Standardizing lighting for ideal-scope/reflector type photography

[denverappraiser]

The lights here are 5100K.

 

[denverappraiser]

Date: 4/4/2005 6:26:26 PM
Author: noobie

I guess no one would want to try to hawk a poorly cut stone with an IS photo

It''s true that the people selling stones may not be interested in publicizing images of less than ideal diamonds but I deal with them all the time. I''m extremely interested in delving into what can be learned from these images. As John points out, standardization is the key for this to be useful in analyzing the stone instead of analyzing the skill of the photographer.

Neil Beaty
GG(GIA) ISA NAJA
Independent Appraisals in Denver

 

[denverappraiser]

Date: 4/4/2005 5:42:44 PM
Author: JohnQuixote

How are you getting the alignment right so that the image is taken from a direction that is perpendicular to the table? It seems like a few degrees of variation can have a profound affect on the image.

This is like the answer given by the corner musician when the young boy asked him “Sir – how do you get to Carnegie Hall?” “...Practice, son. Lots and lots of practice.”

John.
As you know, I’ve been putting some thought into this whole process. I made a tool for lining up unmounted stones from a flat watch crystal glued to the end of a straight cylinder that’s about 10 inches long. I’ve been using the same lighting sytem as Wink although I’ve ordered the trays. Anyway, I set the stone in the little hole, put the face of the crystal against the table of the stone and give the rod a little spin. This seems to line up the stone to the extent that the rod is vertical. I’m just eyeballing it and I didn’t put a lot of effort into making the tool precisely so this only works to within a few degrees but a jig doesn’t seem all that difficult to arrange with this sort of approach that would get it to well under 1 degree of variation from perpendicular to the table. I haven''t really pursued this because it''s tangential to what I’ve been working on but it does show promise. I would hoping you had something better and I could abandon this tactic.

Neil Beaty
GG(GIA) ISA NAJA
Independent Appraisals in Denver

 

[noobie]

Neil,

Those are nice pics. Far better than anything I have been able to take. This is OT, and maybe for another thread, but if you still have those stones, it would be really helpful to get your visual commentary of them. It would give consumers an idea of what stones perform like when they see perfect IS''s and not so perfect IS''s.

 

[denverappraiser]

Noobie,

Those are both gone, Here's another one that's around for a bit. This one isn't mounted. Picture with the little flat light that Garry sells for $25.

1.94cts. RBC
8.21x8.34x4.75
Table: 64.4
Crown 28.7
Pavilion 41.1
Culet 1.2%

AGS 9
HCA 3.5

the full sarin report can be viewed here. it report number - sample9

 

[WinkHPD]

Date: 4/4/2005 10:19:59 PM
Author: JohnQuixote


Ack!

No no. No one is trying to make a bad stone look good or vice versa.

We are simply promoting natural lighting for performance assessment.

Ideal-scope photos have become common parlance...Soon AGS ASET photos will enter our lexicon as well...There is an increasing need for standardization...To this end we would like to share our approach.

I do think that if we had standardization we would all learn more from comparisons, and an ideal-scope photo with CCT is repeatable - the biggest criticism of other performance assessment tools is that they are not.

John,

Thanks. From your Ack! response I hope you did not think I was inferring that you were making bad stones look good, I absolutely know better!

What I am saying is that there have already been instances of people putting photos of stone A with stone B and saying it was a picture of stone B. I think that is a bigger potential issue than what light temperature was used in the photo.

I sincerely mean that I want to learn to take photos as nice as yours though, as they do show the stone in its proper light. I think I see another major investment in photographic equipment in my future...

Wink

 

[WinkHPD]

Date: 4/5/2005 1:19:12 AM
Author: Rhino
Interesting pic Wink. I was wondering how Paul''s princess cuts would look. That image gives an understanding of why it didn''t get ideal grade. Being a photo junkie of these sorts I have some thoughts to contribute when I get up to work.

Cool,

Gary suggested taking a photo against a black background for Princess cuts to better see the contrast. I will start a thread on that and await yours and Gary''s and John''s comments. I have never seen a better looking princess to the eye, it is straining my patience to wait for an AGS 0, which is what we thought this one would be...

