Me and the boys have been discussing that I should have done this guide years ago. Well, I got some time to pass on some knowledge, so that's what I'm gonna do. It's also in my best interest, I'd like to retire in peace where art is concerned, so might as well write something that can stick for those willing to pick it up ;D
Disclaimer: If VU forums has a character limit to forum posts, this one is gonna break it. This post will be excessively long, but by the end of it all, I hope to bring you guys up to a level where you can make your own art for your own needs your own time if you practice at it, without the need of requesting for unavailable people.
Also, the images illustrated here are coming from my Imgur account. I believe there is some country-wide block of this website and its assets by Pakistan. If someone can provide a permanent go around solution (without abandoning my Imgur account because it's the best place to upload pictures for me thus far) that would be great.
Dispelling some Fallacies
First order of business is to completely rewrite the way you think about art and how to get it done, cuz a lot of you self-limit yourselves.
You have to have the talent for art (No)
Art is not a talent, it is a skill, and a skill, no matter what it is, can be taught and can be acquired. Talent is like a natural calling, a way your brain is wired to understand the material easier. You can have all the talent in the world though and still suck because talent is not the driving force of any skill, learning through repetition is.The person who has a talent for art still has to practice. They may improve faster than you I'm not denying that, but zero talent does not equal you're incapable in art, everyone is capable of art. Hell, you guys can get even better than me, all in how much time you willing to put into it.
There is a systemic approach to art (not really)
Since many of you are coders of some light in the community, to put it in a perspective you can understand, art is as free as writing shit scripts. But how do you know your script is shit? How optimal it is when run right? It's the same concept in art. You can practically do anything you wish to achieve the desired result, but there are advisory approaches to ensure that what you did was sound and reasonable and won't cause you a lot of headaches later down if you made a mistake. So as a result, there is no real conventional way to make an art piece, but there is a method to the madness. Art is all about capturing and reproducing an image in whatever creative way you can construe; the end justifies the means.
You need Photoshop (NO YOU DO NOT FFS)
You write a Java script in Notepad++ versus writing it in Visual Studio Code vs writing it in IntelliJ or some other IDE like Netbeans. Different tools make life easier sure, but every single one of them can produce the same damn HelloWorld.java. Not one of these programs gonna make your shit code any better. They all just got the job done which was to write your code as you wished. The program didn't make you any better of a coder than a baby mashing buttons at the keyboard. Refer back to fallacy number 1, build skill; the program is only as good as the user.
The concept applies to Photoshop. Photoshop ain't going to make you a better artist. Sure, it has its reputation and feature set that makes life easier to produce the final product, sure, but it's no miracle worker.
The only art that's still surviving from the years back I use to use Photoshop is VW3 Banner and my signature. I have not used Photoshop for over 9 years now. The only two essential requirements you need of any program you plan on using is:
- The ability to have multiple layers to put drawings on
- The ability to install custom brushes (especially .abr brushes)
If the program of choice has that then you are free to use anything. I, for example, have been using GIMP since 2010/11 roughly after I retired my pirated Photoshop copy. Haven't turned back since.
Another point that exists within this fallacy is that not everything Photoshop is suitable for, meaning you have to understand the category of program it is. Photoshop is an image editor that rasterizes the image. I'll talk more about this term later but just know it makes unsuitable for logo work or symbols. Any program you use to replace Photoshop is equally unsuitable for logo work and symbols. You need another type of program that does vector image processing.
Now that that's been said, the lesson begins here.
What is Art?
Art is an expression of creative imagination that is pleasing to the eye and evokes an emotive response. In this case, it's a visual expression of these two features. How do you capture things visually? Building on the elements of art:
- Line
- Shape
- Form
- Space
- Texture
- Colour
- Value
In the context of VC:MP, the more important ones are colour, value, texture and shape. Let's start with shape.
Shape
A shape is a 2D boundary/outline of space. It's what we all practice the moment we put pencil to paper, and then self-assess ourselves on that we can't draw for shit. But here's a perspective I can show you to immediately improve yourself. Think of the basic shapes (triangle, square, rectangle, pentagon, hexagon, circle etc.). Everything you look at in the environment around you can be broken down to basic shapes. Try to represent an object using just shapes, whether regularly shaped or irregularly shaped (stretched, morphed etc. to fit the need). Do not try to aim for accuracy at this level. Not yet. You're practicing changing the way you think in order to transcribe items around you into the form of visual symbols. So let's change your mindset.
