Difference between revisions of "Documentation:Tutorial Section 3.2"
Jholsenback (talk | contribs) m (heading chages) |
m (Indexentries fix) |
||
(38 intermediate revisions by 3 users not shown) | |||
Line 7: | Line 7: | ||
<br> | <br> | ||
<!--</wikitalk>---> | <!--</wikitalk>---> | ||
− | + | ====Isosurface Object==== | |
+ | {{#indexentry:isosurface, tutorial}} | ||
+ | <p>Isosurfaces are shapes described by mathematical functions.</p> | ||
+ | |||
+ | <p>In contrast to the other mathematically based shapes in POV-Ray, isosurfaces | ||
+ | are approximated during rendering and therefore they are sometimes more | ||
+ | difficult to handle. However, they offer many interesting possibilities, like real deformations and surface displacements</p> | ||
+ | |||
+ | <p>Some knowledge about mathematical functions and geometry is useful, | ||
+ | but not necessarily required to work with isosurfaces.</p> | ||
=====Simple functions===== | =====Simple functions===== | ||
− | <p> | + | <p>Let's begin with something simple. In this first series of images, let's explore the <!--<linkto "Functions, user-defined">user defined function</linkto>--->[[Reference:Function|user defined function]] shown as <code>function { x }</code> that we see in the code example below. It produces the first image on the left, a simple box. The container, which is a requirement for the isosurface object, is represented by the box object and the <code>contained_by</code> keyword in the isosurface definition.</p> |
− | |||
− | |||
− | |||
− | <!--<linkto "Functions, user-defined">user defined function</linkto>--->[[ | ||
<pre> | <pre> | ||
Line 22: | Line 27: | ||
</pre> | </pre> | ||
− | <p> | + | <p>You should have also noticed that in the image on the left, only half the box was produced, that's because the <code>threshold</code> keyword was omitted, so the <em>default value</em> 0 was used to evaluate the x-coordinate.</p> |
+ | <p>In this next code example <code>threshold 1</code> was added to produce the center image.</p> | ||
− | < | + | <pre> |
+ | isosurface { | ||
+ | function { x } | ||
+ | threshold 1 | ||
+ | contained_by { box { -2, 2 } } | ||
+ | } | ||
+ | </pre> | ||
− | <p> | + | <p>It is also possible to <em>remove</em> the visible surfaces of the container by adding the <code>open</code> keyword to the isosurface definition. </p> |
− | + | <p>For the final image on the right, the following code example was used. Notice that the <em>omission</em> of the <code>threshold</code> keyword causes the x-coordinate to be again evaluated to zero.</p> | |
− | |||
− | < | + | <pre> |
− | + | isosurface { | |
− | + | function { x } | |
− | + | open | |
+ | contained_by { box { -2, 2 } } | ||
+ | } | ||
+ | </pre> | ||
− | <p>< | + | <table class="centered" width="770px" cellpadding="0" cellspacing="10"> |
+ | <tr> | ||
+ | <td> | ||
+ | [[Image:TutImgIso_01.png|center|220px<!--leftpanel--->]] | ||
+ | </td> | ||
+ | <td> | ||
+ | [[Image:TutImgIso_02.png|center|220px<!--centerpanel--->]] | ||
+ | </td> | ||
+ | <td> | ||
+ | [[Image:TutImgIso_03.png|center|220px<!--rightpanel--->]] | ||
+ | </td> | ||
+ | </tr> | ||
+ | <tr> | ||
+ | <td> | ||
+ | <p class="caption">function { x }</p> | ||
+ | </td> | ||
+ | <td> | ||
+ | <p class="caption">function { x } with threshold 1</p> | ||
+ | </td> | ||
+ | <td> | ||
+ | <p class="caption">function { x } with open</p> | ||
+ | </td> | ||
+ | </tr> | ||
+ | </table> | ||
− | <p>to | + | <p class="Hint"><strong>Hint:</strong> The checkered ground plane is scaled to one unit squares.</p> |
− | <p> | + | <p>For the last series of images in this section, let's try something different. These next two code examples were used to show the results of changing the user defined function to <code>function { x+y }</code> and <code>function { x+y+z }</code> respectively. They describe planes going through the origin, the function just describes the normal vector of the plane.</p> |
− | < | + | <pre> |
+ | isosurface { | ||
+ | function { x+y } | ||
+ | max_gradient 4 | ||
+ | contained_by { box { -2, 2 } } | ||
+ | } | ||
+ | </pre> | ||
− | <p> | + | <p class="Note"><strong>Note:</strong> To properly render these examples <code>max_gradient 4</code> was added to the isosurface definition, and will be explained later.</p> |
− | |||
− | < | + | <pre> |
+ | isosurface { | ||
+ | function { x+y+z } | ||
+ | max_gradient 4 | ||
+ | contained_by { box { -2, 2 } } | ||
+ | } | ||
+ | </pre> | ||
− | <p> | + | <table class="centered" width="460px" cellpadding="0" cellspacing="10"> |
− | + | <tr> | |
− | + | <td> | |
+ | [[Image:TutImgIso_04.png|center|220px<!--leftpanel--->]] | ||
+ | </td> | ||
+ | <td> | ||
+ | [[Image:TutImgIso_05.png|center|220px<!--rightpanel--->]] | ||
+ | </td> | ||
+ | </tr> | ||
+ | <tr> | ||
+ | <td> | ||
+ | <p class="caption">plane function { x+y }</p> | ||
+ | </td> | ||
+ | <td> | ||
+ | <p class="caption">plane function { x+y+z }</p> | ||
+ | </td> | ||
+ | </tr> | ||
+ | </table> | ||
− | + | <p class="Note"><strong>Note:</strong> When appropriate, to better visualize the difference between the isosurface and the container object, the images in this tutorial have been color coded.</p> | |
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | <p class="Note"><strong>Note:</strong> | ||
− | |||
− | |||
− | |||
− | |||
=====Several surfaces===== | =====Several surfaces===== | ||
− | <p> | + | <p>Now that you're starting to become familiar with <code>isosurface</code> syntax, there really isn't any need to show a code example for each and every image. You can always look back at the earlier examples when needed. The image captions will most often contain additional keyword hints when appropriate.</p> |
− | + | <p class="Note"><strong>Note:</strong> The user defined function portion will <em>always</em> use this color coded format: <code>function { x+y+z }</code></p> | |
− | |||
− | <p>< | ||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | <p><code> | + | <p>For the first image on the left, these two functions lead to identical results: <code>function { abs(x)-1 }</code> and <code>function { sqrt(x*x)-1 }</code> because both of these formulas have the same solution where the function value is 0, specifically <code>x=-1</code> and <code>x=1</code> in this example.</p> |
+ | <p>You can easily mix any of these elements in different combinations, but the results always produce planar surfaces. The last two images in this series used <code>function { abs(x)-1+y }</code> and <code>function { abs(x)+abs(y)+abs(z)-2 }</code> respectively.</p> | ||
− | < | + | <table class="centered" width="700px" cellpadding="0" cellspacing="10"> |
+ | <tr> | ||
+ | <td> | ||
+ | [[Image:TutImgIso_06.png|center|220px<!--leftpanel--->]] | ||
+ | </td> | ||
+ | <td> | ||
+ | [[Image:TutImgIso_07.png|center|220px<!--centerpanel--->]] | ||
+ | </td> | ||
+ | <td> | ||
+ | [[Image:TutImgIso_08.png|center|220px<!--rightpanel--->]] | ||
+ | </td> | ||
+ | </tr> | ||
+ | <tr> | ||
+ | <td> | ||
+ | <p class="caption">identical results with open</p> | ||
+ | </td> | ||
+ | <td> | ||
+ | <p class="caption">linear functions x & y axis</p> | ||
+ | </td> | ||
+ | <td> | ||
+ | <p class="caption">linear functions x, y & z axis</p> | ||
+ | </td> | ||
+ | </tr> | ||
+ | </table> | ||
=====Non-linear functions===== | =====Non-linear functions===== | ||
− | <p>Curved surfaces of many different kinds can be achieved with non-linear | + | <table class="centered" width="570px" cellpadding="0" cellspacing="10"> |
− | functions.< | + | <tr> |
− | + | <td> | |
− | + | [[Image:TutImgIso_09.png|center|220px<!--leftpanel--->]] | |
