Home PageAbout MindatThe Mindat ManualHistory of MindatCopyright StatusWho We AreContact UsAdvertise on Mindat
Donate to MindatCorporate SponsorshipSponsor a PageSponsored PagesMindat AdvertisersAdvertise on Mindat
Learning CenterWhat is a mineral?The most common minerals on earthInformation for EducatorsMindat ArticlesThe ElementsBooks & Magazines
Minerals by PropertiesMinerals by ChemistryAdvanced Locality SearchRandom MineralRandom LocalitySearch by minIDLocalities Near MeSearch ArticlesSearch GlossaryMore Search Options
The Mindat ManualAdd a New PhotoRate PhotosLocality Edit ReportCoordinate Completion ReportAdd Glossary Item
Mining CompaniesStatisticsUsersMineral MuseumsMineral Shows & EventsThe Mindat DirectoryDevice Settings
Photo SearchPhoto GalleriesNew Photos TodayNew Photos YesterdayMembers' Photo GalleriesPast Photo of the Day GalleryMineral Photography
╳Discussions💬 Home🔎 Search📅 Latest
Popular Sections & GroupsGeneralOpen discussion area.Identity HelpOpen discussion area.Improving Mindat.orgOpen discussion area.Fakes & FraudsOpen discussion area.Field CollectingOpen discussion area.PhotosPrivate group.Techniques for CollectorsOpen discussion area.
Recent Images in Discussions
Mineral PhotographyPerfect Bench Photo Lighting Setup by Alan Scalone
After considerable research and even more trial and error, the following is a detailed description of a perfect bench lighting photography setup:
16th May 2012 10:44 BSTAlan Scalone
Materials To Purchase:
3 - 3 foot black 12 volt ceiling track lighting tracks
2 - 90 degree angle track lighting connectors
1 - track lighting end cap
1 track lighting end cap electrical cord connection
6 Solux 12 volt 50 watt 100 CRI 36 degree 4700 color temperature bulbs
6 - 30 inch black 12 Volt track lighting goose neck fixtures
6 - 6 inch hose clamps and 1/2 inch metal strapping strips with holes
6 - 1 inch hose clamps
2 sheets of film industry gel diffuser sheets
Screw track lighting to table top in a U shape
install 2 goose necks to each of the 3 sides of the U
drill 8 inch metal hose clamps and nut and bolt each to a piece of metal strapping
use 1 inch hose clamps to attached metal strip (with 6 inch hose clamps attached) to the goose-necks
Attach a sheet of diffuser gel over each of the 6 hose clamps
So, goose necks allow for unlimited light placements for front,top, rear, side lighting
Solux is the ONLY 100 CRI (color resolution index) bulb on the market - does not change color of minerals
Hose clamps allow for positioning of diffuser gels away from the light bulbs so you do not smell the gel material burning while operating the lights.
Please contact Alan Scalone at alscalone at earthlink dot net for a photo of the setup or for other assistance
Why not post the photo here?
16th May 2012 13:12 BSTJolyon Ralph Founder
Here you go - Two photos showing the lighting setup
19th May 2012 21:37 BSTAlan Scalone
That is one funky looking setup there Alan, but I can see where it would work beautifully!!
24th May 2012 18:31 BSTPaul Brandes Manager
Thanks for sharing. (tu)
I am envious, but it must be kind of hot working around 6 halogen lights. I use two, and it raises the room temperature noticeably (and I switch them off between specimens).
24th May 2012 22:11 BSTKelly Nash Expert
25th May 2012 01:06 BSTAlan Scalone
Yes, setup certainly looks strange! The 6 goose necks are especially useful for photographing solid and transparent crystals. You can back light transparent crystals and position the front lights to reflect off a few crystal faces to have both the transparency and crystal shape represented in the image.
25th May 2012 01:09 BSTAlan Scalone
I actually do not get much room heat off the lamps. Not sure if they are what would be called halogen lamps. They are 12 volt, 2-pin, track lights and are very small lamps as well, like 2 inch diameter or so.
Very nice setup, Alan. What brand goosenecks did you use? I was out trying to recreate this today--we live in the same area, so I'm also curious where you got the lamps.
6th Jul 2012 03:18 BSTJeffrey Hapeman
7th Jul 2012 12:48 BSTAlan Scalone
Yes, there are quite a few goose necks out there and then there is finding the lowest price provider as well.
I purchases the goose necks from Lighting Direct.com - Cal Lighting HT-257 Contemporary / Modern 1 light flexible goose neck track head for HT series track systems. I went with black to avoid any color reflecting off the fixture surfaces and onto the rocks.
I also forgot to mention that there are a few different types of track - I used the "HT" track which is also one of the 3 most common type of track. You will notice the goose necks are a model "HT" as well.
That model is the longest ones I could find at 27". They do have perforated metal on the lamp shade which does allow the heat to not buildup on the bulb, extending bulb life. However, the perforated shade does allow the light to hit you in the eye when you are working around the lights. In the photos I posted you will notice the black flat plastic attached to the diffuser frames - stops light from hitting you in the eye - just attached with double sided sticky tape.
I also started with just 4 lamps but increased to 6 lamps to gain ultimate flexibility to light various specimens.
26th Nov 2012 20:45 GMTJim Robison
Found this thread and have a couple of questions for you about the two photographs of the setup.
1) It looks like there are 8 lights instead of 6. What am I missing here?
2) It appears that you have a sheet of clear glass for suspending your pieces above the black background cloth. How much space between the two?
3) How thick is the glass to span almost three feet of space?
4) It's now been some months since the thread was active. Have you made any changes since then or decided that something else would have enhanced the setup?
Forty bucks a light, bulbs a few bucks, surprisingly cheap.
28th Nov 2012 18:12 GMTTim Jokela Jr
I recall a bad case of sticker shock when I got the bill for my two solux goosenecks from some company out of NY.
