Chrod Style Fractal Manipulation

About a year ago, I developed a method (or methods) for artwork that combine 3D models and fractals. Because it was written a year ago, it is now a tad bit outdated (in technique and grammar). However, for the most part, it is correct and will get you going. Any modifications to the method will probably be put in this post rather than in the pdf files (sorry), though I’ll probably update them anyways.

Direct file downloads (Mediafire):
Tutorial
Tutorial with Images (you probably want this one)

Note that the images are specifically for Gimp 2.6 to 2.8 and Metasequoia 2.4 to 3.0 and I have no intention of updating those images for future versions of those programs.

THIS IS A VERY LONG TUTORIAL!

~~~—————~~~

3D Fractal Art Tutorial – Chrod Style

By Nicolaus Anderson
Completed Dec 17, 2012

Table of Contents

i) Definitions
1) Introduction
2) Pre-design Considerations
a) 3D modeling considerations
b) Preparing the fractal art
3) Part 1 – Making the Model
a) Model creation
b) Rendering
4) Part 2 – Coloring
a1) Branch A – Cartoon Borders
a2) Branch B – Nonintrusive Borders
b) Highlighting and Shading
5) Assembly Techniques

Definitions

Fractal – a structure (in an image) generated on the basis of one or more mathematical functions.

Flame – or “fractal flame” – a fractal whose appearance is designed to mimic flicks of fire or string in one or more aspects. (Update: I’ve read it’s something else, but it’s irrelevant for our purposes.)

Activity of a fractal – the color variations, curvature, etc. – things that make the fractal look unique in every sector (usually 50×50 pixels for the work in this tutorial, but its depends on the size of what you are applying the fractal to).

Highlighting – The brightened areas of a model that indicate the source of the light on the object.

Shading – The darkening colors that indicate areas of an object that the light reaches the least.

Introduction

The method I’m going to describe is one that I’ve developed as a result of figuring out how to make professional-looking buttons from 3D models. It is a method for applying fractal imagery to three-dimensional objects without using model skins and at the same time retaining a three-dimensional appearance. While you may be able to figure out how to replicate my methods using other computer
programs, the tools used in this tutorial are:

> Metasequoia – (free version available, Windows only)
> Gimp – (free, all operating systems)
> fractal editor – (free ones are available for any OS)

For Metasequoia, version 2 or later will do. I prefer version 3 because of improvements in rendering (which will be required in this tutorial).

For Gimp, I suggest version 2.6 or later. As long as the interface and shortcut keys have not been changed, you could probably use earlier versions just as well.

For a fractal editor, I suggest using JWildfire for the largest regions (you’ll find out what this means later). You can also use Apophysis (1,2, or 7x). All of these I’ve mentioned are free. You could also consider using IncendiaEX (which is also free), but there are some limitations I’ll discuss in the next section. JWildfire is made using Java, so it is cross platform but requires Java. It has optional external renderers – one for 32-bit Windows and one for 64-bit Windows, and these do work faster than the internal Java
renderer.

Pre-Design Considerations

3D Model Considerations

A note on Metasequoia:

Metasequoia is a free 3D modeler with certain important features that we will be using in this tutorial. While it is not required that you use Metaseqoia, you must be able to do the following:

1) Create a model.
2) Render the model with a solid color or transparent background.

Metasequoia renders its models with a white background by default.
Furthermore, another feature of Metasequoia used in this tutorial is it’s toon rendering. This rendering creates a border and smooth regions that will be used for the fractals. These regions are not required if you want a single region for all of the fractals. By “region”, I’m not referring to a connected region, I’m referring to all of the sections of the model that will share the same set of fractal patterns as their background. You can attempt to create these regions yourself, applying “Posterize” (to the image of your rendered model) in Gimp to guide you if you wish to determine these regions on the basis of their exposure to the light source (which is what I do in this tutorial).