Wink

 

[JohnQuixote]

Date: 4/5/2005 9:28 AM
Author: denverappraiser

John & Garry,

Unlike the diamond dealers here, much of my interest in this topic has to do with getting a decent image on mounted stones. This prevents use of the tray system or the little light tray that you’re selling. Decent tools for analyzing and documenting the optics of mounted stones are few and far between and this seems like it has tremendous potential. It would be especially useful if it were possible to take a sufficiently standardized image that it could be overlaid or compared with a previously taken pic by someone else and have a reasonable chance of recognizing it as the same stone.

Neil, imaging for mounted stones is a good ancillary topic - and another tally mark for ambient lighting. Differences in mountings will be a concern, but if you’re not trying to split hairs there's merit to your madness.

The photos you took through mountings at 5100K are nicely done. What kind of environment do you have?

What I really want is a reasonable way to convert these images into a DiamCalc or GemCalc wireframe diagram. Standardized pictures seems like a giant step towards this. While I’m at it, I wish I were taller.

Well-met. We have heard these chimes at PS, yes? I agree and, thusly, am hammering the standardization anvil in a shower of sparks with great aplomb.

BTW, doesn’t the thin air in Colorado make you appear taller?

 

[JohnQuixote]

Date: 4/5/2005 4:47:59 PM
Author: Wink

John,

Thanks. From your Ack! response I hope you did not think I was inferring that you were making bad stones look good, I absolutely know better!

What I am saying is that there have already been instances of people putting photos of stone A with stone B and saying it was a picture of stone B. I think that is a bigger potential issue than what light temperature was used in the photo.

I sincerely mean that I want to learn to take photos as nice as yours though, as they do show the stone in its proper light. I think I see another major investment in photographic equipment in my future...

Wink...As I re-read the thread last night I said to Angela: "I typed 'Ack' in a post today." ( I have ack'd her before).

I know you didn't mean it like that, and thanks for the clarification

Ever since \"Rhinogate\" happened we've been brainstorming about what can be done to avoid that, as well as what you describe. We've been victims of image nickery in many public instances...I worry about private situations. Watermarks notwithstanding, the digital age is a two-edged sword.

But - I'm just here to gently muse lighting standardization for now.

Thanks for the comments on our photos - the concept of natural light is simple, but it does take some doing to bring it 'inside.' I hope this discussion is helpful - please share any more thoughts you have.

 

[denverappraiser]

Date: 4/5/2005 7:21:37 PM
Author: JohnQuixote

The photos you took through mountings at 5100K are nicely done. What kind of environment do you have?

I'm sure it's nowhere near as tricked out as yours. I took a promotional IS that Dave Atlas gave me at a show somewhere (thanks Dave!) and broke out the cone. I cut off a piece of a plastic film can to make a half inch long extension tube and attached it with electrical tape where the eyepiece on the IS used to be. This fits nicely around the lens of my Nikon 4500 at just the right distance for focusing the macro mode to the biggest view. I then set the camera on the desk pointing straight up and hand hold the stone at the focusing plane and line it up using the reflected glare between the table and lens as a guide. The 5100K lights are the ceiling lights in my lab. When I'm at customer sites, I use do this under an ottlight and I rest half of a ping pong ball on the back of the ring as a diffuser. The loose stone photo is made with the same arrangement of the camera and the cone but I take the picture from the top and I just press the camera against the IS and against the light box for stability. It works pretty well and it's VERY fast. I suspect that if I swapped out the light box with something that has a better color mix, I would be getting pretty comparable photos to yours. (at least IS photos, I still covet your h&a pics and your advertising style face up pics but that's a different discussion.)

Neil Beaty
GG(GIA) ISA NAJA
Independent Appraisals in Denver

 

[Garry H (Cut Nut)]

Guys this talk about light temperature is way over the top in the complete wrong direction.

Even Neils shot reporduced exactly as was here (with the only background removed) still adequately represents this diamond.

And that was taken with a light I do not recomend, with very flat batteries.

It sounds real cool to talk about all this high tech stuff - but it is the wrong high tech stuff.

 

[denverappraiser]

Garry,

Can you please explain why it is useful to be close to the light diffuser? John seems to be disagreeing on this point and my best pics seem to come from using the ceiling lights, which are 8 feet away.

John,

How about this for a random idea. Use a lap to polish the end of a glass rod to be perfectly square. Pick a rod that’s exactly the size of the eyehole in the IS. To align the stone, put the stone on the light tray, place the IS cone over the top, poke the rod through the eyehole and press the flat end against the table of the stone. Twist the rod to cause the stone to level up. Even if the original placement is a little off center, this will square the table to the eyehole, which is really the objective anyway.

