how do i zeen chin?

>>2602036
try using a mirror

>>2602036
looks like he works from greyscale values and then adds the color

>>2602061
i found his WIPs, its not like this

>>2602070
link the wips

>>2602070
link pls

get lot's of ram so you can use fucking huge canvases(all of the chinese digital painters do this), use the chalk brush with transfer on, keep your value range fairly shallow, don't let your saturation get too high, and git gud at blending and lighting forms/ casting shadows.

File: zeen-chin-selfie.jpg (418KB, 1920x1920px) Image search: [Google]

418KB, 1920x1920px

Not OP but what I don't understand is how they are able to control colors so "effortlessly"
(I say effortlessly because they probably spent years working on color)

I understand value completely, I know how to render different things in value. And I know that values are more important than colors.

But once you start introducing hue, chroma, saturation, etc I get confused. I know the basics of color theory, but things like "Are shadows more or less saturated than the light plane?" or "How much does reflected light influence the value of the shadow plane?" or in pic related how did the artist determine how 'bright' the shadow part of the shirt is (the part where light comes through and creates that "glow" effect)

Zeen Chin is all about design. It's obvious that Zeen has researched enough things to be able to manipulate different objects within the composition in order to create an interesting image.

>>2602110
Its easy desu

>>2602110
can you post your art
(see if youre actually trash..)

File: picking colors via curves.png (182KB, 1000x570px) Image search: [Google]

182KB, 1000x570px

>>2602095
>keep your value range fairly shallow, don't let your saturation get too high
>>2602110
>>how they are able to control colors so "effortlessly"

it's specifically a combination of two things:
- "curved color ramps"
- "tone-locking"


for tone-locking, you can do this automatically. there's a color picker that already does it for you:
https://www.youtube.com/watch?v=3zAfu7xE-Jw

for curved color ramps, you literally just pick your colors in the shape of a half-circle, on the color picker, and then shift the hue around for light and dark variants. that's it.

the reason this works so well is, you paint the majority of the picture using tone-locked colors, and then for things you want to stick out the most, you go "off the rails" and just hand-pick some more vibrant colors.

eg. look at how well the tones of the hair and skin work together, and then compare them to the lips/necklace/earrings.

also yeah, huge as fuck canvas sizes help a lot.

>>2602110
Why don't you check and see the values of his paintings? If you understand values you should be able too how he does it.

Reverse engineer that shit senpai.

>>2602156
kek good one

>>2602192
>Finally I can accurately temperature shift without changing value
You are amazing. I can do this with oil paints fairly easily, but digital painting makes it tougher!

>the reason this works so well is, you paint the majority of the picture using tone-locked colors, and then for things you want to stick out the most, you go "off the rails" and just hand-pick some more vibrant colors.
Is this why artists like Ruan Jia and other asian artists of similar skill level have such high key value structures? Most of the colors they're picking are just really similar in value?

As for color ramps, I saw a video from imagine fx that basically has someone painting in value only. When they want to add color, they just add a color ramp adjustment, then continue from there. Is there a benefit from doing it this way in comparison to the image you posted (that you know of, of course)?
Here's the video https://youtu.be/xRSYupRW8wM?t=13m19s
Also is there a source for that tutorial? If you don't mind me asking.

Zeen is a cute grill?

>>2602192
I always thought tone-locking wasn't an accurate term because tone suggests color. Locking a color would be keeping a single color and changing its value. I think a more accurate term would be value-lock.
There was a tutorial online somewhere by a background artist (from Road to El Dorado I think) showing this method. Cant find it now.

>>2602250
Tone=light and dark values. I am retard.

>>2602110

He likely picked that color because it felt just right, rather than by thinking and trying to calculate it by some kind of a formula. A lot of decision making in art comes from a place beyond the intellect.

File: 1468517132.png (24KB, 494x175px) Image search: [Google]

24KB, 494x175px

>>2602095
>tone-locking

why not just use a HSY' color wheel? it looks like it achieves the same effect

File: 1468520656128.png (8KB, 494x175px) Image search: [Google]

8KB, 494x175px

>>2602360
>looks the same
Well guess what, it isnt.
Id love to have a colorwheel that would keep the value the same!

>>2602432

Get MagicPicker. It is in discount atm, I think.

>>2602520
>buying a fucking color wheel.

>>2602070
Still waiting on those links, senpai

>>2602036

How do you describe this texture? His rendering makes his subject look like they're made of clay or something, a really nice smooth fragile substance.

>CAPTCHA: statues

>>2602580
I really like that almost tactile clay-like texture that artists like Zeen and Kawayoo use, even though I can see why others wouldn't at all.

File: 12487156_1244800078867181_1196281069982592608_o.jpg (197KB, 2048x1447px) Image search: [Google]

197KB, 2048x1447px

>>2602036

File: 12493753_1244800085533847_2290605664014745679_o.jpg (256KB, 2048x1447px) Image search: [Google]

256KB, 2048x1447px

>>2602679

File: 12492020_1244801832200339_883092798402971948_o.jpg (214KB, 2048x1447px) Image search: [Google]

214KB, 2048x1447px

>>2602682

File: 10644187_1244801875533668_3437445475951776822_o.jpg (231KB, 2048x1447px) Image search: [Google]

231KB, 2048x1447px

>>2602684

>>2602684
>just paint the rest of the owl

>>2602192
This looks useful.

Where can I get this plugin (legitimately or otherwise)?

>>2602688
the guy just showed his step by step process. I don't think he's making any fucking tutorial, anon.

>>2602520
>$20 for a color picker
Nope.

>>2602192
just make your palette from said colors

>>2602682
>please make sure the lighting is correct
I wasn't going to, but since you asked nicely...

>>2602679
>>2602682
>>2602684
>>2602687

>Huge difference between 2 and 3, minimal difference between 3 and 4.

Real useful :-/

File: Screenshot_3.png (389KB, 442x442px) Image search: [Google]

389KB, 442x442px

>>2602036

File: d6d.jpg (27KB, 600x488px) Image search: [Google]

27KB, 600x488px

>>2602682
>>2602684

File: HSY'.png (84KB, 741x875px) Image search: [Google]

84KB, 741x875px

>>2602432
How did you do that? Converting a color to it's value is not a straightfoward process, it can't be straightforward, as the "value" of a color is a construct of human perception.

Indeed GIMP seems to think the HSY' wheel is giving colors of different values, contrary to what I suggested.

File: magic picker.png (304KB, 897x3025px) Image search: [Google]

304KB, 897x3025px

>>2603437
>>2602432
Likewise, GIMP also think MagicPicker's colors are also giving different values, contrary to what you suggested.

File: krita.png (54KB, 604x875px) Image search: [Google]

54KB, 604x875px

>>2602432
Krita seems to think both are correct: colors that are the same value are displayed with the same value.

Try doing what you did to my screenshot with the colors from the magicpicker demo.

File: 32.jpg (41KB, 600x477px) Image search: [Google]

41KB, 600x477px

>>2602679
>>2602682
>>2602684
>>2602687

>That jump

Jesus fucking christ I'm dying. That jump is so ridiculous. Literally tells nothing.

Is there a fashionable way to deal with this in Manga Studio? I feel the color picker is a bit simple, but MS doesn't destroy make my comp make jet noises so I prefer to use it to photoshop or most other things (I'm also a shitter and can't draw consistent good lines without stabilization).

>>2602192
Oh my god that pic perfectly explains what i've desperately tried to convey to people on this board when giving color advice.

File: value.jpg (835KB, 1030x2190px) Image search: [Google]

835KB, 1030x2190px

Those WIPs might not seem much, but it's fun analysing them.

If you desaturete it, you'll see that a lot of elements almost disappear, like flowers on her dress, lips , accessories and so on.

While the creatures skin and the background are pretty much the same.

Values are pretty close all over the picture, but if we do some clamping you'll see that while they're close, there's also a lot of range to them, with highest values being out on the light and darkest being on the background.