Everyone on Windows, practice this in Paint. You don't need special software for this. Also, mastering this practice helps immensely with 3D modeling because it operates off of the same concept, breaking things down into basic shapes.
(https://previews.123rf.com/images/wimage72/wimage721501/wimage72150100061/35701403-empty-drinking-glass-cup-on-white-background-.jpg)
This empty drinking glass for example. At the very basic level (not caring much about the 3D form itself) I can see two shapes, a circle (ellipse) at the top and an irregularly proportioned rectangle for the body, wider at the top and narrower at the base. The rectangle was drawn using lines. This is my basic result and can be yours too:
(https://i.imgur.com/hLzovPP.png)
As you can see I've shown a joint and an disjoint version of the two shapes used to accomplish this.
Now the astute among you would realize that the base isn't exact to the glass cup, the narrowed base of the rectangle is actually curved, making it more of a cylinder than a rectangle. If you were able to see this, then congratulations, you have just recognized Form. Form especially continues into 3D art such as 3D modeling. We don't give a shit about it right now. Focus --- Shapes.
If you would want to represent the cup more accurately by just using shapes you have two options:
- Use a circle at the base
- Use a polygon with many sides that you will manipulate into the shape you need
For the example of the cup, the circle base serves better, but the polygon option has its place and will have to be used eventually as you practice with other objects. Here's an example:
(https://i.imgur.com/AZbggRI.png)
As you can see, I can have two circles now and a irregularly proportioned rectangle to get the job done on the left, whilst on the right, I chose a circle up top and a irregularly proportioned octagon for the remainder of the body. You can use any number of sides you wish but the more you use, the closer you get to a circle. A circle is just a polygon with an infinite number of sides afterall.
Finally, you would clean up the lines that don't belong, like those lines running through the ellipses that are not representative of the real image. Practice that basic manner of handling shapes and lines and you are well on your way already. Remember it applies to everything. There's nothing in this world that can't be broken down into a basic shape.
Colour
Colour is a visual sensation for the eye, as sound is to the ears. Colour in the real world is produced from whatever visible light waves an object emits or reflects to our eyes. This information we perceive as the colour of the object. Different colours are called hues of colour, and so, the colour wheel that we normally refer to, is really a hue colour wheel. When mixing colour, we mix different hues of colour to produce, new colour, in-betweens. Mix with enough hue variety between all the different spectrum of colours and you will get the rainbow gradient of colour. Note well though that hue is pure colour, it does not include any form of pale or darkened colours, no whites, greys or blacks.
Now in general, people don't have much issue here replicating colours, but in the context of VC:MP, this lesson is more on two things, blends of colour and emotive properties of colour.
Blends of Colour
(https://www.realmenrealstyle.com/wp-content/uploads/2016/03/color_wheel.gif)
The basic colour wheel. When we talking about colours that blend, we are really talking about colour harmony, colours that work together and contrast each other as well. This is based on the type of colour scheme used in your project. An example I can grab on comes from one of Vice War 7 flags, Hindutva.
(https://i.imgur.com/iK9VCA0.png)
Hindutva's flag more represents a blend of colours that are analogous, similar. Check the colours used for the entire flag and compare where they all generally land on the colour wheel. You will see that every colour is right beside each other from red to orange to yellow to green to blue. That is the natural order of the visual spectrum, making it visually pleasing with a nice natural flow that is balanced and sound.
Analogous colours is the use of any three colours that are subsequently adjacent on the colour wheel. So for example, starting from orange , the other two colours would be yellow and yellow green to be used, and so on. Analogous colours creates harmonic flow.
But note that there's also an elements of contrast. By the time we hit the blue, we have established a direct complementary colour to the orange present. This also enhances the image - contrast makes things pop and stand out. This colour scheme relationship is what we call complementary colours.
Complementary colours is the use of a colour that is directly opposite to the base colour on the colour wheel. What constitutes it being the opposite colour? If a colour is upper left on the wheel, its direct opposite is exactly across from it flowing down and right, so for example, the opposite of red-violet would be yellow-green based on the colour wheel.
In the case of the blue used, it is directly complementary to the orange present, bringing out contrast in the symbol that is very pleasing. The red of the Om symbol is also directly complementary to the green in the flag as well.
Even more powerful is the establishment of a triadic colour scheme in the flag of blue, orange and yellow green.
Triadic colours is the use of any three colours that are equally spaced on the colour wheel.
From a design perspective on colour, this flag is very sound.
