Home

Please support mflenses.com if you need any graphic related work order it from us, click on above banner to order!

SearchSearch MemberlistMemberlist RegisterRegister ProfileProfile Log in to check your private messagesLog in to check your private messages Log inLog in

Please double check me - Fluorescent light
View previous topic :: View next topic  


PostPosted: Sat Jun 22, 2024 1:26 pm    Post subject: Please double check me - Fluorescent light Reply with quote

Not sure this is the right place for this question. Please double check me: If you use daylight balanced film under fluorescent light then people will come out looking green (and hence the purple FL filters)? Or does it depend on the color temperature of the fluorescent light that is installed?

Regards, Christine


PostPosted: Sat Jun 22, 2024 4:57 pm    Post subject: Reply with quote

Generally, yes.

The green effect is somewhat (not entirely) reduced under 5000-5200k lamps.
The worst seem to be the 3500k soft white fluorescent tubes.
Be aware that most fluorescent tubes are being replaced with LED bars these days for their energy conserving nature.

Get the Kodak pocket photo guide from the mid to late 80's. Excellent beginner's guide to color balancing filters...

ETA: It can be difficult at times to determine what the light source is coming from fluorescent fixtures.
Generally, the lower color temperature lamps will have a low pitched hum emitted from the ballast/driver section of the fixture.
The LED's will create moire patterns on low shutter speeds at the monochrome setting on digital cameras, with a high contrast subject, usually around 1/30th sec, or slower.

You don't mention your film type- the filter requirements differ between transparency and color print film.

-D.S.


PostPosted: Sat Jun 22, 2024 7:06 pm    Post subject: Reply with quote

Worst case if you have a room in a National Trust/English Heritage property that is part-lit by daylight through a window, part-lit by incandescents, part-lit by cheap (100Hz flicker) LEDs, and part-lit by fluorescents, and no tripods allowed (flash is prohibited anyway) Rolling Eyes


PostPosted: Sat Jun 22, 2024 10:21 pm    Post subject: Reply with quote

Thank you. The film type is Kodak ColorPlus, ISO 200. The space a hall where one of my bands rehearses. Under normal circumstances there is little or no daylight there.

Good point about the fluorescents being replaced by LED. I am pretty sure these are still fluorescent. To my eye the color temperature is on the warmer side. At least it isn't factory hall stark white as you sometimes see.

Thank you for the tip on that book, I will see if I can find a copy.

On the subject of LEDs. In the same space when we have a concert we will have colored LED lights hanging from the ceiling. While they are much more energy efficient than the incandescent stage lights we used to have, they have a nasty flicker at 50Hz (Europe). The human eye doesn't really pick that up but filming with a telephone or digital camera for example the light intensity goes up and down in waves.

Regards, Christine


PostPosted: Sun Jun 23, 2024 12:36 am    Post subject: Reply with quote

https://imaging.kodakalaris.com/sites/default/files/files/resources/E7022_Gold_200.pdf

Scroll down for fluorescent filter values. They do differ for different lamp color temperatures.
Data sheet is for 200 Gold C-41 film. I can't turn up a data sheet for your film, which should be close for filter requirements.

-D.S.


PostPosted: Thu Jul 18, 2024 1:08 pm    Post subject: Reply with quote

Even if you use dedicated filters with the correct values for the given lamp, you will never get correct colors.

Fluorescent lamp have a discontiuous spectrum - meaning that a few single wavelengths come at a much higher intensity than the rest of the spectrum:

https://en.wikipedia.org/wiki/Fluorescent_lamp#/media/File:Spectrum_of_halophosphate_type_fluorescent_bulb_(f30t12_ww_rs).png

It's impossible to remove these high intensity peaks by common photographic filters.

S


PostPosted: Thu Jul 18, 2024 3:47 pm    Post subject: Reply with quote

stevemark wrote:
Even if you use dedicated filters with the correct values for the given lamp, you will never get correct colors.

