bghazzal Posted June 22 Share Posted June 22 (edited) Hello all, I've been using UR-Pro filters for a while on GoPros, and really like working with them, notably for the consistency the UR-Pro (Shallow Water Cyan) gives in tropical ambient light. - More info on this and ambient light filter video experimentations in this GoPro/filters thread, as well as on Dreifish's thread here , and Interceptor121's test in his blog posts, notably here and here Alas, UR-Pro is a thing of the past, and the company is off the map... The aging company ghost-website that was still online but unresponsive a few years back is now gone, and the filters are no longer manufactured. We still find old info like this 2005 review of UR-Pro filters, but not much more info. This is a long shot, but would anyone have more info on the filters themselves, especially the UR-Pro SW Cyan? I have the Keldan SF -1.5 gels, the original Magic Filter gels and the UR-Pro SW Cyan (acrylic and glass). Magic and Keldan work great on my manual white-balancing compact, but for my GoPro flat-profile + grading purposes, I still find that the UR Pro gives the most consistent, gradable ambient light results, and would love to find a way to use it with the AOI UWL-03 lens I now have... It would be fantastic to recreate the UR-Pro SW CY filter in gel form for easy use in more recent GoPro and action cam models, especially now that accessory lenses like the INON/AOI/Backscatter wide lenses are available... Any leads? cheers ben Edited June 22 by bghazzal Link to comment Share on other sites More sharing options...
bghazzal Posted June 23 Author Share Posted June 23 (edited) Digging around, I found these historical documents to add to the pile : a very heated 😅 Polar Pro / UR-Pro comparison with some basic data on the UR-Pro SWCY https://interceptor121.com/2013/01/13/underwater-video-tips-polar-pro-red-filter-for-gopro/ and an 2005 X-Ray article: Photography_UsingFilters_15.pdf Otherwise coordinates for the historical, US-based UR-Pro company here: email no longer works. Checking my emails, in 2016 it was forwarding to SRP... Edited June 23 by bghazzal Link to comment Share on other sites More sharing options...
bghazzal Posted June 25 Author Share Posted June 25 (edited) Following up on the info posted in these two threads 1 & 2, and to try to better understand how UR-Pro filters work, I ran some tests today, both with the GoPro 7 and the Lumix LX10. The test setup isn't as precise as Dreifish's - I don't have a strobe, and no light meter to check the kelvin values of the light used. I also don't have a calibrated colour checker. However what I do have a Whibal calibrated greycard slate, which I used on the tests. I used what I had for the colour references, ie an AOW slate, a printed book cover, but nothing calibrated, all setup against a dirty Balinese grey wall... The light source used is a Backscatter MW4300 video light, which is supposed to be calibrated to 6000K in wide mode (for matching when used as a pair) - I had no way of checking if this was accurate - which was used in a dark room. So nothing too rigourous or fancy, but enough to highlight the differences in behaviour between the filters tested. Tests were as follows: - Light source: 6000K calibrated (?) Backscatter MW4300 video light - Cameras: GoPro7 Black / Lumix LX10 (stills and video clips for both) - Filters: none, UR-Pro 55m acrylic filter (SRP Blurfix convex design), UR-Pro 55m glass filter, UR-Pro 55mm glass filter (no brand markings, so not 100% sure this is a UR-Pro), Keldan Sepctrum Filter SF -1.5 gel, Original Magic Filter Gel For the LX10 tests, I also added a generic Howshot filter I have designed for the Inon UWL-H100 Test sequence: GO PRO 7 Camera White Balance set to 6000K, GoPro colors photo/4K clip without filter (paper no filter, 6000K) photo/4K clip with UR-PRO acrylic filter (paper UR-PRO ACRY 6000K) photo/4K clip with UR-PRO glass 55mm filter (paper UR-PRO GLASS 6000K) photo/4K clip with UR-PRO (?) glass 2 55mm filter (paper UR-PRO GLASS 2 6000K) photo/4K clip Keldan SF -1.5 (paper Keldan SF-1.5 6000K) photo/4K clip with original Magic filter (paper Magic 6000K) Camera White Balance set to Native, GoPro colors (flat profile) photo/4K clip without filter (paper no filter, Native) photo/4K clip with UR-PRO acrylic filter (paper UR-PRO ACRY Native) photo/4K clip with UR-PRO glass 55mm filter (paper UR-PRO GLASS Native) photo/4K clip with UR-PRO (?) glass 2 55mm filter (paper UR-PRO GLASS 2 Native) photo/4K clip Keldan SF -1.5 (paper Keldan SF-1.5 Native) photo/4K clip with original Magic filter (paper Magic Native) Lumix LX10 Camera White Balance set to 6000K raw photo/4K clip without filter (paper no filter, 6000K) raw photo/4K clip clip with UR-PRO acrylic filter (paper UR-PRO ACRY 6000K) raw photo/4K clip with UR-PRO glass 55mm filter (paper UR-PRO GLASS 6000K) raw photo/4K clip with UR-PRO (?) glass 2 55mm filter (paper UR-PRO GLASS 2 6000K) raw photo/4K clip Keldan SF -1.5 (paper Keldan SF-1.5 6000K) raw photo/4K clip with original Magic filter (paper Magic 6000K) raw photo/4K clip with generic Howshot filter (paper Howshot 6000K) Edited June 25 by bghazzal Link to comment Share on other sites More sharing options...
