@OldMan As per @cegaton, North America is always [FRAMERATE * 1000]/1001, where PAL and other European based media is flat.

Although I disagree with @cegaton regarding slippage; you will in fact see slippage over much less than one expects.

@OldMan If you are in North America, even the "My camera shoots 24p!!!" is always 24000/1001. There are no North American cameras that will shoot flat 24p, even though the specification and such says it.

@Georges It is exactly as per @cegaton; hertz rate based. Projection off of a digital projector is indeed 23.98. Timecode? Yep, same!

@troy_s I think that for shots under a minute with no sound, that would fall on what @OldMan is trying to put on stock libraries, the difference would not matter at all...

@TARDISMaker And in fact, the channel is really a byproduct of folks that care. You can't really push conceptual stuff forward without having a support structure. It's one thing to go "Hey this is great but has some complex cascading things..." and leave it at that.

Imagers need to support each other for things to take foot. (Just look at how long it took some of the independent motion picture peeps to get a grasp on "linear" light blends, and even then, they hit the ceiling on display linear without going deeper.

@cegaton it does.

@cegaton You'll notice a frame of slippage.

(For operating a camera remotely, a frame of slippage is bearable, two almost unworkable, and three unbearable.)

Always better (and frankly easier) to simply get the blasted frame rates correct.

@cegaton mikemost.com/?p=299#more-299

@troy_s nice article!

@cegaton Mike Most is a very knowledgeable peep. He has a wonderful article explaining logs etc.

Should really merge the two CDL answers as well.

your's covers the basic formula well, as I should have, whereas I was more pointing at how certain formulas absolutely break in scene referred models.

I'd add that many folks don't realize how many of the blend modes are absolutely useless now.

Of the AdobePDF spec blend modes, I bet over 70% are worthless.

Not only do they break, they are largely hacks to get around display referred nonlinear imaging limitations.

Greetings @Lukaash

@troy_s Evening (Or whatever for you), was just checking out this chat room, either i've not seen it before, or i'm blind.

It's gone through a few transformations recently.

@cegaton Was looking for good visual effects RSS feeds to link to the room's scheduled dumping.

Can't really tell what this room is about, but at glancing at the favourites and looking around, it looks like a more professional version of the renderfarm, with topics.

@troy_s re: CDL post. feel free to edit whichever post you want. I wouldn't mind deleting my answer just to have a more coprehensive and well documented one.

@cegaton Just merge mine into yours

Done.

@Lukaash as stated: Questions and discussions about scene referred, display referred, color management, pipelines, and rendering. It all begins here: blender.stackexchange.com/questions/46825/…

@Lukaash It is largely a discussion (as there are many questions about the subject) about what RGB means.

@Lukaash Which of course, is the deceptively simple and stupid looking question from which all interesting things flow out...

@troy_s Ah, i think i understand now, sounds interesting of a chat room.

@Lukaash If you understand, you can explain it to us. We don't.

@troy_s It is largely a discussion (as there are many questions about the subject) about what RGB means.

I think i explained that well

LOL

@Lukaash In all seriousness, it's a deep rabbit hole. What starts off as such a simple, no-brainer question ends up melting one's head.

@troy_s Yeah, i get what you mean, i've experienced a few of them questions with family members, they don't often end.

@Georges We have a new color managment system?

@TARDISMaker he is probably referring to the LUT pack I drafted up

Okay

See sidebar

@TARDISMaker please start your journey here: blender.stackexchange.com/questions/46825/…

Ah, okay

That's pretty interesting...

@TARDISMaker Deep hole...

Rewarding, but very deep.

Yeah

I'm going to stand on the ledge just inside :P

@TARDISMaker Bah. Try it.

@troy_s What do spi stand for?

@Georges StupidlyIncludedInBlender, Please Ignore.

@Georges Sony Pictures Imageworks. Included because some folks don't understand crap about colour transforms.

lol

I was just reading spi1d files and spi3d

Interesting stuff

So you're saying these shouldn't be included?

Coz in Config.ocio the looks are linked to them too

@Georges Please read back where I try to clearly state that transforms are based on assumptions.

Sure, sorry for that @troy_s

The assumptions from the SPI LUTs included in Blender (not the formats spi1d and spi3d) are based on different assumptions than Blender uses. As a result, many of the transforms are incorrect to apply.

@Georges LOL. Don't be sorry.

But I cannot stress that LUTs are all about input and output.

If you wrote a function that was based on the assumption that the values were between 1.0 and 14.0, and someone fed values from 0.0 to 3.0, the output would be entirely wrong.

Same concept for LUTs.

If you write an algorithm that is used to manipulate AdobeRGB expected values, and you feed it ACES RGB values, completely wrong outcome.

And that is all Look Up Tables are, they look up one value and output the resultant value (1D) or output all three values (3D) based on that single input.

(In the case of a 3D LUT, all three values are mapped to a set of three values, and the closest matching set of three values (interpolated) spits out three values. In this way, a single channel value can adjust all three output values. Hence to do a desaturation LUT, where one channel must control all three, only a 3D LUT can be used.)

(Open up those LUTs in a text editor and get a learning on how they work.)

I already did

Did open them, didn't learn yet :)

TBH, this should be standard in Blender

I mean, you don't have to be an electrician to use electricity

Same here, everyone can use it, I've been experimenting with it for the last couple of days, the renders look amazingly .. natural

@Georges but you can be an electrician or a layman and still get electrocuted..

@cegaton what's the worst that could happen?

@Georges Part of the "natural" is the unwinding of what someone says is "natural" or "photorealistic". If you study art and design you begin to understand how cultural emergent phenomena influence how we read imagery.

