PBR Color Reference List for Materials - by Grzegorz Baran

[@alky333]

You sir, are an absolute legend

[@GrzegorzBaranArt]

Thanks Brian :)

[@360VRStudios]

Learning about double polarization, though me about the albedo color of an object.

[@bDwS27]

This is amazing! I've spent a lot of time looking for this info , I've already shared this with people I know!

[@GrzegorzBaranArt]

Thanks, veeeeery appreciated

[@gamalgad218]

Great resource, Thanks for sharing your hard work!

[@GrzegorzBaranArt]

Thanks, you are welcome :)

[@artofebelle]

Thanks a lot for sharing! This is really useful and I love the Substance studies series you're doing as well!

[@GrzegorzBaranArt]

Thanks

[@Barnyz]

That's alot of colours and numbers, clearly you have been very very busy 👍👍👍 nice London shot at the end 😁

[@GrzegorzBaranArt]

yeah, but I finally made useful version I am really happy with , yay! ;)

[@Mikeygrecordings]

The Lords work. Very nice 👌

[@GrzegorzBaranArt]

Cheers! :)

[@gamingneeded1902]

amazing work

[@GrzegorzBaranArt]

Cheers!

[@davideagan1620]

Fantastic!

[@GrzegorzBaranArt]

Cheers!

[@TheLazyJAK]

Why are some of these nothing like the picture, like carrots, brick, or radish?

[@GrzegorzBaranArt]

These are images of subjects I measured. I hoped they are gonna help to visually navigate through the list but also help understant what state of certain substance was caltured. Also since its next to the X-Rite Color Checker it should help to understand its actual brightness etc. When I use this list it helps me a lot and with that I dont even use substance names while browsing through. Lets say you are looking for a concrete color.. and there are many different concretes out there and each color is correct for certain type.. these images should help you to know is this the type of concrete you were looking for. They do for me at lest. Refarding carrots.. there are old carrots, weathered, fresh, dirty, clean.. thanks to this image you know exactly what carrot was used for color spectrometer's measurement

[@GrzegorzBaranArt]

Dmt.. I just understood the question.. sorry.. here is the answer. I covered it in details in this guide on my gumroad but basically.. every substance is different and responds to the light in a diffrent way. Color Spectrometer measures surface value. Sometimes color we see comes from more things.. lets say.. what is the color of mirror which reflects sky? Is it blue or is it black? Because glass is usually black and silver/chrome is almost white.. but when you look in the mirror reflecting sky it looks blue. When you look on the tangerine.. its skin is dark grey, but what you see is the water behind illuminated by the light which adds to the perceived color (subsurface scattering).. therefore what I measured are color values of surfaces but more complex surfaces needs deeper treatment and light simulation to reflect actual color as it is made of many components. This is why I put it all in this GUIDE I mentioned in a comment as its a pretty tricky subject. I plan to record entire video about this.. to be honest I work on it already since about a year now and I hope it to be my next one

[@GrzegorzBaranArt]

Its a bit like trying to estimate color of glass when you look through the window seeing trees behind. Actual glass color isnt the view behind the glass.. its a glass itself.. which is in practice very dark but very transparent due to its structure. Would you say that glass is black when looking through it? Not to mention invisible spectrums and light re-emission etc. Hope that makes sense.. this is why it took me so much time to make this list since I had the same questions you have and need to find an answer for them before the list was ready

[@kamilgrzelak4370]

Can you share a descent e-shop in Europe where you bought or I can buy the color meter?