− | + | </td> | |
− | < | + | <td> |
− | + | <p class="tabletext">Curved surfaces of many different kinds can be achieved with non-linear | |
− | < | + | functions. A square function creates the parabolic shape:<br><code>function { pow(x,2)+y }</code></p> |
− | + | </td> | |
− | <p> | + | </tr> |
− | + | <tr> | |
− | < | + | <td> |
− | + | <p class="caption">a parabolic shape</p> | |
− | + | </td> | |
− | + | <td></td> | |
− | + | </tr> | |
− | + | </table> | |
− | <p> | + | <table class="centered" width="570px" cellpadding="0" cellspacing="10"> |
− | + | <tr> | |
+ | <td> | ||
+ | <p class="tabletext">If you describe a circle in 2 dimensions with a constant in the 3rd dimension you get a cylinder:<br><code> function { sqrt(pow(x,2)+pow(z,2))-1 }</code></p> | ||
+ | </td> | ||
+ | <td> | ||
+ | [[Image:TutImgIso_10.png|center|220px<!--rightpanel--->]] | ||
+ | </td> | ||
+ | </tr> | ||
+ | <tr> | ||
+ | <td></td> | ||
+ | <td> | ||
+ | <p class="caption">the cylinder shape</p> | ||
+ | </td> | ||
+ | </tr> | ||
+ | </table> | ||
− | <p><code> | + | <table class="centered" width="570px" cellpadding="0" cellspacing="10"> |
− | <p> | + | <tr> |
+ | <td> | ||
+ | [[Image:TutImgIso_11.png|center|220px<!--leftpanel--->]] | ||
+ | </td> | ||
+ | <td> | ||
+ | <p class="tabletext">It's easy to change a cylinder into a cone, we just need | ||
+ | to add a linear component in y-direction:<br><code>function { sqrt(pow(x,2)+pow(z,2))+y }</code></p> | ||
+ | </td> | ||
+ | </tr> | ||
+ | <tr> | ||
+ | <td> | ||
+ | <p class="caption">the cone shape</p> | ||
+ | </td> | ||
+ | <td></td> | ||
+ | <td></td> | ||
+ | </tr> | ||
+ | </table> | ||
− | <p> | + | <table class="centered" width="570px" cellpadding="0" cellspacing="10"> |
− | + | <tr> | |
− | < | + | <td> |
− | < | + | <p class="tabletext">No worries, creating a sphere is easy too. In this example <code>2</code> specifies the radius:<br><code> function { sqrt(pow(x,2)+pow(y,2)+pow(z,2))-2 }</code></p> |
− | + | </td> | |
− | < | + | <td> |
+ | [[Image:TutImgIso_12.png|center|220px<!--rightpanel--->]] | ||
+ | </td> | ||
+ | </tr> | ||
+ | <tr> | ||
+ | <td></td> | ||
+ | <td> | ||
+ | <p class="caption">the sphere shape</p> | ||
+ | </td> | ||
+ | </tr> | ||
+ | </table> | ||
=====Specifying functions===== | =====Specifying functions===== | ||
− | <p> | + | <p>Until now, we have seen, the functions used to define the isosurface were literally written in the <code>function {...}</code> block:</p> |
+ | <pre> | ||
+ | #declare Threshold = 1; | ||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
isosurface { | isosurface { | ||
function {pow(x,2) + pow(y,2) + pow(z,2)} | function {pow(x,2) + pow(y,2) + pow(z,2)} | ||
Line 139: | Line 233: | ||
</pre> | </pre> | ||
− | <p> | + | <p>Let's expand on that concept, and add some flexibility. Remember that user defined functions (like equations), all float expressions and operators which are legal in POV-Ray can be used, and that functions should be declared first, and then used in the isosurface. See the section <!--<linkto "User-Defined Functions">User-Defined Functions</linkto>--->[[Reference:Function|user defined function]] for more information.</p> |
− | + | ||
+ | <p>This next example takes the above equation, and rewrites it as a user defined function. By default a function that takes three parameters (x,y,z) does not require you to explicitly specify the parameter names when declaring it, however when <em>using</em> the identifier, the parameters <em>must</em> be specified.</p> | ||
+ | |||
<pre> | <pre> | ||
− | #declare Sphere = function {pow(x,2) + pow(y,2) + pow(z,2)} | + | #declare Threshold = 1; |
+ | |||
+ | #declare Sphere = function {pow(x,2) + pow(y,2) + pow(z,2)}; | ||
+ | |||
isosurface { | isosurface { | ||
function { Sphere(x,y,z) } | function { Sphere(x,y,z) } | ||
Line 150: | Line 249: | ||
</pre> | </pre> | ||
− | <p> | + | <p>However, if you need more or less than three parameters when declaring a function, you will also have to explicitly specify the parameter names.</p> |
− | |||
− | |||
− | |||
− | to explicitly specify the parameter names.</p> | ||
<pre> | <pre> | ||
− | #declare Sphere = function(x,y,z,Radius) { | + | #declare Sphere = function (x,y,z,Radius) {pow(x,2) + pow(y,2) + pow(z,2) - pow(Radius,2)}; |
− | + | ||
− | |||
isosurface { | isosurface { | ||
function { Sphere(x,y,z,1) } | function { Sphere(x,y,z,1) } | ||
Line 166: | Line 260: | ||
=====Internal functions===== | =====Internal functions===== | ||
− | <p>There are a lot of internal functions available in POV-Ray. For example | + | <p>There are a lot of internal functions available in POV-Ray. For example a sphere could also be generated with <code>function { f_sphere(x, y, z, 2) }</code>, for these and other functions, see the <code>functions.inc</code> include file. Most of them are more complicated and it is usually faster to use them instead of a hand coded equivalent.</p> |
− | a sphere could also be generated with <code>function { f_sphere(x, y, z, 2) }</code> | + | <p>See the <!--<linkto "internal functions">complete list</linkto>--->[[Reference:Functions.inc|complete list]] for details.</p> |
− | |||
− | Most of them are more complicated and it is usually faster to use them instead of a | ||
− | hand coded equivalent. See the | ||
− | <!--<linkto "internal functions">complete list</linkto>--->[[ | ||
<p>The following makes a torus just like POV-Ray's torus object:</p> | <p>The following makes a torus just like POV-Ray's torus object:</p> | ||
Line 184: | Line 274: | ||
</pre> | </pre> | ||
− | < | + | <table class="centered" width="570px" cellpadding="0" cellspacing="10"> |
− | + | <tr> | |
− | <p>The 4th and 5th | + | <td> |
− | just like the corresponding values in the <code>torus{}</code> object.</p> | + | [[Image:TutImgIso_13.png|center|220px<!--leftpanel--->]] |
− | + | </td> | |
− | <p>The parameters x, y and z are required, because it is a declared function. | + | <td> |
− | You can also declare functions yourself like it is explained in the | + | <p class="tabletext">The 4th and 5th parameters are the major and minor radius, just like the corresponding values in the <code>torus{}</code> object.</p> |
− | <!--<linkto "Declaring User-Defined Float Functions">reference section</linkto>--->[[ | + | <p class="tabletext">The parameters x, y and z are required, because it is a declared function. You can also declare functions yourself like it is explained in the <!--<linkto "Declaring User-Defined Float Functions">reference section</linkto>--->[[Reference:Function#Declaring User-Defined Float Functions|reference section]].</p> |
+ | </td> | ||
+ | </tr> | ||
+ | <tr> | ||
+ | <td> | ||
+ | <p class="caption">the torus function</p> | ||
+ | </td> | ||
+ | <td></td> | ||
+ | </tr> | ||
+ | </table> | ||
=====Combining isosurface functions===== | =====Combining isosurface functions===== | ||
− | <p>We can also simulate some Constructive Solid Geometry with isosurface functions. If | + | <p>We can also simulate some Constructive Solid Geometry with isosurface functions. If you do not know about CSG we suggest you have a look at <em>[[Documentation:Tutorial Section 2#What is CSG?|What is CSG?]]</em> or the corresponding part of the [[Reference:Constructive Solid Geometry|reference section]] first.</p> |
− | you do not know about CSG we suggest you have a look at | ||
− | < | ||
− | <p> | + | <p>For this next group of images, consider the two functions for a cylinder and a rotated box:</p> |
<pre> | <pre> | ||
Line 205: | Line 302: | ||
</pre> | </pre> | ||
− | < | + | <ol> |