Apparently market forces at work, driving prices down!
A very interesting and informative post.
Brilliant setup. (And yes folks, you do need 6 lights... believe me, two work for TN's, but not bigger.)
29th Nov 2012 07:21 GMTAlan Scalone
Yes, minimum 6 lights. I have since added two more to increase my flexibility for creating reflections on crystal faces. Other than that, no I have not enhanced the setup.
You can view photo work with the setup on my site http://www.kidzrocks.com I have done post work on most photos to 1) Sharpen filter, 2) increase brightness only if photo was a bit too dark. 3) black paint out white debris specs visible on background.
I use back plexiglass over the black cloth - not glass. You can use the thinnest plexiglass to keep costs down - the solid wood desk is under the plexi so. The plexi does scratch so be careful when you place and remove rocks from the surface and use a vacuum to remove debris. You can use both sides of the plexi and move it around as area become scratched.
The black plexi makes a very subtle and nice reflection of the rock on the surface and also of course makes a nice black background to show off your specimen. Obviously any color besides black or white will change the color of your specimens.
29th Nov 2012 7:19am
Nice work, Alan!! (tu)
4th Dec 2012 02:10 GMTPaul Brandes Manager
To much chaotic for me
17th Dec 2012 13:57 GMTMatteo Chinellato Expert
Although my photos are mediocre I do come ahead of at least half of other posters. It should not be so since I use an old 8MP Olympus camera (that can focus very close to the specimen) and a very simple setup. Focusing rack is there because of lack of room for the tripod legs.
15th Jan 2013 17:26 GMTJonathan Levinger Expert
So here is what I use; it is inexpensive and readily available. I rarely use the Halogens that come with the kit I bought on eBay and I added a frosted glass.
Here is a kit on eBay:
Looks like the link to eBay page is not allowed so you can look it up if you want under:
15th Jan 2013 17:32 GMTJonathan Levinger Expert
New 100w Photo Studio 16" Photography Lighting Tent Kit Backdrop Cube In A Box
Two $10 LED lamps from IKEA are great and one can use more than two if needed.
This are great as a side lamps for your microscope to.
Two or four way focusing rack is available on eBay, best 4 way sells for $35, frosted picture glass, PVC tubes and Velcro strips are from Home Despot or some other hardware joint.
the side walls of your photo box can act as a diffuser to.
Johnathan, very nice addition to this thread topic!
15th Jan 2013 21:56 GMTAlan Scalone
Since first posting this topic I have noticed it is quite a popular topic which confirms what I found when trying to figure out how to best setup to photograph minerals. There was all the standard light cubes and packaged setups out there that are just worthless for mineral photography and very little legitimate information published out on the net on how to create your own setup that actually works properly.
The one clear benefit of Johnathan's configuration that is missing in my setup is the ability to light a specimen from the bottom as well. I had considered cutting out the wood desk and installing a glass surface so I could also light from the bottom when needed and I may still make that modification.
Good stuff everyone and keep contributing!!
I don't have much of a photo setup, and I don't take many photos, but one thing I have found helpful with specimens hat have reflective crystal faces is to make sure a few selected faces are lit or partially lit in the photo. The idea is mentioned above.
16th Jan 2013 04:00 GMTKeith Wood
One thing that has worked is to use a light shining almost directly away from the specimen onto pieces of paper that I can position using string and a clip. The idea is to light up an individual face. When you look through the lens you can see the light reflecting off the face you are highlighting and no other face typically gets any light from that piece of paper. It is extra work, but allows a high degree of control over the lighting of individual faces - both for which faces you light up, and the degree to which you light them up. You have to have the rest of the room pretty dark, and you have to use a remote trigger to take the shot sometimes or you get in the way of your strategic reflectors. I've used up to three such reflectors to get a good effect, but it was a lot of work. If I built a better setup in the first place I could do it more easily.
This picture used two paper reflectors, one of which I was holding. The focus wasn't what I hoped, but the lighting was. The paper reflectors provided lighting to the two front faces, center and right. I especially like how I could make the face on the right be both reflective and transparent. It is the best aspect of the picture, capturing both form and depth.
out of focus in the termination and high grain visible in the photo
16th Jan 2013 08:07 GMTMatteo Chinellato Expert
One question I have is if there are any blue prints and instructions to build a more stable mineral photo box? My father and I would like to build one out of wood or something that can be broken down, moved, and put back together and still be sturdy. I do like the pictures and ideas on here, but still looking just the right setup for me. Thanks
11th Nov 2013 04:10 GMTAaron Cross
Thanks for shared your light setup but i think is excesive lamps. See my mineral gallery:
11th Nov 2013 22:02 GMTFernandez Oscar
Only use two Ikea lamps.
Nice work, Fernandez!! (tu)
12th Nov 2013 01:17 GMTPaul Brandes Manager
I personally have four Ikea leds and one larger overhead led. I typically don't use all at once, but it's nice to have them at my disposal, just in case......
Look like Oscar is photographing micros, while Alan's setup is for macrophotography.
12th Nov 2013 04:13 GMTSteve Stuart Expert
Fernandez, Paul and Steve,
12th Nov 2013 12:42 GMTAlan Scalone
Thanks for checking out my post on a lighting setup. It was not my design but a write up I found on the internet by one of the sites that sells specimens in $5,000 and up range - incredible images.
Not familiar with LED's but key important point of the setup I posted is the 100 CRI Solux bulbs, Color Rendering Index http://en.wikipedia.org/wiki/Color_rendering_index 100 CRI = zero change in the colors of the specimen which is critical in mineral photography especially if selling high end specimens if not all specimens. Solux is the only bulb out there with a full 100 CRI
Second, as Paul mentioned you can always turn off lamps individually and use 1, 2,4 or all 6 depending on the need. Where they do come in handy is back lighting clear crystals while at the same time lots of front lighting to light lots of crystal faces to better show the shape of the crystal formation.