Here's a fuller picture of the total colour schemes that can be used (and explanation for the remaining colour schemes):
(https://d2jv9003bew7ag.cloudfront.net/uploads/Color-Harmony-Schemes-Image-via-Msbixler-weebly-com.jpg)
- Split Complementary - the use of analogous colours of the direct opposite of a base colour; 3 colour scheme
Example:
- Base Colour to use: Red
The direct opposite of red is green, but you do not use the green, you use the colours that are beside the green on each side. This would be cyan and yellow-green. Therefore, based on the colour wheel, the 3 colours to be used are:
Red, Cyan, Yellow-green.
- Tetradic - to use the analogous colours of each of the two directly opposite base colours; 4 colour scheme
Example: - Base Colours that oppose each other: Red and Green
You don't pick the base colours, you use the two colours adjacent to each side of the base colour respectively. Looking at the colour wheel as a guide, you will therefore use magenta and orange for the red side, and cyan and yellow-green for the green side. And so you will get:
Magenta, Orange, Cyan, Yellow-Green
- Square - to use four colours equally spaced on colour wheel; 4 colour scheme
Example:
There are many other colour schemes, but they involve the topic Values, and some of those won't be stated until later down in the post.
These mix and blends of colour based on different colour schemes really help with logo and signature work to make what you want to stand out pop but at the same time not be so chaotic to look like you don't know what you're doing. For example you don't use analogous colours as the theme for your text and background, chances are you will not be able to see the text. Complementary/Split complementary would suit well. Inversely, you don't use complementary/split complementary on the background of an image as it will take away from the symbol, the eye-catcher to your work, or what you wish to be legible.
We have one more thing to discuss though regarding blends of colour and that is Colour Context.
(https://cdn2.hubspot.net/hubfs/53/color-context.png)
Colour Context is a play on similarity and contrast. A colour that's perceived to be in the foreground will experience a slight colour shift depending on the colour in the background. Take for example the inner dots of each pair of circles close together. They are the exact same colour within, but your eyes might notice a slight colour shift. Due to differences in lightness, shade and hue, you can create contrast even between analogous colours. Colour context provides a good way to highlight things and split colours apart when things are crowded.
What do I mean by crowded? Monochromatic, another colour scheme, involves the topic of Values so I won't go too in depth yet on it. Basically, an example of a monochromatic work is the very forum banner I made for Vice Underdogs.
(https://i.imgur.com/BI9ucNq.png)
One base colour is used throughout this work, but the value of that colour is different to create the split between the palm trees, the environment background and the text in the foreground. Overall you get an image that both flows but is still contrasting to be legible.
With all that said, now we move on from blends of colour to the emotive properties of colour.
Emotive Properties of Colour
As stated before, art is an expression of beauty and emotion, and the emotive properties of art is expressed via colour almost exclusively. This topic however, is where things get subjective, as different colours mean different things depending on the culture. I'll try to stray from that as much as possible, but where I can't, then what I'm about to explain represents more or less the Western and European viewpoint.
Referring back to the colour wheel, let's go in a clockwise fashion.
(https://www.realmenrealstyle.com/wp-content/uploads/2016/03/color_wheel.gif)
The wheel is split evenly into 12 sectors. Starting from Red, I am skipping the colour that is immediately after, so my list will therefore be Red, Orange, Yellow, Green, Blue, Violet. i.e. Primary and Secondary colours.
Quick note, what are primary colours? Colours (hues) that cannot be achieved by mixing with another colour.
Secondary colours? Colours (hues) achieved by mixing two primary colours together
Tertiary colours? Colours (hues) achieved by mixing a primary and secondary colour together
Let's begin:
- Red
Second only to yellow, red is 2nd most luminous colour of all the colours of the visual spectrum. In terms of mixing paints of colours, red is the 2nd most potent colour (it shifts the hue of colours very easily when mixed in), 2nd only to black. Red globally tends to represent one of three things:
- Aggressive & Sinister (rage, danger, assertive critical warning)
- Pride & Boldness (fashion to entice etc.)
- Femininity (lighter tones)
- Orange
Orange is a hot colour that is typical of fruits in the natural environment, a symbol of good health. However, orange is usually connotated in a polarizing fashion to have one of two meanings:
- Excitement & Flamboyance (thrill, energetic, full of vitality)
- Nonchalance & Dullness(calm, boring)
People either love or hate orange, and I personally am not fond of it myself. It's my most rarely used colour of all my years in art.
