[a / b / c / d / e / f / g / gif / h / hr / k / m / o / p / r / s / t / u / v / vg / vr / w / wg] [i / ic] [r9k] [s4s] [vip] [cm / hm / lgbt / y] [3 / aco / adv / an / asp / bant / biz / cgl / ck / co / diy / fa / fit / gd / hc / his / int / jp / lit / mlp / mu / n / news / out / po / pol / qst / sci / soc / sp / tg / toy / trv / tv / vp / wsg / wsr / x] [Settings] [Home]
Board
Settings Home
/3/ - 3DCG



Thread archived.
You cannot reply anymore.



File: metalness02.jpg (380 KB, 1280x1090)
380 KB
380 KB JPG
Working on my own little 3D game engine

What standard should I use for my materials? Metalness or Specular?
>>
Why do you have to put exagerated and unrealistic dents and imperfections everywhere? Are dents the new lens flares? Also my dog is working on his own 3D engine too. He's predicting he's going to be able to publish a mediocre game in a 12 years time frame.
>>
>>570533
> no one is allowed to be happy in my presence
>>
>>570528
https://disney-animation.s3.amazonaws.com/library/s2012_pbs_disney_brdf_notes_v2.pdf
>>
>>570533
they are not that exaggerated, and everything is imperfect. i don't see a problem there.
>>
>>570533
only an idiot wouldn't know that everything in life has scratches and dents, even your asshole
>>
>>570557
im not that guy but, too exaggerated imo

its becoming a bit of a meme, yeah some thing have scratches but i can look on my desk, my phone, tv controller, mouse, none are really that new and none are covered in scratches like they would be if they were on a substance timelapse
>>
>>570564
I'm also not these guys but I agree, small scratches mayybe but dents this size on a fucking lens? Lens are 99% sit in a protector container.. if lenses are that damaged they either don't work or are in really shitty hands
>>
>>570528
metalness is the standard. specular is for cucks
>>
>>570528
Metalness sounds incredibly gay. Go with the industry giants and use specular.
>>
File: IMG_5912.jpg (213 KB, 1000x1000)
213 KB
213 KB JPG
>>570557
For a Leica lens, that's horribly exaggerated, what kind of fiend would abuse their Leica to that extent?
The surface finish is wrong anyway.
>>
File: IMG_5913.jpg (138 KB, 1000x1000)
138 KB
138 KB JPG
>>570573
Actually now that I look at it, the modeled lens has features of different 50mm crons, but the shell is still the wrong finish.
>>
>>570573
>>570575
He's probably modelling a used lens and not an out-of-factory brand new one. I look around my desk and I can't find anything that don't have dents, scratches or dust on them.
>>
the fuck guys

this shit is from joe "earthquake" wilson from marmoset as a showcase for marmoset toolbag

obv you guys dont do shit on this board, besides bitch and moan.
>>
>>570572

you dont into PBR? get ahead of the curve senpai
>>
>>570627
Why do they call him "earthquake"? Is he horribly obese?
>>
>>570528
If you love all your non-metals looking like plastic and if you love nasty artifacts, go with metalness.
>>
>>570643

Didn't he have thing for the samurai showdown character.

Also fuck Marmoset agreeing with the 15mb upload limit on Artstation.
>>
>>570661
Specular does have those artifacts too but they're black not white thus not that noticeable. And getting rid of those is as easy as making high quality maps to later lower the size down to what you need, that shit happens because your maps are small.
>>
specular have nothing to do with metalness. if you use a metallic map on a model it will disable specular on parts where the metallic exists.