This is basically a good example of working a limited pallete in digital form, with modern color pickers it's supposed to be easier, since you can straight up more saturated color and leave the value the same.

Limited palette, good drawing fundamentals, edge control, color knowledge, and value control.

It's all very fundamental stuff at a high level.

File: hornel_1.jpg (3MB, 1534x2100px) Image search: [Google]

3MB, 1534x2100px

So this reminded me or Hornels work and i've decided to do the same checkup on his pictures, and lo and behold, he used the same technique basically, to achieve the same result!

Here's one

File: hornel_2.jpg (3MB, 2040x1648px) Image search: [Google]

3MB, 2040x1648px

And another one with more apparent colors.

Notice that the value range is pretty narrrow, but wide in that narrowness.

File: hornel_3.jpg (531KB, 1020x824px) Image search: [Google]

531KB, 1020x824px

And a bonus, with the painting on top of a flat 50% grey.

As you can see for yourself, if you open it in photoshop and go around with the color picker, the HSV value is going all over the place, but the LAB L stays in the same 50% range.

So with the same luminosity, you can get a veriety of colors still.

After reading some links from a thread around the corner and stumbling upon Munsell color system, they say that the closest you can get to Munsell is Lab.
I've also read some actual photo retouchers talk about how they only use Lab for retouching. So maybe the key is to stepping out of RGB conformity and try painting in Lab.

File: op_2.jpg (896KB, 1920x1358px) Image search: [Google]

896KB, 1920x1358px

And as a bonus number, here's the OP picture turned into Lab and with lightness set to 50%

This thread made me think about how I don't even have a color wheel, I use HSB sliders. It could be detrimental but I've gotten used to it.

>>2603683
How can you painting in lab ? I mean that would be tricky to work with 3 color weels and stuff

lol, this thread pretty much exemplifies why the vast majority of IC never has or will make it. Anons spend more time looking at someone else's work trying to find "tricks" or answers rather than just painting to figure them out yourself.

You don't need to see the guys steps, it's not going to help you. His paintings look impressive because he can draw really well, understand 3d forms and how to light/render them. No tutorial or process shot is going to teach you that, only drawing and painting a fuck ton will. Even if you had a magic tool that gave you his color pallet and told you where to put the colors your work would be shit because up can't draw.

People need to just go fucken draw instead of endlessly swooning, analyzing and wasting time on here picking apart someone else's work. If you want to become generic asian artists #23757 you need to be spending your days studying from life and painting personal work, then continue that for 5+ years.

It's seriously not brain surgery.

>>2602238
Nah, dude from Malaysia.

>>2603701
You are a massive faggot. Any good artist has studied the work of other artists he admires and tried to figure out the techniques being used. You also assume we don't do any actual practice but like, fuck you man.

File: KyXPhvq.png (155KB, 685x557px) Image search: [Google]

155KB, 685x557px

>>2603443
Ok, turns out I have absolutely no idea how any of this works.

>>2603685
How did you do that? The setting lightness to 50% part? Also what exactly is lightness and how does it relate to luminosity, value and brightness?
Could you maybe also analyze some of Krenz and/or Nesskain work please? I feel like your insight could be really helpful for me (and others) that struggle to understand this. Write up why they d stuff the way they do.


btw, this loosk like a pretty useful thingy http://davidjohnstone.net/pages/lch-lab-colour-gradient-picker

>>2603774
>Ok, turns out I have absolutely no idea how any of this works.
No worries I don't really understand it either. The Y' in HSY' is called "Luma", I don't really understand what Luma is and why HSY' Just Werks (more or less) for picking colors of the same value, it's arcane.

File: krenz.jpg (706KB, 1200x1698px) Image search: [Google]

706KB, 1200x1698px

>>2603802
Is this some sort of level 5 irony?

File: krenz_2.jpg (936KB, 2400x1015px) Image search: [Google]

936KB, 2400x1015px

>>2603774
If anyone knows this moon dialect, they could try translating it, because as you can see in this example, his values stay almost literally the same on all steps.

File: cushartprocess.jpg (304KB, 1200x1015px) Image search: [Google]

304KB, 1200x1015px

>>2603845
Here's the non moon version.

i really want to learn more about his process, cushart has patreon and/or gumroad. but his tutorials seem to be more about on anatomy etc. can i find more about his coloring technique? maybe different artists that use the same technique?

How many layers am I supposed to use?
Should I make another layer for lighting and/or different objects?

>>2603948
>supposed to

Not gonna make it famalam

>>2603948
I've seen great paintings with 1 layer and 50+ layers. Use as many or few as you feel you need to make it look good.

>>2603683
>>2603685
What am I looking at here? Just pure color?
If i put a value study would I just have to put this layer on "color" and it would look "correct"

>>2603701
Why aren't people allowed to analyze and learn something new to start trying out consciously in their practice, instead of having to hash out everything on their own when there are resources and learning opportunities in other artists' work? Looking at an artist's work like this takes only a little time. Seems like you're projecting unnecessarily.

>>2604021
>If i put a value study would I just have to put this layer on "color" and it would look "correct"
Exactly.

If you use Lab, you can paint in greyscale directly on the L layer to set your value and then just edit a and b to add color.

>>2604028
Hoooly shit bro

This changes everything. Thanks for sharing too.

>>2604028
Could you explain more please ?

>>2604028
Could you explain that in a way a laymen like me could understand? Just a bit of elaboration would be greatly appreciated, I don't even understand what "Lab" refers to.

File: Lab.jpg (1MB, 1264x1080px) Image search: [Google]

1MB, 1264x1080px

>>2604047
Well you just change Image Mode into Lab and edit lightness channel. It'll keep the hue/saturation, but you will be able to edit the lightness. that's basically it.

You have to poke around with the effect but the process is basically it.

File: ureOcVZ.jpg (438KB, 2500x1211px) Image search: [Google]

438KB, 2500x1211px

>>2604275
This also seems like a backwards way to do it (but maybe a nice way to fix up your image)

If anyone has this tut translated, it might shed some light on the similar process.

>>2603955

would you stop being a gatekeeping faggot

>>2602360
It doesn't achieve the same effect, that's why people don't use it.

>>2602796
>>from said colors
yes, and how do you pick "said colors" hmm?
i can't pick tone-locked manually by eye at all, especially in warm colors (thanks brain!).

>>2603483
It's one of the best pics I've come across for explaining color palette stuff.
I only wish I'd known of it sooner.

>>2602691
I don't know about non-legitimately, but just searching MagicPicker should get you it.
I really do wish it didn't cost so much just for a color picker who's only really good features are tone-locking and always staying open.

>>2602235
>>You are amazing
>>is there a source for that tutorial
You're welcome! The source is _paperabbit, a sprite artist (worked on Tree of Savior I think).

>>high key value structures
Yup that's exactly the reason why.

>>Is there a benefit from doing it this way in comparison to the image you posted
I think the main benefit of it would be that you get a good sense of the lighting before applying color, especially contrast.
I think it's certainly a good way to do things, and I know that quite a few artists and designers do this (designers often make logos without color first, to make sure the contrast sticks out between graphics and text, foreground and background, etc).

>>2603685
This picture gave me an erection.

File: r1GNHWZ.png (117KB, 556x474px) Image search: [Google]

117KB, 556x474px

>>2604363
Not him, but why doesn't this seem to work also? The blue seems waaay lighter even though its the same Lightness?

>>2604378
>>The blue seems waaay lighter even though its the same Lightness

That's because it's positioned between two warm colors (orange, red) which tricks our eyes. Warm colors are very attractive to the brain compared to cold colors.

If it was between green and gray it wouldn't look much different.

>>2602192
>>2604375
You're very helpful, anon. Thank you.

File: color.jpg (215KB, 962x1950px) Image search: [Google]

215KB, 962x1950px

Does anyone know where this method came from? Or if there is a non-jpeg version?