Fluorescent lamp have a discontiuous spectrum - meaning that a few single wavelengths come at a much higher intensity than the rest of the spectrum:

https://en.wikipedia.org/wiki/Fluorescent_lamp#/media/File:Spectrum_of_halophosphate_type_fluorescent_bulb_(f30t12_ww_rs).png

It's impossible to remove these high intensity peaks by common photographic filters.

S


I'll agree with this up to a certain point. Although results will not be 100% perfect color, using a color temperature meter will get color much better corrected than simple charts and tables referencing~ perhaps close enough for easier final correction in the darkroom.

That articles skimps a bit on voltage versus color stability. The originators must be aware of it- they picked a middle of the road voltage to set their examples by.

-D.S.


PostPosted: Thu Jul 18, 2024 5:08 pm    Post subject: Reply with quote

Doc Sharptail wrote:
stevemark wrote:
Even if you use dedicated filters with the correct values for the given lamp, you will never get correct colors.

Fluorescent lamp have a discontiuous spectrum - meaning that a few single wavelengths come at a much higher intensity than the rest of the spectrum:

https://en.wikipedia.org/wiki/Fluorescent_lamp#/media/File:Spectrum_of_halophosphate_type_fluorescent_bulb_(f30t12_ww_rs).png

It's impossible to remove these high intensity peaks by common photographic filters.

S


I'll agree with this up to a certain point. Although results will not be 100% perfect color, using a color temperature meter will get color much better corrected than simple charts and tables referencing~ perhaps close enough for easier final correction in the darkroom.

-D.S.


I have pretty extensive experience with LED lamps (MUCH better than classical fluorescent lamps) used for illumination of artwork, e. g. in museums and churches. Very difficult to get an "accurate" (="acceptable for our eye") color reproduction. I have been documenting the restoration of a major church in Rome, and after having completed the job using the freshly installed LED light we decided to start all over again, using the natural light only. MUCH better / nicer colors with natural light (indirect sunlight) than with LEDs! And of course using grey charts and color metering all the time ...

S


PostPosted: Thu Jul 18, 2024 6:51 pm    Post subject: Reply with quote

stevemark wrote:
...
I have pretty extensive experience with LED lamps (MUCH better than classical fluorescent lamps) used for illumination of artwork, e. g. in museums and churches. Very difficult to get an "accurate" (="acceptable for our eye") color reproduction. I have been documenting the restoration of a major church in Rome, and after having completed the job using the freshly installed LED light we decided to start all over again, using the natural light only. MUCH better / nicer colors with natural light (indirect sunlight) than with LEDs! And of course using grey charts and color metering all the time ...

S


Fascinating! Any guesses/theories as to why? LED light tends to be concentrated within a small wavelength range. Do you think pigment reflectivity skews the color balance? Would combining images with differing white balance help with obtaining more accurate colors?

I did have similar problems reproducing a painting -- I simly could not get the colors correct! The artist pointed out to me part of the painting was missing from the photo!


PostPosted: Thu Jul 18, 2024 7:32 pm    Post subject: Reply with quote

visualopsins wrote:


Any guesses/theories as to why?


Neither guess nor theory - simply look at the emission spectra of common sunlight, LEDs and fluorescent lamps:


Of course you can adjust ("calibrate") the three colors "red" / "green" / "blue" on your sensor (or choose corresponding filters for film) - but if you 1) use LED light and 2) reduce the blue sensitivity of your sensor to compensate for the intense LED emission @ 450nm, you will 3) also reduce the sensitivity in the cyan range (about 480 nm) where the LED emission is weak. Guess what happens if a pigment of a painting mainly reflects around 480 nm? It will look pretty dark. Ifthe whole image consity of cyan tones, that's not a problem - you simply can prolong the exposure. But if you have a wide array of colors, the cyan will look rather dull. Similar problems arise with the low red / NIR emissions of the typical LED. Even after calibration with grey cards the reds AND the greens would look unnatural. everything has a greenish tint, and the reds are just "plain reds" - not much differentiation between the different shades of red ... (same for the greens BTW). I'll dig out a few examples later on.