Neil

 

[JohnQuixote]

Neil – Your ideas are very resourceful. Necessity is the mother of invention.

Keep ‘em coming – anyone who has tried to do this is nodding in agreement I suspect. For now we depend on Brian's commitment to quality and a very patient (did I say very?) photographer. We are hoping the equipment we're testing will help 'level' the playing field (ha ha). I sent you a PM.

I had a conversation with Garry this evening - afternoon for him - and we discussed the two different approaches. In short, he feels that the diamond in the tray sitting on the light source lets light come in from the side, and the photos may appear more accurate to the parametric models generated in DiamCalc/GemAdviser (Garry, I hope the short summary is ok).

Our philosophy, in short, is predicated on the idea of replicating natural daylight, which is ambient, thus some separation from the light source. For our purposes it has been consistent, resulted in a clean & natural look to the images and agrees with what our eyes tell us.

Brian has been the force behind our methodology, and I'll share the notes from these convos with him.

 

[Garry H (Cut Nut)]

Neil the further the light is from the perspex tray - the less light shines on the pavilion from angles other than the axis of the diamond.

8* had white reflective prespex walls in the original FireScope (which I have had since 1975). The new version has blackened walls, pressumably because this reduces the amount of leakage you see through the scope.
That is why i like the light to be a flood light resting right under the tray.

Sorry John - edited to add - I was posting about the same time I guess.

Yeah - a 10 second explanation of 1/2 hour chat about the world, wine and a bit about diamonds too

 

[Garry H (Cut Nut)]

Dear all,

This is as good a way as any to photograph diamonds in jewellery.
The Cabin light costs about $150 and takes up very little room. Its color temperature is perfect - especially if you use a better quality camera that allows you to set the white balance (i.e. adjust for the color temp of the lighting).
If you are having a problem with yellowish back light - then manually overdie the exposure time or set the camera to over expose - then your background will be white.

Neil and Wink - if you are using the ideal-light - and the batteries are a bit low, then set to Tungsten light, and definitely over expose the picture.
My little pocket sized manual Canonixus 40 does not let me do some of this stuff - but it does let me change to indoor lighting - which works great.

 

[Garry H (Cut Nut)]

The photo on the left is with a flattish battery in the ideal-light - on normal camera settings, and on the right I just changed the setting to ''indoor''.
A better camera will give you more control. i did it quick - never lined up etc

 

[Garry H (Cut Nut)]

this shows why the light must be up close and broad enough - the same holds true for both set stones, and stones in the tray

 

[denverappraiser]

Garry,

The settings on my camera make it difficult to be in both Macro mode and Indoor mode. I’m reasonably confident that this is a problem with the operator and not the camera.

Your expert model includes a diffuser rather like my ping pong ball that effectively expands the angle of the light source to a full 180 degrees of relatively even light but this doesn’t seem to be one of your recommended tools for taking pictures.

I agree that the color of the light doesn’t really have much to do with the ability to analyze the merits of the stone but it sure makes the pics prettier. There’s something to be said for that too. It also has an effect on the color saturation in the image. As a stand-alone thing this is not so important but it seems like it would be important for a consumer who is comparing photos taken by several different people. Lastly, it seems to have an effect on the sharpness of the different areas. For my long term objective of making the wireframe model from the picture, this seems like an important goal. No?

Neil

 

[denverappraiser]

John,

Two more random thoughts on the alignment topic:

It seems like the light tray could be modified to include a fitting of some sort so that the IS cone fits into it in exactly one way. This will lock the alignment of the eyehole to the stone location.

A second IS cone could perhaps be modified to invent a leveling tool. I doesn’t need to be the same cone that the image is taken through since the cones are readily available and are identical to each other. This also frees up the opportunity to drill additional holes, add extra parts and the like that would otherwise mess up the image.

Neil Beaty
GG(GIA) ISA NAJA
Independent Appraisals in Denver

 

[Garry H (Cut Nut)]

Date: 4/6/2005 9:23:19 AM
Author: denverappraiser

Garry,

The settings on my camera make it difficult to be in both Macro mode and Indoor mode. I’m reasonably confident that this is a problem with the operator and not the camera. Do you have access to a 12 year old boy?