File: values.jpg (121KB, 1193x624px) Image search: [Google]

121KB, 1193x624px

>>2604378
The red is a darker value

File: helpforanon2.jpg (875KB, 1913x1463px) Image search: [Google]

875KB, 1913x1463px

>>2604445
Oh yeah, Erik Olson gave that to our class

Basically, when you work in oil, you have direct control of if you want to change the temperature of a color without changing its value (much). Resulting in "octopus pools" of gradients from color a to color b. Olson was HUGE on making sure you have set up your palette even before you touch the canvas. Anyway, in digital painting, it's quite a bit tougher in order to achieve a subtle temperature shift without altering the value. It's possible, but it takes time to make sure everything is the correct value, etc etc.

In order to do this easily, find the color you want using this chart and color-picking it...taking note of its value position (left to right). Make a cross shape using this base color. Then take another color and put an opaque brush stroke on one of the ends of the cross. Take a soft brush with low opacity and get gradient making.

You can even just use the gradient tool for better accuracy, but you have more control with a soft brush.

Hopefully this image helps describe the possibilities.

File: helpforanon3.jpg (507KB, 1913x1463px) Image search: [Google]

507KB, 1913x1463px

>>2604508
Almost forgot
Look at the "skin tone" cross.
When I go towards red, it doesn't seem very "saturated". Technically, that red is fully saturated...only in that value step unfortunately. The Saturation is High, the Value is High, but the Chroma is Low.
If I wanted the full force of red (highest value in chroma), I'd have to make a gradient from the skin tone all the way down to the red at the 9th value step (left to right) and color-pick a High Saturation and High Chroma, Again the process is the same with the cross and soft brush...but be aware that you are changing the value AND the Chroma.

It's weird and it takes some time getting used to it, but it's worth it.

The lines are to help you keep the color-picking fairly similar in saturation. Typically, you want to keep the saturation the same in a gradient of color, but again "it doesn't matter, really"

Also, this pic shows that, while not perfectly, the chart does what it's supposed to in terms of keeping values constant.

IMHO it's way more useful to learn how to mix color, be it digitally or traditionally, than to use all those weirdass, plastic programs

Not only you'll be able to get any color you want, you'll also be able to understand most colors you see and paint everything on an even palette

It also helps to try still lifes but I know you won't do that

File: Confusing Bit.jpg (585KB, 1920x1358px) Image search: [Google]

585KB, 1920x1358px

>>2604521
I'm an idiot
That pic was taken in Greyscale Mode.


What I am having trouble understanding is:
When I look at >>2603685, am I looking at color which is all on the same Value?
If I turn off B and Lightness channels, I'm left with something like this.

Now, is what I'm looking at the range of "saturation" in the sense that anything lighter is more saturated and anything darker is less saturated red?
>As you can see for yourself, if you open it in photoshop and go around with the color picker, the HSV value is going all over the place, but the LAB L stays in the same 50% range.

Is this a visualization of the HSV going all over the place?
I'm making this super complicated because I know I'm realllly close on making a breakthrough...There's just like, one more obstacle I need to jump through.

File: value_vs_lightness.jpg (202KB, 1021x376px) Image search: [Google]

202KB, 1021x376px

>>2604535
I think you're confusing Value, Brightness, Luminance and Lab Lightness.

Here's a pic i've spent too long too make.
Increasing Value/Brightness in RGB moves your color picker straight up.
Increasing L of Lab moves your color picker in a curve.

>>2604363
How can you tell? Are you just eyeballing it?

File: value_vs_lightness3.jpg (140KB, 1016x374px) Image search: [Google]

140KB, 1016x374px

"But anon, the left is nice and bright, while keeping the saturation, right?"

If you turn them black and white you'll see the difference.

Now, if you're really next level, you can try separating your hue/sat on separate layers, but this would probably need a whole script for better worksflow, or only would really work for a finished image.

http://retouchpro.com/tutorials/?m=show&id=152

>>2604472
Then what the fuck is the point of the L in LAB if not to keep the grayness constant in grayscale?

File: lab_approximation.jpg (147KB, 1364x780px) Image search: [Google]

147KB, 1364x780px

>>2604472

>>2604767
Are you by any chance work in RGB but just clicked on L while colorpicking? Because they'll all be one Lightness when in Lab and not, when in RGB.

I can't upload the file in a format that supports Lab, but here's a screepcap of colors, and their lightness channel and the color picker. You can replicate that yourself while using Lab image mode.

You can also see perfectly that some color shift while switching from Lab to RGB.

File: 1468680511954.jpg (464KB, 1920x1200px) Image search: [Google]

464KB, 1920x1200px

>>2604774
Im getting lost here, thought I was onto something, but what the shit?
First, my L: 56 do3snt seem to be exactly the same as yours.
Second strip of pic related is having just the L channel on. How come its not uniform gray when its all supposed to be L:56 ?
Third strip is previewing it in the gray mode for conpariaon.

File: chroma_lab.jpg (239KB, 2602x1500px) Image search: [Google]

239KB, 2602x1500px

>>2604833
are you taking my RGB image and putting it into Lab?

There's a shift happening when you convert the color modes, if you want to get the same result, create a clean document, switch it to Lab, fill it with one color, check the Lightness.

Then turn on the color picker and select the colors with the same lightness. After putting them down, you'll see that the lightness is uniform (the same as the original color you've put down)

>>2604833
It's also impotant to note that Value and Lab Lightness are not one and the same.

On a somewhat related note.
Iif you look at this >>2602192
and the look at this>>2604672
you'll see that Lab lightness is the same curve colorpicking as the "secret asian technique".

The hilarious thing about this entire thread is that this chinese guy has most likely never even heard of any of this color wheel tone locking bullshit in his entire life. He probably just did a lot of cast studies and understands how color and light works, which allows him to pick appealilng palettes and consciously control his value range.

Why do so many people always try to find the most complicated shortcuts in the hopes of skipping the very simple, yet time intense solution to a problem? Maybe you want to start drawing some golden ratio grids on top os this guy's paintings too?

File: linranpallette.jpg (68KB, 690x571px) Image search: [Google]

68KB, 690x571px

>>2602834
>>2603457
>>2603033

he can show you how to paint but not how to think.

tl;dr
learn how to utilize temperature, value control in your painting by studying how light works in theory and in real life

use rgb or cmyk sliders, they make subtle variations much easier than hsb

https://s-media-cache-ak0.pinimg.com/564x/28/ce/ec/28ceec0109cb3ebf242bfc2071e0c8bf.jpg

http://www.huevaluechroma.com/index.php

File: 1458076955972.png (4MB, 1984x1698px) Image search: [Google]

4MB, 1984x1698px

>>2604858
>cast studies
>related to relative digital saturation
Sure thing bub

>>2604871
cast studies developed his ability to control value ranges. Lots of asian / chinese artists do them and you can clearly tell in their work because it's usually one strong directional light source + fill light, executed just like they would in a cast study, which is why their work often has this clay-ish, sculptural look to it.

When it comes to saturation, you need to understand how light works and why colors become more or less saturated depending on the exposure and color of the light source. There is no magical slider mode that lets you automatically pick the right colors that will make your work look good.

>>2604858
90% of op's pic is Zeen's design knowledge.
From what we've learned is that his value structure is very high-key, yet with lots of variation within the high-key range.
We also have learned that most of the picked colors in op's pic are desaturated with the red being the only thing saturated enough to pop-out. We've already figured out the image...There's nothing really special about it.

Some other related topics popped up and now we're talking about how we can pick colors that are of the same value. Using Lab makes it easier as both a and b control the color, and L allows to shift the value of said color. With Lab, we can pick colors of the same value in order to shift temperatures rather than shift values, which is crucial in painting. Using standard rgb is okay, but using lab allows for ease of access to the colors you want in the value you want (albeit slight variations).