bghazzal Posted June 25 Author Share Posted June 25 (edited) GO PRO 7 Camera White Balance set to 6000K, GoPro colors photo/4K clip without filter (paper no filter, 6000K) photo/4K clip with UR-PRO acrylic filter (paper UR-PRO ACRY 6000K) photo/4K clip with UR-PRO glass 55mm filter (paper UR-PRO GLASS 6000K) photo/4K clip with UR-PRO (?) glass 2 55mm filter (paper UR-PRO GLASS 2 6000K) photo/4K clip Keldan SF -1.5 (paper Keldan SF-1.5 6000K) photo/4K clip with original Magic filter (paper Magic 6000K) Camera White Balance set to Native, GoPro colors (flat profile) photo/4K clip without filter (paper no filter, Native) photo/4K clip with UR-PRO acrylic filter (paper UR-PRO ACRY Native) photo/4K clip with UR-PRO glass 55mm filter (paper UR-PRO GLASS Native) photo/4K clip with UR-PRO (?) glass 2 55mm filter (paper UR-PRO GLASS 2 Native) photo/4K clip Keldan SF -1.5 (paper Keldan SF-1.5 Native) photo/4K clip with original Magic filter (paper Magic Native) Edited June 25 by bghazzal Link to comment Share on other sites More sharing options...
bghazzal Posted June 25 Author Share Posted June 25 (edited) The video clips of the filter tests follow the same sequence - idea was to check if there were differences in the way WB was handled when shooting video as well (which is what I do exclusely), maybe grabing some captures to check that as well. Now I'll need to wipe the cobwebs off an old copy of lightroom and see if I can extract some values. Visually, there is a massive difference between Magic and UR-Pro, which is much yellower. As expected, the Keldan SF-1.5 is the weakest - and it seems the 2nd non-branded glass filter (which was sold to me as a UR-Pro) is not the same build, as suspected... Probably not a UR-Pro - Cheeky vendor... The acrylic and glass UR-Pros seem identical at first glance - other than the fact that the acrylic is not in a good state, (it lived mounted on a GoPro in a BCD pocket, albeit with a lens cap on, for thousands of dives over the years...). But I have three acrylic UR-Pro 55mm, including one which is in a much more preserved state... Edited June 25 by bghazzal Link to comment Share on other sites More sharing options...
bghazzal Posted June 25 Author Share Posted June 25 (edited) rats... realised I only transfered the jpegs for the GoPro tests, and deleted the raw files... But the LX10 raw tests are valid. So, after importing and rebalancing in Llightroom, I'm getting the following white balance values for the LX10 raw stills No filter, raw, camera WB set to 6000K to match light supposed to be 6000K - white balance "as shot" 5300K, +19 Magenta - corrected white balance 5000K, +28 Magenta UR-PRO acrylic corrected WB 2350K, -14 Magenta UR-PRO glass 55mm corrected WB 2250K, -8 Magenta UR-PRO (?) glass 2 corrected WB2300K, -23 Magenta Keldan SF -1.5 corrected WB 3050K, -27 Magenta original Magic filter corrected WB2950K, +24 Magenta Howshot corrected WB2200K, -32 Magenta Edited June 25 by bghazzal Link to comment Share on other sites More sharing options...
bghazzal Posted June 26 Author Share Posted June 26 (edited) Ran the test again on the GoPro raw files to check the difference, testing in WB set to 6000K then to "native" (flat), testing only the UR-Pro acrylic and the Original Magic filters. Results are as follows for GoPro RAW still files WB 6000K NO FILTER, WB set to 6000K As shot WB 5200K Magenta +6 Corrected WB 4900K Magenta +9 UR-PRO acrylic, WB set to 6000K Corrected WB 3100K Magenta -45 Original Magic, WB set to 6000K Corrected WB 3450K Magenta +22 WB NATIVE NO FILTER, WB set to NATIVE As shot WB 5500K Magenta +13 Corrected WB 4900K Magenta +11 UR-PRO acrylic, WB set to NATIVE Corrected WB 3100K Magenta -44 Original Magic, WB set to NATIVE Corrected WB 3500K Magenta +20 Edited June 26 by bghazzal Link to comment Share on other sites More sharing options...
bghazzal Posted June 26 Author Share Posted June 26 (edited) To develop the comparison between the UR-Pro SW Cyan (55mm acrylic SRP-Blurfix model) and the original Magic filter gel, based on the GoPro raw still files (LX10 raw files gave similar results, we can make the following observations: When it comes to white-balance, other than the UR-Pro being slightly warmer, the main difference between the two filters comes from the handling of magenta. As observed on the GoPro (and the LX10), rebalancing with UR-Pro gives a negative Magenta reading, whereas rebalancing with the Magic gives a positive Magenta reading. GoPro raw files, WB set to Native, 6000K light (results are similar with camera WB set to 6000K) UR-Pro Cy, unbalanced WB UR-Pro Cy, rebalanced WB: 3100K Magenta -44 Original Magic filter gel, unbalanced WB Original Magic filter gel, rebalanced WB: 3500K Magenta +20 *** Based on these readings, there is a 400K warming effect difference between the two filters, which is not much, however when it comes handling of Magenta, the UR-Pro increases magenta by +31 (+13 to -44) whereas the the original Magic decreases Magenta by -31 (+13 to +20) As an experiment, if we try to match the two filters by warming up the Magic filter and boosting the Magenta, we get a very similar result between the unbalanced UR-Pro filter and the modified unbalanced original Magic filter: UR-Pro Cy, unbalanced WB Modified Magic filter gel, unbalanced but with WB warmed to 5900K and Magenta increased to 76: Results are now visually quite close for such rough measurements. But while the results are visually similar, the histogram still shows significant differences in colour distribution. To fine-tune this, it would be interesting to get readings on main individual colour channels (squares) as well, see how both filters affect them, but I'm not sure how to approach this. Edited June 26 by bghazzal Link to comment Share on other sites More sharing options...