Do not forget about the light source. For a consistent piece of art, you need to make sure that the light source on your models comes from the same place. If you are using Metasequoia, you can set the light source very easily using the light source ball. Make sure to save your models, especially since they contain light source information. To maintain light source information between models, you can open the Metasequoia file (extension “mqo”) in a text editor and copy everything between the curly brackets after the keyword “scene” to the same section in your other Metasequoia model files in the project.

An important consideration for your models is the surface area. The more surface area that is shown, the more area there is for the fractals to appear. Naturally bulbous objects such as ocean creatures, mushrooms, and pipes are good choices. You can make just about anything bulbous. Also, things will large surface area with few fine details will work. Plants with smooth surfaces can be done ( tree trunks shouldn’t have bark – instead, use a pattern (possibly a fractal one) to generate the bark). Cars can work, although, if you’re doing the front, the car should be made a single region (see note above) to make the fractals easier to identify.

Another important consideration is smoothing. If you’re model editor does not render curved surfaces as being smooth, it will makes things much more difficult to achieve the same slick results that I’m getting. This is fine if that’s the style you’re trying for, but it can make coloring the model much more tedious than it would normally be.

Preparing the Fractal Art

A note on Apophysis:

The default fractals for any version of Apophysis aren’t usually as variant in shape and structure than those of JWildfire (by that I mean that shape and structure variance with respect to size). They also tend to not be as smooth. The reason for this is the rendering techniques of both programs. I suggest creating a large version of the images and then applying a 1×1 pixel Gaussian blur to them using Gimp. I tried applying a post-gaussian blur inside of Apophysis, but the results were worse than if I hadn’t applied the
blur at all. Another technique, aside from blurring, is to enlarge the image in Gimp without full-scale (100%) resolution on resize. The default is 72% resolution. Then, shrinking it again (using 72% resolution as before) results in a very softly-blurred image that will effectively give you a slightly more defined image than if you had used a Gaussian blur. However, this tends to take longer and there’s no easy way of identifying the optimal size to enlarge your image in order to get the desired blurring effect. Another issue with Apophysis fractals is that the color mixture goes out of whack when you try to change the hue using Gimp’s hue modifier. This modifier is found in the menu “Colors” and labeled “Hue-Saturation”. I frequently use “Colorize”, but this also tends to diminish the strength of the coloring of the Apophysis fractals. To fix this problem, it is best to change the coloring within the program
(Apophysis) itself by changing the gradient color scheme. Make sure you save the parameters of the fractals so that you can reload them and change their color if you don’t like your previous results.

A note on JWildfire:

The fractals of JWildfire, while they come out very nice, always tend to have a dark cloud around them from the fadeout of the fractal (the fractals tend to be light in the middle and then fade out to darkness around them). This is most prevalent with flowers, but it occurs in other fractal structures as well. To get the strongest coloring in your models, you’ll want to reduce or remove these dark regions since they will interfere with any fractals you may layer on each other.

A note on IncendiaEX:

To create fractals using this program requires alot of processing. It’s best to use only if you have a powerful enough computer to handle the work. Otherwise you’re just melting your hardware. Futhermore, fractals from this program tend to be rather large. To use them at full scale (without image resize), the models you create will need to be such that the regions where the fracal image is applied are composed of large enough clumps to see the “activity” of the fractal.

Preparation of any fractal:

While most fractals you create should already be very large on the basis of program output, the large scale is not necessarily required. I tend to make 1024×768 size fractals because then they fill the full
image region (even though the final image can be larger). By “filling the full image region”, I’m referring specifically to the output image of the rendering of our model from Metaseqoia (noting that this image’s size will not be modified until the final image is auto-cropped and output, as I will describe later).

Resizing your fractal ought to be determined on the basis of the fractal’s activity. The more activity in the fractal, the less likely it needs to be resized. If you do decide to resize it, you will need to consider the issues that arise with smaller fractals. Small images of fractals tend to have the problem of hard-edges. This occurs because part of the fractal extends outside of the image borders and is thus cut off. This issue can be eliminated entirely by making
the fractal image a seamless pattern via the following method:

1) Open the fractal image in Gimp.