- Yellow
Yellow is the most luminous (brightest) colour of all the singular colours of the visual spectrum, (white is not a colour in the mixing paint perspective, and is not a primary nor secondary colour in light spectrum perspective). It has three main global meanings:
- Happiness (warmth and sunshine)
- Naivety & Youth
- Vigilance (warning/caution signs)
- Green
Green is the 2nd most abundant colour of the natural environment, 2nd only to blue. Green also has polarizing meanings:
- Trust (bright tone greens)
- Hope & Renewal (brighter tone greens)
- Envy & Jealousy (mid tone greens)
- Fear & Shame (lower tone greens)
- Blue
The most abundant colour of the natural environment and has the most diverse meanings of all the colours:
- Trust (lighter tones)
- Serene & Peaceful (lighter tones)
- Pride & Eminence (lower tones)
- Grief & Sadness (brighter tones)
- Violet
The rarest colour of the natural environment. As a result, violet tends to evoke feelings of:
- Pride & Eminence
- Mystery
- Sacred & Spiritual
- Femininity (lighter tones)
In general, that's what the colours tend to evoke, but the meaning can be shifted somewhat based on the context of the image. Example, violet can look sinister and threatening if affixed with a shape that resembles something fierce. Red can look happy and joyful in a similar fashion. But overall, warmer tones are more exciting whilst cooler tones are more calming
What are warmer tones:
Reds to Oranges to Yellows
What are cooler tones:
Greens to Blues to Violets
Values
Values are a sub-property of colours. Values of a colour are properties that shift colour one of three ways:
These shifts can be categorized under three headings of mixing:
- Tint - adding white to a colour
- Tone - adding grey to a colour
- Shade - adding black to a colour
Now some terminologies to clear up and consider:
Bright vs Light
When we say a colour is bright, we are saying that the hue of the colour is pure, 100% saturated. When we say a colour is light (or pale), we are saying that the hue of the colour is dull, not 100% saturated. The hue (colour itself) did not shift (meaning we didn't move from red to orange etc., that is a hue change), what happened was that the colour got pale due to adding white. No greys or blacks must be added for tinting.
Light colours is what you refer to as pastel colour swatches.
Neutral
A colour becomes neutral when grey is added. So when we say neutralize a colour or tone it down, we mean to add grey to the colour. A colour can be neutralized three ways:
- Adding pure grey to the base colour
- Adding both white and black to the base colour
- Adding the complementary colour of the base colour (for a more true to life result)
Shading (Darkening)
You shade (darken) a colour by adding black to the colour. No greys or whites are added to darken a colour.
Now look at this picture of a colour chart for the hue Red.
(https://image.shutterstock.com/image-vector/color-palette-mixer-3-black-260nw-700482016.jpg)
This picture is really depicting tones of red, with the extreme edges of the left and right side depicting shades and tints of red respectively. Whether tinting, toning or shading, they all kill saturation of the colour, that vibrance of the hue at max brilliance. These are the technique used to develop monochrome colour schemes mentioned previously above in Blends of Colour - Colour Context.
And so to complete the point previously on colour schemes, you can reference this link here (http://www.theinformedillustrator.com/2012/10/color-schemes-defined_16.html) to gain a full understanding of each that exist that I haven't mentioned yet but rely on both shifts in hue and values.
That will cover all the knowledge theory needed where it regards to how to practice and why certain art practices are carried out for the best results. This is honestly the bulk of the information you need to know to get started, as you can practice in any program, even Paint. But now we will get into the programs aspect of things.
Clearing Up Photoshop Misconceptions Part 2
Now, near the beginning of the post, I did say that you do not need Photoshop and that Photoshop isn't suitable for every task. These two statements are correct, but why exactly?
A more detailed explanation of what Photoshop is is that it is a photo-image editor that processes and saves rasterized images. What exactly are rasterized images?
Rasterized images are your everyday image file that you preview on your devices or capture with your smartphone / cameras. The PNGs, JPGs, GIFs, TIFFs, they're all rasterized images. The term rasterize is used to describe the fact that the image was made by using a grid array of pixels to capture and represent the image. It's convenient and saves a ton on processing power, but it has one big limitation - it is a lossy compression style that's not flexible.
It is because of that reason, it is absolutely ill-advised to create logos and symbols in any editor that does rasterized image editing. When you create a project file within these image editors to create such work, the fate of the image quality is sealed the moment you choose the resolution you're working at. No matter what resolution you save in, you're losing quality with every upscale/downscaling that needs to occur, and working in big resolutions eats system resources (RAM) very easily for little reward as use cases require smaller images typically, until one day comes where high resolution images are required (like the evolution of video files 720p to 1080p to 4k now, it's just gonna keep going).