Your expert model includes a diffuser rather like my ping pong ball that effectively expands the angle of the light source to a full 180 degrees of relatively even light but this doesn’t seem to be one of your recommended tools for taking pictures. not absolutely recomended for taking large quantites, but adequate for your purposes i think Neil.

I agree that the color of the light doesn’t really have much to do with the ability to analyze the merits of the stone but it sure makes the pics prettier. There’s something to be said for that too. It also has an effect on the color saturation in the image. As a stand-alone thing this is not so important but it seems like it would be important for a consumer who is comparing photos taken by several different people. then you might want to go down to a camera shop and try your camera with one of the Cabin type lights - the one we like costs $150 from the on-line vendor Lastly, it seems to have an effect on the sharpness of the different areas. For my long term objective of making the wireframe model from the picture, this seems like an important goal. No? That reminds me - we were working on a ''manual'' for DiamCalc users to do that - have you become proficient yet Neil?

Neil

 

[denverappraiser]

Date: 4/6/2005 3:58:39 PM
Author: Garry H (Cut Nut)

Date: 4/6/2005 9:23:19 AM
Author: denverappraiser

Do you have access to a 12 year old boy?

I can do better than that. I have a 14 year old granddaughter. She is much smarter than me. She is much smarter than any boy. Just ask her.

That reminds me - we were working on a ''manual'' for DiamCalc users to do that - have you become proficient yet Neil?

Proficient? Hah! I’m dreadful at it.
I actually suspended my practice because I was doing so poorly and it was getting frustrating. With a fair bit of effort I can get a pretty good model of very well cut stones but I’ve found several areas that cause me trouble if the stone is less than excellent and even with super-ideals I''m not all that satisfied with my results. For example, are you able to make a Diamcalc model the 1.94 in the sample photo?

Neil

 

[WinkHPD]

Date: 4/5/2005 7:39:24 PM
Author: JohnQuixote

Wink...As I re-read the thread last night I said to Angela: ''I typed ''Ack'' in a post today.'' ( I have ack''d her before).

I know you didn''t mean it like that, and thanks for the clarification

Ever since ''Rhinogate'' happened we''ve been brainstorming about what can be done to avoid that, as well as what you describe. We''ve been victims of image nickery in many public instances...I worry about private situations. Watermarks notwithstanding, the digital age is a two-edged sword.

I agree, it is a pain. Here is a photo I took a few minutes ago with both a watermark and a copywrite on it. Both have their problems. The copywrite is too easy to crop and the watermark can interfere with seeing the item. Make it too dark and it is a total pain, make it too light and no one can figure it out anyway, plus it is REALLY annoying. Sort of like saying, "Hey, I don''t trust you to look at my WONDERFUL photo you thief." Gee, now there is one impression I would like to avoid, while at the same time avoiding "Rhinogate". (Heck, my photos aren''t THAT good, yet, and if they want to steal it they will anyway.)

There is a way to put an invisible watermark on them in photoshop, and then their spider runs round and tells you whereever it can find your pictures, then if you did not authorize the use there you can go administer an email spankiing to the offending webmaster, but so what?

Sorry if this is slightly off topic, but it does merit its own discussion sometime...

Wink

 

[Garry H (Cut Nut)]

Date: 4/6/2005 5:20:37 PM
Author: denverappraiser

Date: 4/6/2005 3:58:39 PM
Author: Garry H (Cut Nut)

Date: 4/6/2005 9:23:19 AM
Author: denverappraiser

Do you have access to a 12 year old boy?

I can do better than that. I have a 14 year old granddaughter. She is much smarter than me. She is much smarter than any boy. Just ask her.
he he

That reminds me - we were working on a ''manual'' for DiamCalc users to do that - have you become proficient yet Neil?

Proficient? Hah! I’m dreadful at it.
I actually suspended my practice because I was doing so poorly and it was getting frustrating. With a fair bit of effort I can get a pretty good model of very well cut stones but I’ve found several areas that cause me trouble if the stone is less than excellent and even with super-ideals I''m not all that satisfied with my results. For example, are you able to make a Diamcalc model the 1.94 in the sample photo?

Neil

This was a qucky Neil - about 6 minutes
and not real accurate because I did not model the mm sizes etc (could nt open your Sarin files either - crashed me)

 

[lostdog]

"Since the color temp of the lighting is fixed the saturation is the same edge-to-edge no matter what size."

