2016 Share Posted October 5, 2016 With a narrow-bandpass filter, there will be little or no real chromaticity. I suspect that these images are largely playing with cast gradients, i.e. you could create similar images by starting with the monochrome version and playing with color balance controls. Link to comment
Andrea B. Posted October 5, 2016 Author Share Posted October 5, 2016 Yes, I think this is a valid observation.I'm going to take a shot at demonstrating it. Here is a copy of the raw composite of the UV Rudbeckia photo in Post #2 above. I sampled a small area on the ray (petal) tip and on the ray base and made larger circles painted with those sample values so we could see the colour better. Next to those petal colour circles are two fully saturated, fully bright versions of the petal colours. The two values from petal tip and petal base have hue designations of 31° and 34° (regardless of saturation or brightness) which are obviously quite close together in a 360° circle. So, yes, the recording does seem to be monochrome. :D When we apply a white balance tool to the photo and wrench the hue values of these false colour into click-white submission, it seems we can "pull them apart" enough to get the RGB complementary blue/yellow false colour scheme. (For the record: the orange/green false colour scheme above was forced and not a result of either raw colours or white balancing.) Added: This Rudbeckia photo's raw colours were recorded under a combination of sunlight and UV-flash. Under sunlight only the raw colours would likely be slightly different. Link to comment
Andrea B. Posted October 5, 2016 Author Share Posted October 5, 2016 Here is the same type of analysis applied to the sunflower from Post #4 above which was recorded only in sunlight (and with a lot of reflection problems). The light and dark samples taken from the petal tip and petal base are in this case 14° apart in hue. This is still not a lot of distance apart in a 360° circle. But in this particular case the 14° difference takes us from a yellow colour to an orange colour which appear to be further apart than they really are. Input always welcomed for these little experiments and demonstrations. Let me know where I've made wrong turns, please! :D Link to comment
Recommended Posts
Please sign in to comment
You will be able to leave a comment after signing in
Sign In Now