2) Apply the “Make Seamless” filter. You will find it in the menu “Filters”->”Map”.
3) Use “Edit”->”Copy Visible” to copy everything showing.
4) Select “Edit”->”Paste As”->”Pattern” and give it a name.

The pattern should now appear in the patterns box. I’ve applied the tag “fractal” to my fractal patterns to make them easy to find. This is easy to do in Gimp 2.8 (though I don’t recall if you can do that in 2.6).

Part 1 – Making the Model

Model Creation

Using Metasequoia

I expect you to already be skilled in making models using the model editor of your choice. If you are new to Metasequoia, fear not! it is a fairly easy program to figure out, but I’ll give you few steps here to get you going:

1) Open the program. It’s a stand-alone executable, meaning you can double click on it and it will load right away.

optional – Click “File”->”Basic Mode” to add other buttons. DO NOT USE BLOB! Blobs created with this
tool do not show up in the rendering (at least, not in the free version).

2) Click “Primitive”.
You can now select the primitive of your choice from the window that appears.

3) In the semi-translucent menu that appears near the model, select “create”.

What you should appear is a solid gray primitive.

I usually begin with a simple primitive (rather than making my own) and dividing it into segments. This can be done prior to creation using “Segment” (in the same menu as “Create”) and clicking and dragging the bar that appears. It can also be done using the knife tool (labeled “Knife” in the main menu bar) and the sub-menu option “Cut with a knife”. Click and drag the mouse from a point outside the model to a point on the other side of the model. Notice that a line is generated, starting at the point where you first clicked an ending at the current location of your mouse pointer. When you release the mouse button, wherever this line intersects an edge of the model, a vertex will be created, accompanied by its own set of edges and faces to keep the model as solid as it once was (Keeping the model as solid as it once was depends on what you are trying
to apply the knifing mechanism on; some surfaces simply can’t be cut without distorting the model or making it discontinuous along the cut).

Using other Modeling Tools

Some modelers cannot create renderings of the model apart from the workspace. This can be rather inconvenient, but if you insist on making your models this way, you might consider loading it with an
engine (for example, irrlicht) or converting it to the format of a modeler that will make renderings of the model.

Pre-rendering Note

Before you render your model, take advantage of any smoothing techniques your modeling software has to offer. This will ensure ease of manipulation of the image. Metasequoia provides Catmull-Clark smoothing, which, for simple models, creates bulbous components that are slick and professional-looking when combined with the smoothing. The steps to apply this are as follows.

1) In the menu bar, under “System”, click “ObjPanel”.

2) In the menu that appears, there will be a list of objects (usually one for me). Double click on the object. This will bring up a panel that displays special information about that object, including its color.

3) Under “Patch”, click the radio button next to “Catmull-Clark” and raise the segmentation as high as it will go (16, usually). Depending on how complex your model is, this can make rendering very slow.

4) Under “Smoothing”, click the check-box next to “Apply Smoothing”.

5) Now click “Ok”.

Rendering

In Metasequoia, to create a rendered image, go to “File”->”Rendering” or hit Ctrl+Shift+R. In the menu for “Effect”, there will be a check box labeled “Toon” and a box with a number in it labeled “Edge Wd.” (edge width). The edge width should be set to 0.5 unless you really want a cartoon look. You need the following:

1) One rendering with “Toon”. This is NOT the same as a normal rendering with a toon line, as you might discover.

2) One normal (or “plain”) rendering (with “Toon” box unchecked). This rendering should have a nice smooth gradient color to it. I repeat myself: this is the “plain” rendering.

Note, unless you’re trying to be experimental, I suggest leaving the texture of the model as the default material. This gives the model the proper glare for shading and highlighting (discussed later in this
tutorial).