>Cast studies
Bargue 101, nothing special there. Separate light and shadow and then add hard/soft edges as well as reflected light.
>How color and light works
Simple plein air studies or still lives work best. But Gurney's books as well as Alla Prima and other stuff works well too.

Master copies are important for studying composition, value structures, and color usage. Is that not what we're not doing here?

File: gl_52cfeac8-2adc-4aa8-8420-458c0af4b6c2.jpg (838KB, 1658x1050px) Image search: [Google]

838KB, 1658x1050px

>>2604858
it's just different ways of achieving the same goal, not everyone works the same. the phrase "no rules, only tools" comes to mind, things aren't as black and white as you seem to think.
all people are doing here is discussing a particular technique, not trying to arrive at a conclusion of "this is the best way to do it, no other way should matter".

secondly, something being "complicated" is entirely subjective. what's complicated for one person may be a walk in the park for another.

an example of this is software UI:
- some people find dumbed down, simplified UI's good, while others find them over-engineered and complicated or tedious to use because so many things are hidden away behind menus and extra clicks (Material Design from Google, Windows 8 aka "AOL Kids Only").
- just as much, some people prefer UI's to have everything super-accessible and laid out on the screen all at once in an orderly fashion (think of audio mixing boards), while others stare at it and just see a complete cluster-fuck of buttons everywhere and become overwhelmed by it all.

both ways of achieving this technique have their cons, and require patience:
- some people might find the brute-force cast study method a waste of their valuable time. "time" is the con here.
- just as much, some people might find the tools to automate such color pickings to be really complicated and tedious to master. "complication" is the con here.

the whole "Why do so many people" approach is just generalizing and counter-productive.
it's like walking into a bar dressed as captain obvious and pointing out that everyone is sitting around drinking cheap beer and slowly getting drunk, when they could just go to a nightclub instead for a quick shot of some monstrosity that looks like it's about to go sub-atomic.
we already know we're sitting around drinking cheap beer, we're in a damn bar...duh.

a bit of both would be ideal. not a jack of all trades, just an understanding of both cast studies and tools.

>>2604870
>>huevaluechroma
thanks for posting that link. i should've included it earlier

>>2604275
>ing really special about it.
>Some other related topics popped up and now we're talking about how we can pick colors that are of the same value. Using Lab makes it easier as both a and b control the color, and L al


Apologies, can you clarify a bit more.

So you paint like normal, and the head to the lightness channel and then use for examples curves to adjust ?

>>2604275
Apologies so you paint in the lab channel grayscale, then add color to a and b. Can I ask please is there any rules to painting on a or b.

For example do you paint shadows on a , and highlights in b for example.

Thanks

>>2604848
But if Im in Lab mode, have the colorpicket set to L, then printscreen what I see and look at it in only L channel, its still not uniform. Like wtf...

>>2604928
>Rules in art
Not gonna make it mate-o

File: lab.jpg (7KB, 273x221px) Image search: [Google]

7KB, 273x221px

>>2604908
>>2604928
Lab consists of:
L as in Lightness
a how green or red the color is (White in a channel is Red; Black in a channel is Green)
b is how blue or yellow color is (White in b channel is Yellow; Black in b channel is Blue)

It follows the idea that one color can't be red and green or blue and yellow at the same time.

In Lab your Lightness is separated from your color, so painting in (greyscale) the Lightness layer will preserve any hue/saturation information you have on your image.

You can in theory use that popular some time ago way to paint via AO and lock that layer as Lightness, so when you color in it won't break your values, when blending or whatever process going on.

File: prtscr.jpg (108KB, 1504x858px) Image search: [Google]

108KB, 1504x858px

>>2604944
My guess is is that printscrening converts into RGB by default?
Here's a screenshot of original Lab and the bottom part is prtscreend and pasted in. You can see some difference with the naked eye, i guess there's more loss going on any time you convert.

>>2602036
Another useless waste of time thread with a bunch of plebs spouting bullshit. If talking about paintings with other unskilled artists actually made you better IC would be full of masters.

Unless you morons can draw/render as well as zeen no amount of picking apart their work will help you. These threads are all just distractions from actually doing work. It's a way of convincing yourself you're accomplishing something while actually doing fuck all.

>>2604950
is there any easier way to use a, b channel? like, having one "color channel" ?

File: lum_layer.jpg (100KB, 1351x656px) Image search: [Google]

100KB, 1351x656px

>>2604957
either save your lightness layer separately and then periodically paste it in your picture

or use a layer in luminosity mode on top to lock in you luminosity.

>>2604964
oh so that is a purpose of luminosity layer mode. thanks.

>people who can't even understand photoshop spend days trying to figure out secret 'tricks' into painting like someone
Why don't you just start mixing colors until you get it right? This thread is nonsense

>>2604995
>>2604956
Imagine if you faggots had your own thread where you can bitch and whine about other people bot doing what you think they shouild be doing?

This would improve the board immensely, why not do that?

>>2604964

So paint in the lightness layer, then copy to a new layer and then change to lumonisity?.

Or do you mean copy the lightness layer, paste in a new layer and then create another layer with lumonosity?. Do you paint on this layer, or do you create a layer below to paint?.

THanks btw

>>2604998
Imagine if you actually listened and invested some time mixing colors, this would improve your skills immensely, why not do that?

>>2605023
>tfw did more studies today than you did all week
At least you shitposted, right?

>>2605023
you're not even good, what's up with that hubris anon

>>2605033
>>2605029
>desaturate colors is studying
Merc me

File: 1468497552075lum.jpg (470KB, 920x651px) Image search: [Google]

470KB, 920x651px

>>2605012
I just tried >>2604964 's method.
top layer is Luminosity ( from L channel ). and below that it's normal layer ( from a+b channel ).
It seems working, but weird thing is that resulting image ( left side ) is lighter than original ( right side ). tried adjusting level and curve but it doesn't look right either. maybe I did something wrong extracting from Lab channel.

>>2605035
You mad?

>>2605041
oh i just found out. using original image as Luminosity mode & a+b channel on normal mode works perfectly. But... isn't it suppose to work with Lightness channel too?

>>2604995
>>stop discussing things that i don't think you should discuss
why do you care so much?

>>2605035
>>stop studying color techniques that i don't like

>>2605107
You aren't studying any techniques here. You are overthinking a simple painting process. The artist you are "studying" doesn't do any of this lab color wheel luminosity layer bullshit. He just understands to pick harmonizing colors that stay within controlled value ranges.

>>2605137
And how do you know what he does?

>>2605137
Youre right, he probably doesn't paint like this.

If you actually read the thread you'd understand that the topic has changed to whether or not actually using Lab color modes is efficient. Again if you read the thread, lab allows you to maintain value while shifting temperatures. It's relatively tough to do so when using the rgb color picker.

It's called actually reading the thread anon. Who are you to decide what we talk about?

>>2605053

anychance of screencap of the layers in ps mate please?

File: 2016-07-16 23_42_24-Untitled-1 @ 100% (Lab_8) _.png (37KB, 314x346px) Image search: [Google]

37KB, 314x346px

>>2602192
Also coolorus have option "luminosity lock" too.

File: Color.jpg (175KB, 1502x1064px) Image search: [Google]

175KB, 1502x1064px

Okay, after looking at this image again >>2604280 and translating what the brush settings look like

It seems they're painting with Color mode on the brush (the one you set up on the toolbar up top).

Now if you do this in RGB you'll get some Lightness changes, but if you use this method in Lab, you get a perfect perfectly preserved Lightness level. No need to hassle with the layers and so on.

>>2605320

What is the sat 20 about in that photo?

So in lab mode, do you still paint in grayscale in the channel lightness?. Then copy that lightness channel into a layer and then it on lumonsity right at the top.

Then create a layer underneath that and set brush mode to color?.

>>2605137
>>stop over-thinking things that i can't grasp as being simple and easy to think about

>>2605235
Does this support older software, or do I have to use Adobe's shitty subscription based crap?