Davide DB Posted June 26 Share Posted June 26 Thanks Ben, Very interesting. Looking at the original unbalanced and white balanced histograms I see that the two filters are very different. Even it seems that the UR-Pro filter leaves a lot of green compared to the Magic Filter. You should photograph a color scale so you can better compare the two photos. Even something not accurate thst you can print by yourself: https://upload.wikimedia.org/wikipedia/commons/4/4f/Color_Checker.pdf 1 Link to comment Share on other sites More sharing options...
bghazzal Posted June 26 Author Share Posted June 26 (edited) Yes, there's clearly a lot going on. I couldn't find a way to do this with the raw files in Lightroom, and my Photoshop CS6 can't seem to handle raw files (probably some camera raw issue), but to give an rough idea of what's going on with the colours with the two filters, I worked on the JPGs in Photoshop, and grabbed the colour data from each square, but from the GoPro files set to 6000K WB (theoretical calibration of the video light used) If anyone knows how to do this with raw files (I'm guess my image software is too old), it would be more accurate... The (very rough) data I have is as follows: GOPRO UNFILTERED - WB 6000K RED SQUARE: R187 G100 B93 ORANGE SQUARE: R193 G118 B87 YELLOW SQUARE: R200 G212 B0 GREEN SQUARE: R62 G153 B96 BLUE SQUARE: R105 G110 B168 SLATE-GRAY SQUARE: R111 G100 B117 GOPRO + URPRO - WB 6000K RED SQUARE: R243 G132 B113 ORANGE SQUARE: R244 G149 B117 YELLOW SQUARE: R252 G208 B49 GREEN SQUARE: R171 G144 B91 BLUE SQUARE: R183 G129 B143 SLATE-GRAY SQUARE: R197 G140 B123 GOPRO + MAGIC - WB 6000K RED SQUARE: R229 G125 B96 ORANGE SQUARE: R234 G139 B95 YELLOW SQUARE: R237 G213 B29 GREEN SQUARE: R153 G168 B87 BLUE SQUARE: R168 G142 B127 SLATE-GRAY SQUARE: R176 G142 B117 If we reorder this per square to track the global changes the filtration induces, we get: RED UNFILTERED R187 G100 B93 UR PRO R243 G132 B113 MAGIC. R229 G125 B96 ORANGE UNFILTERED R193 G118 B87 UR PRO R244 G149 B117 MAGIC R234 G139 B95 YELLOW UNFILTERED R200 G212 B0 UR PRO R252 G208 B49 MAGIC R237 G213 B29 GREEN UNFILTERED R62 G153 B96 UR PRO R171 G144 B91 MAGIC R153 G168 B87 BLUE UNFILTERED R105 G110 B168 UR PRO R183 G129 B143 MAGIC. R168 G142 B127 SLATE-GRAY UNFILTERED R111 G100 B117 UR PRO R197 G140 B123 MAGIC R176 G142 B117 This, combined with the histogram and white balance / temp data, should give a rough ideal of how each of the two filters is affecting the colour spectrum (and, hopefully, what needs to be adjusted on the Magic gel to more closely match the UR-Pro). Edited June 26 by bghazzal 1 Link to comment Share on other sites More sharing options...
Davide DB Posted June 26 Share Posted June 26 1 hour ago, bghazzal said: GREEN UNFILTERED R62 G153 B96 UR PRO R171 G144 B91 MAGIC Humm difficult to understand with numbers and we are approaching "radius and entrance pupil" overthinking 🙂 As example this is the green channel, from the RGB value you wrote above. Unfiltered UR PRO MAGIC 1 Link to comment Share on other sites More sharing options...
Davide DB Posted June 26 Share Posted June 26 3 minutes ago, Davide DB said: Humm difficult to understand with numbers and we are approaching "radius and entrance pupil" overthinking 🙂 As example this is the green channel, from the RGB value you wrote above. Unfiltered UR PRO MAGIC So it seems that the Magic has a lower filter on the green channel. 1 Link to comment Share on other sites More sharing options...