No different than literature. There are learned things, via repetition, that we begin to expect that become "natural". What folks frequently forget is that when it comes to "photorealistic", you are attempting to recreate a convention. To imitate it, you have to peel apart what the darn convention is.

What is film? Why is it unique? How can we simulate it with CGI?

There is much more to photorealistic than lighting and textures, but the medium of capture itself. Of course, the medium in CGI, or the dynamic range in this case, also bring much more "photorealistic" calculations with it because the light levels are so much more close to simulating a typical scene.

@troy_s I understand you post documents on the sidebar of this room ? Where are these posts and how to read these posts ? I mean the "feeds posted to this room"

@OldMan google Troy's name and you'll find the white rabbit^^.

I wanted to put a feed of ACES and other bits there. Tested with the other Academy RSS feed. Need to figure out how to turn a Twitter search into an RSS feed without a script ideally.

@troy_s I agree, this is a subjective matter, each person has his own comprehension of visual and sensual material in general, created by huge variety of conditions that turns it into more of a fingerprint variety,

still we are discussing a norm, that provides a base ground that has a significant amount of shared comprehension, which is film quality renders, I can't think of any reason this isn't added to Blender by default, specially that it's not a destructive procedure "right?", people can enable it in the scene menu, if for some reason they don't want to, they just have to leave it to sRGB.

@Georges Comes down to figuring out why. Imagers that care will care. Imagers that don't, won't. More important that the imagers that care can get the information and leverage the concepts than worrying about some goofy struggle arguing with folks that have no clue.

@troy_s I visited this room following a link on a Q&A that I found interesting, and got lucky (and thankful) coz you were here and explained this to me, so I had to get lucky twice in a raw in order to know about it :) still imagers who know about it and just don't care, they wont use it anyways. I guess when I get enough knowledge about the procedure, a tutorial might be a good start, but this still doesn't explain why this "option" shouldn't be available in Blender!

The option sort of is. I'd prefer it if more imagers that call themselves imagers care about it, and that requires explaining things to people and being patient as well as having a support infrastructure.

@troy_s Yes, you're right

@Georges I consider it an act of gravity; more imagers that understand the concepts and generate compelling work leveraging the concepts equates to a larger mass. Greater gravity. Greater pull.

@Mareck Why exactly would you?

Fresnel is just another shader. They try to mimic reality (some more effectively than others) based on the modelled reality.

When I'm looking just the fresnel node in -10+6.5 view it's strange, and the color grading is after the render, so I ask myself if it change something

So must I do my own fresnel with math nodes and the fresnel formula?

@Mareck Your material probably stinks.

@Mareck Remember when I said that it is quite rare for real-world objects to exceed an albedo of 80%? :)

yes

so even the simple diffuse and glossy are wrong?

@Mareck Materials behave differently when faced with proper levels of illumination.

The bottom line is that when you look at a texture / an image

Those values are not albedo for example.

Despite the fact that when someone puts say, a "white" surface on an object. Let's pretend it is a photograph in TIF or whatever. That value is 1.0.

The problem is, when you start to think that the "white" surface means 1.0 albedo, it's completely wrong.

The image itself is a "baked spectral" that is, it takes in all sorts of facets you can't see like surface angle of reflection, the properties, the wavelengths etc. They all come off and are transformed by the camera into some notion of "white"

But as we discussed, white doesn't exist.

So the bigger question is "Well if white doesn't exist, what is this material? How much does it reflect? How am I going to make this end up looking white in my creative display transformed view?"

Those questions are harder and deeper.

Sometimes it is reflective angle (think car commercials) sometimes it is the surface illumination intensity, sometimes it is a creative grade that pushed the value up to white, etc. etc., etc.,

So when you begin to see things "wrong" you have to apply all of your knowledge and try to diagnose what exactly is wrong. Is it the surface texture? Is it the amount of light? What particular facet is being "wrong" and what exactly, in clear terms, are you seeing that you are defining as "wrong".

Until you do that, it's going to be very challenging and you'll start blaming random things, when it likely is a complex series of lesser things.

Ok I got it, I'm not going out of the hole.

Ok more challenging as expected, but when it will be ok, it'll be nice I know

@Mareck I learn quite a bit in bursts.

@Mareck The very subject we are discussing is relatively new to me - understanding albedo etc.

It has roots in where everything is heading - PBR, but I'm pretty sure most PBR folks don't realize the depths of PBR until they light scenes and materials using proper light levels.

Things behave badly when you design materials to what you see, without testing them in differing lighting scenarios etc.

Little more like sculpting than drawing; check your angles, adjust, try different lighting angles, etc.

I don't do materials but the folks I have been trying to help out with wide dynamic range stuff do, and that has led down the path of albedo research etc.

The best advice is to use the mathematical values for albedo from charts, and design your materials according to the charts.

If things look odd, then you know its your current setup.

@gez has been helping a couple of folks with materials.

@Mareck No, because it doesn't get any easier.

The bottom line is your materials are broken.

So either tackle that, or don't. Up to you.

That can work... The HSV formula in Blender sets V to max(RGB) which will hold the value no brighter than the highest.

I will i will, but for the moment I haven't even got a fresnel node working and I have to make a image for a client.

Uh

Why isn't it working?

It's math.

It works.

So figure out why what you are seeing isn't right.

Screenshots help.

custom one

Custom what?

Fresnel node?

There's your first mistake.

Wait I need to try something, my brain is active ^^

No it's nothing, but do you use the default fresnel node in Cycles?

I was thinking it wasn't working correctly

but yes in rgb it wasn't black in the middle, but what is black, ok, ok, need to dig.