There comes the program that is suited for such work - Adobe Illustrator. Now I'm not saying you need Adobe Illustrator, you don't either, but Illustrator represents the most popular program that processes vector images. What is a vector image? It is an image that is entirely represented my mathematical calculations of shapes. As a result, it is impossible to lose quality when upscaling or downscaling any image, it is a lossless format, very fit for a project file. When the image needs to be used for the web (converted into a rasterized image), you can quite easily export to what you need, whether it was smaller than the original project image size or way larger, image quality does not degrade.
So why do people use Photoshop?
Rasterized images have its plus if you treat it for what it is, an image editor. Consider an image that has all the possible colours of the 8 bit digital spectrum, that's 16.7 million colours blended together. It is not computationally intensive to load a grid of pixels in such manner, but it is computationally intensive to load a bunch of layers or shapes in such a manner. This is the negative of Illustrator, and it's also why work from vector images are relatively simple looking. You will not see a picture painting coming from Illustrator, the art is more cartoon-fashion rather than realistic. Gradients used in these programs are also actually rasterized effects where the resolution of said effect is controlled. But to blend colour like you can in Photoshop, to truly do it in Illustrator you would need millions of boxes in the shape of pixels to achieve it which is impractical.
So final verdict then:
- For realistic images and paintings / drawings - use a raster image editor software
- For logos, symbols, maps and other simplistic work that can be layered and have minimal colour gradients - use a vector image editor software
Examples of raster image editing softwares:
Adobe Photoshop, GIMP, Corel Painter, Affinity Photo
Examples of vector image editing softwares:
Adobe Illustrator, Inkscape, CorelDraw, Affinity Designer
All these examples satisfy the two requirements needed to make working in either a robust setup for you, i.e.:
- They support multiple layers for objects / drawings
- They support custom brushes that can be installed (category for raster image software specifically)
I personally use GIMP for my image editing and drawing needs and Inkscape for my logo, badge, map and even texture needs for some of my 3D models where it's applicable.
GIMP (Raster Images)
Employing the practices above, when I work in GIMP, every object is given it's own layer. This minimizes destroying the entirety of the work when I'm erasing a mistake.
(https://i.imgur.com/hLzovPP.png)
For example, in this previous picture above where we drew the glass cup in Paint, there's no support for layers, and so, to remove the lines between the joint shapes, I would have to infringe on both the rectangle (what I'm aiming for) and the circle (what I don't want troubled). No matter how skilled you are, you're gonna damage the ellipse in some way.
When working in GIMP (or any other image editing software), the workflow would more look like the disjoint example on the right - one layer would be for the ellipse and the other layer would be for the rectangle. I would move each layer in such a way where they are aligned to be joint, the illusion, but technically they are not. This gives me freedom to make changes to the rectangle and remove the visible lines cutting through the circle layer that could be above/below the rectangle's layer.
The same is true of colouring. Each base colour for the object gets its own layer. These layers can be grouped to be collectively called whatever the final object is suppose to be. Then, colours and sketches / outlines can be subgrouped into their own categories. Colours could then be subgrouped even further to represent layers of Highlights, Shades and Tones. Sketch groups could be layered out to be hand sketches and finished outlines (drawn and traced with the Path Tool). All of this comes together to represent what you see here for example:
(https://i.pinimg.com/originals/ab/bd/91/abbd91fbabb79792c90654a7a62d631c.jpg)
- The background is its own group and layer
- The details of the buildings is its own layer subgrouped within the background grouping
- The sky is its own subgroup and layer that is behind everything
- The water and beach is its own group and layer
- The details of the water and the beach are their own layers
- The outline of the person is its own group and layer that's on top of every other group and layer
- The rim lighting and shading are their own subgroup and layers
- The glasses of the man is its own subgroup and layer
- The gun is its own group and layer
- The details of the gun is its own subgroup and layer
Point is, everything gets it's own layer as much as possible where feasible in order to maximize flexibility of editing without deleterious effects.
Images are usually created at 200 DPI and at around 2000 x 2000 pixel resolution (taxing on RAM).
Inkscape (Vector Image)
The same concept of layers apply for Inkscape (vector image processing) as well. Everything gets its own layer. Take for example this recent badge I made:
(https://i.imgur.com/Vylgg2z.png)
- The background colouring is its own grouping and layer that's behind every other grouping and layers
- The emblem is its own grouping and layer
- The text is its own grouping and layer that is above every other grouping and layers
- "VU of the Year" is its own subgrouping and layer
- The year number "2020" is its own subgrouping and layer
Here's the example of how the layering looks in Inkscape:
(https://i.imgur.com/oOASUUo.png)
This allows me to quickly make changes to the colour of the background (creating the silver and bronze badges), change out the category of award (new text as new layers within the subgrouped layer), and change the year for every subsequent year needed.