bghazzal Posted June 26 Author Share Posted June 26 (edited) Yes definitely -now that the WB is clearer, the idea is just to try to see what parts of the color spectrum are the most affected by each filter, and how. The example you just posted fits nicely with the grading data - as Dreifish noted in his mixed lighting / ambient filter tests, and also as I remarked when grading the footage, the greens are much less affected with the Magic Filter than the UR-Pro. The brownish tint is also consistent, with the look and feel of the ungraded footage posted here For sure, these numbers are difficult to deal with, but they also do show roughly how the filters affect the colour spectrum, beyond the white balance / magenta temperature reading. When grading, the UR-Pro and Magic are really quite different in the rendering of certain parts of the colour spectrum. It would be more accurate if I could work on the raw files instead of the JPGs, and the LX10 test shots are more standardised Better still would be using testing with a standardised colour checker and a 100 CRI strobe, instead of an 70 CRI video light and an AOW slate, but I think there's still workeable data to be gained. These are screeshots of the lightroom WB and histogram data from the LX10 raw files, WB set to 6000K (like the light), exposure ISO 125, f/8, 1/60 (some exposure change due to the vignetting caused by the filter mount) WB is taken on the same spot on the Whibal calibrated grey card - LX10, unfiltered, WB uncorrected (temp 5350K, tint Magenta +19) unfiltered, wb rebalanced in post to check (temp 5000K, tint Magenta +27) UR-Pro acrylic filter, WB uncorrected (temp 5350K, tint Magenta +19) UR-Pro acrylic filter, WB corrected (temp 2350K, tint Magenta -14) Magic filter, WB uncorrected (temp 5350K, tint Magenta +19) Magic filter, WB corrected (temp 2950K, tint Magenta +23) ******* These are extras, but here is an original glass UR-Pro (not damaged like the acrylic UR-Pro), which shows a reading very close reading to the acrylic UR-Pro, and also a Keldan -1.5 Spectrum Filter gel. UR-Pro glass, WB uncorrected (temp 5350K, tint Magenta +19) UR-Pro glass, WB corrected (temp 2250K, tint Magenta -10) Keldan SF -1.5 gel, WB uncorrected (temp 5350K, tint Magenta +19) Keldan SF -1.5 gel, WB corrected (temp 3000K, tint Magenta -28) It would be nice to grab the colour square data from these raw files instead, but I don't know how to do this in the version of lightroom I have (there's no "colour pipette") and my photoshop CS6 is too old to handle Panasonic RW2 files... EDIT - I found a way to display the RGB info in lightroom (switching to "soft proofing"), and will work on the raw files filtered RGB values, which should be more precise. Edited June 26 by bghazzal Link to comment Share on other sites More sharing options...
DreiFish Posted June 26 Share Posted June 26 Basically, I'd interpret those results (similar to Interceptor121's conclusions in this 2013 post) to conclude that the Magic filter is nothing more than a 1 1/2 strength CTO filter (orange filter, which filters out blue light), whereas the UR PRO is a combination of a slightly stronger CTO filter (perhaps a CTO 2) combined with a minusgreen (magenta) filter since it also filters out some of the greens. My hypothesis is that you can get a very similar result to the UR PRO filter by taking a Magic Filter and stacking it with a 1/8 or 1/4 or maybe 1/2 minusgreen filter to it. Or, indeed, if just working with Lee or Rosco gels, you could replicate the attenuation properties of the UR PRO filter by stacking the appropriate strength CTO and Minusgreen gels. I ordered some Minusgreen gels from B&H Photo Video that should arrive later this week and I have CTO gels on hand. I also have Magic and UR Pro filters lying around somewhere in my closet. Time permitting, I can try to validate my hypothesis with some test data later in the week. Other random observations: You can test the color temperature of the Backscatter 4300 torch just as you did -- by illuminating a white target and taking the measurement in lightroom. Looks like it's actually closer to 5000k than 6000k, which is good. The brownish color cast the UR Pro produces (rendering green as brown) is probably what makes it aesthetically appealing. It eliminates unnatural color cast from skin tones and sand, which makes the whole image appear more neutral (if desaturated) and natural. I actually try for a similar end result when processing my photos in Lightroom by desaturating the aqua channel and shifting the hue of green towards yellow-brown. 2 Link to comment Share on other sites More sharing options...
bghazzal Posted June 27 Author Share Posted June 27 (edited) Thanks, that makes total sense, the green to brown conversion of the UR-Pro CY just works as a base for grading. On one hand, a filter like the Keldan spectrum seems more accurate in its colour handling, but on the other, the UR-Pro offers an easy to work with base-colour palette, which just works for grading to visually pleasing results in most blue water situations. The Magic's handling of magenta (+23 on a readjusted WB) also explains the spike I was seeing at depth, and makes it difficult to handle imo - for video at least... Really looking forward to seeing the gel test results! One other thing i've been considering, given that the Keldan gels are -1.5, is to stack two (the standard glass filter is -2, the next strength gel is -3.5) This should give a rough -3, which probably can be handled by the GoPro in sunny tropical conditions, and maybe offer a good base as well, since the Keldan doesn't rebalance with magenta boost like the Magic filter does. I've worked on the LX10 raw files to extract colour data, and will post a summary of the observations for the 3 filters. Edited June 27 by bghazzal Link to comment Share on other sites More sharing options...