Also, I have used the setup to photograph large 14 inch specimens down to tiny specimens in micro mode.
Were I to redo the design I would have the setup on wall mounted sliding mechanism so I could slide the whole thing up and down vertically to allow for straight over head shots without the need to use a ladder to see through the viewfinder! I would also make the table bottom center photography area with a clear glass window bottom to apply bottom lighting.
Just a really good flexible setup while maintaining perfection in the actual colors of the specimens.
12th Nov 2013 13:12 GMTFernandez Oscar
Remember... If you setup correctly the white balance in the camera the source of light not is a problem.
Some people pay big money for expensive solutions but they're not necessary. Of course, this is my personal opinion.
P.D.: I don't know what happen with my before message that answer to Steve Stuart, I say that in the flickr gallery you can see macro-minerals not micro-minerals like blue aragonite (FOV of 4,4cm), acicular aragonite (from Pantoja), celestine (from Torà), etc. I use ikea light and in some ocassions energy-saving lights.
While you don't need perfect lighting, you also can't solve the problem in the camera--if the spectrum of the lighting is deficient, its just going to be a problem. You need to use broad-spectrum lighting to properly take photos.
13th Nov 2013 01:47 GMTJeffrey Hapeman
I have not checked this article I had written for quite a long time. I am glad to see that so many have found it interesting and hopefully useful.
19th Jul 2015 08:40 BSTAlan Scalone
Aaron Cross - The rig I detailed above is now mounted on a 24 inch by about 30 inch piece of 3/4 inch plywood so it is now mobile unit that is also very sturdy with the thick plywood base.. The goose neck lights all fold down fairly flat to about 3 inches high. I just grab it. fold the lights down some and slide it into the truck of my car.
Kelly Nash - Halogens do put off some heat but in a room with either A/C or windows open on a cool night, the heat has never been an issue. The flexibility and great photos are well worth a bit of heat!! I used to use it in a closed small basement so I setup a box fan in the doorway which worked fine to move away the heat.
Jim Robison - Yes, I made a mistake when I wrote the original post. I have a total of eight lights, not 6. Thanks! I use 3 on each side and two on the front track.
I actually suspend my pieces above the bottom backdrop using a black plexiglass stand I had a plexiglass shop make for me. It has about a 3 inch square 1/4 inch black plexi base with 1/2 inch square upright plexi post that is about 3 inches high and a 2 inch square 1/4 inch plexi flat top on top of the post that I set the specimen on. With the black plexiglass bottom background and black cloth rising up the back that is out of focus most of the time, the black plexi stand pretty much disappears and the specimen appears to float in the air.
I still use the setup exactly as I originally configured it. Have not discovered a need to enhance it. And I can report that I have not yet had to replace even one of the Solux bulbs, they seem to last forever! The only thing I did do was to remove it from the large table and mount it on a separate piece of plywood so it is easily portable.
Tim Jokel, Jr. - I kept the cost down by buying the goose necks and the bulbs separately and not as a combined unit from a single company.
NOTE to everyone: One of the key elements to the design that no other kits or other lighting rigs offer is the "100 CRI" bulbs. CRI = Color Resolution Index (Google it) Basically, 100 CRI means it will not change the actual original colors of your specimens at all. Solux is the only bulb I could find that has a 100 CRI rating.
Incandescent bulbs have a CRI rating of 100 as well.
19th Jul 2015 09:47 BSTHenri Koskinen Expert
The problem with CRI is that it measures the performance of the light source against a reference source with the same color temperature, not against full spectrum light source. So you can have CRI 100 but very bad color reproduction. Candles with 1700 Kelvins have CRI 100 but you can't really photograph minerals in candlelight. Same applies, to a lesser extent, to incandescent bulbs and halogen bulbs, they have CRI 100 but are mostly too warm and weak in reproducing colors in the blue end of the spectrum. You can do color correcting but if some part of the spectrum is missing or very weak not much can be done about these missing colors.
That said, it is true that halogen bulbs are very good light sources. And so are IKEA leds (with a measured cri of 91). They are just not the best or perfect lights for minerals photography, which, I believe, was the subject of this thread.
Best lights are any full spectrum bulbs with color temperature matching natural sunlight (5000 - 6000 K).
Discussions like this come up every now and again and always leave me puzzled. There is 'proof by assertion' but the whole argument is unconvincing to me.
19th Jul 2015 14:29 BSTOwen Melfyn Lewis
Come at it from another angle. Colour has no existence independent of the human mind and can have no absolute value. Scales created by humans to assign some quantification to colour are useful for some purposes but are artificial and are mainly a means to train humans to say the same thing about what they see - which is not at all the same as them all seeing the same thing.
Rather, 'colour' it is a word used to describe an analogue created and stored in the human brain (and, possibly, some other life forms) for the detection of some (mix of?) wavelengths of electro-magnetic radiation that is detectable to receptors in the human eye which and we call, collectively, visible light.
The lowest wavelengths of visible radiation (perceived as red) deliver less energy to a detector than do the the highest (perceived as blue). Also.the sensitivity of human eyes is not uniform across the range of visibility. Peak sensitivity in the human eye is in the lower-middle range (perceived as yellow) and falls away quite sharply towards the red and blue ends of the visible spectrum. Oh what a muddle! To make it even worse, we know (excepting the clinically colour-blind) we humans do not share in common a uniform colour perception - reasonably close but not identical. Finally and as generalities, colour perception (differentiation) is more acute in females than in males and in both sexes the acuity of perception reduces markedly with age.
Now fit that mess into a pedantic view as to the quantification or even quality assignment of colour. Other than in very broad terrns I think all such attempts have serious shortcomings.