File: Untitled-1.jpg (865KB, 2500x1211px) Image search: [Google]

865KB, 2500x1211px

>>2604280
I translated the previous tutorials so I figured I might as well

>>2605505
You're a superstar. Danke.

File: LabFormat.jpg (167KB, 948x1428px) Image search: [Google]

167KB, 948x1428px

Holy shit. Lab Mode makes it to much easier to "lock in" your values.

All I did was paint a greyscale ball. Then in a Hard Light (left) or a Color (right) layer, I painted in the local values. Then I color picked the reflected light area, and it gave me a color. In order to get a color I wanted, I would raise and lower the arrow on the vertical bar. (if you are on the A channel, this bar determines how much red or green there is in your color. B channel, red or blue) In doing so, I can change the temperature of the color in the reflected light area to whatever I want, without affecting the value.
I then picked a more saturated color to go before the terminator, added a bit of glow. added a few colors to the "skin" ball to give it variation in the light and voila! you have colorful balls!

This really helps as I can easily change temperature on a face (red cheeks, yellow forehead) without affecting the value.

>>2605505
Thanks chinesebro

File: Untitled-1.png (133KB, 920x699px) Image search: [Google]

133KB, 920x699px

>>2605216
either way works.
I suspect that I did something wrong extracting Lightness channel from Lab mode.

>>2605594

I'm still trying to understand this. But essentially why is this different from saying just creating painting in RGB mode and do all your work in grayscale and then adding a color mode layer, normal om top of the grayscale layer to color?.

Really confused?. Is this the same thing?.

>>2605802
Thanks, I still don't get this.

I thought you were meant to mean to use the lightness layer to draw your image in grayscale?. Then extract that place in the layer and change layer mode to lumonisty?. Then create a layer on top and paint on top for color?

Where does the a+b channel come from?

>>2606120
>>2606122
read the thread:

>>2604950
>>2604678
>>2604521
>>2604508

File: A-man-Sitting.jpg (17KB, 378x317px) Image search: [Google]

17KB, 378x317px

fuck i dont understand anything anyone is saying in this thread lmao

File: lab_blend.jpg (108KB, 902x916px) Image search: [Google]

108KB, 902x916px

i also did a quick test in Lab to see how using Color blending mode there is different from RGB and the result is pretty significant.

Even when converting from Lab to RGB the loss is pretty small, compared to drawing in on Color mode layer in RGB.

This thread is fucking gold , I hope it won't delete before I've found a moment to read everything

before this thread deletes, could someone give a quick summary of what all it entails? it looks like valuable information but i cant wrap my head around it yet.

File: lab_blend2.jpg (180KB, 2372x822px) Image search: [Google]

180KB, 2372x822px

And another one to drive the point home.

image TL;DR, painting in one Lightness in Lab adn using Color mode in Lab will yield the same result, but painting in Color mode in RGB will give you some great Lightness steps.

Lab to RGB conversion is lossy, but still closer to the original Lightness.

>>2606129

I'm still confused after re reading it.

File: f9FW2.gif (2MB, 240x180px) Image search: [Google]

2MB, 240x180px

This thread.

I don't understand any of this.

seeing all this conversion loss happening in Photoshop makes me wonder, What software does experts use when it comes to this specific subject? like, color proofing or, some scientific shit.

>>2606178

Photoshop.

>>2606175


Same here

>>2606178
Every photo editing/painting software has this functionality, you jsut have to be aware of that.

Most of the time it barely matters, but since we're delving into this subject it seems more important than it probably is.

>>2606140
>>2606175
good colors n sheit niggas

>>2606189
It's important to note that not EVERY editing/painting software will have this, but the ones that are trying to be more than just "your photo editor that came with the printer" usually will.

I've seem some really shit photo editing software out there.

File: colourmappath_rgb_lab.png (208KB, 587x394px) Image search: [Google]

208KB, 587x394px

>>2606175
>>2606161
Lab, unlike RGB is designed with human perception of color in mind, so unlike RGB Value, Lab Lightness is uniform across colors and you can use Lab to keep your Ligthness intact, directly editing color, which allows you to use Helmholtz–Kohlrausch effect to make colors seem brighter than their actual Lightness.

You can also use this mode to color greyscale/value studies without getting "muddy" colors, achieve the look in OP image
and god knows what else.

It's pretty fun to poke at it, i'll hit up some articles and maybe get some more info on how to use it in digital painting.


Should i compile this thread into one post/article/tutorial for future reference?

>>2606194
Oh, i wasn't regarding those at all, good point.

early archive, because this thread is useful and i can never have enough color stuff
http://archive.is/QYula

File: lab vs rgb vs cmyk.png (138KB, 325x336px) Image search: [Google]

138KB, 325x336px

>>2606198
>>Should i compile this thread into one post/article/tutorial for future reference?
yes definitely, would make things easier to read without the shitposting.

>>2606175
>>2606183
>>2606161

it's basically discussing two types of color, "Lab" vs "RGB":
- Lab is a range of color that's more "real-world" than the primitive format used on a computer screen. It's like the human perception of color.

- the other one, RGB, is a more primitive, artificial form of color used on our computer screens. it has far less colors than what we see.
in order to make pictures look really nice, and feel more like real world colors, we take RGB convert it to Lab (which is basically a bunch of nifty math), and it makes illustrations or photos on our monitor a bit nicer to look at, because it feels like there's a better range of colors going on (it tricks the eyes).

this thread was basically discussing easy ways of doing this with the color pickers and color channels in Photoshop.

if you look at this diagram, you'll see just how many extra colors Lab has, compared to RGB.
the third one in there, "CMYK" is the color format we use when sending things to printers, it also has a slightly different (but still primitive) range of colors too (since it has to make them using ink instead of monitor pixels).

File: CLv9rzn.jpg (44KB, 960x955px) Image search: [Google]

44KB, 960x955px

>>2606211
Guys, I'm a genius. I get what they're talking about now. They're talking about color!

File: CORRECT.gif (1008KB, 320x240px) Image search: [Google]

1008KB, 320x240px

>>2606216
Correct!

>>2606198
>Should I compile this thread into one post/article/tutorial for future reference?

FUCK YEAH

>>2606200
I love you

File: Rgb vs Lab.jpg (511KB, 2996x916px) Image search: [Google]

511KB, 2996x916px

>>2606140
>>2606146
Hopefully this clears some things up.

Note, that this most likely isn't how Zeen or Ruan or any asian artist paints. But, you can be damn sure it's a really good way.

A value structure is what creates a great painting. And if you can get that, Lab WILL NOT affect your values. Allowing you to experiment and get nice color variations without affecting the value.

Just save your image as a .tiff or a .pdf and then find a converter to make it .jpeg

File: CK3.gif (2MB, 498x279px) Image search: [Google]

2MB, 498x279px

holy shit,we actually managed a productive thread, good job /ic !

>>2606240
I love you too

>>2602036
Why does this woman have a baby coming out of her chest? Is it some sort of Chinese legend?

>>2606246
No, this is SWAG

Can author of those walls of text post his work/sketches?

>>2604395
You're welcome anon, glad to help

>>2606198

Thank you for responding and answering my questions.

I understand that lab preserves the values, however, I don't get how I can use that in painting my images. I been re reading the thread but I'm confused about how to start.

Like I said do I paint in the lighness layer in grayscale to do my values.

Then do I copy and paste into a layer, and then change to lumonosity?.

Do I then create a layer below or above to color?.

If you could do a small tutorial I be insanely grateful.

>>2606250
>>walls of text post
which one? quite a few of us have contributed some delicious walls of text. throw up some post quotes

File: GreytoColor.png (48KB, 788x328px) Image search: [Google]

48KB, 788x328px

>>2606256
Paint in greyscale in any mode, RGB, Greyscale, Lab

Then change the color mode to Lab and create a new layer ontop of the values. Set that layer to Color and paint away.