− | + | <li>If we combine them the following way, we get a <em>merge</em>:<br> | |
− | < | + | <code>function { min(fn_A(x, y, z), fn_B(x, y, z)) }</code></li> |
− | + | <li>An <em>intersection</em> can be obtained by using <code>max()</code> instead of <code>min()</code>:<br> | |
− | < | + | <code>function { max(fn_A(x, y, z), fn_B(x, y, z)) }</code> |
− | + | </li> | |
− | + | <li>A <em>difference</em> is possible, by adding a minus (-) before the second function:<br> | |
− | < | + | <code>function { max(fn_A(x, y, z), -fn_B(x, y, z)) }</code> |
− | <code>max()</code> instead of <code>min()</code>:< | + | </li> |
− | + | </ol> | |
− | |||
− | |||
− | < | ||
− | |||
− | < | ||
− | |||
− | <p>< | + | <table class="centered" width="700px" cellpadding="0" cellspacing="10"> |
+ | <tr> | ||
+ | <td> | ||
+ | [[Image:TutImgIso_14.png|center|220px<!--leftpanel--->]] | ||
+ | </td> | ||
+ | <td> | ||
+ | [[Image:TutImgIso_15.png|center|220px<!--centerpanel--->]] | ||
+ | </td> | ||
+ | <td> | ||
+ | [[Image:TutImgIso_16.png|center|220px<!--rightpanel--->]] | ||
+ | </td> | ||
+ | </tr> | ||
+ | <tr> | ||
+ | <td> | ||
+ | <p class="caption">merge example</p> | ||
+ | </td> | ||
+ | <td> | ||
+ | <p class="caption">intersection example</p> | ||
+ | </td> | ||
+ | <td> | ||
+ | <p class="caption">difference example</p> | ||
+ | </td> | ||
+ | </tr> | ||
+ | </table> | ||
− | + | <p>Apart from basic CSG you can also obtain smooth transits between the different surfaces, for instance the <!--<linkto "blob object">blob object</linkto>--->[[Documentation:Tutorial Section 3.1#Blob Object|blob object]]:</p> | |
− | |||
− | <p>Apart from basic CSG you can also obtain smooth transits between the | ||
− | different surfaces | ||
<pre> | <pre> | ||
− | #declare | + | #declare Blob_Threshold=0.01; |
isosurface { | isosurface { | ||
function { | function { | ||
− | (1+ | + | (1+Blob_Threshold) |
− | -pow( | + | -pow(Blob_Threshold, fn_A(x,y,z)) |
− | -pow( | + | -pow(Blob_Threshold, fn_B(x,y,z)) |
} | } | ||
max_gradient 4 | max_gradient 4 | ||
Line 243: | Line 354: | ||
</pre> | </pre> | ||
− | < | + | <table class="centered" width="570px" cellpadding="0" cellspacing="10"> |
− | + | <tr> | |
− | <p>The <code> | + | <td> |
− | the transit between the shapes. | + | [[Image:TutImgIso_17.png|center|220px<!--leftpanel--->]] |
− | + | </td> | |
+ | <td> | ||
+ | <p class="tabletext">The <code>Blob_Threshold</code> value influences the smoothness of | ||
+ | the transit between the shapes. A lower value leads to sharper edges, and it's function looks like:</p> | ||
<pre> | <pre> | ||
− | function{fn_A(x,y,z) + pow( | + | function{fn_A(x,y,z) + pow(Blob_Threshold,(fn_B(x,y,z) + Strength))} |
</pre> | </pre> | ||
+ | </td> | ||
+ | </tr> | ||
+ | <tr> | ||
+ | <td> | ||
+ | <p class="caption">smooth transitions using blob</p> | ||
+ | </td> | ||
+ | </tr> | ||
+ | </table> | ||
=====Noise and pigment functions===== | =====Noise and pigment functions===== | ||
− | <p>Some of the <!--<linkto "internal functions">internal functions</linkto>--->[[ | + | <p>Some of the <!--<linkto "internal functions">internal functions</linkto>--->[[Reference:Function#Internal Pre-Defined Functions|internal functions]] have a random or noise-like structure</p> |
− | have a random or noise-like structure</p> | ||
− | <p>Together with the pigment functions they are one of the most powerful tools | + | <p>Together with the pigment functions they are one of the most powerful tools for designing isosurfaces. We can add real surface displacement to the objects rather than only normal perturbation known from the <!--<linkto "normal">normal</linkto>--->[[Reference:Normal|normal]] statement.</p> |
− | for designing isosurfaces. | ||
− | rather than only normal perturbation known from the | ||
− | <!--<linkto "normal">normal | ||
<p>The relevant internal functions are:</p> | <p>The relevant internal functions are:</p> | ||
Line 265: | Line 383: | ||
<ul> | <ul> | ||
<li><code>f_noise3d(x,y,z)</code><br> | <li><code>f_noise3d(x,y,z)</code><br> | ||
− | uses the | + | uses the [[Reference:Pattern Modifiers#Noise Generators|noise generator]] specified in <code>global_settings</code> and generates structures like the bozo pattern.</li> |
− | <code>global_settings | ||
− | |||
<li><code>f_noise_generator(x, y, z, noise_generator)</code><br> | <li><code>f_noise_generator(x, y, z, noise_generator)</code><br> | ||
generates noise with a specified noise generator.</li> | generates noise with a specified noise generator.</li> | ||
− | |||
− | |||
<li><code>f_ridged_mf(x, y, z, H, Lacunarity, Octaves, Offset, Gain, noise_generator)</code><br> | <li><code>f_ridged_mf(x, y, z, H, Lacunarity, Octaves, Offset, Gain, noise_generator)</code><br> | ||
generates a ridged multifractal pattern.</li> | generates a ridged multifractal pattern.</li> | ||
− | |||
<li><code>f_ridge(x, y, z, Lambda, Octaves, Omega, Offset, Ridge, noise_generator)</code><br> | <li><code>f_ridge(x, y, z, Lambda, Octaves, Omega, Offset, Ridge, noise_generator)</code><br> | ||
generates another noise with ridges.</li> | generates another noise with ridges.</li> | ||
− | |||
<li><code>f_hetero_mf(x, y, z, H, Lacunarity, Octaves, Offset, T, noise_generator)</code><br> | <li><code>f_hetero_mf(x, y, z, H, Lacunarity, Octaves, Offset, T, noise_generator)</code><br> | ||
generates heterogenic multifractal noise.</li> | generates heterogenic multifractal noise.</li> | ||
</ul> | </ul> | ||
− | < | + | <table class="centered" width="570px" cellpadding="0" cellspacing="10"> |
− | + | <tr> | |
− | + | <td> | |
− | + | <p class="tabletext">Using this simple noise3d function results in the image on the right. The value <code>-0.5</code> matches the default <code>threshold</code> value of zero. The <code>f_noise3d</code> function returns values between 0 and 1:</p> | |
− | < | + | <p class="tabletext"><code>function { f_noise3d(x,y,z)-0.5 }</code></p> |
− | + | </td> | |
− | <p class=" | + | <td> |
− | threshold value of | + | [[Image:TutImgIso_18.png|center|220px<!--rightpanel--->]] |
− | 0 and 1 | + | </td> |
− | + | </tr> | |
− | + | <tr> | |
− | + | <td></td> | |
− | + | <td> | |
− | + | <p class="caption">simple noise3d function</p> | |
− | <p><code> | + | </td> |
− | + | </tr> | |
− | < | + | </table> |
− | |||
− | < | ||
− | |||
− | |||
− | < | ||
− | |||
− | <p> | ||
− | |||
− | < | ||
− | |||
− | |||
− | |||
− | |||
− | |||
− | < | + | <table class="centered" width="570px" cellpadding="0" cellspacing="10"> |
+ | <tr> | ||
+ | <td> | ||
+ | <p class="tabletext">In these next two images the noise function was added to a plane function. The x-parameter was set to 0 so the noise function is constant in x-direction. This way we achieve the typical heightfield structure.</p> | ||
+ | </td> | ||
+ | </tr> | ||
+ | </table> | ||
− | < | + | <table class="centered" width="570px" cellpadding="0" cellspacing="10"> |
+ | <tr> | ||
+ | <td> | ||
+ | [[Image:TutImgIso_19.png|center|220px<!--leftpanel--->]] | ||
+ | </td> | ||
+ | <td> | ||
+ | <p class="tabletext">With this and the other functions you can generate objects similar to heightfields, having the advantage that a high resolution can be achieved without high memory requirements:</p> | ||
+ | <p class="tabletext"><code>function { x + f_noise3d(0,y,z) }</code></p> | ||
+ | </td> | ||
+ | </tr> | ||
+ | <tr> | ||
+ | <td> | ||
+ | <p class="caption">a noise3d heightfield</p> | ||
+ | </td> | ||
+ | <td></td> | ||
+ | </tr> | ||
+ | </table> | ||
− | <p> | + | <table class="centered" width="570px" cellpadding="0" cellspacing="10"> |
− | + | <tr> | |
− | + | <td> | |