Be sure to turn uncheck “Anti-alias”, otherwise you will end up with a speckle effect. If you forget, it’s fine – I’ll show you how to clean it later in this tutorial.

IMPORTANT NOTE: The images you’ve created will be the final size and cropping of the image until the very last step of this process. Unless you’re intentionally trying to make the model larger than the
picture, you need to have zoomed out from the model so that it fits within the frame for rendering.

Saving Your Renderings

Metasequoia allows you to save you images in any size you want (see the menu “Size” in the rendering screen). Consider taking advantage of this if you want huge images.

Part 2 – Coloring

Now for the fun part and, depending on how much work you put into the model, probably the longest part. Figuring out this method from scratch took me several hours, but I can now do it in about fifteen
minutes or less.

If you’ve created your model in Metasequoia, you should have two images: the “plain” rendering and the “toon” rendering. These images should have been taken at the same time so that they can be super imposed.

If you did not use Metesequoia, the image you will still need an image I call “toon”. This image is the one that determines the regions the fractal patterns will occupy as well as the overall outline of the model itself.

Note, both the “plain” and the “toon” renderings, whether the toon be artificial or not, are required. I use a 3D modeler to save myself work.

If you decide to create a mimic “toon” rendering, you can use the plain rendering as a guide. Steps to creating a mimic “toon” rendering are:

1) Open the plain rendering in Gimp.

2) Select “Colors”->”Posterize” and set it to 3 colors. Then hit “Apply”.

3) Select “Filters”->”Edge-Detect”->”Edge” and choose the algorithm from the dropdown menu. I suggest one with the fewest gaps in the lines (and I’m not referring to the dots that often show up in the lines – those are irrelevant). Slide the bar labeled “Amount” to “10”. I believe this is supposed to thicken the edge, but it only works some of the time. Now hit “Ok”.

4) Select “Colors”->”Invert” to change the outline to black and everything else to white.

5) Select “Colors”->”Threshold” and click “Ok”. This ensures all other colors have been removed.

There you have it – regions for your fractals. You can now follow the steps in “Branch A – Cartoon Borders: Creating the Colored Regions” with the exception of the fact that you will need to use the Fuzzy
Select tool instead of Select-by-Color.

(Actual) Step 1 – Clearing the Border

The model itself needs to be isolated. In the plain rendering and the toon rendering (if you did not do it via the prescribed method above), you may need to clear the surroundings and clean the edges via the following method. If you created the too rendering as given above, you only need to follow the first three steps.

1) Select “Layer”->”Transparency”->”Add Alpha Channel”.

2) Use “Fuzzy Select” (the wand tool in the main toolbar; shortcut key is “U”) and click on the region surrounding the model in the center of the image. In addition to this, you will need to select other
regions that are not part of the model.

3) Select “Edit”->”Clear” or hit “Delete” on your keyboard. The area surrounding the model should now be transparent.

At this point, your renderings may have edges that are speckled with white or light gray dots (which can come from trying to use Metasequoia’s anti-alias for rendering). EVEN IF THIS IS NOT THE CASE FOR YOUR PLAIN RENDERING, the following steps will give it a smooth edge that is noticeable in the final result. Toon renderings do not require the following procedure if you are trying to make a cartoon border.

4) Turn on Quick Mask (shortcut key Shift+Q). To do this, click on the little gray box in the lower left hand side of the main window. This box turns red when selected.

5) Select “Filters”->”Blur”->”Gaussian Blur”. Set both the horizontal and vertical values to 2 pixels. This should be done automatically when you adjust one of them, assuming the values are locked (as indicated by the chain next to them). Then click “Ok”.

6) Turn off the quick mask by clicking the same button.

7) Select “Edit”->”Clear” or hit “Delete” or your keyboard. For images from Metasequoia, you may need to repeat this step more than a couple of times to fade the bright colors to a satisfactory amount. It doesn’t matter if they disappear entirely – just as long as they aren’t noticeable when the zoom is at 100%.