>>2606266
Can't we already do the exact same thing with RGB colors ?

>>2606270
See
>>2606241

File: wip.jpg (58KB, 433x623px) Image search: [Google]

58KB, 433x623px

>>2606250
Here you go, my dude

>>2606250
flash me

>>2606266
I did the same thing in both RGB and LAB to see the difference and I got the same result for the two
What did I do wrong ?

>>2606296
Looks like a SIMS 1 skin

File: Capture.png (728KB, 1648x911px) Image search: [Google]

728KB, 1648x911px

>>2606266
>>2606266

Is this correct?, still can't see the difference really?.

>>2606323
Why do you have 2 same greyscale ? is this important

what is going on in this thread

>>2606325

I was following this, not sure if this is correct.
>>2606266

>>2602684
literally draw the rest of the owl

>>2606339
I'm sorry, that was my bad
You only need one greyscale layer

File: stock-photo-17880219-confused-senior-man.jpg (69KB, 556x370px) Image search: [Google]

69KB, 556x370px

>>2606140
Same here. I didn't know understanding color was so complicated. It sounds almost like a science

>>2606357

So what I done here is correct?>>2606323

File: Comparison.jpg (861KB, 2692x2184px) Image search: [Google]

861KB, 2692x2184px

>>2606307
>>2606323
Ok hopefully this is clearer.
To be honest, it doesn't matter what mode you use in the end. As long as you are aware of what each mode does, you can adjust your workflow to adapt to these situations.

Although, to me, Lab seems like a much faster, more accurate way to put colors over a greyscale rather than RGB.

File: Capture.png (700KB, 1035x716px) Image search: [Google]

700KB, 1035x716px

>>2606379

You're right, lab definitely looks better.

File: MoreComparison.jpg (2MB, 3920x2828px) Image search: [Google]

2MB, 3920x2828px

>>2606379
I even did it with the OP image.
I used >>2603685 for the color layer on top of the greyscale.

Again, we see that in RGB some colors make the value lighter than the original. Just look at the red thing just below her boobs.

>>2606389
>>2606379

Fuck man, thank you for helping out.

I understand it

I finally get it.

Thanks

>>2602036
so is anyone going to actually post any work using this new discovery?

>>2602036
>mfw not sure if people are just trying to troll me with this thread

File: fuck.jpg (435KB, 1050x1050px) Image search: [Google]

435KB, 1050x1050px

>>2606394
Welcome to the glory that is color anon!

>>2606400
if people are only just getting the hang of it, you expect them to have already finished a piece with it?
good things take time anon.

>>2606402
no troll

Updated the archive if anyone was interested
http://archive.is/TyVsp

File: greyscaletocolorlab.jpg (488KB, 1500x972px) Image search: [Google]

488KB, 1500x972px

Im not any of the previous posters but ive been lurking the post since it started, I made this quick and rough example for the people that think this is a troll, found a random image from the internet that had different diferent values, then used the coloring technique that was described before

File: hqdefault.jpg (14KB, 320x180px) Image search: [Google]

14KB, 320x180px

>>2606241

As someone who has never given a thought or understood the benefit of LAB, holy shit.

Thank you, ive been looking for a good method of keeping values like this for a long time.

>>2606443
thanks anon

is it possible to use lab in clip studio, in which software's is it available?

>>2606241


The problem I have with this workflow is that it only works well with tightly rendered, solid objects like a figure. It wouldn't work with a loose painterly style since you aren't going to have every brushstroke on a separate layer or mask. Coloring over loose brushwork can only lead to that ugly digital glaze look. Why not simply aim to nail color and value in one go and skip the middle step? Please explain.

>>2606650
Look at any master painting. The strength that comes from these paintings are three things. These three things can be listed from most important to least important.
1. Composition (Rule of 1/3s, golden ratio)
2. Value Structure (High/Low key, light on dark, dark on light)
3. Color Scheme (Limited palette, complementary, triadic, etc)

When you consider a workflow, in it's basic form, starts with an idea > line drawing/composition sketches > value studies > color studies > final piece.
In traditional media, the masters were very methodical when approaching a painting. As long as you are careful, you can be in control every step of the way. There is no guessing unless you add an element of chaos (splattering paint, destroying and bringing it back. Look at golucho's process...you'll see what i mean).
As time went on, painters like Zorn or Sargent became the forefathers of mastering value, color, and form in a single stroke. Although they were able to achieve this mastery, they still had a concrete understanding of the first two elements : Composition and Value. Look at Sargent's drawings or Zorns self portrait etchings. Mastery of value. Even with the "loose" look, they know what value they need and where it needs to go. It's actually not quite difficult in traditional media as you were limited by the tubes of paint you had.

>Why master color and value at the same time
Go for it, no one is stopping you. But understand it is more approachable when learning them seperately.

>loose digital painting styles won't work with this
You're exactly right, but understand that in the digital media, it is much harder to get the stroke you want at the value you at the color you want. With a solid value study, you take away 90% of the guessing, which allows for artists to slowly learn how color and value work together. Over time, the idea is to achieve what Zorn and Sargent did, It brings back methodical creation rather than disruptful guessing in the digital medium.

>>2606685
>baaby's first critique

>>2606686
You have to be 18 or older to post on 4chan

>>2606706
Should be raised to 21.

>>2606722
It should be invite only

>>2606724
That's pretty much how it was in the early days.
>hey, come and check out this retarded site
And then everyone gets invited and you get what you have now.

>>2606685
>>loose digital painting styles won't work with this
I'd actually say this depends on the user. Painting value and color separately might not be loose, but this mode allows you for a tighter control of you Lightness, since when blending color it won't go into RGB weirdspace, and if you're having a hard time not geting muddy colors in your digital (which most beginners do, just look at any beginner drawthread), this might be a better way to control your colors.

>>2606782

Imagine a grassy field painted with a gritty dry brush texture brush with a lot of different hues. The overlapping brush strokes simply wouldn't be possible to color in solidly without having all brush strokes on hundreds of individual layers. Yes, you can add color with blending modes but the outcome would be a different style.

In the end it comes down to personal preference I guess. Different artists seem to lean towards either seeing in line, value or color primarily.

>>2602036
>draw anime
>erase the line art
>mute the colors

Wow so hard

>>2606986
le post work xd

>>2606650
>>this color method doesn't work for some other specific scenario, so it shouldn't be used at all

no one cares, this is fine for the type of images being discussed in this thread.

All this talk because you don't want to learn how to mix your god damn colors

>>2607064
>>stop talking about doing colors one way instead of another

>>2606731
there was no /ic/ back then, weeb

Good explanation:
https://www.youtube.com/watch?v=ZN6tzaCj68k&ab_channel=PhotoshopTutorialsbyPhlearn

File: LINEARVSGAMMA.png (29KB, 493x350px) Image search: [Google]

29KB, 493x350px

>>2603437
What's responsible for the discrepancy here is gamma-awareness. To convert between RGB and total luminance you have to convert from sRGB gamma to linear, which photoshop and krita seem to be doing.

Top here is photoshop->adjust->grayscale (gives similar output in linear and sRGB gamma), bottom is with a linear profile assigned (ie tricking photoshop into ignoring gamma)

>>2607078
they were probably meaning the entire website in general, not any specific board.

>>2607166
no

>>2603701
this thread is actually one of the better ones /ic/ has from time to time because it is a god damn discussion and study. Threads of "get gud" "more loomis" don't mean shit and help no one, /ic/ is at its best when it is breaking down specifics and figuring out things. Not when its just bitching. The tldr of the thread so far is
1) color has both an actual value related to brightness and light, and a value related black and white.
2) good painters take into account the value of the brightness and realize keeping the brightness of colors similar but varied can give an image pop merely by spacing out colors hue.
3) there is software to simplify the process of recognizing a colors brightness that doesn't just change its saturation and value but there is also lab pickers which work similar but aren't as inherently intuitive.