+ | <p class="tabletext">The noise function can of course also be subtracted which results in an <em>inverted</em> version:</p> | ||
+ | <p class="tabletext"><code>function { x - f_noise3d(0,y,z) }</code></p> | ||
+ | </td> | ||
+ | <td> | ||
+ | [[Image:TutImgIso_20.png|center|220px<!--rightpanel--->]] | ||
+ | </td> | ||
+ | </tr> | ||
+ | <tr> | ||
+ | <td></td> | ||
+ | <td> | ||
+ | <p class="caption">a noise3d heightfield - inverted</p> | ||
+ | </td> | ||
+ | </tr> | ||
+ | </table> | ||
− | <p><code> | + | <table class="centered" width="570px" cellpadding="0" cellspacing="10"> |
+ | <tr> | ||
+ | <td> | ||
+ | [[Image:TutImgIso_21.png|center|220px<!--leftpanel--->]] | ||
+ | </td> | ||
+ | <td> | ||
+ | <p class="tabletext">Of course we can also add noise to any other function. If the noise function is very strong this can result in several separated surfaces.</p> | ||
+ | <p class="tabletext"><code>function { f_sphere(x,y,z,1.2) - f_noise3d(x,y,z) }</code></p> | ||
+ | </td> | ||
+ | </tr> | ||
+ | <tr> | ||
+ | <td> | ||
+ | <p class="caption">noise3d on a sphere</p> | ||
+ | </td> | ||
+ | <td> | ||
+ | </td> | ||
+ | </tr> | ||
+ | </table> | ||
− | <p>[[Image:TutImgIso_22.png| | + | <table class="centered" width="570px" cellpadding="0" cellspacing="10"> |
+ | <tr> | ||
+ | <td> | ||
+ | <p class="tabletext">This is a noise function applied to a sphere surface, we can influence the intensity of the noise by multiplying it with a factor and change the scale by multiplying the coordinate parameters:</p> | ||
+ | <p class="tabletext"><code>function {</code><br><code> f_sphere(x,y,z,1.6) -</code><br><code> f_noise3d(x*5,y*5,z* ) * 0.5<br> }</code></p> | ||
+ | </td> | ||
+ | <td> | ||
+ | [[Image:TutImgIso_22.png|center|220px<!--rightpanel--->]] | ||
+ | </td> | ||
+ | </tr> | ||
+ | <tr> | ||
+ | <td></td> | ||
+ | <td> | ||
+ | <p class="caption">noise3d on a sphere - scaled</p> | ||
+ | </td> | ||
+ | </tr> | ||
+ | </table> | ||
<p>As alternative to noise functions we can also use any pigment in a function:</p> | <p>As alternative to noise functions we can also use any pigment in a function:</p> | ||
Line 340: | Line 507: | ||
</pre> | </pre> | ||
− | <p>This | + | <p>This is a vector function, it returns a color vector for use in isosurface functions. They <em>must</em> be pre-declared first. When using the identifier, you have to specify which component of the color vector should be used.</p> |
− | + | <p>To do this, the dot notation is used. Refer to the above example: <code>fn_Pigm(x,y,z).red</code></p> | |
− | |||
− | |||
− | </p | ||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | < | + | <p>A color vector has five components, their supported dot types to access these components are:</p> |
− | |||
− | </ | ||
− | </ | + | <ol> |
+ | <li><code>fn_Pigm( ).x</code> | <code>fn_Pigm( ).u</code> | <code>fn_Pigm( ).red</code><br> | ||
+ | to get the red value of the color vector </li> | ||
+ | <li><code>fn_Pigm( ).y</code> | <code>fn_Pigm( ).v</code> | <code>fn_Pigm( ).green</code><br> | ||
+ | to get the green value of the color vector</li> | ||
+ | <li><code>fn_Pigm( ).z</code> | <code>fn_Pigm( ).blue</code><br> | ||
+ | to get the blue value of the color vector</li> | ||
+ | <li><code>fn_Pigm( ).filter</code> | <code>fn_Pigm( ).f</code><br> | ||
+ | to get the filter value of the color vector</li> | ||
+ | <li><code>fn_Pigm( ).transmit</code> | <code>fn_Pigm( ).t</code><br> | ||
+ | to get the transmit value of the color vector</li> | ||
+ | </ol> | ||
− | </ | + | <p>And two special purpose operators, their supported dot types to access these operators are:</p> |
+ | <p class="Note"><strong>Note:</strong> The <code>.hf</code> operator is experimental and will generate a warning.</p> | ||
− | <li> | + | <ol> |
− | + | <li><code>fn_Pigm( ).gray</code> to get the gray value of the color vector<br> | |
− | < | + | <em>gray value</em> = Red*29.7% + Green*58.9% + Blue*11.4% </li> |
+ | <li><code>fn_Pigm( ).hf</code> to get the height_field value of the color vector<br> | ||
+ | <em>hf value</em> = (Red + Green/255)*0.996093</li> | ||
+ | </ol> | ||
− | < | + | <table class="centered" width="570px" cellpadding="0" cellspacing="10"> |
− | + | <tr> | |
− | < | + | <td> |
− | + | [[Image:TutImgIso_23.png|center|220px<!--leftpanel--->]] | |
− | < | + | </td> |
− | + | <td> | |
− | + | <p class="tabletext">There are quite a lot of things possible with pigment functions. However, it should be noted that, some functions can cause longer render times:</p> | |
− | + | <p class="tabletext"><code>function {<br> f_sphere(x, y, z, 1.6) -<br> fn_Pigm(x/2, y/2, z/2).gray*0.5<br> }</code></p> | |
− | + | </td> | |
− | + | </tr> | |
− | + | <tr> | |
− | + | <td> | |
− | + | <p class="caption">noise using a pigment function</p> | |
− | + | </td> | |
− | </ | + | <td></td> |
− | + | </tr> | |
− | < | + | </table> |
− | |||
− | < | ||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | </ | ||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | <p><code> | ||
− | |||
− | <p> | ||
− | |||
− | < | ||
− | |||
=====Conditional directives and loops===== | =====Conditional directives and loops===== | ||
Line 493: | Line 593: | ||
=====Transformations on functions===== | =====Transformations on functions===== | ||
− | <p> | + | <p>Transforming an isosurface object is done like transforming any POV-Ray object. Simply use the object modifiers, scale, translate, and rotate. However, when you want to transform functions within the <code>contained_by</code> object, you have to substitute parameters in the functions.</p> |
− | Transforming an isosurface object is done like transforming any POV-Ray object. Simply use the object modifiers | ||
− | |||
− | |||
− | |||
− | However, when you want to transform functions within the contained_by object, you have to substitute parameters in | ||
− | the functions. | ||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | </p | ||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | <p> | + | <p>The results <em>seem</em> inverted to what you would normally expect, here's why:</p> |
− | < | ||
− | |||
− | |||
− | </p> | ||
− | < | + | <p>Remember the sphere function we created earlier in this tutorial: <code>Sphere(x,y,z)</code></p> |
+ | <p>We know it sits at the origin because <code>x=0</code>. If we want to translate it 2 units to the right to <code>x=2</code> we need to write the second equation in the same form: <code>x-2=0</code>. Now that both equations equal zero, we can replace the parameter <code>x</code> with <code>x-2</code>, call our function as: <code>Sphere(x-2,y,z)</code> and it's translated two units to the right.</p> | ||
− | <p> | + | <p>Let's look at how to scale our test sphere by <code>0.5</code> in the <em>y direction</em>. Given the default value of <code>y=1</code> <em>one unit</em> we'd want <code>y=0.5</code>. To do this we need to have the equation in the same form as the first one, so we'll multiply both sides by two: <code>y*2 = 0.5*2</code> which gives <code>y*2=1</code>.</p> |
− | < | + | <p>Now we can replace the <code>y</code> parameter in our sphere: <code>Sphere(x,y*2,z)</code>. This scales the <em>y-size</em> of the sphere by half.</p> |
− | |||
− | |||
− | </p> | ||
− | < | + | <p>Here is an overview of some useful substitutions, we'll be using a pseudo-object designated as <code>P(x,y,z)</code> in the following examples:</p> |
+ | <p><strong>Scale:</strong></p> | ||
+ | <p> To scale <code>x</code> replace <code>x</code> with <code>x/scale</code>:<br> <code>P(x/2,y,z)</code></p> | ||
− | <p><strong> | + | <p><strong>Scale Infinitely:</strong></p> |
