I Googled like mad for the reflection coefficients of gold at various angles. I could not find it.

So I've spent all week photographing gold in very careful ways and measuring the colors. (Also copper, brass, and a few others.)

Then I built a shader. Here it is demonstrated.

Gold is 255, 215, 147 when viewed straight on. It changes in surprising ways as the viewing angle becomes more shallow.

Click the pic for bigger.

PS: I know my shadow doesn't match the environment - I was just trying to anchor the piece on the ground so it would not be floaty/distracting.

cool beans, thanks for the data. Especailly the angle info. Tried to make my own gold bars about 2 years ago and never could make them look right. Have to go looking for that scene and see if I can bring it back to life.

wonder how much that thing weights times 1510.00USD an ounce. (today's closing gold spot price)

So, BB, you gonna share your shade on this one? Enquiring minds wanna know! hehe

But seriously, I use (or try to use at least) gold a lot, and when we were talking in the caht room, it got me a thinking, and you too apparently...  lol

So if you took the red and green of traditional gold on an rgb color scale (255,215) and added in the blue of pale gold (138) you'd have something very close to the real thing.

Quote - It changes in surprising ways as the viewing angle becomes more shallow.

How so?  And do any other metals change in the same way?

Also a dumb question I feel like asking so please forgive me:  you used 24 karat gold, right?

Ah Bagginsbill you truly are a digital alchemist!

Wow! That truly is amazing! And it sure does look like real gold.  I've got so many metal materials and the gold ones look yellow and really nothing like gold.

Excellent job BB!!!

Now, where is Harrison Ford? He seems to be missing from that render!

dph

BB, you need to consider selling this shader (along with some others)! Along with an instruction booklet (i.e., pdf - easy to make, and just the thing!)... just throwing ideas out there.

This shader is worth its weight in... er... :blink:

@RobynsVeil

Sssssssssssssssssssssssssssst, silence is Golden....
Don't give the man "bad" idea's.....
But he deserves all our respect.

:-) Oeps, what would the RGB values for Silence be ? ? ?  :-)

Give our shader master a week, he might come up with some :-)
He must eat, drink, and sleep in that mat room...

We also must change the name of that room.
From Material room to Math class.
But then, I was never good at Math :-(

Whilst I'm a reasonably good speller, I was told when I was young that I would never be good at maths, because I couldn't do equations in my head.

I wish parents wouldn't do things like that!

Even if matmatic sounds like it involves a lot of complex mathmatics (and it can, it has the capability, but it is not a requirement), I've found it more of an automation tool than anything else. I can write stuff that makes things. This is cool. Writing:

def makeSurface():
  myColour = IColor(255,0,0)
  s = Surface(myColour,.8,0,0)
  return s

makeSurface()

makes an incredibly simple shader... but yes, it makes something. I can make all sorts of stuff this way - much more complex things - without really doing hardly any math. Well, no math, really. Yes, I'm shocking at maths. :blink:

Copper (the pipes) is 255, 189, 162.

There is yellow and red copper.

My central heating copper 1/2" and 1" pipes are rather dull, and yes they do age......and are not that shiney. 8-(

The stainless steel is awsome.... Whaw. That is pretty close to photo quatily.....

Red copper pic included.

ooo, thank you.  shiny gold to go with the diamonds.

Thankyou for the data

This is great info, BB.  Many thanks.

I have to confess I was never happy with the gold you made before so I came up with one of my own.  I'll have to see how your colour compares with my version.

The copper and stainless look superb, too.  My dad was a plumber so I've seen more than my share of copper pipe. :)

The copper seems a little too brown to me for some reason.  At least if we were talking about untarnished copper.  The steel looks right though.

Quote - "It changes in surprising ways as the viewing angle becomes more shallow."

Fresnel in action. In metals the amount of reflected light is a function of the incidence angle and the wavelength. To achieve this effect you need to have a Fresnel function for each RGB channel with different parameters.

Dielectric materials don't change the color of the reflected specular light, so a single Fresnel function for the three RGB channels is enough.

Quote - I was told when I was young that I would never be good at maths, because I couldn't do equations in my head.

I'll have to try your forumla for matmatic.    Is that all I have to put into the notepad file?

I'm not good at higher math, but I can do the basics pretty easily.

When I started school it was kind of at the tail end of the imperial system and moving into metric.  We were still being taught fractions and such, but I just couldn't wrap my head around it.  I figured out how to convert a fraction into a decimal all on my own, but I couldn't figure out how to get it back to ratio form.  At first they were marking my answers wrong because they weren't in fraction, but eventually  because I was getting the right answer, just in a different format, they started to mark me based on the metric answers I gave.

So after having dealt with metric for most of my life, imagine my surprise when I applied to nursing school in the mid 1990's when I was in my 30's and told that we had to do an entrance exam that contained math questions.....all dealing with the imperial system and especially fractions and ratios!