bghazzal Posted June 27 Author Share Posted June 27 (edited) This is a summary of observations / data gathered from filter tests based on raw files shot on the LX10. Parameters: LX10 raw files, WB set to 6000K (like the theoretical value of the MW4300 light), exposure ISO 125, f/8, 1/60 (some exposure change due to the vignetting caused by the filter mount) - WB point is taken on the same spot on a Whibal calibrated grey card. - Light source is a Backscatter MW4300 video light used in wide mode, theoretically calibrated to 6000K (in reality, seems closer to 5000K), CRI 71.1 - colour data is samled in Lightroom Classic (colour slate used isn't a calibrated colour checker) LX10, unfiltered, WB uncorrected (temp 5350K, tint Magenta +19) wb rebalanced in post to check (temp 5000K, tint Magenta +27) Unfiltered colours observed: LR Histogram: UR-Pro acrylic 55mm filter, WB corrected (temp 2350K, tint Magenta -14 - warms by 3000K) Filtered colour modification observed: LR Histogram: - Known filter light loss values for UR-Pro Cyan (measured by Interceptor121): -1.6 Ev Original Magic filter, WB corrected (temp 2950K, tint Magenta +23 - warms by 2400K) Filtered colour modification observed: LR Histogram: - Known filter light loss values for the Magic Filter (measured by Interceptor121): -1.6 Ev Keldan SF -1.5 gel, WB corrected (temp 3000K, tint Magenta -28 - warms by 2350K) Filtered colour modification observed: LR Histogram: - Known filter light loss values for the Keldan SF -1.5 gel: -1.5 Ev (Keldan SF -2: -2 Ev) UR-Pro glass filter, WB corrected (temp 2250K, tint Magenta -10 - warms by 3100K) (note: filter in better condition than the acrylic UR-Pro) Filtered colour modification observed: LR Histogram: - Known filter light loss values for UR-Pro Cyan (measured by Interceptor121): -1.6 Ev Side-by side filtered colour modifications observed in LR: Unfiltered: Ur-Pro Glass : Ur-Pro Acryl: Orig. Magic: Keldan SF1.5: Based on these observations, most significant variations occur on the yellow (warmer on the UR-Pros), green (stronger filtration by UR-Pro, as expected) and also on the handling of the slate-gray tones by each filter. Magic handles the blue channel quite differently - which probaly relates to the magenta boost when rebalancing the WB. Keldan and Magic are actually quite close, but differ on the cooler end of the spectrum, greens, blues and slate-grey. Temperature-wise, the UR-Pro Cyan has the strongest warming effect (roughly 3000K and 3100K for -1.6 Ev), followed by the original Magic Filter (2400K for -1.6 Ev) whereas the Keldan SF-1.5 has the weakest (2350K for -1.5 Ev, in the same range as the Magic). Only the Magic filter requires boosting the Magenta tint to rebalance the WB, whereas all other filters require a reduction of the Magenta tint to rebalance. Edited June 27 by bghazzal Link to comment Share on other sites More sharing options...
bghazzal Posted June 29 Author Share Posted June 29 (edited) Digging into the archives, i found this interesting 2003 article comparing UR-Pro filters to fluorescent filters, and giving "recipes" for some combo filters including the UR Pro ones. https://wetpixel.com/articles/filters-and-ambient-light-photography/ Here is an extract: Color Conversion filters. Color conversion filters come in two varieties, warming and cooling. For this discussion the warming filters are valuable since they can be used to counteract the cooling effects of the water. White light is characterized by its color temperature, specified in Kelvin, where warm light has low values around 3000K and cool light has higher values around 5500K. Although Kelvin is the measure photographers are accustomed to, it is more useful to speak of light in terms of mireds, or micro-reciprocal degrees. You can convert Kelvin to mireds by inverting and mutliplying by 1000000. color temperature equivalent mireds 2800K 357 3000K 333 3200K 313 3800K 263 4100K 244 4700K 213 5500K 182 6500K 154 There are two series of warming filters, the 81 series and the 85 series. They are characterized as follows: filter mireds 81 9 81A 18 81B 27 81C 35 81D 42 81EF 53 85C 81 85 112 85B 131 There are two analogous series of cooling filters, the 80 and 82 series, and they have matching but negative mired values. For our application they are uninteresting but they can be used to adjust the color of strobes or halogen lights. When applying a filter to light simply add the filter value to the light value to get the result. Water generally needs 2-5 mireds per foot to compensate for its cooling effects with turbid water requiring less compensation. Fluorescent Filters. Up until now we've talked about filters that are primarily available as gels that you have to combine and mount yourself. As it turns out, photographers have had to deal with another light source that is overly green and that's fluorescent lighting. Because of that we have at our disposal a limited range of thread-on and rectangular filters that are combinations of the filters that we desire. Additionally, many underwater photographers are familiar with UR Pro, a company that specializes in filters for underwater use. The UR Pro filters are functionally like fluorescent filters. Below is a table that I've compiled characterizing these filters. filter mireds CC M UR Pro glass CY 140 75 UR Pro glass VLF 145 60 UR Pro resin GR 25 30 Tiffen glass FL-B 100 50 Tiffen glass FL-D 40 65 B+W FL-D glass 20 20 Singh-Ray resin FL-B 147 50 HiTech resin FL-B 96 75 HiTech resin FL-D -12 30 Hoya glass FL-D 34 30 Hoya glass FL-W 9 60 Lee resin FL-B 3600K 35 40 Lee resin FL-B 4300K 69 45 Lee resin FL-B 5700K 134 40 Lee resin FL-D 3600K -50 70 Lee resin FL-D 4300K -16 70 Lee resin FL-D 5700K 14 60 ---- This 2007 discussion is also worth a glance: It that the UR-Pros were sold as plastic that you can cut to size or glass, both are exact same underwater but plastic gets scratched really easy. and also describing UR Pro filters as simply FL-B filters repurposed. If you can find FL-B filters in your size they will work well and be much cheaper than UR PRo. UR-Pro was apperently run by Kirk Kreutzig (and were not always easy to reach) It also turns out Nick Hope was looking into the makeup of UR-Pros at the time as well: A few years ago I trashed a big UR-Pro plastic external filter on it's first ever dive (staghorn coral). In subsequent discussions with them about re-polishing etc. I was genuinely interested in what UR-Pro had managed to patent about their filters, and what makes them so unique. I had a search for the patent (was involved with patent stuff in my old career) but couldn't find it, and I could never get a specific answer or a patent number out of them. Just that they "sill hold the patents" and that other filters will not give the same performance. Would still love to know what's so special about them. Other old threads discussing the UR-Pro / Magic filters worth a read include this one here as as well as this one here I'll keep digging around, see if anything else comes up. cheers b Edited June 29 by bghazzal 1 Link to comment Share on other sites More sharing options...