− | + | <p> To scale <code>y</code> infinitely replace <code>y</code> with <code>0</code>:<br> <code>P(x,0,z)</code></p> | |
− | |||
− | |||
− | </p> | ||
− | |||
− | <p> | ||
− | |||
− | y | ||
− | |||
− | </p> | ||
− | <p> | + | <p><strong>Translate:</strong></p> |
− | + | <p> To translate <code>z</code> replace <code>z</code> with <code>z - translation</code>:<br> <code>P(x,y,z-3)</code></p> | |
− | z | ||
− | z | ||
− | </p> | ||
− | <p> | + | <p><strong>Shear:</strong></p> |
− | + | <p> To shear in <em>xy-plane</em> replace <code>x</code> with <code>x + y*tan(radians(Angle))</code>:<br> <code>P(x+y*tan(radians(Angle)),y,z)</code></p> | |
− | y* | ||
− | y | ||
− | </p> | ||
− | < | + | <p><strong>Rotate:</strong></p> |
− | + | <p class="Note"><strong>Note:</strong> These rotation substitutions work like normal POV-rotations, they already compensate for the inverse behavior.</p> | |
− | |||
− | |||
− | |||
+ | <p>To rotate around the X-axis:</p> | ||
+ | <p> replace <code>y</code> with <code>z*sin(radians(Angle)) + y*cos(radians(Angle))</code></p> | ||
+ | <p> replace <code>z</code> with <code>z*cos(radians(Angle)) - y*sin(radians(Angle))</code></p> | ||
− | <p> | + | <p>To rotate around the Y-axis:</p> |
− | + | <p> replace <code>x</code> with <code>x*cos(radians(Angle)) - z*sin(radians(Angle))</code></p> | |
− | + | <p> replace <code>z</code> with <code>x*sin(radians(Angle)) + z*cos(radians(Angle))</code></p> | |
− | with | ||
− | </p> | ||
− | <p> | + | <p>To rotate around the Z-axis:</p> |
− | + | <p> replace <code>x</code> with <code>x*cos(radians(Angle)) + y*sin(radians(Angle))</code></p> | |
− | with | + | <p> replace <code>y</code> with <code>-x*sin(radians(Angle)) + y*cos(radians(Angle)) </code></p> |
− | </p> | ||
− | <p> | + | <p><strong>Flip:</strong></p> |
− | flip X - | + | <p>To flip X - Y:</p> |
− | with | + | <p> replace <code>x</code> with <code>y</code> <em>AND</em> replace <code>y</code> with <code>-x</code></p> |
− | </p> | ||
− | < | + | <p>To flip Y - Z:</p> |
+ | <p> replace <code>y</code> with <code>z</code> <em>AND</em> replace <code>z</code> with <code>-y</code></p> | ||
+ | <p>To flip X - Z:</p> | ||
+ | <p> replace <code>x</code> with <code>-z</code> <em>AND</em> replace <code>z</code> with <code>x</code></p> | ||
− | <p> | + | <p><strong>Twist:</strong></p> |
− | <strong>Twist</strong>< | + | <p>To twist N turns/unit around the <code>x</code> axis:</p> |
− | + y*cos(x*2*pi*N)</code>& | + | <p> replace <code>y</code> with <code>z*sin(x*2*pi*N) + y*cos(x*2*pi*N)</code></p> |
− | y*sin(x*2*pi*N)</code> | + | <p> replace <code>z</code> with <code>z*cos(x*2*pi*N) - y*sin(x*2*pi*N)</code></p> |
− | </p> | ||
=====Improving Isosurface Speed===== | =====Improving Isosurface Speed===== | ||
− | <p>To optimize the approximation of the isosurface and to get maximum | + | <p>To optimize the approximation of the isosurface and to get maximum rendering speed it is important to adapt certain values:</p> |
− | rendering speed it is important to adapt certain values | ||
− | <p><code>accuracy</code></p> | + | <p><strong><code>accuracy</code>:</strong></p> |
− | <p>The accuracy value influences how accurate the surface geometry is calculated. | + | <p>The accuracy value influences how accurate the surface geometry is calculated. Lower values lead to a more precise, but slower result. The default value of <code>0.001</code> is fairly low. We used this value in all the previous samples, but often you can raise this quite a lot and thereby make things faster.</p> |
− | Lower values lead to a more precise, but slower result. | ||
− | The default value of <code>0.001</code> is fairly low. We used this value in all the | ||
− | previous samples, but often you can raise this quite a | ||
− | lot and thereby make things faster.</p> | ||
− | <p><code>max_gradient</code></p> | + | <p><strong><code>max_gradient</code>:</strong></p> |
− | <p>For finding the actual surface it is important for POV-Ray to know the | + | <p>For finding the actual surface it is important for POV-Ray to know the maximum gradient of the function, meaning how fast the function value changes. We can specify a value with the <code>max_gradient</code> keyword. Lower max_gradient values lead to faster rendering, but if the specified value is below the actual maximum gradient of the function, there can be holes or other artefact's in the surface.</p> |
− | maximum gradient of the function, meaning how fast the function value changes. | ||
− | We can specify a value with the <code>max_gradient</code> keyword. Lower | ||
− | max_gradient values lead to faster rendering, but if the specified value is | ||
− | below the actual maximum gradient of the function, there can be holes or | ||
− | other | ||
− | <p>For the same reason functions with infinite gradient should not be used. | + | <p>For the same reason functions with an infinite gradient should not be used. This applies for pigment functions with brick or checker patterns for example. You should also be careful when using <code>select()</code> in isosurface functions because of this.</p> |
− | This applies for pigment functions with brick or checker | ||
− | You should also be careful when using <code>select()</code> in isosurface | ||
− | functions because of this.</p> | ||
− | <p>If the real maximum gradient differs too much from the specified value | + | <p>If the real maximum gradient differs too much from the specified value POV-Ray issues a warning together with the found maximum gradient. It is usually sufficient to use this number for the <code>max_gradient</code> parameter to get fast and correct results.</p> |
− | POV-Ray | ||
− | It is usually sufficient to use this number for the <code>max_gradient</code> | ||
− | parameter to get fast and correct results.</p> | ||
− | <p>POV-Ray can also dynamically change the <code>max_gradient</code> when you | + | <p>POV-Ray can also dynamically change the <code>max_gradient</code> when you specify <code>evaluate</code> with 3 parameters in the isosurface definition. Concerning the details on this and other things see the <!--<linkto "evaluate">evaluate</linkto>--->[[Reference:Isosurface|evaluate]] keyword in the reference section.</p> |
− | specify <code>evaluate</code> with 3 parameters the isosurface definition. | ||
− | Concerning the details on this and other things see the | ||
− | <!--<linkto "evaluate">evaluate</linkto>--->[[ | ||
− | <p><code>contained_by</code></p> | + | <p><strong><code>contained_by</code>:</strong></p> |
− | <p> | + | <p>Make sure your <code>contained_by</code> object fits as tightly as possible. An oversized container can sky-rocket the render time. When the container has a lot of empty space around the actual isosurface, POV-Ray has to do a lot of superfluous sampling: especially with complex functions this can become very time consuming. On top of this, the <code>max_gradient</code> needed to get a proper surface will also increase rapidly, almost proportional to the oversize! You could use a transparent copy of the container (using exactly the same transformations) to check how it fits. Getting the <code>min_extent</code> and <code>max_extent</code> of the isosurface is not useful because it only gives the extent of the container and not of the actual isosurface.</p> |
− | Make sure your <code>contained_by</code> | ||
− | sky-rocket the render time. | ||
− | to do a lot of superfluous sampling: especially with complex functions this can become very time consuming. On top of | ||
− | this, the <code>max_gradient</code> needed to get a proper surface will also increase rapidly | ||
− | the oversize! | ||