Step 2 – Creating the Regions

Everything here on after will be with regards to the toon rendering unless directly specified. Regardless of what images you have now, follow the next two steps:

1) Select “Select”->”None” (shortcut keys Ctrl+Shift+A) from the main menu.

2) Rename the layer “base”. If you right-click on the layer, a menu will appear. Selecting “Edit Layer Attributes” brings up a dialogue box that allows you to name the layer. If you’ve designed your outline outside of Metasequoia, then you don’t have the option of nonintrusive borders and you can skip these subsections all together.

Branch A – Cartoon Borders

For our work, cartoon borders are thick, distinct borders around and sometimes throughout the image. Cartoon borders will appear stronger or weaker depending on whether you put this border on top of all of the other layers. Its standard placement is directly below the layers containing the fractals (as will be addressed later in this tutorial). To create cartoon borders, we apply an edge-detect to the toon rendering.

1) Select “Layer”->”Duplicate Layer” (shortcut key Ctrl+Shift+D). Now select the new layer in the layers panel. Rename this layer “border”.

2) Select “Filters”->”Edge-Detect”->”Edge”. I prefer Prewitt-Compass since it usually creates a thick line with fewer holes than the other algorithms. The resulting image should be black with a white border. If it is not, apply threshold via “Colors”->”Threshold” to ensure that the image contains only black and white.

3) Use the “Select by Color” tool in the main toolbar and click on a black region. All of the black regions should now be selected.

4) Select “Edit”->”Clear” or hit “Delete” or your keyboard.

5) Select “Layer”->”Duplicate Layer” (shortcut key Ctrl+Shift+D). Now select the new layer in the layers panel. Rename this layer “border blurred”.

6) Select “Select”->”None” (shortcut key Ctrl+Shift+A).

7) Select “Filter”->”Blur”->”Gaussian Blur”. Set both the horizontal and vertical values to 2 pixels. This should be done automatically when you adjust one of them, assuming the values are locked (as
indicated by the chain next to them). Then click “Ok”.

You should now have three layers. The bottom layer (the original layer, named “base”) will be used for color extraction. The topmost layer (“border blurred”) is the primarily outline of the final image. The second layer (“border”) is used for creating the backdrop.

Creating the Background

1) Click on the layer labeled “border” (the second layer).

2) Use the Fuzzy Select tool to select the regions that are not part of the model.

3) Select “Select”->”Invert” (shortcut key Ctrl+i) from the main menu.

4) Create a new layer. You can do this via “Layer”->”New Layer” (shortcut key Ctrl+Shift+N) or via the piece of paper button at the lower left corner of the layers panel. The background should be set to transparent.

5) Select the new region in the layers panel.

6) Select the Fill button (the paint bucket) in the main toolbar.

7) Set the foreground color to whatever color you want for behind the fractals. I usually use black (sometimes white) because it makes the colors of the fractals stand out.

8) Click within the selected region. This should change the color.

9) Select “Select”->”None” (shortcut key Ctrl+Shift+A).

10) Apply a 2×2 Gaussian blur. This will ensure that the background edges are not sharp enough to bleed through the border.

Creating the Colored Regions

Using cartoon borders has the benefit of separating the outline creation from the colored region handling. Since this is the case, we can now select those regions without regard as to whether some of
the edge is included.

For each region (area or areas containing the same color or intended to have the same fractal pattern background), repeat the following steps.

1) In the layers panel, click on the layer named “base” (the original image).

2) Use the “Select by Color” tool in the main toolbar and click on a region of color. You should now have one or more areas selected.

3) Create a new layer. You can do this via “Layer”->”New Layer” (shortcut key Ctrl+Shift+N) or via the piece of paper button at the lower left corner of the layers panel. The background should be set to transparent.