bghazzal Posted June 30 Author Share Posted June 30 (edited) A little snippet on fluorescent filters (especially the UR-Pro) by Dr. Mustard from this 2005 thread: I have noticed quite a big difference between CC filters and Fluorescent filters is in the cameras ability to auto white balance. Fluor filters (specifically the UR-Pro series that I have used) tend to work really well on AUTO-WB (even when they actually don't look too good on the LCD screen when they are underwater). Whereas CC filters tend to need custom white balancing either in situ with a grey/neutral card or in RAW conversion in post processing. This shot, for example, was shot on Auto WB and can out of the camera looking pretty much like this. The only way I can get such good colour balance straight from the camera with a CC filter is in situ WB. The obvious answer to this is that fluor filters are better. But I am not convinced about this because at certain depths the camera uses more extreme white balance (colour temp and tint) settings for the UR Pro than are needed for the CC filter. It is just that the camera (well my cameras D100 and D70) have found it easy to get the right white balance (in Auto) with the Fluor filter as opposed to the CC filter. And also from this 2006 thread where Nick Hope is pondering UR-Pro vs. Magic for video The Magic Filter does work with Video, but requires manual white balance - which is not always advantageous with moving images. I can only comment on their use on stills, but the Magic filter is actually quite different from the UR Pro. I can tell a UR Pro shot from a Magic shot straight away (on stills). On stills the UR Pro tends to produce more Magenta blue water, while the Magic tends to more to cyany blues. On general reef topography I find the Magic much more neutral coloured (I find the UR Pro CY more muddy looking). The UR Pro also renders Anthias much more red rather than orange (I have just been in the Red Sea shooting them both - hence the Anthias comment). On the plus side for the UR Pro, I think that the UR Pro filter gives more contrasty images than the Magic Filter. Which tends to help moving images. All that said, I am reluctant to persue the video market with the Magic Filter. We set up Magic Filters to provide a solution for photographers than didn't already exist (namely a gel format UW filter). And it is not our intention to take on UR Pro (or others) in the video market. That said we are happy to supply filters to those who want to try them. Alex and also here in 2006 You cannot judge filters just by looking at them. Filters are rather more complicated that the colour that they look. For example UR Pro CY and SW-CY look identical but they are not. It is the details of the transmission spectra that are important. Another example is that the original magic filter looks orangy, but the transmission spectra is nothing like a standard orange filter. http://wetpixel.com/forums/index.php?act=A...pe=post&id=2619 This is the magic transmission spectra and it lets through lots of red, orange and yellow light, some green and very little blue. And orange filter, which may look quite similar, has a much more defined peak in the orange (although I don't have an example). The 2006 UWP article on the inception of Magic Filters is also definitely worth a read: http://www.magic-filters.com/download/UWP26a.pdf Here's an extract I found particularly interesting (highlights by me) : I have used many filters over the last few years and firm favourites have been Kodak’s Wratten CC (Colour Compensation) Red series and UR Pro’s CY and SWCY filters. Both have strengths but neither was the perfect solution for me and this is what started me on the path to Magic. The main limitation of CC Red filters is that only work in very narrow depth range, for example I have found that the CC0R only produces pleasing images between 2 and 6 metres, even with adjustable white balance). This makes them impractical unless we know the exact depth of the subject before we dive. UR Pro’s filters are designed specifically to counteract the filtering effect of seawater and work very well. They are an excellent choice for videographers and work over a wider depth range (I find the CY works best on stills between and 12 metres). Frustratingly, for the still photographer UR Pro filters are only available as glass or thick optical plastic, which means that they cannot be physically fitted on the most popular SLR wide angle lenses such as fisheyes (which can only take gel filters). I designed the Magic Filter to work over as wide a depth range as possible. The Magic filter differs from other UW filters as it is not designed to perfectly counteract the filtering effect of seawater because this is highly variable. Instead it adjusts colours to produce a colour balance that is easily corrected by the camera’s white balance as is this 2007 article And fast-forwarding to a very interesting 2010 thread with a bit of spectrometer action: Nick Hope: The Lee, Rosco etc gel filters are designed to just change the colour of electric lights and are allegedly not "optical" quality (whatever that means), although I have seen point-and-shoot users get away with them. You can get a swatch book from them with loads of samples. Apparently the Magic Filters are somehow different in a way that makes them "optical" quality, but exactly what that is, I have no idea. The archived UR-Pro site (last archived in 2019) is also accessible here: http://web.archive.org/web/20191209011451/http://www.urprofilters.com/content.do?region=FilterInstructions#tips This dive into this pre-Keldan, film to digital, SD to HD history is pretty fascinating - to me anyway 😅 Really worth looking into discussions/input by Craig Jones (author of the "filters and ambient light photography" article I posted extracts of above), Nick Hope and Alex Mustard. We've come a long way but also, in others, full-circle (with non-manually white-balancing cameras like action cams) Technological advances (sensors, RAW, and LED technology, with the availablity of powerful UW light sources) have changed certain aspects (MWB was a hot-topic back in the days, and I hadn't heard of the Expodisc...), but some technical questions remain when it comes to handling the UW colour spectrum, especially for UW video. Just look at Nick Hope's UW-MWB tests in 2007 Trying to figure out what the now extinct UR-Pro CY/ SWCY were feels a little like trying to figure the coca-cola recipe 😝 It's a little sad that such valuable information has seemingly vanished with the company and its patent (?). cheers Edited June 30 by bghazzal 1 Link to comment Share on other sites More sharing options...