The copy layers are just for boosting some elements that weren't too visible on its own with just one layer.
Vector image processing are more forgiving though because everything present is already an object. However, had I not taken the approach to layering, I would have to change the background colour, the text and the name to these single objects each and every time I want to save a new category.
With the concept of layers, I hide the layer that I don't want to use, expose the layer I desire and make only one new addition which is to add a new layer for the new year number. Everything else remains common between the previous years, making the workflow far easier and quicker to export and have ready.
This particular work could've also been done in GIMP, but the plus about using Inkscape instead is the freedom to export the image to whatever required size you need. For example, the requirement for this image was to be of size 120x35. If I had worked at 120x35, then enlarged the image, this is the result:
(https://i.imgur.com/drRvlGs.png)
As you can see, it looks very pixelated, blurry and overall shitty, quality was lossed on the upscaling. But if I worked the image at a size of 1200 x 350, then all is fine....for now. But what if I need it for displaying for a market of high resolution 4K monitors, having the freedom to easily upscale and natively export to say even 12,000 x 3,500 resolution or hell, 1,200,000 x 350,000 pixels, I can easily do that without trouble nor quality loss. But the moment I click in a resolution that's not high enough in GIMP as the project resolution size, that's it for the image. Upscaling or downscaling, it only goes downhill from there where quality is concerned (downsampling still degrades quality, but not as bad as upscaling; always best to produce an image at the native resolution needed)
Textures
The next big topic that really can't be taught without going over all the previous first. Texture refers to the way things feel, but in this case, visually representing how it would feel if touched. Texture is all about capturing the look of a material. Colour, once again, is the driving point that almost exclusively defines how a texture comes out.
To capture texture properly in art (if you're aiming for high realism), you have to therefore do a lot of material studies. Again, texture defines material. So for example, it is good to accept that polished metals reflect well. Understanding that chopper bike handles, engines and frames are all chrome plated with chromium that reflects only 70% of the visual spectrum, tapering off reflecting the reds at only 60% or so, that is a totally different ball game of achieving a sense of realism to your art. It comes easier when you dabble into 3D art, with working with what we call physically based rendering (PBR) shaders, but its not required for this. I'm just stating how far it can go.
For this, I'm gonna speak of textures in two ways, the drawing art perspective and the 2D image perspective (to map onto 3D models). Both can be created without really stressing to download some pre-made thing that doesn't meet your needs (and for the 3D models, the 2D images that aren't tiled which is very annoying).
Drawing Perspective
When it comes to portraying various textures in a drawing/painting, it's really just about capturing the unnecessary details. For example, this cinder block:
(https://images.homedepot-static.com/productImages/ae65c7d3-3968-4d3f-a016-3ba9fec08257/svn/cinder-blocks-100002879-64_1000.jpg)
If we were drawing this without a care of texture, then all you do is break up each standout feature of the edges into its basic shape, create your outlines off of that premise and then just slap on a grey colour within the space and just like that you're done. With textures however, you're accounting for all those little cracks, chips and crevices that you can see. You're adding that otherwise unnecessary detail to the drawing to really sell it that this material is concrete based on giving it the texture of concrete.
In your image editor, you would create a separate layer for this detail. One for the base colour of the concrete block that would be the bottom layer, and then another for the details, that would be above the base colour layer.
The good thing about this approach to it is again, minimizing deleterious actions to only one object as much as possible, but two, is that you have access to a feature called layer blending.
What is layer blending?
A fusion of the upper layer with the lower layer (or layers below the top layer) via a mathematical algorithm expressed onto every pixel's individual colour. If you didn't understand that, then another explanation can be that it's a bunch of modes used to create different kinds of overlays by the upper layer to the rest of the image. It's a powerful tool to use to quickly express how colours will add up (or even subtract), especially when dealing with art that involves lighting and shadows.
Use Case:
You finished drawing and colouring an entire scene, and you want to add a red global light to it. You have two options:
- Overlay the scene with layer blending of the red colour for the light you wish to use
- Try and recalculate all the colours you used in the scene
And if you've used many colours, it's a no brainer, just overlay the damn thing with some layer blending.
Now there are different layer blending modes, and they vary vastly between software, as I believe Photoshop was the pioneer for this method of workflow and so tend to be above the rest of programs with the modes they provide. I will however go over a basic list of them though, that is especially useful for texturing.