I was in an utter panic!  I borrowed a kids book from the library called "Math 123 the Easy Way" or something like that, and studied it and sat down with my Sister-in-Law for a couple of days while she taught me how to do fractions and ratios.

I aced the entrance exam and all of the math tests in nursing school!

Part of my problem with math though is the fact that I'm dyslexic and I don't always process what I'm seeing, correctly, especially when under stress/pressure, such as tests.

Quote - Sorry if mistaking, but your copper looks like bronze?

There is yellow and red copper.

My central heating copper 1/2" and 1" pipes are rather dull, and yes they do age......and are not that shiney. 8-(

The stainless steel is awsome.... Whaw. That is pretty close to photo quatily.....

Red copper pic included.

Quote - The copper seems a little too brown to me for some reason.  At least if we were talking about untarnished copper.  The steel looks right though.

Pipes which you know are not copper - it is always alloy and colour depends of ingridiens and production process.

Copper forms a major part of the alloy, however.

Interestingly, in MentalRay (the 3DS render engine) the IOR for gold is about 25.

I don't know if Firefly will use IOR values the same though but who the hell decided that metal is refractive ??? :biggrin:

This looks perfect, BB! Any chance we can get our hands on those full shaders? If you released stuff as merchant resources I'd be more than willing to buy.

Quote - but who the hell decided that metal is refractive

It is not, but it is a way to solve problems. A perfect metal will have an infinite index of refraction and in metals or conductors the refraction index is a complex number (real + imaginary part). Ok, something imaginary is something that doesn't exist, it is only an abstraction, but this abstraction has the physical meaning of how much is reflected and to where and the imaginary part produce a phase shift of the incoming wave (polarization efects and blah, blah, blah)

Matmatic files?

Quote - Matmatic files?

Scripts that can be used with BB's matmatic program to generate the coded in texture/shader.

Acadia, when I'm asked to do like how many mls/hr for a given volume of IV fluid, I so seriously still have to break out pen and paper and do the maths manually, with a lot of head-scratching and self-doubt "nah, this can't be right!"...

... which is what is so cool about matmatic. It sounds like it would be all about maths, but it's much more about making things: shaders, to be specific. And as far as the maths goes (in matmatic), it'll do the lot for you. You just plug in a formula. BB's happy to share those: he's got heaps for all sorts of cool materials.

So, it's more about learning a language - Python - and what that language works like. At this point, I'd say I'm conversational, if not exactly fluent. :blink:

Was away over night. Very interested in this discussion! Great examples BB!

Work fine with this http://www.renderosity.com/mod/forumpro/showthread.php?thread_id=2812046 , thx !

Anthanasius rightly points out these colors work with my BBGlossy shader for a decent metal and you could certainly use that and 99% of folk would never notice it wasn't perfect.

However, the BBGlossy assumes a constant color from all angles, whereas that's not true of real life. Kawecki rightly brings up the notion of a complex IOR and that is certainly a factor. However, I built a shader with a 3-channel IOR and it doesn't look right. Maybe I did it wrong.

After building the 3-channel IOR and seeing it produce pink at the edges of gold, I decided that maybe in this case the physics department has an over-simplified answer and I'd be better off actually measuring the colors and then apply theory, instead of the other way around.

It was after I did that, I found the realism of the gold increased. So did my stainless steel. I did not have a good sample of copper, and may have to revisit that one. For gold, steel, and brass, I have flat shiny pieces. For copper all I have are curved and/or dull pieces, so accurate measurements are very difficult.

My plan is to sell this shader, not give it away. My wife is really mad at me for never selling anything, always giving it away.

Note that this thread is not about my shader - it's not a commercial. I honestly wanted to share the news of what the ding-dang color of gold actually is! I could not find it anywhere.

Your wife is right. Good on you for listening to her.

You realise, of course, this is going to alleviate community guilt tremendously... we've been taking and taking for years! Finally, we'll have an opportunity to give back.

Thank you, "BB's wife" ...

Quote - After building the 3-channel IOR and seeing it produce pink at the edges of gold,...

He,he,he. Here is the problem. Pink is red + blue, two colors of the extreme of the spectrum and are subjective colors. In physics we measure and tabulate the parameters by wavelength, but we don't render by wavelength. We render using only three primaries and a red primary, for example, it is not only a red light, it has the peak in the red spectrum, but has also orange, yellow, green and even blue colors, even with smaller intensities. If we change the R intensity we are changing also the blue intensity. We are not able to change the red color without changing the other colors that makes part of the primary. The same thing happens with Green and Blue primaries.

I don't know how to solve this problem to create images. Maybe use two gold colors, one for normal incidence and one for great incidence angle and then use the Fresnel as the blending factor for these two colors.

ADDENDUM: Use for refraction index glass and not metal for the blending Fresnel. In metals the Fresnel factor changes very little, so the blend would be useless.