4) Select the new layer and label it “inner region”.

5) Create a layer mask via “Layer”->”Mask”->”Add Layer Mask”. You can also doing this by right clicking on the layer in the list and select “Add Layer Mask” from the menu that pops down. The mask background should be black.

6) Select the Fill tool (the bucket) in the main toolbar and set the Affected Area to “Fill whole selection”.

7) Set the foreground color to white.

8) Click within the selected area. Depending on the thickness of the selected area with respect to the image size, this may change the thumbnail representing the mask. If it does, that’s good. If it doesn’t, it may just mean your area is too small to show up in the thumbnail. No big deal.

9) Apply a 2×2 pixel Gaussian blur. This time, it will be done to the mask.

Now you’ll repeat all of these steps (yes, even 1).

Notice that the new layer(s) has not been filled in – it is completely transparent. Within this layer, you may now apply whatever fractal pattern you like. I recommend making copies of the layer and
repositioning these layers to above the border layer.

At this point, reselect the layer labeled “border” and select “Colors”->”Colorize”. Changing the lightness until the border becomes colored. Now you can give your border a color that best compliments the fractals you used in your image.

Branch B – Nonintrusive Borders

Nonintrusive borders are very thin, almost non-existent borders. These are best when dark shadows are used (as will be explained later) to help shape the image.

Nonintrusive borders require immediately handling the colored regions. Furthermore, because the color of the toon border produced by Metasequoia can be a dark grey similar to the interior, you can’t use the Select-by-Color tool (as in the Cartoon Border Method) for every region. Instead, some of the regions must be selected by the fuzzy select. This makes it tedious to work on images containing models with several regions.

1) Use the Fuzzy Select tool to select the regions of the same color (or those that are meant to contain the same fractal imagery). You may need to use the quick mask to ensure that none of the border is
ripped away. It’s not a big deal if you don’t care about the border, but having a border, even a tiny one, can make things look much more professional (at least for this style of artwork).

2) Cut the selection using “Edit”->”Cut” (shortcut key Ctrl+X).

3) Create a new layer.

4) Select the new layer and label it “inner region”.

5) Create a layer mask via “Layer”->”Mask”->”Add Layer Mask”. You can also doing this by right clicking on the layer in the list and select “Add Layer Mask” from the menu that pops down. The mask background should be black.

6) Select the Fill tool (the bucket) in the main toolbar and set the Affected Area to “Fill whole selection”.

7) Set the foreground color to white.

8) Click within the selected area. Depending on the thickness of the selected area with respect to the image size, this may change the thumbnail representing the mask. If it does, that’s good. If it doesn’t, it may just mean your area is too small to show up in the thumbnail. No big deal.

9) Apply a 2×2 pixel Gaussian blur. This time, it will be done to the mask.

Now repeat all of these steps for each layer. Notice that the new layer(s) has not been filled in – it is completely transparent. Within this layer, you may now apply whatever fractal pattern you like. I
recommend making copies of the layer and repositioning these layers to above the border layer.

Note that step 2 is the fundamental difference between this method and the cartoon border creation method. This way, we remove the undesired colors and leave residual lines that are too thin to make a
noticeable difference in the image.

The border of the image is what remains in the layer “base”. To convert this to its final form as a border, follow the next steps.

1) Select the layer “base” in the layers panel.

2) Apply a threshold, “Colors”->”Threshold”, and set the slider such that all of the colors turn out white or black. You may need to try both to see which works best for your image (black usually works for
Metasequoia toon borders) because one color may leave a speckling of the color at the other extreme.

3) Apply a 2×2 pixel Gaussian blur.

Now for coloring:

4) Select “Colors”->”Colorize”. Change the lightness until the border becomes colored. Now you can give your border a color that best compliments the fractals you used in your image.

Highlighting and Shading

Highlighting the image is what brings out the true three-dimensional appearance of the model.