Davide DB Posted June 30 Share Posted June 30 A quick search on FLB filter brought to this Tiffen filter: https://www.amazon.com/Tiffen-58FLB-58mm-FLB-Filter/dp/B001V5J3H4/?th=1 An "ancient" 2011 comment: Quote Great for Under water!! Bought this lens as a substitute for the URPRo CY filter to use for underwater video. I don't have any direct comparisons, but I will say that with ambient sunlight at 20-30 feet of water this lens performed well. The reds are red, the yellows are yellow (as opposed to a pale cream color). The URPro CY filter runs over $70 so this is a better buy. Plus, you can do a little color correction in post if needed, but it will put you that much closer to true colors. I bought the 55mm because it fit the GoPro flat lens mod from Snake River prototyping. 1 Link to comment Share on other sites More sharing options...
bghazzal Posted Monday at 05:43 AM Author Share Posted Monday at 05:43 AM Thanks for this, interesting product - in the discussions, Craig Jones and Nick Hope also bring up the glass vs plastic issue. To quote Craig's article again: First, your choice of filter depends on whether you are deep or shallow and whether the water is blue or green. The stronger FL-B filters are only suitable for blue water and moderate to deep depths (20-70 feet). Use the FL-D type filters for green water or shallow depths. For any given lens or housing, you may be able to use thread-on filters, rectangular filters (externally) or gels. Some combinations may be supported only by a UR Pro filter in which case your choice is fairly obvious. For those that can use thread-ons I prefer the Singh-Ray FL-B, the Hoya FL-D and the B+W FL-D. I have not used the Hoya FL-W but it may be desirable for very green water. For those who can mount a 4 inch/100mm Lee filter holder externally, the Lee FL-B filters provide essentially a shallow/medium/deep solution. I have also adapted a UR Pro CY resin filter to the Cokin X-Pro filter system. I personally avoid the Tiffen filters and the UR Pro glass filters due to their sandwich glass construction. Sandwich filters have many more light boundries that make them more susceptable to flare. For those lenses that have rear gel holders your choices are more interesting since you can roll your own. First choose your magenta gel. I would use 20 to 50 units depending on depth and clarity of the water. You can then optionally add a warming gel. I would bring an 81C, and 81EF, and a 85 for example. The benefit of the warming gel is that it brings the balance of the light closer to matching the optimal balance of the imaging sensor. The downside is that you give up light sensitivity. On bright days and deep dives use a strong warming gel. Otherwise use a weaker gel or leave it out entirely. Note from the editor - According to Craig, a good starting point is to purchase two strengths of magenta (say, 30 and 50 or 20 and 40), and then add two strengths of warming (81EF and 85). For shallow shots use the weak pair. For deep the strong pair. For green water use the strong magenta with or without the 81EF. You can combine the weak pair or the CC30M alone with strobes if you carefully control strobe power. Reduce strobe power by 1-2 stops to start. Filtering strobes is ideal, but we have some more experimentation to do before anything conclusive can be said. Link to comment Share on other sites More sharing options...