- Dissolve - the upper layer is dithered into the lower layer when blended together. It produces a noisy dusty blend of dots and irregular shapes like this:
(https://photoshoptrainingchannel.com/wp-content/uploads/2017/04/sample-02-dissolve-blending-mode-800x235.jpg)
It only comes into effect when you change the transparency of the upper layer. If the opacity is 100%, you won't get an effect.
By far the easiest way to add the chips to the cinder block for example. All you do is create a layer with dark brown/black (your call), set the layer mode to Dissolve and change the Opacity percentage to a level where there is sufficient amounts of dots present to give the result (in GIMP, that was roughly about just 5% for me)
- Multiply - multiplies the luminosity of the colours on the lower layer by the colours of the upper layer. The resulting colour is always a darker colour. White produces no change, while the black pixels remain.
(https://photoshoptrainingchannel.com/wp-content/uploads/2017/04/sample-04-multiply-blending-mode-800x235.jpg)
Very useful for shades / shadows. I need not say anymore.
- Screen - produces many different levels of brightening depending on the luminosity values of the upper layer
(https://photoshoptrainingchannel.com/wp-content/uploads/2017/04/sample-09-screen-blending-mode-800x235.jpg)
Very useful for highlights and brightening a light emission source in your scene (creating sun glares and so on).
- Overlay - a combination of Multiply and Screen with the base layer always shining through. Uses the Screen blending mode at half-strength on colours lighter than 50% grey and the Multiply blending mode at half strength on colours darker than 50% grey. 50% grey itself becomes transparent.
(https://photoshoptrainingchannel.com/wp-content/uploads/2017/04/sample-13-overlay-blending-mode-800x235.jpg)
I abuse this a lot when I want to tone and saturate colours to a particular hue as it adds both brightness and contrast. I also use it if I have a rather complicated upper layer to blend with (like a radial gradient of black and white). The white portions gets Screened while the black portions get Multiplied.
Also useful for drawing a bump mapping effect.
What is bump mapping? This:
(https://universalbri.files.wordpress.com/2016/11/bump-map1.jpg?w=602&h=355)
It's what makes the crevices, cracks, holes etc. in the texture come to life, based creatively using lines of tints and shades to create a beveling effect. This term is more for 3D software though where its an effect of 3D modeling itself regards to how to express the texture, but it can be captured in drawing. We call it shading. Like this:
(https://cms-assets.tutsplus.com/uploads/users/108/posts/28318/image/how-to-draw-scales-final-min.jpg)
Did you recognize that these 3 examples are scales? Exactly why capturing textures are important to sell the material.
2D Image Perspective
This part is geared more for 3D modelers, and so you could skip this section if you solely only care for the drawing aspect of things.
Ensure that you square your image beforehand.
This section employs all the previously mentioned techniques where it regards to layer blends and so it's important to note well that section of the post. The creation of textures for 3D models does not solely rely on a rastor image editor. In some scenarios, the vector image editor will need to be used, and so for this section, it requires some skill in both programs. For example, this work I did once:
(https://i.imgur.com/c9uXOTx.png)
(https://i.imgur.com/GSmuMbO.png)
The holes within the model were not created through geometric means in order to keep polygon count low. This is as a result of making the texture this way, however, patterns are rather difficult to create in GIMP. And so the alternative was to create this texture in Inkscape, where patterns like these are very easy to create. Another problem I ran into was trying to figure out the resolution needed to make this texture look good. I had originally thought a resolution of 512x512 would've done the job but when I tested it out, it looked horrible; the holes lacked definition. For Inkscape, exporting at a higher resolution is an easy piece of cake. If it originally done in GIMP, I would've had a massive headache trying to sort out the quality because upscaling is gonna degrade to hell.
There are benefits to using GIMP for textures however. For example, there is no easy way in Inkscape to add effects to the image. So say for example I wanted the texture to be more realistic, adding in rust or so to the texture, there is no feasibly easy way to do that in Inkscape where as processing such in GIMP is simple enough.
And so creating these texture images depends on what you need and where it is easier to accomplish it. Patterns, grids, and resolution bumps are far easier to accomplish in vector image editors while realistic detailing is far easier in raster image editors.