>>2607078
>>this retarded site
Now I wonder if I was talking about 4chan or /ic/?

>>2602192
>>2604672
Am I understanding correctly that RGB mode messes with lightness when painting over grayscale, and lightness is a combination of saturation and brightness that changes S and B in a curve; and tone-locking is basically picking a color, figuring out its value and painting over it with desired hues (in Lab mode)?
Why use half-circle/seemingly random curves when making a palette?
How exactly lightness "curves" S and B? Why it changes hue slider slightly?
And why do you need RGB mode at all?
I am a complete beginner in photoshop, digital painting and color, so I feel like an idiot asking these questions, but my head already hurts from re-reading posts and googling every single term, it's all too new and confusing

>>2607480
the curve color picking method is a work around and self imposed limitation to stop too much value (absolute brightness, not whiteness) change when all you want is hue change since extreme changes in color value is what makes colors to disharmonious. RGB thanks to the anons tests, does seem to treat color chroma differently than LAB and although lab isn't 100% brightness then hue based it is much closer in separating a color actual value from their hue. Color blending mode and luminosity blending modes relate better to what some anons are trying to do, and that is vary color hue but not color brightness/value.

>>2607249
>Threads of "get gud", "more loomis" don't mean shit and help no one....
Sentiments like this are why 99% of IC will never get good.

Threads like this are good for people that have enough prerequisite knowledge to do something with the info. Most of IC is still stuck on learning to draw, so the answers you get most often are "Go draw more." It's not like using LAB mode will make people into amazing painters that understand color, rendering, drawing, etc. All those "study loomis you moron" type threads are telling people what they don't want to hear, no amount of photoshop tricks and bullshit will make up for studying and learning how to do something w/o tricks.

I bet there are idiots in here right now that think this new method will solve all their problems and allow them to now make amazing work..... Methods such as LAB mode are little more than a means for an IC pleb to polish a turd. OP asked how to become Zee Chin, I can assure you it's not LAB mode.

>>2605356
Works on CS6. I've had it for a while but don't really use any of the features.

>>2606144
Use ScrapBook for firefox

>>2606986
Why do people like you proudly disclaim that they know literally nothing about the fundamentals of art?

>>2607741
What are you talking about? All I'm saying is that OP's pic is standard ass anime garbage with the lines erased and a nice muted color scheme.

>>2606241
why aren't we all paiting in LAB then, is it a new thing?

>>2607756
but that's wrong you fucking retard

>>2607756
>All I'm saying is that OP's pic is standard ass anime garbage with the lines erased and a nice muted color scheme.

Exactly, which means you know nothing about painting and rendering.

The guy paints as if he was doing cast studies. A clear value structure, clean edge control and simple directional lighting + bounced light. There is nothing "anime" about this painting style whatsoever.

right hand looks weird af tbqhwy

>>2607845
People aren't all painting in LAB because it's not some cure all magic bullet. You still need to know how to draw, you still need to have a grasp on how color works and how rendering works otherwise your work will still look like shit.

It's good alternative for people who already work from b/w underpainting into color. But if you'd just study color it'd probably be faster to start the painting in color and cut out all the extra steps. Ultimately this is just a crutch for people who don't want to learn color and just want to colorize b/w paintings then add a little bit of finishing touches.

Interesting new info but ultimately useless for 99% of IC and probably detrimental to their learning by giving them, what they perceive as, an easy out from learning color.

>>2608010
As long as I know that all I need is a good value painting, I can spend less time thinking about color. Lab gives the artist one less thing to worry about. Which in turn allows for quicker workflows and to pump out more for a possible deadline.

Also, there's nothing stopping someone from painting using a normal layer directly on top. Maybe this is an easier way to get a certain color, but you can fine tune it to get the result you want.

Lab allows for color variance and experimentation without risking your values. I fail to see the negatives for Lab.

It is a new discovery for /ic/ and maybe no one will ever use it, but the idea that there's a color mode that locks your values is a pretty nifty. Crutch has such a negative connotation and will drive away people from experimenting with it. A better word is that Lab is yet another tool in an artist's toolbelt and it encourages experimentation, which is what helps beginning artists.

File: nox.jpg (191KB, 1004x750px) Image search: [Google]

191KB, 1004x750px

I remember Nox Iz Mad used LAB color mode because he said he felt like he could just get better colors with it. He would have a color wheel swatches image open to quickly pick a ballpark color from, then adjust with the LAB sliders. So he used LAB color mode but still just painted directly. He demonstrated it in a video he did for conceptart.org back in the day, just before the big fallout.

as a traditional artist I cringe when I see digital """artists""" looking for shortcuts to avoid developing color by themselves

do you also trace 3D models because you couldn't learn gesture and anatomy properly?

>>2608366
jesus christ, why is this thread such a magnet for whinny faggots

>>2608239
>he said he felt like he could just get better colors with it

1. NOX is a retard
2. LAB was intended for theoretical colors
3. Monitors have come along way since 2004

LAB was made for photographers who wanted a future proof color gamut. Why do you think the sliders work in reference to exposure? Photographers wanted real world colors for their digital photo editing. Monitors back then didn't have a large enough gamuts to replicate real world colors accurately. So LAB filled the void by registering real world colors theoretically. So when the selected color was shown on a LAB quality monitor it would match up to it's real world color.

Here is the kicker, nearly all LCD monitors back then, and most today do not display LAB colors. Instead what you get is the nearest RGB equivalent. It works the same way CMYK is shown on your screen.

>>2607868
The figure is an anime girl. I know you faggots have some sworn duty to protect all your weaboo heroes until death, but don't lie to yourself and say that isn't anime. If you showed that picture to any asshole on the street, they'd probably tell you it's anime. Doesn't matter how well it's rendered or how soft the edges are. If you're painting anime character, you are painting anime.

>>2608397
>If you showed that picture to any asshole on the street, they'd probably tell you it's anime.
Nah, most would be "boobs, hehehe" and have no idea what anime actually is.

>>2608518
>anime isn't mainstream
just wow

>>2606986
wow you have superior artistic taste than these defensive anime fags. I surely can see that. You must be very proud of yourself.

>>2608389
you should've understood that most important thing is that Nox used Lab color slider. which is not about monitors; but personal taste how to determine color usage. just like using RGB slider or HSB slider. ( I believe majority use HSB ) and which is clearly seen by that screenshot.

File: 1468877922089.jpg (25KB, 215x285px) Image search: [Google]

25KB, 215x285px

>>2608518
>and have no idea what anime actually is

>>2608596
Live up-north lad.

>>2608518

I differ, >>2608397 is probably right.


calling a pic "anime" is stupid to start with, anime is pretty much a piece animated, calling it something like manga or oriental styled would be more correct imo

but yeah, >>2608397 is right


cool thread

>>2603532
wtf is going on here?

>>2605505
thanks corean freind you are my hero

oh boy, im gonna need a crash course on color theory please and values, and good reads or videos please im begging

File: color_workshop.jpg (30KB, 500x500px) Image search: [Google]

30KB, 500x500px

>>2610469
Get this bad boy and do the studies that are in the book.

>>2610507
And for values?

>>2606731
its was such a better place a decade ago.
except for this thread. this thread is the tits.

(even though I still don't know shit about any of this. Fuck me for just doing line art :( )

>>2610783
>>Fuck me for just doing line art :(
nothing wrong with that anon. think about it this way, when you want to do color stuff instead, you'll have plenty of study material to work with.

keep it up!

>>2610469
G U R N E Y
U
R
N
E
Y

File: unnamed.jpg (35KB, 500x407px) Image search: [Google]

35KB, 500x407px

>>2611173
>>2611172
>>2611174
>>2611175
lmao

>>2611194
This whole thread is autism anyway and will lead to nothing. But do you really think you will see an image or read a bit of text and suddenly your art will become very good? Thats not how art works no matter how much information you consume it wont make you a better artist, if it would be about information art schools or other institutions would produce god tier artists 99% of the time. So stop trying to find the magic bullet it won't happen, i spent way too much time doing this kind of autistic shit instead of drawing.Stop.