[@slaypidin]

Grzegorz, it is awesome work. On the technical side, the values are intended to represent albedo, which is the diffuse component. The spectrometer used for collecting these values used a geometry of (0/45) degrees, which would minimize capturing the specular component, as you explained near the end. The question that I have is whether if you had a choice between two instruments, one with 0/45 geometry and the other with an integrating sphere, specular included, which one do you think would provide more correct albedo values? Specifically, the 0/45 geometry is said to capture the "appearance" of the color, which will result in a darker value for shiny materials due to the loss of the energy that was reflected via the specular vs. a matte version of the exact same material that would lose less light via specular reflection. An instrument with an integration sphere, configured in "specular included" mode, would "re-capture" the light from specular and "re-reflect" it again diffusely, resulting in a higher luminance value from the instrument. Since a properly configured PBR system would explicitly model the loss of light via specular reflection and use only the remaining light for the diffuse calculation, the end result will be that specular loss is included twice when using a 0/45 instrument: It was lost once in the instrument and then again during PBR modeling. For this reason I am inclined to think that only the instrument with a integrating sphere would give correct albedo values, including the luminance value. Asking in order to learn from your expertise, or from that of anyone else who would share his.

[@GrzegorzBaranArt]

Thanks, I can tell you that it all depends on the surface type. As far as I am aware, there is no universal color spectrometer and each has its pros and cons. Also there is no uniform surface and each has its own micro nuances which affect measurment. So basically each color spectrometer gives slightly different measurements. I believe that sphere spectrometers are better to 45degree ones as they cover wider spectrum of materials. Currently I am saving money to get a £3.000 worth one which isnt a 45% not sphere, but which measures full spectral coverage as I believe this infomration should help me way more in my research and give a better picture of what I get from my measurments. But I would say that nothing in universe is uniform and any value +/-10% should be considered as correct. I described more details on this regard in my book about color calibration. I plan to release it soon when its finished. So far I share it on my gumroad (gbaran.gumroad.com/l/rjthlk) and update it from time to time, but its still very in-progress at this point. To be honest color is way more trickier then I thought at the beginning of my research, therefore cannot ever be measured with 100% accuracy.

[@slaypidin]

@@GrzegorzBaranArt Purchased the book on gumroad (and can recommend to eveyone interested in the subject). You mention that you are rewriting your PBR texture book and would release it in 2023. Would customers of the current PBR book automatically get the rewritten version or it will be a separate purchase? You mean you plan to buy a spectrophotometer that would measure the energy in each 10 nm band of the visible spectrum, and calculate the sRGB values from that? Hopefully that would give more accurate estimates than the lower cost device. But I think that does not exactly answer the doubt that I have. Per principle, a 45 degree device is going to give different result on a surface that has a lot of specular factor than one with an integrating dome and (and keeping the specular hole closed). The 45 degree device will give a darker color (again, limiting to a shiny surface to emphasize the differences) because it reflects a large amount of light via specular reflection away from the direction of the camera. Only the light that did not get reflected away vis the specular mirror effect will have a chance to enter the material and reflect the color in the direction of the sensor. With the integration dome, all the specularly reflected light would get back into the dome and the surface, and eventually contribute to the diffuse reflection, making it appear "lighter". Since the two methods of measurement are different per principle, only one of them can be compatible with what the PRB albedo/diffuse color setting is expecting. So it would be really great to understand which of the two methods is "correct" / compatible, in theory with the PBR model. My expectation would be that it is the measurement with the integration dome. Using the lighter color value from the measurement with the dome + a dedicated model for specular reflections in the PBR model (which would reduce the amount of light available in the PBR model for use during the diffuse calculations) should, hopefully, together produce a final render value similar to how the original object/color appeared in real life. With the 45/0 model the device would measure the actual appearance (with the reduced amount of light available for diffuse reflection). If the PBR model further reduce the incoming light (via the PBR models specular component) then the final render would appear darker than in real life. That's my current logic but I do not have confidence and would be great if someone with the knowledge could correct or confirm this logic.