Yes Kawecki! You describe exactly what I did. I measured the color at low and high incidence angle and use the single Fresnel to blend between them.

when I saw the title of this thread I thought it meant that finally BB has been paid what he is worth for his work.

Love esther

YES , he deserves the Math room in Gold..... :-)

But how to weight the Math Room??
In PNU?
Poser Native Units?

Gonna be a hell of a job......... :-)

I'm gonna write to SM.
Change the name of that "Math room" into the ""BB room"".... and make it shiney. with a golden glossy background...

Oh, boy my fantasy is drifting today.......

Quote - ... Gold is 255, 215, 147 when viewed straight on. It changes in surprising ways as the viewing angle becomes more shallow.

...

As far as i understand it, the 255,215,147 are the rgb values of something golden, when viewed straight on, normalized to R=255; but comparing with the e.g. wikipedia entry (Reflectivity), i noticed the  blue component being higher (147/255 = .58) than the nominal value (< 0.4), so i guess this is either for impure gold (e.g. with silver) or for some different (unknown) wavelengths. The green (215/255=0.84) is very close to an srgb-green of 550nm). Not that i know what the real value is or if anybody would ever build a chest of pure gold :-) How did you measure it?

oh this looks like reay Gold, very very well done.

Your wife is right, you should sell what you work so hard for occasionally.

I for one can not wait to buy those shaders.

Wonderful.

ha-ha-ha-
All woman support the selling part.... a funny ting to see this happen.... :-)

Quote - ha-ha-ha-
All woman support the selling part.... a funny ting to see this happen.... :-)

We are thankful to him and want to see him earning something off his (lifesaving) work. Save the macho crap for a teen community, will ya?

how-how-how- Hold your horces, it was meant to be a rather funny observation that put a smile om my face.  :-)

Quote - Yes Kawecki! You describe exactly what I did. I measured the color at low and high incidence angle and use the single Fresnel to blend between them.

Good, if it worked this is the way to solve the problem.

All is a big confusion, I am working on different models for metals, there are so many models, Beckmann-Kirchoff, modified Beckmann-Kirchof, Cook-Torrance, Sparrow-Torrance, He-Torrance, Ward and many others, and you can add several different shadowing functions, but ..... which is the best ??????

Assumption is specular color is 586 nm, with ±35 nm either way.

I make conversion from wavelength to RGB but maybe method not so accurate

Quote - how-how-how- Hold your horces, it was meant to be a rather funny observation that put a smile om my face.  :-)

Which should also make you think how much better off we'd all be if there were more women economists, bankers and stockbrokers ;)

IMO we'd all be better off if there were no economists, bankers or stockbrokers of any gender but that's dangerously close to the No Politics rule so I'll shut me gob.

BB, about time you got paid for all your shader work.  Hope you make a fortune.

Have a little faith SamTherapy.  Things will get better and years from now you'll be sitting in your 600.000 pound home telling your grandchildren about the great depression of '08.

BB don't forget to spread the wealth.  Like Adam Smith says horded wealth is non-productive labour.

No politicians would be even better, avoids us from breaking the no politics rule.

But I see a golden future laying ahead of us, at least for poser.

Bopper.

Metals are not easy. With gold, there is the 24 carat variety of modern jewelry, and the various hues of ancient golds found in Egypt, Peru, and Asia, some much darker than others. Gold thread is much browner. Copper and Bronze can look very similar, depending on the forging, polishing, and age. The hardest metal for me is pewter.

Now throw refraction, reflection, and anistrophy into the mix, and you know why a believable metal shader is worth its digital weight in gold. ;)

The image on the left is a brick poured at a mine roughly 45 minutes before this photo was taken (back in 2002). It's about 960 troy ounces (30 kilos, about 66 lbs), about 88% gold (the rest silver, copper and few other impurities). The image on the right is from a different mine (taken last year) where the gold, of slightly less purity, was poured into a mold (for the inukshuk) and polished up a bit to make it more shiny. The gold bar laying on its side underneath it was what would have been normally poured, but that was also shined up a bit to make it look prettier.

And yes, that inukshuk is 49.8 kg (110 lb) of raw gold.

Quote - but we don't render by wavelength. We render using only three primaries and a red primary, for example, it is not only a red light, it has the peak in the red spectrum, but has also orange, yellow, green and even blue colors, even with smaller intensities. If we change the R intensity we are changing also the blue intensity. We are not able to change the red color without changing the other colors that makes part of the primary. The same thing happens with Green and Blue primaries.

Which is why renderers like Maxwell render in spectra instead of RGB (part of the reason for their ridiculous levels of photorealism)

jeez, lookit all that gold!  keith is definitely the go-to guy if one needs a few bars.

I volunteer to clean Keith's shoes after work every day! ;) 

but more importantly, what is the color of chocolate? 

milk chocolate bunnehs, cadbury butterscotch, swoon dovebar