bghazzal Posted Monday at 06:12 AM Author Share Posted Monday at 06:12 AM (edited) Here is some news -following up on Nick Hope's idea, I ran a patent search for UR-Pro's Kirk Kreutzig in the US. patent data base and got 3 hits for optical filters. Among the 3 three filter patents, 2 actually concerned night-vision devices, but one was of particular interest: patent US-4542959-A "Color correction filter and method of taking underwater photographs therewith", filed on July 25, 1983 and published (patented) on September 24, 1985 Here is the downloadable patent file: 4542959.pdf Not sure if this is UR-Pro Cyan filter patent or not, but maybe there's some related information within. Here is the abstract: 6 pages of rather interesting reading, and while it's difficult to know if it is the UR-Pro Cyan we're familiar with, it does sound like a blue-water/cyan underwater photo filter... To summarize in copy-pastable form, the light wavelength (nm) to % of incident light transmitted is as follows: Below 370 nm: 0% Below 400 nm: 25% Below 410 nm: 27% Below 450 nm: 12% Below 470 nm: 8% Below 500 nm: 4% Below 520 nm: 7% Below 550 nm: 18% Below 570 nm: 50% Below 600 nm: 87% Below 700 nm: 90% EDIT- looking into it, it seems like this is indeed the UR-Pro Cyan, or at least one of its early versions Here is another related patent filed by W TUCKERMAN BIAYS, refering to U.S. Pat. No. 4,542,959 as covering the "UR/Pro Marine CY Filter" https://patents.google.com/patent/US5719715A/en It is well known that light is made up of electromagnetic radiation having a range of wavelengths usually considered to be between 0.4 and 0.7 microns. Underwater color correction is described in U.S. Pat. Nos. 4,542,959; 3,588,215 and 3,929,487. In addition, the following publications discuss the "UR/Pro Marine CY Filter" (which may be described in U.S. Pat. No. 4,542,959) Frink, "Warmer U/W Photos", Skindiver, December 1983; Herbst, "New Filter Banishes the Undersea Blues", May 1986 and Holland, "Filters May Improve your U/W Photography", February 1988. Murphy, in "Shark Eye", appearing in Skindiver, May 1987, apparently describes a filter which may be related to U.S. Pat. Nos. 3,588,215 and 3,929,487. Adams, in "Exposure Control for Underwater Photography", appearing in ElectroOptical System Design Conference, 1971, at page 336, discusses the use of color correcting red filters, as does Rowlands, in The Underwater Photographer's Handbook, pages 74-75, Van Nostrand Reinhold Co. Cardone, in "Clearoptic Lenses", in Skindiver, May 1989, describes a yellow filter, sometimes known as "shooters" glasses. As a bonus, here is a link to a local Chicago Tribune article published in 1999 on Kirk Kreutzig's various filter-related inventions, titled "FILMING 100 FEET UNDER SEA " cheers ben Edited Monday at 02:07 PM by bghazzal 1 Link to comment Share on other sites More sharing options...
Davide DB Posted Monday at 09:14 AM Share Posted Monday at 09:14 AM When filter hunting meets investigative journalism 😜 1 Link to comment Share on other sites More sharing options...
bghazzal Posted Tuesday at 05:34 AM Author Share Posted Tuesday at 05:34 AM (edited) Following up, I tried to find a blank light spectrum file I could plot values onto to give a visual rendering of the filter data, but couldn't find anything to work with... So I made the most of my fragile photoshop skills and adapted the filter curve given above by scaling it to an existing filter spectrum graph... (if anyone knows practical way to render the available data in a more accurate manner, do let me know!) Anyway, here are the results: UR-Pro Cyan filter spectrum transmission, based on available patent data As a reminder, known values for light wavelength (nm) to % of incident light transmitted are: Below 370 nm: 0% Below 400 nm: 25% Below 410 nm: 27% Below 450 nm: 12% Below 470 nm: 8% Below 500 nm: 4% Below 520 nm: 7% Below 550 nm: 18% Below 570 nm: 50% Below 600 nm: 87% Below 700 nm: 90% **** While this quick-and-dirty visual is far from perfect, it does give a good idea of how the filter is acting in blue water, and affects spectrum transmission and colours making it to the camera sensor. Some observations: - As noted in empirical tests, filtering of the blue-green spectrum is quite pronounced, with only 4% transmission at 500 nm (which removes the greenish hues, pushing them to yellow-orange) - There is a strong transmission of the orange-red spectrum, 90% transmission at 700nm+ (warming effect, to compensate for the filtering effect of water and restore colour balance) - The filter does let some deeper blues hues through (25 to 27% transmission in the 400 nm zone, likely to avoid losing water-column blue tones), while also strongly filtering out darker blues and,importantly, almost all the purple/magenta hues (0% transmission below 370 nm!) This strongly marked section of the curve seems to be the UR-Pro's most characteristic spectrum filtering feature, and would be the one to focus on when looking for alternatives: cheers! ben Edited Tuesday at 12:16 PM by bghazzal 1 Link to comment Share on other sites More sharing options...
bghazzal Posted Tuesday at 08:25 AM Author Share Posted Tuesday at 08:25 AM (edited) Looking into alternatives, on the Lee Filter side, the Lee 166 Pale Red is somewhat similar: Lee filter's customer service suggested this gel could be combined with the 226 Lee UV to block the UV spectrum, combination which would give the following curve: There are differences - the 600nm wave length with its warm yellow/oranges, is already at 87% transmission on the UR-Pro, as opposed to 64% on the Lee Pale Red 166 + 226 Lee UV combination. In the blues the UR-Pro's 410nm peak at 27% transmission is moved to a greener 420nm at 22% transmission, but otherwise it's quite close (with some blues but also a strong filtering of the 400nm greens, 5% for the UR-Pro for 4% for the combo). I'm guessing it will be a little cooler and with a bit more green, which doesn't sound too bad! Anyway, should be worth a try... b Edited Tuesday at 08:58 AM by bghazzal 1 Link to comment Share on other sites More sharing options...
Recommended Posts