The brain is clever and adjusts the colour perception it stores and uses to compensate largely for differences in the wavelength mix (colour balance) of different light sources, which is why, within a few seconds of adjustment, we see with a colour balance approximating to daylight although the source (say incandescent) has a strong bias in its wavelength mix to the lower end of the spectrum.
As a light source that has been filtered through the Earth's atmosphere, sunlight, the light that human eyes evolved to detect , is actually very impure, making it useless for accurate observation of the absorptive or luminescent effects of passing light through a transparent material. For a whole range of scientific purposes it is important to make optical measurement using only a very well corrected (uniform output across the complete visual spectrum) light source. I do not believe that this is necessary for photographic work where any rendering is subjective. For photographic purposes, the white balance system in a good modern camera, sensibly employed is capable in perfectly adequate colour balancing. Photographic images are stored perceptions and are a means of communicating those perceptions to others. For me, there can be little point in striving at considerable expense to create images balanced in a way that stores some perception that cannot be commonly shared.
The use of light sources of near-perfect uniformity of output across the vis spectrum is not even required for optical diffraction spectroscopy, for which purpose an incandescent light is the light source of choice. Only for optical spectrometry are nearly uniform output sources - with careful calibration at each session - an essential. But spectrometry does not concern itself with colour at all. Rather, it measures and records the energy levels presented at its detector across a series of very small spectrum shifts (typically =<1nm). These measurements are in no way qualitative but strictly quantative and the uniformity of the light flux across the entire spectrum is essential.
Owen - a clear and complete explication of the shortcomings of human eyes and the brains they are attached to.
19th Jul 2015 18:26 BSTTony Peterson Expert
But our color perception is not the issue when photographing minerals. If a light source (fluorescent, LED, etc.) has an incomplete spectrum, with large gaps, and has been manufactured to provide what only appears to be "white" light to the human eye, it will remain deficient for mineral photography. Crystalline substances are of course notorious for absorbing, reflecting, and transmitting specific colors and even being variable for different transmission properties in different directions, and often strongly polarizing the light that is returned to the camera. To my knowledge, most "white" LEDS are blue LEDS with a yellow phosphor: the balance of blue and yellow light is managed to excite the RGB retinal cells of our eyes equally and hence appear white. But a green mineral (unless the green is itself composed of a combination of B and Y, not impossibe, but not common) will appear quite dark under such lighting. As usual, I point to a pair of photos I uploaded years ago, cinnabar photographed in halogen and fluorescent light. What a difference!
It may be that I am out of date, and better quality LEDS now consist of RGB arrays, but that only mitigates the problem, it doesn't solve it.
In my opinion, a serious mineral photographer will always use continuous-spectrum light sources. Just my opinion. These can be quite yellow: the spectrum of the image can be shifted and massaged afterward to produce an accurate result.
19th Jul 2015 22:27 BSTOwen Melfyn Lewis
Thanks - and interesting too. Looking at the description of your excellent pics, you seem to have omitted (to make your point?) mention of any use of white balancing that you were using (I guess none?).
Here are some shots - using white balancing - and also using the output of non-continuous spectrum or uneven across the spectrum light output.
1. A 7 ct blue spinel, fairly included with other crystals - apatite, I think, plus others. The single light source was a multi-folded fluorescent tube with a CRI index of 91. The spectral output has pronounced absorption bands. This image was made with the stone supported in air and with the light being transmitted through the stone from underneath. The transmitted light was polarised to avoid most of the unwanted reflections (a bane when imaging cut and polished gemstones) and also to emphasise the strain (by differential colour display) set up by some of the included crystals and the spinel host crystal. There was also an unquantified low level of unpolarised daylight present. The aim was to create (one of several) images that emphasise the inclusions rather than the host. Stacked shots, using Helicon Remote and a macro lens rather than a microscope (needs must when the devil drives).
2. Pollen grain encapsulated in Madagascan copal. FOV 2 mm. Light was quartz halogen, polarised and transmitted from behind . Specimen suspended immersed in a glass water-filled cell. Ambient and unpolarised room lighting (daylight) also. Use of automatic white balance setting has removed the yellow cast that the host copal would otherwise introduce and permit an image in which the colours approximate to those seen directly through the microscope. This is not a stacked image..
3. Lighting and immersion as for 2. The specimen is Burmese amber containing plant material and the polished specimen has a deep red colour. The consequent deep red colour had to be removed by white balancing before the vivid display of strain colouration can be properly recorded. Burmese amber - of which there is very little - is is older than most amber, at about 100Ma. The presence and extent of strain colouration in polarised light is a helpful indicator - if a very rough one - to the age of the material one is examining. There is normally no strain colouration to be found in copal - or in amber that has been heat treated and maybe also reconstituted. By about 15Ma (e.g. the Caribbean/Mexican ambers) fourth order interference colours are normally present, and by 100 Ma one is seeing first order interference colours set up by physical strain within the amber. Certain plant resins can flow copiously and on exposure to light and air, polymerise and hardenn. In that condition the material can be called copal and is useually one year old or upwards. If the environmental conditions are favourable, over a long period of time the composition of the material continues to change by the loss of volatile organic compounds and a continuation of the hardening process . At about 1Ma some is fit to be considered amber (a stable and workable ornamental material. At about 30Ma, amber is about good as a workable material as it gets. The older it gets the more brittle it becomes and stress in the material increases At c 100Ma and in pieces of amber of gem size, considerable strain can be set up internally that will lead, eventually, to the material cracking apart. This high stress is shown by the first order interference colours in the pic below. This long-winded explanation is given as it is necessary to understand why it is necessary to control or eliminate the body colour of the material whilst observing and capturing the interference colours created within the material when illuminated by polarised light. No expensive lamp (would make no difference) - only white balance control.