there was a short period of time where you could use gloss maps with specular maps and it was sort of filled the demand until metallic workflow came along.
specular is used more like a fresnel effect but its not quite the same. its best to have a specular map than just let the engine decide it by geometry.
to your question. metallic for sure. its the closet thing you can have far as realism goes
>>
>>570581
>He's probably modelling a used lens
Time and use will not wear down the spaces around the imprints like OP's faux-enamel surface.
>>570528
OP stop being a nigger, look at the source object and model the finish it actually has, not the shitty chipped paint you THINK it has.
>>
>>570697
I've never got this - I'm used to Vray wirth the spec+gloss workflow, and to me the metalness workflow just looks like you make a map where you point out wether materials are metallic or dielectric and sets the fresnel values accordingly, but how does it know what parts are supposed to be reflective or not and how glossy they should be?
>>
>>571476
with color intensity. full white is fully metallic. full black is none metallic. if you don't use any maps and make metallic maps proceduraly (or from a picture) you would use something like color ramp or some kind of gradient tool to set the metallic values
>>
>>571483
So what if you have a shiny plastic object? You'd want the fresnel to be around 1.6 or so because it's a dielectric material, and I guess if you paint it white in the metallic map it would turn out looking like chrome, with a much higher fresnel value, no?
>>
>>571486
there are a few differences here:
metal is not dielectric like you said. not even metallic paint which goes on cars and guitars.

if you want to do a glossy plastic you have to lower your roughness values to something like 0.2 or 0.1 and leave metallic at 0 (none). the IOR value should be 1.460 - 1.55.

keep in mind that there are few types of plastic materials. rubberized plastic would have different values and so does billiard balls
>>
Part 1/2

Seriously - you idiots need to learn the difference between metalness and specularity/glossiness/roughness once and for all.

Metalness and Specularity are NOT mutually exclusive.
Specularity literally just means 'reflectivity'. Metalness is a very specific kind of reflectivity which is only observed in electricity-conducting materials.

Non-metal objects are all bound by Fresnel's observations that everything is shiny when viewed at an appropriately oblique angle - everything. The skin on the back of your hand, the road outside your house is actually as reflective as a mirror, just at such a tiny angle that you can't notice it.
For this reason, in something like VRay, all dilectric materials should technically have pure white in the reflection slot, and Fresnel reflections should always be on, and locked at 1.6.
Yes - Pure white should be used every time in the reflection slot, regardless of how dirty and matt your object is supposed to appear. This is because as stated previously, everything reflects everything. White = everything.
By putting in grey or something other than white in your reflection slot, you are limiting the range of colours that the object can reflect and that is not accurate, again, *regardless* of how dirty and matt it's supposed to be - *everything* is a mirror at the right angle.
The level of gloss should *always* be controlled with the glossiness value/map, not the reflection colour.

ctd....
>>
>>571550

Part 2/2

Metalness is another bag of cats entirely.

Electricity-conducting mats in real life reflect almost 0% diffuse light. That is to say, nothing of the light you see bouncing back off a piece of copper actually penetrated the metal surface and then got back out again. Almost all of the light hit the surface of the metal and bounced straight back at you.
But here's the weird thing - these direct reflections from metal surfaces have colour characteristics, unlike everything else in the world that returns direct reflections of every colour that hits it (which is why, again, you should use pure white).
Metal surfaces cause the light that hits them to 'break apart' almost, and certain wavelengths (colours) of light reflect at different angles and different intensities, causing the reflections of say, a piece of gold, to appear gold in colour.

SO - a metal object should have almost absolute black in it's diffuse slot (because it returns no diffuse reflections) and the colour information should be in the reflect slot, so when a direct light hits the object it returns a coloured, metallic reflection.
As well as this, fresnel reflections (direct reflections which only occur at grazing angles) should be turned OFF. Metal surfaces return direct reflections at EVERY angle.
Again - the level of glossiness is determined by the glossiness value/map, nothing else.

If, however, your object is part metal and part non-metal - you need to add a Fresnel IOR map to designate metal areas where Fresnel is off (black) and where everything else has Fresnel on at IOR 1.6 (which is described by a grey made of 187, 187, 187). You also need to UNLOCK Fresnel values in order for the Fresnel IOR map to work.

So,,,, NOW YOU KNOW.
Faggots.
>>
>>571483

No. In VRay the Fresnel IOR map is;

Black = Fresnel OFF = Metallic
Mid-grey = Fresnel 1.6 = Non metallic.