− | check how it fits. Getting the <code>min_extent</code> and <code>max_extent</code> of the | ||
− | is not useful because it only gives the extent of the container and not of the actual isosurface. | ||
− | </p> | ||
====Poly Object==== | ====Poly Object==== | ||
− | + | {{#indexentry:poly, tutorial}} | |
− | <p>The polynomial object (and its | + | <p>The polynomial object (and its <em>shortcut</em> versions: <!--<linkto "cubic">cubic</linkto>---><code>[[Reference:Cubic|cubic]]</code>, <!--<linkto "quartic">quartic</linkto>---><code>[[Reference:Quartic|quartic]]</code> and <!--<linkto "quadric">quadric</linkto>---><code>[[Reference:Quadric|quadric]]</code>) |
− | + | of POV-Ray is one of the most complex and mathematical primitives of the program. One could think that it is seldom | |
− | of POV-Ray is one of the most complex and | + | used and more or less obsolete, but we have to remember that for example the torus primitive is just a shortcut for the equivalent <code>quartic</code>, which is just a shortcut for the equivalent <code>poly</code> object. Polys are, however, seldom used in scenes due to the fact that they are so difficult to define and it is far from trivial to get the desired shape with just a polynomial equation. It is mostly used by the most mathematically oriented POV-Ray users.</p> |
− | mathematical primitives of the program. One could think that it is seldom | ||
− | used and more or less obsolete, but we have to remember that for example | ||
− | the torus primitive is just a shortcut for the equivalent <code>quartic</code>, which | ||
− | is just a shortcut for the equivalent <code>poly</code> object. Polys are, however, seldom | ||
− | used in scenes due to the fact that they are so difficult to define and | ||
− | it is far from trivial to get the desired shape with just a polynomial | ||
− | equation. It is mostly used by the most mathematically oriented POV-Ray | ||
− | users.</p> | ||
<p>This tutorial explains the process of making a polynomial object | <p>This tutorial explains the process of making a polynomial object | ||
in POV-Ray.</p> | in POV-Ray.</p> | ||
− | <p class="Note"><strong>Note: </strong>Since version 3.5, POV-Ray includes the new <code>isosurface</code> object | + | <p class="Note"><strong>Note:</strong> Since version 3.5, POV-Ray includes the new <code>isosurface</code> object |
− | which makes the polynomial object more or less obsolete. The isosurface | + | which makes the polynomial object more or less obsolete. The isosurface is more versatile (you can specify any mathematical function, not |
− | is more versatile (you can specify any mathematical function, not | + | just polynomials) and easier to use. You can write the function as is, without needing to put values in a gigantic vector. Isosurfaces also often (although not always) render considerably faster than equivalent polys.</p> |
− | just polynomials) | ||
− | without needing to put values in a gigantic vector. Isosurfaces | ||
− | often render considerably faster than equivalent polys.</p> | ||
<p>However, the most mathematically oriented still like polys because | <p>However, the most mathematically oriented still like polys because | ||
Line 686: | Line 686: | ||
are more than good enough for most applications, though.</p> | are more than good enough for most applications, though.</p> | ||
− | <p class="Note"><strong>Note:</strong> | + | <p class="Note"><strong>Note:</strong> A maximum of 35th degree polynomial can be represented with the poly object. If a higher degree polynomial or other non-polynomial function has to be represented, then it is necessary to use the isosurface object.</p> |
− | the poly object. If a higher degree polynomial or other non-polynomial function has to be represented, | ||
− | then it is necessary to use the isosurface object.</p> | ||
=====Creating the polynomial function===== | =====Creating the polynomial function===== | ||
Line 694: | Line 692: | ||
You will need some (high-school level) mathematical knowledge for this.</p> | You will need some (high-school level) mathematical knowledge for this.</p> | ||
− | <p><strong>1)</strong> Let's start with an easy example: | + | <p><strong>1)</strong> Let's start with an easy example, a sphere:</p> |
<p>The sphere function is:</p> | <p>The sphere function is:</p> | ||
− | |||
− | < | + | <table class="centered" width="190px" cellpadding="0" cellspacing="10"> |
− | < | + | <tr> |
+ | <td> | ||
+ | [[Image:TutImgPolyfunc1.png|center|170px<!--centered--->]] | ||
+ | </td> | ||
+ | </tr> | ||
+ | <tr> | ||
+ | <td> | ||
+ | <p class="caption">sphere function</p> | ||
+ | </td> | ||
+ | </tr> | ||
+ | </table> | ||
+ | |||
+ | <p>Now we have to convert this to polynomial form, we will need a polynomial of the 2nd degree to represent this:</p> | ||
− | <p> | + | <table class="centered" width="205px" cellpadding="0" cellspacing="10"> |
+ | <tr> | ||
+ | <td> | ||
+ | [[Image:TutImgPolyfunc2.png|center|185px<!--centered--->]] | ||
+ | </td> | ||
+ | </tr> | ||
+ | <tr> | ||
+ | <td> | ||
+ | <p class="caption">sphere polynomial</p> | ||
+ | </td> | ||
+ | </tr> | ||
+ | </table> | ||
<p><strong>2)</strong> A more elaborated example:</p> | <p><strong>2)</strong> A more elaborated example:</p> | ||
<p>Let's take the function:</p> | <p>Let's take the function:</p> | ||
− | < | + | |
+ | <table class="centered" width="130px" cellpadding="0" cellspacing="10"> | ||
+ | <tr> | ||
+ | <td> | ||
+ | [[Image:TutImgPolyfunc3.png|center|110px<!--centered--->]] | ||
+ | </td> | ||
+ | </tr> | ||
+ | <tr> | ||
+ | <td> | ||
+ | <p class="caption">function</p> | ||
+ | </td> | ||
+ | </tr> | ||
+ | </table> | ||
<p>Converting this to polynomial form we get:</p> | <p>Converting this to polynomial form we get:</p> | ||
− | < | + | |
+ | <table class="centered" width="215px" cellpadding="0" cellspacing="10"> | ||
+ | <tr> | ||
+ | <td> | ||
+ | [[Image:TutImgPolyfunc4.png|center|195px<!--centered--->]] | ||
+ | </td> | ||
+ | </tr> | ||
+ | <tr> | ||
+ | <td> | ||
+ | <p class="caption">polynomial</p> | ||
+ | </td> | ||
+ | </tr> | ||
+ | </table> | ||
<p>Although the highest power is 4 we will need a 5th order polynomial to | <p>Although the highest power is 4 we will need a 5th order polynomial to | ||
Line 717: | Line 761: | ||
<p><strong>3)</strong> And since we talked about the torus, let's also take it as an example.</p> | <p><strong>3)</strong> And since we talked about the torus, let's also take it as an example.</p> | ||
+ | <p>A torus can be represented with the function:</p> | ||
+ | |||
+ | <table class="centered" width="295px" cellpadding="0" cellspacing="10"> | ||
+ | <tr> | ||
+ | <td> | ||
+ | [[Image:TutImgPolyfunc5.png|center|275px<!--centered--->]] | ||
+ | </td> | ||
+ | </tr> | ||
+ | <tr> | ||
+ | <td> | ||
+ | <p class="caption">torus function</p> | ||
+ | </td> | ||
+ | </tr> | ||
+ | </table> | ||
− | |||
− | |||
<p>where r<sub>1</sub> is the major radius and r<sub>2</sub> is the minor radius.</p> | <p>where r<sub>1</sub> is the major radius and r<sub>2</sub> is the minor radius.</p> | ||
<p>Now, this is tougher to convert to polynomial form, but finally we get:</p> | <p>Now, this is tougher to convert to polynomial form, but finally we get:</p> | ||
− | < | + | |
+ | <table class="centered" width="700px" cellpadding="0" cellspacing="10"> | ||
+ | <tr> | ||
+ | <td> | ||
+ | [[Image:TutImgPolyfunc6.png|center|680px<!--centered--->]] | ||
+ | </td> | ||
+ | </tr> | ||
+ | <tr> | ||
+ | <td> | ||
+ | <p class="caption">torus polynomial</p> | ||
+ | </td> | ||
+ | </tr> | ||
+ | </table> | ||