Do I have use for a expensive lamp with near perfect uniformity of flux output right across the vis+ spectrum? Yes. I have a xenon arc lamp with fibre feed that costs about 5 bucks and hour to run and is capable of damaging sight (or camera sensor plates) permanently :-) This (or similar) is essential for spectrometry. I also use it (very carefully!) for microscopy and spectroscopy where a specimen is too dark or translucent to blast light through otherwise. I have yet to use it for photography.
I like the spinel - you know that would work well in stereo......;-)
20th Jul 2015 00:04 BSTTony Peterson Expert
Of course there's a use for ANY light source. I should have clarified that continuous sources are needed to produce the best results when accurate color is desired. They can be quite inconvenient to use!
The issues concerning color and its perception are very complex. Tony and Owen raise some very good points. But the reproduction of color has been a long standing problem and much has been done about standardizing the process as much as possible. The International Commission on Illumination (CIE) in 1931 standardized the “Standard Observer” spectral responses based on man y measurements of many individuals with normal color vision. They also standardized predefined illuminants, e.g., D50, D65, A, etc., with tables of values at intervals of wavelength. This standardization was to enable consistent calculation of color values based on measurements that would allow for metrics to be defined to aid in defining how close two colors match. They also enabled the definition of “color spaces’ that are perceptually more uniform to better evaluate color matching.
20th Jul 2015 00:12 BSTRonald J. Pellar Expert
There was as much art as science in the days of film photography. Most of the color science was in the manufacturing and development of film itself. The only white balance was tungsten or daylight film with supplementary filters for fine tuning for the professional. Digital cameras has changed this considerably, sensor responses are designed to approximate the visual response of the standardized human observer, but cannot match it exactly. The ability of human brain to adapt to different illuminants, ignoring all the other color, contrast, sensitivity, etc., adaptations, can be approximated by providing a “white balance” control in the camera. There are usually several predefined white balance settings that invoke various internal algorithms to try to duplicate what the human eye and brain can accomplish. These algorithms differ between camera manufacturers and the intended use for the camera. Critical color work requires a better white balance than the provided presets. This where the “custom” white balance setting comes in. Using a spectrally neutral gray card with the actual light source to be used in the photographic process allows the camera to properly set the R, G, and B responses so that true grays look gray to camera and to the eye in its output. The use of white or gray papers and/or cloths can be worse than using the preset, or automatic, white balances in the camera. An item that looks gray under a given illuminant may not be spectrally flat. This will result in it not looking gray when the light source is changed, but more importantly, the camera response does not match the human response and a gray item will not appear gray camera in the output. It is important to use a “gray card” specially constructed to have a flat neutral reflectivity with wavelength so that the camera can be set to render this gray card as gray.
The mismatch between camera response and eye response can cause errors in color reproduction even after a proper and accurate setting of white balance. These color differences are due to camera sensor responses convolved with the subject, e.g., mineral specimen, spectral reflectivity and convolved with the light source spectral content being different than the result with the camera response replaced by the eye response. This color perception difference is sometimes referred to as metamerism and can be a real problem for accurate color reproduction.
The photographer has no control over the spectral response of the subject. He has some control of the spectral response of the camera through his choice of camera purchase and white balance setting. The photographer does have control of the illuminant spectral quality by a wise choice of light source. The important thing is a full spectrum, i.e., all visible wavelengths represented. The distribution of the amount of light at each wavelength is not as important as this affects the perceived color temperature of the light source. Differences in color temperature can be accounted for in the proper white balance setting in the camera.
Beware of using “color temperature” as a measure of illuminant quality. It is possible to take two lasers of appropriate complementary wavelengths, mix them in the proper proportions and create a white light source of any color temperature! The color temperature of light source is simply an indication of the relative amounts of blue and red in the perceived white.
The CIE defined the Color Rendering Index (CRI) to help in color reproduction by providing a simple measure of the ability of a light source to render the color of certain predefined color patches in comparison to a standard wide spectrum light source of the same color temperature. The standard patches are Munsell chips that are widely used in the printing industry for predefined color standards. The CRI of CIE defined Illuminant A (essentially a tungsten light source) is defined to be 100. The same is true for electronic flash and the CIE predefined D50, D65, and etc., standard illuminants. Sources having a predefined CRI of 100 are all full spectrum light sources. Fluorescents light sources are constructed using phosphors that are stimulated by the UV and/or blue line in the mercury vapor emission spectrum. The mercury vapor emission spectrum also has very narrow band lines in the green and red portion of the spectrum. These narrow line emissions can have a big impact on color perception of objects as seen by the eye and the camera. Most fluorescent light sources have a CRI in the 60-70 range. The exceptions are special fluorescents that are designed for better color rendering at the expense of efficiency. Some of these are referred to “Daylight” fluorescents and can have a CRI in the 90 range.
Most white LEDs are made from a blue LED with a phosphor coating that the blue excites. These phosphors are in general the same as used in fluorescents and can a poor CRI. It is hoped that in the near future white LEDs with much better CRIs will become available.
How does this relate to mineral photography? A good white balance, by means of a good gray card, can solve 95% of most color mismatches, provided a wide spectrum light source is used, i.e., a CRI > 90 is a good measure of color rendering capability. Any remaining mismatches can be the result of the camera response not matching the eye response, i.e., camera metamerism. This can be corrected by careful and sophisticated manipulation of color in a photo editing software program like Photoshop. The biggest factor in obtaining an acceptable image, however, is the photographer’s tolerance for color differences.
LED technology has advanced very much and there are many quality LEDs available already. Check this one as an example of these new high CRI LEDs
20th Jul 2015 09:14 BSTHenri Koskinen Expert
This one uses a violet LED and R+G+B phospors and has a cri close to 98 through out the visible spectrum. Unfortunately these are still quite expensive. Cheaper ones with 90+ CRI are readily available, like the IKEA gooseneck ledlamp costing about 10 euros.