>>571486

A shiny plastic object has white reflections (always) and Fresnel on at 1.6. The level of shiny-ness is controlled with glossiness values.

>>571487
Practically everything is Fresnel 1.6
There are so, so few instances where you need values of anything other than 1.6 (non-metal) and just turning Fresnel off entirely (metal).

You guys are overthinking everything. The default settings of a VRay mtl are basically all set correctly for 99.9% of all matierals in the known universe. You just need to worry about the diffuse map, the level of gloss, and the bump and that's it.
>>
It depends on the engine inplementation. Metalness is faster and specular givesmore control. Arnold metalness implementation is a good hybrid. It allows for colored speculars on dielectrics like gemstones whereas games usualy just asume 4% gray and alows for colored diffuse on metals like gold
>>
File: image.jpg (100 KB, 944x590)
100 KB
100 KB JPG
>>570573
>>>what kind of fiend would abuse their Leica to that extent?

Er... Robert Capa? (pic related)
>>
>>571552
i don't even use Vray. im just telling you what i would do if i had no uber shader or material library with pre-made values.

>Black = Fresnel OFF = Metallic
Mid-grey = Fresnel 1.6 = Non metallic.

well that's sounds ok
>>
>>570573
People who actually use their cameras for work not just jerk off over them
>>
File: 1495068673473.jpg (39 KB, 534x598)
39 KB
39 KB JPG
"Metalness" isn't even a word.
Goddamn retards
>>
>>571456
Artists generally have a hard time understanding wear and tear damage.
Give them a break. Either the product looks unnaturally perfect, or they have to spend unreasonable amounts of time simulating and predicting wear for an object that's going to take up an insignificant amount of screen space for an insignificant amount of time
>>
Why is everyone so fixated on the wear level of this lens model?
1. There is fuck all to say it was anything to do with OP
2. It definitely doesn't have anything to do with the thread
3. The level of wear is absolutely not excessive anyway. I have a 35mm prime lens in a drawer next to my desk right now that I got from my dad that has way more wear and damage to it than this model.

Fucking tards on this board can't give much insight to the question at hand and probably don't even know the difference between metalness and spec/gloss/roughness so they just wade in with the 'hurr too much edge wear' patter to make it look like they know something.
>>
>>571779

There are also these things called 'fiction' and 'artistic license' and 'dramatic exaggeration' and 'hyper-realism'.
No one goes to see an action movie or play an AAA FPS for the realism, they want an exaggerated, comic-book version of realism, so, when you show, say, a fast car in those movies - you don't show a Ferrari, you show an insane rocket-booster version of a Ferrari. When you show a 'big gun' in those games, you don't just show a Desert Eagle - you show a Desert Eagle on steroids with a grenade-launcher attachment.
And so yeah, when the production asks for a 'used camera lens' you don't show the lens that's sitting in my drawer - you show a fucked-up, scratched-to-shit lens that looks like it's been used in war zone.

Fuck's sake. You guys make Rain Man look like Jay Z sometimes.
>>
>>571779
no, its quite simple

theres absolutly perfect, no wear and tear mode

then theirs average wear and tear mode, depending on the kind of object, where small scratches and wear are added in common sense locations

then theres OP wear and tear which is overdone and exaddurated
>>
>>571781
>Why is everyone so fixated on the wear level of this lens model?
Autism
>>
>>571781
because most wear and damages are done with generators and sliders and very few artists take the time to go over detail by hand
>>
>>571808
I kinda realised that a moment or two after I posted it, which precipitated remark about Rain Man.

What I don't get about autists is how they're apparently incapable of abstract thought and therefore have a pitiful lack of imagination - but at the same time they're super into fiction, like sci-Fi and cartoons about talking ponies and shit...

How do those two things work together?




Delete Post: [File Only] Style:
[Disable Mobile View / Use Desktop Site]

[Enable Mobile View / Use Mobile Site]

All trademarks and copyrights on this page are owned by their respective parties. Images uploaded are the responsibility of the Poster. Comments are owned by the Poster.