<p>A 4th order polynomial is enough to represent this.</p> | <p>A 4th order polynomial is enough to represent this.</p> | ||
− | <p class="Note"><strong>Note:</strong> | + | <p class="Note"><strong>Note:</strong> Not every function can be represented in polynomial form. Only functions that use addition (and substraction), multiplication (and division) and scalar powers (including rational powers, eg. the square root) can be represented. Also, the poly primitive supports only polynomials of the 35th degree at max.</p> |
− | functions that use addition (and substraction), multiplication (and | ||
− | division) and scalar powers (including rational powers, eg. the square root) | ||
− | can be represented. Also, the poly primitive supports only | ||
− | polynomials of the | ||
<p>Converting a function to polynomial form may be a very laborious task for | <p>Converting a function to polynomial form may be a very laborious task for | ||
Line 738: | Line 802: | ||
=====Writing the polynomial vector===== | =====Writing the polynomial vector===== | ||
<p>Now that we have the function in polynomial form, we have to write it | <p>Now that we have the function in polynomial form, we have to write it | ||
− | in POV-Ray syntax. The syntax is specified in the | + | in POV-Ray syntax. The syntax is specified in the sections on [[Reference:Polynomial|polynomial]] and [[Reference:Quadric|quadric]] of the reference section. There is also a table in this chapter which we will be using to make the polynomial vector. It is easier to have this table printed on paper.</p> |
− | |||
− | |||
− | table in this chapter which we will be using to make the polynomial vector. It is | ||
− | easier to have this table printed on paper.</p> | ||
<p class="Note"><strong>Note:</strong> It is also possible to make a little program with your favorite | <p class="Note"><strong>Note:</strong> It is also possible to make a little program with your favorite | ||
Line 748: | Line 808: | ||
function, but making a program like this is up to you.</p> | function, but making a program like this is up to you.</p> | ||
− | <p>< | + | <p><strong>1)</strong> Let's start with the easy one, ie. the sphere.</p> |
<p>Since the sphere can be represented with a polynomial of 2nd degree, we | <p>Since the sphere can be represented with a polynomial of 2nd degree, we | ||
− | look at the | + | look at the column titled <em>2nd</em> in the [[Reference:Polynomial|table]]. We see that it has 10 items, |
− | ie. we need a vector of size 10. Each item of the vector will be the factor | + | ie. we need a vector of size 10. Each item of the vector will be the factor of the term listed in the table.</p> |
− | of the term listed in the table.</p> | ||
<p>The polynomial was:</p> | <p>The polynomial was:</p> | ||
− | < | + | |
+ | <table class="centered" width="205px" cellpadding="0" cellspacing="10"> | ||
+ | <tr> | ||
+ | <td> | ||
+ | [[Image:TutImgPolyfunc2.png|center|185px<!--centered--->]] | ||
+ | </td> | ||
+ | </tr> | ||
+ | <tr> | ||
+ | <td> | ||
+ | <p class="caption">sphere polynomial function</p> | ||
+ | </td> | ||
+ | </tr> | ||
+ | </table> | ||
<p>Writing the poly in this way we get:</p> | <p>Writing the poly in this way we get:</p> | ||
Line 793: | Line 864: | ||
</pre> | </pre> | ||
+ | <table class="centered" width="340px" cellpadding="0" cellspacing="10"> | ||
+ | <tr> | ||
+ | <td> | ||
+ | [[Image:TutImgPolypic1.png|center|320px<!--centered--->]] | ||
+ | </td> | ||
+ | </tr> | ||
+ | <tr> | ||
+ | <td> | ||
+ | <p class="caption">sphere polynomial image</p> | ||
+ | </td> | ||
+ | </tr> | ||
+ | </table> | ||
− | <p> | + | <p></p> |
− | + | <p class="Note"><strong>Note:</strong> There is a shortcut for 2nd degree polynomials: The <code>[[Reference:Quadric|quadric]]</code> primitive. Using a shortcut version, whenever possible, can lead to faster | |
− | <p class="Note"><strong>Note:</strong> | + | renderings. We can write the sphere code described above in the following way:</p> |
− | <code | ||
− | primitive. Using a shortcut version, whenever possible, can lead to faster | ||
− | renderings. We can write the sphere code described above in the following | ||
− | way:</p> | ||
<pre> | <pre> | ||
Line 810: | Line 889: | ||
<p><strong>2)</strong> Now lets try the second one. We do it similarly, but this time we need | <p><strong>2)</strong> Now lets try the second one. We do it similarly, but this time we need | ||
− | to look at the | + | to look at the column titled <em>5th</em> in the table.</p> |
<p>The polynomial was:</p> | <p>The polynomial was:</p> | ||
− | < | + | |
+ | <table class="centered" width="215px" cellpadding="0" cellspacing="10"> | ||
+ | <tr> | ||
+ | <td> | ||
+ | [[Image:TutImgPolyfunc4.png|center|195px<!--centered--->]] | ||
+ | </td> | ||
+ | </tr> | ||
+ | <tr> | ||
+ | <td> | ||
+ | <p class="caption">5th order polynomial function</p> | ||
+ | </td> | ||
+ | </tr> | ||
+ | </table> | ||
<p>Writing the poly primitive we get:</p> | <p>Writing the poly primitive we get:</p> | ||
Line 859: | Line 950: | ||
</pre> | </pre> | ||
− | < | + | <table class="centered" width="340px" cellpadding="0" cellspacing="10"> |
− | + | <tr> | |
+ | <td> | ||
+ | [[Image:TutImgPolypic2.png|center|320px<!--centered--->]] | ||
+ | </td> | ||
+ | </tr> | ||
+ | <tr> | ||
+ | <td> | ||
+ | <p class="caption">5th order polynomial image</p> | ||
+ | </td> | ||
+ | </tr> | ||
+ | </table> | ||
<p><strong>3)</strong> And finally the torus:</p> | <p><strong>3)</strong> And finally the torus:</p> | ||
<p>The polynomial was:</p> | <p>The polynomial was:</p> | ||
− | < | + | |
+ | <table class="centered" width="700px" cellpadding="0" cellspacing="10"> | ||
+ | <tr> | ||
+ | <td> | ||
+ | [[Image:TutImgPolyfunc6.png|center|680px<!--centered--->]] | ||
+ | </td> | ||
+ | </tr> | ||
+ | <tr> | ||
+ | <td> | ||
+ | <p class="caption">torus polynomial function</p> | ||
+ | </td> | ||
+ | </tr> | ||
+ | </table> | ||
<p>And we get the proper 4th degree poly primitive:</p> | <p>And we get the proper 4th degree poly primitive:</p> | ||
Line 890: | Line 1,003: | ||
<p>When rendered we get:</p> | <p>When rendered we get:</p> | ||
− | < | + | <table class="centered" width="340px" cellpadding="0" cellspacing="10"> |
+ | <tr> | ||
+ | <td> | ||
+ | [[Image:TutImgPolypic3.png|center|320px<!--centered--->]] | ||
+ | </td> | ||
+ | </tr> | ||
+ | <tr> | ||
+ | <td> | ||
+ | <p class="caption">torus polynomial image</p> | ||
+ | </td> | ||
+ | </tr> | ||
+ | </table> | ||
− | <p>There is a shortcut for 4th order polynomials: The | + | <p>There is a shortcut for 4th order polynomials: The <code>[[Reference:Quartic|quartic]]</code> primitive. We can write the torus like this:</p> |
− | <code | ||
− | We can write the torus like this:</p> | ||
<pre> | <pre> | ||
Line 907: | Line 1,029: | ||
1,0,-2*(r1*r1+r2*r2),0,pow(r1,4)+pow(r2,4)-2*r1*r1*r2*r2> | 1,0,-2*(r1*r1+r2*r2),0,pow(r1,4)+pow(r2,4)-2*r1*r1*r2*r2> | ||
pigment { rgb <1,.7,.3> } finish { specular .5 } | pigment { rgb <1,.7,.3> } finish { specular .5 } | ||
+ | } | ||
+ | </pre> | ||
+ | |||
+ | =====Polynomial made easy===== | ||
+ | <p>Since consulting the table in the section [[Reference:Polynomial|Polynomial]] or writing a program to get the right poly vector can be a bit cumbersome, especially when the poly vector is not a write-once-only expression and that you want to get it back, so let's examine how those equations would be rewritten using the <em>simplified</em> syntax.</p> | ||
+ | <p>You should refer to the images in the previous section, as these examples produce <em>exactly</em> the same results.</p> | ||
+ | |||