Do I miss something or are these two different issues? Is saturation only a function of color temperature? Or isn''t there a question of photometric distribution?

Also, would it make any sense to have a reference as part of the frame? The equivalent of the way a color palette is used in calibrating other visual systems, it could be something, let''s say a bar across the bottom of the frame, with various zones that should show a certain range of values. If the lighting is even, then it matches a certain pattern. And then down the road as the image was displayed or printed, you could be sure that if the zones matches the standard pattern, then the actual diamond image is being represented correctly, too. (and maybe build in a registration serial number to each image to allow being sure it comes from the right source?)

CCT and CRI don''t tell the whole story. Each just measures (or approximates) color temperature in a certain manner. They may average out similarly, yet possess or lack "important" parts of the spectrum. (CRI is usually CRI-averaged over 8 tested colors, for example). Spectral Power Distribution plots allow a more complete comparison.

Though as you point out, if you can match CCT then you can compare CRI pretty directly. CRI assumes a continous color spectrum.

http://www.gelighting.com/na/business_lighting/education_resources/learn_about_light/color_specifying.htm is a good refence on the web

 

[Rhino]

My apologies for not being able to participate on this thread and others relating to the subject of photography this week. My assistant Tim has been out sick for 3 days and my work plus his is a little overwhelming for me at the moment and I just do not have the time to post and comment this week. I have alot to say and contribute to this thread and others but my most precious commodity (time) is at a shortage. As soon as I get freed up I plan on writing and contributing to the subject at hand.

Best regards,

 

[lostdog]

"CCT and CRI don''t tell the whole story. Each just measures (or approximates) color temperature in a certain manner."

Okay, I wrote that poorly. CRI doesn''t approximate color temperature. It does attempt to describe the color properties of light in an indirect way. The ability of a source to render colors is related to the distribution of the spectrum of colors that compose the light it is emitting. CRI sums it up, but but a high CRi can still be poor for color matching purposes.

 

[JohnQuixote]

Date: 4/6/2005 9:22:51 PM
Author: lostdog
'Since the color temp of the lighting is fixed the saturation is the same edge-to-edge no matter what size.'

Do I miss something or are these two different issues? Is saturation only a function of color temperature? Or isn't there a question of photometric distribution?

What up, Dog?

In the quote above I was answering Neil’s initial question about whether diamond size makes colors look different in the aperture of the viewer from stone to stone. The distribution you allude to is, of course, a natural function of the diffuser so yes it applies as well – my reference to CCT in that context was to illustrate that the background light will result in the same relative hues in diamonds regardless of size.

Also, would it make any sense to have a reference as part of the frame? The equivalent of the way a color palette is used in calibrating other visual systems, it could be something, let's say a bar across the bottom of the frame, with various zones that should show a certain range of values. If the lighting is even, then it matches a certain pattern. And then down the road as the image was displayed or printed, you could be sure that if the zones matches the standard pattern, then the actual diamond image is being represented correctly, too. (and maybe build in a registration serial number to each image to allow being sure it comes from the right source?)

"CCT and CRI don't tell the whole story.  Each just measures (or approximates) color temperature in a certain manner."

Okay, I wrote that poorly.  CRI doesn't approximate color temperature.  It does attempt to describe the color properties of light in an indirect way.  The ability of a source to render colors is related to the distribution of the spectrum of colors that compose the light it is emitting.  CRI sums it up, but but a high CRi can still be poor for color matching purposes.

Exactly - this is why we begin with a uniform CCT. Match it with CRI near 100. >95 should be fine for uniform purposes, wouldn’t you think? Your idea about having a frame reference bar (a la color palette calibration) is interesting.

I know Garry does not want over-complexity to run rampant here. Our aim is twofold:

1. To endorse a natural real-world lighting condition for ideal-scope performance assessment (what could be more credible and repeatable?).

2. To suggest a uniform, attractive and professional appearance for the photos. Though this may not be a priority for scientists, it will not hurt to have attractive, sharp photos to represent this important aspect of science in the trade to new consumers.

Right now any given IS image must be taken separately and judged individually. Comparisons between setups are possible but difficult for untrained eyes. Hues, brightness of image, cleanliness, natural appearance, etc. vary depending on the setup de jour and who is making the photos... We place value on the time we have invested in support of IS. As ASET (standardized, no doubt) becomes popular we would like to see IS keep pace.