S


Last edited by stevemark on Thu Jul 18, 2024 7:35 pm; edited 1 time in total


PostPosted: Thu Jul 18, 2024 7:34 pm    Post subject: Reply with quote

stevemark wrote:
Even if you use dedicated filters with the correct values for the given lamp, you will never get correct colors.

Fluorescent lamp have a discontiuous spectrum - meaning that a few single wavelengths come at a much higher intensity than the rest of the spectrum:

https://en.wikipedia.org/wiki/Fluorescent_lamp#/media/File:Spectrum_of_halophosphate_type_fluorescent_bulb_(f30t12_ww_rs).png

It's impossible to remove these high intensity peaks by common photographic filters.

S


Stephan; you were absent for a while and may have missed visualopsin's post about how the old phpBB version used here has issues rendering urls with apostrophes (or, relevant here, parentheses):

https://forum.mflenses.com/how-to-use-single-quote-here-in-subject-or-link-t85295.html

If you replace "(" in the url with "(" and ")" with ")" the link will render fine and take the reader directly to the image:

https://en.wikipedia.org/wiki/Fluorescent_lamp#/media/File:Spectrum_of_halophosphate_type_fluorescent_bulb_(f30t12_ww_rs).png

NOTE: if you later want to edit the post and hit the "Edit" button, all the hard works gets undone an you will have to make all the substitutions in the urls all over again. For important posts that you may want to edit, it is worth keeping a plain text version for yourself...

stevemark wrote:

I have pretty extensive experience with LED lamps (MUCH better than classical fluorescent lamps) used for illumination of artwork, e. g. in museums and churches. Very difficult to get an "accurate" (="acceptable for our eye") color reproduction. I have been documenting the restoration of a major church in Rome, and after having completed the job using the freshly installed LED light we decided to start all over again, using the natural light only. MUCH better / nicer colors with natural light (indirect sunlight) than with LEDs! And of course using grey charts and color metering all the time ...

S


The CRI quoted for LED lighting helps a bit to differentiate the bad from the horrible, but it is an imperfect measure (with a few different standards). A relatively small number of test colour samples are used, depending on the CIE standard used (see Wikipedia article below). Also note the "Criticism" section.

For critical works that doesn't suffer from metameric failure, IMO nothing beats natural light or daylight-balanced incandescent (yet).

https://en.wikipedia.org/wiki/Color_rendering_index

https://en.wikipedia.org/wiki/Metamerism_(color)
https://en.wikipedia.org/wiki/Metamerism_(color)#Metameric_failure


Last edited by RokkorDoctor on Thu Jul 18, 2024 7:45 pm; edited 1 time in total


PostPosted: Thu Jul 18, 2024 7:39 pm    Post subject: Reply with quote

BTW, I got best reproduction results by photographing the painting in the same lighting she used for painting.


PostPosted: Thu Jul 18, 2024 8:05 pm    Post subject: Reply with quote

visualopsins wrote:
BTW, I got best reproduction results by photographing the painting in the same lighting she used for painting.


I'm trying to get my head around whether that actually makes sense, from a perceptual / viewing intent perspective...

As an extreme example, assume for argument sake that an artist paints a colour painting using a low pressure sodium-vapor lamp (near monochromatic orange light) for lighting. Hint: this is one weird (original?) artist and the resulting painting will likely look horrible when viewed under daylight Wink .

Should the reproduction than be an orange monochromatic one as the painting would have been observed whilst painting under sodium-vapor lighting, or is the more accurate reproduction one that replicates the (horrible) colour image as it is seen under natural daylight?

Interesting question I think...