[@GrzegorzBaranArt]

@@slaypidin Thanks. I am not sure which book are you talking about. I am currenrly rewriting the book about photogrammetry (gbaran.gumroad.com/l/YanD) since about 4-5 years now. It will be released as totally new and sepatare book as basically but based on the principle of the first one. I want it to be the most comprehensive book possible on this subject. There is still a lot left to cover so it might happen this year, but might also even next one. Its a totally new book. I was thinking about giving some discount tho customers who purchased the first edition, but it all depends on where it is going to be released. If I release it on amazon there will be no option for me to give any discount to anyone for it. I stopped updating original version after about 30 updates as it started to mess with the book's structure. The new edition is totally rewritten and will have totally different form and is going to be much bigger. So its not a simple edit. Regarding the one about Color Calibration for PBR materials (gbaran.gumroad.com/l/rjthlk) I also plan to release it on amazon, but will update its gumroad version so everyone who purchased it there can access it. I will send an email about the update to all current customers. To get it, you will need to re-log into gumroad and re-download it. This book already had some updates since it was released and the number and date of the last one is on main gumroad page. I just dont bother anyone with update emails since these updates aren't significant enough, the next one probably will be as these are 3 full new sections and some relevant findings I discovered. After incoming update I plan to release its version on Amazon and increase the price on Gumroad to match that Amazon's price, but as said, everyone who purchased it already on gumroad will get access to the updated version for fee on gumroad. Hope that answers your question :)

[@henrikdrygas9536]

Hey, can you recommend any good polarization filter for the ar400? :) and where to buy it

[@GrzegorzBaranArt]

Hey, would strongly recommend the one I presented on my video about props capture. It's made by Scanspace. Link to the store and details should be in video description there

[@henrikdrygas9536]

@@GrzegorzBaranArt Hey, I just checked it. The website u linked gives a 404 Error unfortunately. :/ any ideas?

[@GrzegorzBaranArt]

@@henrikdrygas9536 Hey Henrik, are you sure? I have just doublechecked and it seems to work fine, not just for me. I hope you tried the right one.. just to be sure I will paste it also here: gbaran.gumroad.com/l/rjthlk

[@StealthMacaque]

Any chance we can have this as a downloadable? Happy to pay.

[@GrzegorzBaranArt]

Sure, I put that option in video description :). It is available in here: gbaran.gumroad.com/l/rjthlk

[@sankalpchandrta]

Treasure...

[@GrzegorzBaranArt]

Thanks

[@ytronik]

Feed it to an ai and we’ll never have to think about it ;)

[@WinExploder]

Thats very cool! but some of the dont seem correct. For example, the value for carrots is clearly wrong.

[@GrzegorzBaranArt]

I know and thanks for pointing this out. This is why I have added a disclaimer in video description whcih says: These are values which come from measurements and color we perceive in reality aren’t just an albedo but is made of many other components like specular, glare, reflections, micro-shadows, subsurface scattering etc. This is why I also included reference images with the X-Rite Color Checker next to them so we can also evaluate them visually in comparison to reliable color reference. So please use this list as a guide and reference rather then absolute answer to everything. And please don’t be afraid to question material's look as at the end what matters is always the full light response. Not to mention that I can simply be wrong about something as it’s a bit uncharted and much wider territory then it appears to be and I believe the reason why there are no so many reliable reference lists out there already. So I could either remove these measurements I thought look weird and rise questions.. or keep them there investigating why some are so off :D.. and I decided to keep most.

[@GrzegorzBaranArt]

btw. if value for carrots is clearly wrong, can you please tell me what is proper albedo value for carrot? I checked this one and it doesnt feel so wrong. I measured it a few times and this was average and the most common result I had with color spectrometer. Bear in mind its an sRGB, not linear color .. and carrot get some subsurface scattering which I believe might multipy perceived color value when applied. If I am very away from the proper carrot's color calue I will try to remeasure it with different carrot instead but will do some more work to prepare a better sample

[@jimBobuu]

Noy a single reference to Pabst Blue Ribbon! What a clickbaity video! :-)

[@GrzegorzBaranArt]

my bad, maybe next time ;)