There are also newer fluorescent bulbs with very good color rendering. These can have CRIs up to 98, fairly even and continuous spectra and are available in many color temperatures, like 5200 K, and cost about 20 euros for 16/32W versions.
I know this is a personal preference, but I don't use anymore my expensive halogens and light guides, but instead use much cheaper 5200 K high CRI fluorescent bulbs in micro/macro studio photography.
p.s. Owen, if a light source has gaps in emission spectra these gaps will be just transmitted to the reflectance spectra and to the sensor recording this spectra. No amount of postprocessing or white balancing can help here as these colors are just not there. Humans on the other hand are very good at filling these gaps and guessing colors from the context, just as you explained.
Henri Koskinen Wrote:
20th Jul 2015 11:35 BSTOwen Melfyn Lewis
> p.s. Owen, if a light source has gaps in emission
> spectra these gaps will be just transmitted to the
> reflectance spectra and to the sensor recording
> this spectra. No amount of postprocessing or
> white balancing can help here as these colors are
> just not there. Humans on the other hand are very
> good at filling these gaps and guessing colors
> from the context, just as you explained.
Three quick points.
1. We know that daylight is full of 'holes' (and therefore is useless as illuminating source for both spectroscopy and spectrometry). Yet for photography, we also know it is a satisfactory light source - as millions of great photographs made by daylight witness.
2. I think you are mistaken with regard to what can be done with white balancing. In my example images above, the body colour of the copal is a medium saturated yellow and a deep red for the Burmese amber. The images formed at the camera sensor plate have had to be adjusted so that where where interference colour patterns (white in particular) is created in the specimen by interference patterns in polarised light,they can be recorded reasonably accurately (as the eye/brain would see them. This is not an exact process, usually being worked out 'on the fly' by the photographer, specimen by specimen.
3. In the photography of most coloured, anisotropic and transparent monocrystals, the capture of body colour is a moveable feast according the the orientation of the crystal to lights and camera. Change the physical orientation and colour tone will shift or, sometimes, there will be complete colour change. Where is any concept of 'true' color now. Which shade or colour is true? They all are!
There are certainly some important needs for precise lighting standards to enable *fairly* exact colour matching. IMHO, photography only very rarely is involved in one of these. In the case of transparent anisotropic mineral photography, it must be delusory. In photography, as in life, the recording of colour is mostly a catch-as-catch can business. The photographer needs to be skilled and have a good understanding of all the properties of light that, together with his camera, are his main tools.
Owen, what do you mean by holes in daylight? I have not heard about such holes. Do you mean uneven
20th Jul 2015 13:52 BSTHenri Koskinen Expert
spectral power distribution of natural daylights? This distribution varies depending on time of the day, cloudiness etc.
The concept of "true color" does not have much meaning, just as you explained, as sensing colors is subjective and varies from person to person. But the concept of emission spectra or reflectance spectra is well defined and objective. I appreciate your views and don't think there is any real difference of opinion here.
From a robots or camera sensors point of view light with continuous equal energy spectrum through out the visible light range would be just perfect. Unfortunately such lights do not exist, as far as i know. Closest are some natural daylights. Next closest are some halide lamps, some xenon lamps and some full-spectrum fluorescent lamps.
Anyway, 5200 K full-spectrum fluorescent lamps with (advertized) CRI of 96 is my choice for studio micro/macrophotography. I like what I see with these lights, the colors please my eyes.
Henri Koskinen Wrote:
20th Jul 2015 15:08 BSTOwen Melfyn Lewis
> Owen, what do you mean by holes in daylight?
> I have not heard about such holes.
The Fraunhofer aborption lines - lots of them. Point a diffraction spectroscope at the sky (not at the sun) and look through it. The daylight spectrum is a mess :-) when examined critically.
< Do you mean
> spectral power distribution of natural daylights?
> This distribution varies depending on time of the
> day, cloudiness etc.
Well there is that too. Which is why, for around 400 years or so, those making colour images have illuminated their subjects (in the Northern Hemisphere) by a north-facing sky-light. North light has the least spectrum shift across the hours of daylight though, as you say, even north light is prone to shifts in flux strength by the changing of the seasons and weather.
> The concept of "true color" does not have much
> meaning, just as you explained, as sensing colors
> is subjective and varies from person to person.
> But the concept of emission spectra or reflectance
> spectra is well defined and objective.
> I appreciate your views and don't think there is
> any real difference of opinion here.
Certainly not as above. But going to the root of this thread (and others similar) I really am not persuaded to the essentiality of expensive lamps for benchtop photography. Its a given that those who do are keen photographers one and all, producing images that are well above the average standard. My feeling is that the undoubted quality of their images relies essentially on their photographic skills (and te quality of their camera(?) far more than it does on near perfect spectral uniformity of their light sources.
> From a robots or camera sensors point of view
> light with continuous equal energy spectrum
> through out the visible light range would be just
> perfect. Unfortunately such lights do not exist,
> as far as i know. Closest are some natural
> daylights. Next closest are some halide lamps,
> some xenon lamps and some full-spectrum
> fluorescent lamps.
OK, again there is little difference between us, For bench-top photography, I use all of incandescent, Q-H, fluorescent (three types) and white LEDs. Sometimes I select just a single source most appropriate to create the emphasis - and sometimes ( quite frequently) I will pick and mix my light sources for a given task. The one light I have that I have never (yet) used for photography is the one whose output is closest to perfect uniformity in output and spectrum continuity, my xenon arc lamp.
> Anyway, 5200 K full-spectrum fluorescent lamps
> with (advertized) CRI of 96 is my choice for
> studio micro/macrophotography. I like what I see
> with these lights, the colors please my eyes.
And that's at the heart of it, isn't it? One sees something that pleases or interests and want to record the image not for self as much as to be able to share one's pleasure with others.