For realistic images, as a general guideline to capture certain materials, this is my advice:
- Polished Metals
- Find the base colour for the type of metal
- Look up the reflection data for the type of metal (just a basic overview of how much it reflects and how it looks when it reflects)
- Have a colour gradient of at least five different tones from the same base colour that alternates between tints and shades. This serves as the metal's reflection from a source of light, giving that metallic glimmer
- Brushed Metals
- Find the base colour for the type of metal
- Have a colour gradient of at least five different tones from the same base colour that alternates between tints and shades
- Create a new layer of white noise that you will then apply a motion filter to of angle 0 or 90 degrees. If enough motion blur was used, this will create a streaking effect that you can then overlay the colour of the metal with to create the brushed textured look (use Overlay Blend Mode)
- Concrete
- Find the base colour of the concrete
- Create a new layer that will be filled with the colour of choice for the crevices, cracks and chips of the concrete base colour. Blend this layer by changing its blending mode to Dissolve then reduce the opacity to the level needed to represent the amount of detail you require. Blur this layer if needed as a final touch
- Roads
- Find the base colour for the road
- Create a new layer of white noise that you can then overlay the base colour with
- Apply Gaussian blur to the white noise at areas where tire streaks would "soften" the grit of the road.
- Sand
- Find the base colour for the sand
- Create a new layer of white noise (increasing chroma if available) then overlay this layer to the base colour (use Burn Blend Mode)
That's five type of textures you have the capability of making on your own, just by using a raster image editor. A sixth one could be wood, but I've never personally done it. The approach though would be to draw a layer of circle outlines that I would then warp by using a warping tool. And if you manage to do that, then the only thing worth sourcing is clumps of grass and leaves.
But let's say you're too lazy for all that learning or you see a picture that looks worthy of texturing with but you know it's currently not in the best of setup in terms of resolution. And even after you've had your fun of creating your own textures, there are still processes you have to run through before its 100% usable. Let's go over that now.
You would have noticed that I stated that the image should be a square image, i.e. 1024x1024, 512x512 etc. this is a requirement for tiling the texture image.
What is tiling the texture image?
It's when the image is absent of seams and can be continuously tiled as a grid of clones. It is a necessity.
An image that was made with one base colour that flows through the entire frame is already tiled. Example, a completely red, black, brown, yellow, whatever colour picture. That is already tiled. If you keep the edges in such a fashion where your pattern is the same at every edge and every side where it's relevant, then this also satisfies the requirement when mapped onto a 3D object. But most of these blanket images aren't tiled, so how do we fix that?
There is a simple method and an advanced method depending on the image you're working with.
NOTE: Ensure that your image is either a square image already or cropped to be a square image.
Simple Method:
- Offset the image using an Offset Tool for layers, and ensure wrap around is selected. What will happen is that the edges of the image will get wrapped into the image frame as you offset the image. You want to offset the image in a fashion where all four original corner edges are now present within the frame. Your result should look something like this if you used the correct tool and done it properly:
(https://img.glyphs.co/img?q=85&w=900&src=aHR0cHM6Ly9zMy5tZWRpYWxvb3QuY29tL2ltYWdlcy9zZWFtbGVzcy10ZXh0dXJlLXJlZHV4LXR1dC8zLmpwZw==)
- Use a healing / cloning brush tool to get rid of the edges present within the frame. Basically what the healing brush does is takes a sample of the image that you specify and paints it over where you choose to paint. Cloning does the same thing, but is for large mass of objects within the image. As a result, it moves the sample zone in accordance to your painting (in order to copy the entire object).
Usually the simple method is enough for most images that are already decent but just not tiled. The simple method should also be enough for your creation of textures. However, it's not a miracle worker. In the event it isn't (like this wood texture used as the example), the advanced method comes into play.
Advanced Method:
The advance method has two precursor steps before the simple method is implemented. And so the full roster of steps would look be this:
- Duplicate the image you wish to tile, making it the top layer, set its opacity to 50% and flip it horizontally and vertically
- Merge the duplicated image to the original image
- Employ the Simple Method thereafter
- As a final step, this image may need sharpening at the end
The final result for this wood texture came out to be this:
(https://img.glyphs.co/img?q=85&w=900&src=aHR0cHM6Ly9zMy5tZWRpYWxvb3QuY29tL2ltYWdlcy9zZWFtbGVzcy10ZXh0dXJlLXJlZHV4LXR1dC9yZXN1bHQuanBn)
And just like that, shopping for textures is practically a thing of the past. You can now draw them, make them, or process pictures into them. Good job. :thumbsup:
So, that's that for that. Remember art is all about creatively capturing your perception of what you're seeing. Having a keen eye improves the detail but understanding how to break everything down to the basic fundamental level is key to good artistry because we can never capture everything perfectly, we can just give back the best approximation we can produce.