The Age Old Question about Color
I've been collecting for almost 60 years and I still cannot find a way to accurately ID stamp color. I'm hoping some of you might have a solution that actually works. So, I get this classic British Colony stamp that is supposed to be Cat #xxx, perf 12.5 and #1 wmk. I look it up in Scott and it says "1d lake". But hold on a minute, there's also a carmine/lake variety and it makes a big difference in value. Hey, no problem, I dust off my Stanley Gibbons Color gizmo and find lake and carmine/lake. Oops, neither matches the stamp. Damn, I find my Stanley Gibbons catalogue and it says lake and deep carmine. Surely, since Gibbons makes the color gizmo, its colors are accurately noted in the catalog. Well, no. Gibbons lists colors in the catalogue that do not exist in the gizmo. There are also colors in the gizmo that I have never seen on a stamp (and I've seen a lot). Then I think, "The stamp is over 150 years old, perhaps the color has darkened with age, or maybe the color was bleached by sitting out too long in the sunlight?" Nice, if any of those are true, what color do I actually have?
I check on the Internet. It seems that almost everyone who has a Stanley Gibbons Color gizmo thinks it is totally useless. So I call my friend who is an expertizer for the APS in the German area. I know many of the stamps he deals with are very old and there are typically 2-7 color variants, often with more than one color per stamp. Hah! I am sure he has a solution. His answer, "I have the best German color palette money can buy. I never use it. To determine the color, I compare it to colors of stamps that have been certified that I have in my collection."
Well, thanks for clearing that up. So his carmine/lake for country X may be different for carmine/lake for country Y, depending on the expert.
Des anyone have some way to be sure of color that an old fart collector can make work? (BTW the above mentioned stamp above does not actually exist; I made it up for illustration purposes.)
I check on the Internet. It seems that almost everyone who has a Stanley Gibbons Color gizmo thinks it is totally useless. So I call my friend who is an expertizer for the APS in the German area. I know many of the stamps he deals with are very old and there are typically 2-7 color variants, often with more than one color per stamp. Hah! I am sure he has a solution. His answer, "I have the best German color palette money can buy. I never use it. To determine the color, I compare it to colors of stamps that have been certified that I have in my collection."
Well, thanks for clearing that up. So his carmine/lake for country X may be different for carmine/lake for country Y, depending on the expert.
Des anyone have some way to be sure of color that an old fart collector can make work? (BTW the above mentioned stamp above does not actually exist; I made it up for illustration purposes.)
Comments
For a classic collector, like me, there are thousands of $100+ items in my collection. The cost for most certs is a percentage of the catalog value. I just could not afford all those certs. Besides, why cannot Gibbons or Scott or Facit, or whatever have a website with actual stamps of certain colors to be compared with?
Then you run into problems related to the hardware on your computer in rendering colors and displaying them correctly. And then your eyes are different from mine, etc. The problem could get really awful.
Consider this. Your computer can render millions of colors; each is given a number. You scan in a stamp and compare it to your color database. The computer returns a number, + or - and you look up the number to get its name. Now all you have to do is trust your scanner/camera/computer hardware to be fairly accurate.
I also knew a Carmine once and it was a girl. LoL
https://www.hipstamp.com/store/harry-patsalos-philatelics
According to the BBC, carmine is made from bugs:
"The insects used to make carmine are called cochineal, and are native to Latin America where they live on cacti. Now farmed mainly in Peru, millions of the tiny insects are harvested every year to produce the colouring."
Bill, are you sure that was a Carmine, not a Carmen?
I'm not sure how far I will chase this rabbit, but as of now, it looks promising. If it turns out to work well, I'll be glad to share this with you all in the hopes it may get further refined. I'll keep you posted.
look of ink as well as fade it over time.
The best way is to look at a stamp under as close to ideal lighting as possible and compare it with a known color image (preferably a stamp).
I not keen on a leave-it-to-the-experts mentality. And colors do change over time, but that's true for eyeballs as well as spectral analysis.
If this hobby wants to survive in the 21st century, it's going to need to keep up with tech. I would suggest standards that say this color is these wavelengths in these proportions +/-. The tech to prove an expert right or wrong is getting cheaper than the stamp very quickly.
Don't get me started on handheld X-ray fluoroscopy to see if it's really bug juice or synthetic.
I'm just an old fart with failing eyes and the desire and (finally) the resources to peek at more expensive acquisitions, but how do you even start a reference collection without just having to take someone's best guess?
Just askin'. Luv you all. Ideas?
I think this will work, at least for me. My scanner (Canon LiDE 220) scans a stamp and gives me the image. The color picker grabs samples from several parts of the stamp and returns the RGB number. The monitor has nothing to do with these steps. The scanner, however, is absolutely necessary. I get a color from the scan and color pick by assigning the color I know the stamp to be. I average as many stamps with the same color to determine its RGB number and look for a match. How much "slop" in the system I'm willing to accept will be personal and I suspect that my numbers may not be that good on someone else's computer. There are about 10,000,000 colors in the RGB system so there will be some guesswork only if a color falls half way between two known colors. The actual colors should have a pretty big separation. I'm working on that now.