20th Jul 2015 18:52 BSTRonald J. Pellar Expert
1) Your point about daylight not being appropriate is a total "red herring". The "holes" you refer to are spectral lines and they are not holes! They are regions of the spectrum that are reduced in energy. There is plenty of energy at those wavelengths. In addition the dark spectral lines are very narrow, whereas both eye and camera R, G, B responses are broad band and the narrow spectral lines have no effect.
2) White balance in the camera and/or software is trying to emulate what the eyes do naturally to obtain "color constancy" in different illumination conditions. The purpose of photography is to capture an appearance of a subject in a form that looks very much similar to what a person would visualize at the same viewpoint as the camera taking the photograph. This is what is meant by color reproduction. In some cases the degree to which accuracy can be achieved in the reproduction process can be very important (and expensive), particularly in the publishing industry.
3) Yes, we know the vagaries of color perception with lighting, orientation, etc. However, you as the photographer have control over the perception that you want to convey in your own photographs and can apply whatever means at your disposal to achieve that. Color accuracy is not always the goal of a photo session, particularly for artistic intents.
You, as the photographer, have complete freedom to do as you wish to achieve your ends. But some of us photographers want to convey the appearance of their specimens as they see them in their hands with their own eyes. To achieve this objective there are some well established techniques that are used in the color industry that will help the photographer to achieve decent color reproduction. So lets do away the "angels dancing on the head of a pin" type arguments and try to concentrate on tips that are useful to most photographers.
20th Jul 2015 21:53 BSTOwen Melfyn Lewis
Ronald J. Pellar Wrote:
> 1) Your point about daylight not being appropriate
> is a total "red herring".
With regard to photography, I say no such thing. Just the reverse in fact. Please read again? But daylight illumination *is* no good for spectroscopy and spectrometric work.
> The "holes" you refer to
> are spectral lines and they are not holes!
The usage was a simile and actually is as close a one as the commonly used 'lines'. 'Line' in the present context is a synonym for a narrow waveband of energy absent or reduced (or raised!} in level with reference to some threshold of detectability. Where below the threshold of detectability, radiation in such a waveband can be said to be absent. Terms such as 'lines' 'narrow bands' and broad bands are all relativistic and the words are suited to simplistic two-dimension representation of some waveband of electro-magnetic radiation emr.
> They are regions of the spectrum that are reduced in
> energy. There is plenty of energy at those
You risk confusing your readers. Energy of some given waveband is either above or below a level of detectability. If below then that waveband can be said to be absent from the emr band under examination. The threshold of detectability can vary.
> In addition the dark spectral lines
> are very narrow, whereas both eye and camera R, G,
> B responses are broad band and the narrow spectral
> lines have no effect.
No, Wavebands of reduced or absent photon count *may* be very narrow and described as 'fine lines' or else such wavebands may broader - even much broader - adding to the spectroscopist's lexicon, thick lines, narrow bands, medium bands and broad bands - the limits for each description being subjective and without fixed boundaries.
In a closed room, 100% opaque to e-m radiation in the visible light band, there will be no photons to count in the vis spectrum and an spectroscopic examination of that spectrum will show as black from end to end. No vis band radiation. We all know this.
> 2) White balance in the camera and/or software is
> trying to emulate what the eyes do naturally to
> obtain "color constancy" in different illumination
> conditions. The purpose of photography is to
> capture an appearance of a subject in a form that
> looks very much similar to what a person would
> visualize at the same viewpoint as the camera
> taking the photograph.
Yes. But because a photographic image is limited in what it can present to the eye/brain combination, it is often necessary to present a series of photos to convey all that the eye/brain can record in a short space of time by examining the object directly. The eye/brain also makes a series of images, subconciously re-focusing and switching the centre of its attention ( ignoring what is irrelevant to it or it does not want to record for some other reason), seamlessly stitching all into one record to archive in memory.
> This is what is meant by
> color reproduction.
Do you mean colour matching? Great for ensuring that paint from different batches of manufacture all look the same when brushed out onto a wall. or that the colour reproduction in one edition of a book closely matches that of the same photos in earlier editions of the same book.
But, as to the photo to be reproduced in that book... well the photographer is king and decides what suits his purpose. It's then for the printer to keep all of hundreds of thousands of copies made over several years all looking like the original photo.
> You, as the photographer, have complete freedom to
> do as you wish to achieve your ends. But some of
> us photographers want to convey the appearance of
> their specimens as they see them in their hands
> with their own eyes.
I think that with some noteable exceptions (e.g. x-ray photography) what you have written just above is the general aim of all photographers, whether or not they realise it.That said, if you and I photograph the same specimen using the same set of equipment, I'll bet that there would be noticable difference between our images. That's because you and I very probable do not see in quite the same way
> To achieve this objective
> there are some well established techniques that
> are used in the color industry that will help the
> photographer to achieve decent color reproduction.
And we are back to 'proof by assertion'. The essential point of this sub-thread it that it has not been shown that lamps of continuous vis spectrum output with a uniform flux output across the entire band are necessary for good results from a skilled photographer using a good quality modern digital camera. Actually,with skilled dark-room work it was not true even in the days of dye-transfer photographic technology though it's a great deal easier now to get a satisfactory result.
> So lets do away the "angels dancing on the head of
> a pin" type arguments and try to concentrate on
> tips that are useful to most photographers.
Yes, one of which is, 'Hey guys and gals, you really don't need those super-expensive lamps for successful bench-top photography'.
But to each their own. It would be dull world if we all thought the same.
21st Jul 2015 00:19 BSTRonald J. Pellar Expert
Of course daylight is not used for spectroscopy! The absorption lines would interfere with the spectrum you are trying to measure. This spectroscopic use is totally off topic! Photographing minerals has nothing to do with "measuring" spectrums. The spectrum is only important in a light source if there are large gaps in the spectral coverage which can and do cause different interpretation between the eye and camera.
As far as the rest of this topic enough has been said!