>>2611175
He really has a problem with boobs

>>2611194
Whad do you expect? 400 pages tutorial how to volumes?

File: Helmholtz-Kohlrausch_effect_visualized.png (145KB, 1000x412px) Image search: [Google]

145KB, 1000x412px

Okay so i'm still ready the wonderful book about Lab and it's uses for photo retouching, but i already see how it could be useful for digital painters.

I sure as hell don't know why nobody seem to use it, but since Lab is rarely used for photo retouching, the same thing Photoshop was made for, i'm not surprised.

Anyhow, i think the main issue people are having in digital painting is the fact, that while you seemingly pick proper color lotsa times after blending and cleaning your image still looks like a mess, like this post for example >>2611356
A lot of time it's because RGB is not as straightforward as it seems and because, if you know your color theory, colors are affected by nearby colors, so they look brighter or darker. See pic related or google Helmholtz–Kohlrausch effect.

File: labapple_1.jpg (67KB, 599x400px) Image search: [Google]

67KB, 599x400px

Now what's different about Lab compared to RGB or CMYK is that your color and your Lightness are separated, which allows you for a lot more control over your tone values (not RGB values). With L being Lightness layer and a and b being red/green and yellow/blue. With positive values being warm and negative values being cold.

So what you get from this as a digital painter is that your tone values can be corrected at any time.

This also makes Color blend mode work properly, keeping your lightness but editing the color.

File: labapple_2.jpg (93KB, 599x400px) Image search: [Google]

93KB, 599x400px

And what this allows us to do is to paint with color without messing up the values, since your tone value is separated.

Now you can use this for painting over grayscale with color wiuthout that disgusting Multiply blend mode effect.

You can add color variations without messing up the values, for example adding variations into fruit or skin.

It's also pretty close to traditional media, if you have any experience with that, since your neutral grey is in the middle and you warm and you basically have a color wheel in your color picker.

It also allows you to pick color of the same Lightness eliminating guesswork.

File: labapple_3.jpg (117KB, 599x400px) Image search: [Google]

117KB, 599x400px

But the most useful part is that this method makes your workflow a lot more flexible, since editing your Lightness channel is completely non-destructive.

You can add shadow and highlights withouth losing your hue and saturation.

You can edit your contrast and set high and low points without losing color.

You can remove a tone and clean up, or you can shift you color temperature by editing a and b channel, since they're cold or warm.

File: labapple_4.jpg (124KB, 599x400px) Image search: [Google]

124KB, 599x400px

So i encourage you to try this workflow.

I'll try doing a straight color version to see how well Lab handles color mixing and how we could manipulate the result with channel editing and i'm want to finish that book before compiling any kind of guides, but this already seems promising as a separate non-RGB workflow or as a great base for color studies and fixing up your work.

>>2608612
>you should've understood that most important thing is that Nox used Lab color slider. which is not about monitors

What you fail to understand is that regardless of the interface, regardless of using slider, the color shown in the picker, the color shown in the slider or whatever is NOT THE LAB COLOR it is an RGB approximation standing in for the lab color.

>>2611369

Huh? The portrait you linked to looks bad because the anon who painted it needs to train his eyes more. It doesn't have anything to do with any flaws in the RGB color mode

>>2611374
>>2611379
>>2611386
>>2611394
yeah but the color still looks like shit. instead of wasting time git gud.

>>2611379
>It's also pretty close to traditional media
t. someone who never did oils

>>2611211
>>I failed at a certain method, so no one else should try it either

File: c_prc_3okd4dj.jpg (551KB, 900x814px) Image search: [Google]

551KB, 900x814px

I tested LAB mode out with this environment I did.

What I did:

Started with values only one 1 layer, added colours in a new layer on top of first layer and set to colour mode.

When I converted to RGB I the colours have changed and values shifted a little. I'm guessing colour mode works differently in RGB and LAB. In any case, if anyone knows how to preserve the colours when exporting as a common file format (jpg, png) let me know.

>>2604858
It's the superior western way of thinking.

>>2611616
Save as a pdf in LAB
then convert to jpeg using a converter

Or save as a .tiff, too and do the same thing

>>2612249
or just convert to RGB and save asjpg, since it's the same result.

>>2612253
It's not the same result, actually.

You too would know if you read the thread.

>>2611616
wtf is lab even mean? i cant even see a difference

can someone just message this zeen nigga and find out what the fuck his process is and if he even knows about fucking values, i just want to see this thread get BTFO when he says he just does if of the dome

>>2611616
I actually like the RGB version better. Some colors are a bit brighter and it suits the mood.

File: 1469141406912.gif (425KB, 873x396px) Image search: [Google]

425KB, 873x396px

>>2613154
>>2613158

>>2613167
literally 0.1% difference in saturation thats it

the fact you're taking it apart and talking about values is pathetic

just greyscale a picture and colour it simple.

>>2613171
what you meant to say is
>my monitor is shit so i can't see the difference

>>2613171
>>2613172

but it's not supposed to be different

>>2613167
RGB is better though.

>>2613172
no motherfucker i have a mac and do art related work so my screen has been calibrated to optimum industry levels.

bottom line is you guys are looking way to hard, if you cant greyscale properly you will never be able to colour.

morale of the story do more life observation drawings its the only way

File: Screen Shot 2016-07-23 at 11.02.21 AM.png (209KB, 347x444px) Image search: [Google]

209KB, 347x444px

>>2613211
>no motherfucker i have a mac and do art related work so my screen has been calibrated to optimum industry levels.

File: 1406276175077.jpg (284KB, 1400x625px) Image search: [Google]

284KB, 1400x625px

>>2613220

>>2613222
So basically FengSu has taken over the industry and all these art boards now, great

no wonder theres so much brown and grey shit in the industry lacking colour

>>2613228
lel`

>>2613167
LAB is muddy as shit, which doesn't mean the palette is properly adjusted

>>2613234
its not lel he's great but we're slowly going to get saturated now with his stuff and people sucking his cock

File: maxresdefault.jpg (28KB, 568x709px) Image search: [Google]

28KB, 568x709px

>>2613242
Feng's art is pretty ugly. he treats it for what it is, art that communicates concepts. it's pretty soulless and i don't really see anyone but noobs who only do art just to get a place in the industry idolising his work

File: lab_tow.jpg (279KB, 900x406px) Image search: [Google]

279KB, 900x406px

>>2611616
>if anyone knows how to preserve the colours when exporting as a common file format (jpg, png) let me know.
You'll always get some losses no matter what, since Lab has more colors than RGB, but at least relative Lighness will stay the same.

>>2611616
Did you convert to sRGB or Adobe RGB?

>people in this thread report a discovery
>talk about huge potentials in Lab

>several days later shitposters who don't have the decency to read do what they do best

Good thing it was archived before all of these morons came in

File: 12138607_10152748640377537_3716897715859145026_o.jpg (236KB, 1200x1856px) Image search: [Google]

236KB, 1200x1856px

>>2613250
Some of his paintings may fall a bit flat but you have to recognize the dudes skill in his line drawings.

pic is one of the general greivous anatomy sketch. Doesn't fully capture his abilities but i think it's pretty neat from a production point of view.

>>2613211
>>i have a mac
that doesn't mean anything. there are both good and bad monitors across most brands, especially apple.

>>2613705
guess what i meant to say is that he isn't much of an illustrator, it's all pretty boring and cookie cutter style concepts. he's one of the best at that shit but it's pretty boring, and unless that's what you're aiming to do in the industry i don't see there being much there to like apart from the fact that he has the skills to pull it off.

>>2613167
LAB looks more muddy.

>>2614080
Wrong Apple are specifically calibrated