PostPosted: Thu Jul 18, 2024 8:14 pm    Post subject: Reply with quote

RokkorDoctor wrote:

Stephan; you were absent for a while and may have missed visualopsin's post about how the old phpBB version used here has issues rendering urls with apostrophes (or, relevant here, parentheses)


Indeed! Thanks for repeating that information Wink

Here's a sample - two JPGs straight out of the A900 (apart from re-sizing of course). Both taken within a few minutes, one with LED light (left), the other with "natural" light (indirect daylight partially colored by slightly tinted glass windows):



In both cases a grey card was used for adjusting the blue/yellow balance (aka color temperature) and red/green balance (aka R/G balance); the grey card was positioned on top of the altar, roughly were the small image of the bearded monk is sitting.

The differences between the two photographies are quite stunning, especially when looking at the main painting itself. Even more pronounced are the color differences when looking at the marble columns (especially the right one). You can easily see how tuning up the red channel (when using LED light, left image) results in pretty accurate depiction of the dark red spots on the marble column. The more orange-red parts of the column, however, are quite a bit overexposed, resulting in a strange and completely un-natural apperarance of the column - certainly not what you want to publish in a large high quality book ...

S


PostPosted: Thu Jul 18, 2024 8:33 pm    Post subject: Reply with quote

L.E.D.'s can be a pretty unreliable light source- especially with drivers that have been in use for a while- with the blue and purple visible to the naked eye on a driver near the end of it's useful cycle. This seems to occur more at higher voltages btw- 210-220 to 347. Early forms of L.E.D.'s were pretty bad for strange colors. From my somewhat limited reading on the subject, it had something to do with the way the diodes themselves were manufactured.

My own experiences with L.E.D. lighting have led me to avoid it where ever possible. Strange effects on contrast in your samples too, Stephan- was it the same lens?

-D.S.


PostPosted: Thu Jul 18, 2024 8:53 pm    Post subject: Reply with quote

Fluorescents were a royal pain in the film days!
There were color differences even between brands and if there was a mixed light situation it was impossible
I had 3 different filters for Fluorescents and always had to take a reference pic (Kodak Q13) for the lab to try and do magic

I guess it is the same with LEDs these days


PostPosted: Thu Jul 18, 2024 9:03 pm    Post subject: Reply with quote

stevemark wrote:
RokkorDoctor wrote:

Stephan; you were absent for a while and may have missed visualopsin's post about how the old phpBB version used here has issues rendering urls with apostrophes (or, relevant here, parentheses)


Indeed! Thanks for repeating that information Wink

Here's a sample - two JPGs straight out of the A900 (apart from re-sizing of course). Both taken within a few minutes, one with LED light (left), the other with "natural" light (indirect daylight partially colored by slightly tinted glass windows):



In both cases a grey card was used for adjusting the blue/yellow balance (aka color temperature) and red/green balance (aka R/G balance); the grey card was positioned on top of the altar, roughly were the small image of the bearded monk is sitting.

The differences between the two photographies are quite stunning, especially when looking at the main painting itself. Even more pronounced are the color differences when looking at the marble columns (especially the right one). You can easily see how tuning up the red channel (when using LED light, left image) results in pretty accurate depiction of the dark red spots on the marble column. The more orange-red parts of the column, however, are quite a bit overexposed, resulting in a strange and completely un-natural apperarance of the column - certainly not what you want to publish in a large high quality book ...

S


The image on the right looks far better by a country mile. Even if the colour rendering had been OK, the LED lighting still would have been far too harsh IMO; the softer natural lighting gives much more pleasing softer shadows, especially noticeable at the top of the main painting, the missing distracting candle silhouettes, and the bottom of the two round columns.


PostPosted: Fri Jul 19, 2024 1:49 pm    Post subject: Reply with quote

Doc Sharptail wrote:
Strange effects on contrast in your samples too, Stephan- was it the same lens?

-D.S.


Yes, Canon TS-E 4/17mm L in both cases (albeit with a slightly different taking position). I have cropped the original JPGs to some extent, though, to match the size of the main altar & painting (which BTW is oil on slate [shale], by Francesco Salviati)

S