Ronald J. Pellar Wrote:
21st Jul 2015 12:38 BSTOwen Melfyn Lewis
> Of course daylight is not used for spectroscopy!
> The absorption lines would interfere with the
> spectrum you are trying to measure.
Good, then we are agreed.
> This spectroscopic use is totally off topic!
No, IMO it is not; it is the means by which it is shown that natural daylight, used in the making of so many fine photographs, is not one contiguous bandwidth of uniform flux output across the entire bandwidth as is a theoretically perfect light source. It therefore follows, quite clearly, that such expensive, 'perfect' lamps are unnecessary to obtain fine photographic results. The essential place for such 'perfect' light sources in spectrometry only. The much derided incandescent light sources (tungsten filament) are not only quite adequate as a light source for spectroscopy but are the common light source of choice for that purpose, a more critical application than is the requirement for photography.
Of the three techniques for recording information about the behaviour of transmitted and reflected light (spectrometry, spectroscopy and photography), photography is by far the most forgiving of an imperfect light source. As we agree, the photographer tries to emulate in his output images what the human brain does in direct viewing of the same subject, Which is to compensate, approximately, for the skewing (sometimes caused by source lighting and sometimes by the characteristics of material of which the subject is composed) of the radiation transmitted though and reflected off the subject. This is not an exact process but an interpretive one - and different people commonly have slightly different perceptions; this makes the pursuit of perfection an illusory goal. A skillfully made photographic image may be widely applauded by many people. That does not make it a 'perfect' image but one that, by the skill the photographer, can be generally appreciated as a superior piece of workmanship in optical representation.
21st Jul 2015 20:33 BSTRonald J. Pellar Expert
Nowhere above did anyone mention that broad spectrum lamps had to have “uniform” energy distribution across the spectrum. One of the most uneven distributions is for tungsten. The adaptation of the eye and a proper white balance in the camera will compensate for an uneven spectral distribution. The problem in lighting that can affect color rendition, is due to either broadband gaps or energetic line emission in the spectrum.
Most white LEDs, especially the cheaper ones, have a rather large gap missing in the green. This causes a lower green response in both the eye and the camera, but due to the mismatch between the camera’s green sensors and the eye’s green cones the images from the camera will not match the perception of the eye and neither will have a similar color appearance to the item illuminated by daylight or tungsten. Some of this can be corrected in software by manipulating the green channel in the image at the risk of shifting the colors of other objects in the image (however, appropriate masking techniques can alleviate this problem).
Fluorescent lamps have a different problem, the mercury vapor emission lines have significant energy at particular wavelengths, particularly in the green and red. The phosphors fill the intervening spectrum but not enough to compensate for the emission lines. (Exception: special daylight emulation bulbs have better phosphors that help with this problem at the expense of efficiency.) The energy in these emission lines has the effect of over-emphasizing the reflectance value of the item being photographed at that particular wavelength, e.g., skin tones under a cool white fluorescent lamp. The color perceived by the eye and camera will be different for this lamp than it would be for a light source with better behaved spectral distribution. This is due to the sensors, in both the eye and camera, integrating over red, green, and blue broadband regions of the reflected spectrum. This type of color shift is much more difficult to correct for in software. If you like the color rendition that you see using these lamps then by all means use them in your photography!
Spending a little more money to get lamps of higher CRI will pay dividends in the long run in improved color rendition. Tungsten lamps have a broad, but not uniform, spectrum and with good white balance can produce excellent color rendition. But so can daylight since the absorption lines are very narrow and are distributed over the whole spectrum, they do not unduly influence any particular region in the spectrum at the expense of other regions. Every photographer has to decide for himself as to what degree of perfection he is aiming for and can afford to achieve.
1st Apr 2016 15:01 BSTAlan Scalone
It has been quite a while since I visited my post to answer any new questions. Photographing mineral specimens is extremely difficult to produce an image that not only has the exact colors in the specimen but where a small specimen fills the frame, everything is in focus, etc. When it comes to recreating an image of a clear crystal unless you have lights shining through the back of the crystal and lights to light up the front crystal faces, you simply will not be able to properly capture the appearance and shapes of a transparent crystal.
So whatever you end up constructing, I recommend that you plan your setup to have at least 6 independent lights that can be moved around as needed for each specimen. As setup that will produce quality recreations of the specimens is in my opinion the primary consideration I would recommend.
Regarding portability, the setup I created is very portable. The goose-necks fold down fairly flat and it ends up 30 x 30 inches by 6 inches high and easily fits into my car truck or back seat.
Thanks Oscar for contributing!
1st Apr 2016 15:07 BSTAlan Scalone
Absolutely, yes, my setup is quite elaborate. I reviewed your images at the link provided. In some of the specimens that have deeper cavities in the specimen, the additional lights and mounted on long goose-necks would allow you to evenly light the cavities as well as the top surfaces of the specimen.
For photographing clear crystals, it is simply impossible with 2 lights to shine light through the back of the crystal and also reflect light off of 2-3 of the front crystal faces to reveal the shape and transparency of the crystal.
Absolutely correct Henri .. full spectrum light is the first requirement, 100 CRI is the icing on the cake. The Solux MR-16 bulbs are 5000K and 100 CRI - Best of both worlds!
1st Apr 2016 15:15 BSTAlan Scalone
Mindat.org is an outreach project of the Hudson Institute of Mineralogy, a 501(c)(3) not-for-profit organization. Public Relations by Blytheweigh.
Copyright © mindat.org and the Hudson Institute of Mineralogy 1993-2019, except where stated. Most political location boundaries are © OpenStreetMap contributors. Mindat.org relies on the contributions of thousands of members and supporters.
Copyright © mindat.org and the Hudson Institute of Mineralogy 1993-2019, except where stated. Most political location boundaries are © OpenStreetMap contributors. Mindat.org relies on the contributions of thousands of members and supporters.