The procedure for creating shading is almost identical to the one for creating highlighting with one difference: the color being manipulated in black instead of white. Where this difference comes into play will be pointed out in the procedure. It is possible for you to change the color of the shading or highlighting using Colorize, however, the shadow multiple-light highlighting and shading will need to be covered in another tutorial.

The steps to create highlighting are as follows:

1) Open the plain rendering.

2) Copy everything via “Edit”->”Copy Visible”. Note that at this time ,you should have already made the background transparent and cleaned the border as instructed above.

3) In your current image (the one that began with the toon rendering), paste the contents of the clipboard as a new layer via “Edit”->”Paste As”->”New Layer”. Rename this layer “shading” (we will use this layer for both highlighting and shading).

4) Bring the layer to the top of the layers in the layers panel (by clicking and dragging it there). It is important that this layer be on the top.

5) Set the layer opacity to very low (maybe 5%). This will add a very, very soft grey coloring that unifies the interior with the border and adds a slight 3D effect to the model. It’s not much of a 3D effect, but from this layer, we primarily want the unity and the softening of the colors.

6) Duplicate the layer, select the duplicate and label it “highlighting” (or “shadows” if you’re doing shading).

7) Set the visibility of the layer “shading” to off. Do this by clicking the eye next to the layer in the list of layers in the layers panel.

8) Set the opacity of the layer “highlighting” (or “shadows”) to about 70%. You may need to adjust this later, but 70-80% seems to be the usual setting for my work.

9) Select “Colors”->”Color to Alpha”. A window should pop up. Click on the button with the color and another window should appear that allows you to pick the color. For highlighting, set the color to black
(since this color will be removed). For shadows, set this color to white. Click “Ok” for both windows.

10) Select “Layer”->”Transparency”->”Threshold Alpha”. Another dialog box should appear. Set the slider until the white region or regions are about a third of the size of the regions they reside on (but note, this isn’t the best idea for every model). For my models, this tends to mean the slider is right in the middle.

11) Apply a Gaussian blur. 2×2 pixel blurs make for a kind of slick-button look. Any larger of a blur and you will need a mask to keep the white (or black) from showing up outside of the border.

Sometimes a different glare is needed to enhance the image. For a different highlighting, reduce the opacity of the highlighting layer. Then repeat steps 6-11, but in step 10, create thinner white regions,
and in step 11, increase the blur to a 10×10 or an 18×18 pixel blur. Note that you will need a layer mask on the outside (which you can make from the layers “base” or “border”) to prevent bleeding of the
highlighting out of the object border and into the transparent background.

Assembly Techniques

Combining your images into one large image can have a really cool result, but there are a few things to watch out for and a tip or two you should know.

Things to watch out for include border discrepancies. If you’ve made a border incorrectly without realizing it, it will stick out. The best thing to do in this case is shrink that object or model image. In this
way, pixilation hides the error. Another issue is that of border style inconsistency – you’ve made your models using different borders. You might be able to get away with it (via awe-factor of the overall
image), but it will probably be the first thing that stands out to critics.

Putting your sub-images partially off the screen is a great way of making the image seem like it is merely the caption of a larger world rather than the world merely being the size of the caption. It’s also a way to hide flaws in models. Furthermore, you don’t worry about hiding your precious work since the element’s regions were created in bulk (thus, it’s not like you’re wasting effort).

Conclusion

Learning the process takes time. Given how long this tutorial was, it’s hard to believe I can make a model and color it a total of maybe 45 minutes or less, depending on model complexity and how experimental I go.

I’d love to see your results. You can respond on my DeviantArt account.
(ablipintime.deviantart.com or chronologicaldot.deviantart.com)

About chronologicaldot

Just a Christ-centered, train-loving, computer geek.
This entry was posted in art, fractal and tagged , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , . Bookmark the permalink.

Enter the space and time of my little world... Welcome Earthling.

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s