+ | <p><strong>1)</strong> The sphere example can be rewritten as:</p> | ||
+ | <pre> | ||
+ | #declare Radius=1; | ||
+ | polynomial { 2, | ||
+ | xyz(2,0,0):1, | ||
+ | xyz(0,2,0):1, | ||
+ | xyz(0,0,2):1, | ||
+ | xyz(0,0,0):-Radius*Radius | ||
+ | } | ||
+ | </pre> | ||
+ | |||
+ | <p><strong>2)</strong> Let's now see the second one:</p> | ||
+ | <pre> | ||
+ | polynomial { 5, | ||
+ | xyz(2,0,1):1, | ||
+ | xyz(0,4,1):1, | ||
+ | xyz(1,2,0):-2 | ||
+ | } | ||
+ | </pre> | ||
+ | |||
+ | <p><strong>3)</strong> And finally the torus example:</p> | ||
+ | <pre> | ||
+ | polynomial { 4, | ||
+ | xyz(4,0,0):1, | ||
+ | xyz(2,2,0):2, | ||
+ | xyz(2,0,2):2, | ||
+ | xyz(2,0,0):-2*(r1*r1+r2*r2), | ||
+ | xyz(0,4,0):1, | ||
+ | xyz(0,2,2):2, | ||
+ | xyz(0,2,0):2*(r1*r1-r2*r2), | ||
+ | xyz(0,0,4):1, | ||
+ | xyz(0,0,2):-2*(r1*r1+r2*r2), | ||
+ | xyz(0,0,0):pow((r1*r1-r2*r2),2) | ||
} | } | ||
</pre> | </pre> | ||
Line 913: | Line 1,075: | ||
<table width=100% border=0 cellspacing=0 cellpadding=5> | <table width=100% border=0 cellspacing=0 cellpadding=5> | ||
<tr><td width=50% bgcolor=#EEEEEF> | <tr><td width=50% bgcolor=#EEEEEF> | ||
− | [[Documentation:Tutorial Section 3.1# | + | [[Documentation:Tutorial Section 3.1#Height Field Object|Height Field Object]]</td> |
<td width=50% bgcolor=#EEEEEF align=right> | <td width=50% bgcolor=#EEEEEF align=right> | ||
[[Documentation:Tutorial Section 3.3#Superquadric Ellipsoid Object|Superquadric Ellipsoid Object]]</td></tr> | [[Documentation:Tutorial Section 3.3#Superquadric Ellipsoid Object|Superquadric Ellipsoid Object]]</td></tr> |
Latest revision as of 14:20, 16 December 2016
This document is protected, so submissions, corrections and discussions should be held on this documents talk page. |
Isosurface Object
Isosurfaces are shapes described by mathematical functions.
In contrast to the other mathematically based shapes in POV-Ray, isosurfaces are approximated during rendering and therefore they are sometimes more difficult to handle. However, they offer many interesting possibilities, like real deformations and surface displacements
Some knowledge about mathematical functions and geometry is useful, but not necessarily required to work with isosurfaces.
Simple functions
Let's begin with something simple. In this first series of images, let's explore the user defined function shown as function { x }
that we see in the code example below. It produces the first image on the left, a simple box. The container, which is a requirement for the isosurface object, is represented by the box object and the contained_by
keyword in the isosurface definition.
isosurface { function { x } contained_by { box { -2, 2 } } }
You should have also noticed that in the image on the left, only half the box was produced, that's because the threshold
keyword was omitted, so the default value 0 was used to evaluate the x-coordinate.
In this next code example threshold 1
was added to produce the center image.
isosurface { function { x } threshold 1 contained_by { box { -2, 2 } } }
It is also possible to remove the visible surfaces of the container by adding the open
keyword to the isosurface definition.
For the final image on the right, the following code example was used. Notice that the omission of the threshold
keyword causes the x-coordinate to be again evaluated to zero.
isosurface { function { x } open contained_by { box { -2, 2 } } }
Hint: The checkered ground plane is scaled to one unit squares.
For the last series of images in this section, let's try something different. These next two code examples were used to show the results of changing the user defined function to function { x+y }
and function { x+y+z }
respectively. They describe planes going through the origin, the function just describes the normal vector of the plane.
isosurface { function { x+y } max_gradient 4 contained_by { box { -2, 2 } } }
Note: To properly render these examples max_gradient 4
was added to the isosurface definition, and will be explained later.
isosurface { function { x+y+z } max_gradient 4 contained_by { box { -2, 2 } } }
Note: When appropriate, to better visualize the difference between the isosurface and the container object, the images in this tutorial have been color coded.
Several surfaces
Now that you're starting to become familiar with isosurface
syntax, there really isn't any need to show a code example for each and every image. You can always look back at the earlier examples when needed. The image captions will most often contain additional keyword hints when appropriate.
Note: The user defined function portion will always use this color coded format: function { x+y+z }
For the first image on the left, these two functions lead to identical results: function { abs(x)-1 }
and function { sqrt(x*x)-1 }
because both of these formulas have the same solution where the function value is 0, specifically x=-1
and x=1
in this example.
You can easily mix any of these elements in different combinations, but the results always produce planar surfaces. The last two images in this series used function { abs(x)-1+y }
and function { abs(x)+abs(y)+abs(z)-2 }
respectively.
Non-linear functions
Curved surfaces of many different kinds can be achieved with non-linear
functions. A square function creates the parabolic shape: |
|
If you describe a circle in 2 dimensions with a constant in the 3rd dimension you get a cylinder: |
|
It's easy to change a cylinder into a cone, we just need
to add a linear component in y-direction: |
||
No worries, creating a sphere is easy too. In this example |
|
Specifying functions
Until now, we have seen, the functions used to define the isosurface were literally written in the function {...}
block:
#declare Threshold = 1; isosurface { function {pow(x,2) + pow(y,2) + pow(z,2)} threshold Threshold ... }
Let's expand on that concept, and add some flexibility. Remember that user defined functions (like equations), all float expressions and operators which are legal in POV-Ray can be used, and that functions should be declared first, and then used in the isosurface. See the section user defined function for more information.
This next example takes the above equation, and rewrites it as a user defined function. By default a function that takes three parameters (x,y,z) does not require you to explicitly specify the parameter names when declaring it, however when using the identifier, the parameters must be specified.
#declare Threshold = 1; #declare Sphere = function {pow(x,2) + pow(y,2) + pow(z,2)}; isosurface { function { Sphere(x,y,z) } threshold Threshold ... }
However, if you need more or less than three parameters when declaring a function, you will also have to explicitly specify the parameter names.
#declare Sphere = function (x,y,z,Radius) {pow(x,2) + pow(y,2) + pow(z,2) - pow(Radius,2)}; isosurface { function { Sphere(x,y,z,1) } ... }
Internal functions
There are a lot of internal functions available in POV-Ray. For example a sphere could also be generated with function { f_sphere(x, y, z, 2) }
, for these and other functions, see the functions.inc
include file. Most of them are more complicated and it is usually faster to use them instead of a hand coded equivalent.
See the complete list for details.
The following makes a torus just like POV-Ray's torus object:
#include "functions.inc" isosurface { function { f_torus(x, y, z, 1.6, 0.4) } contained_by { box { -2, 2 } } }
The 4th and 5th parameters are the major and minor radius, just like the corresponding values in the The parameters x, y and z are required, because it is a declared function. You can also declare functions yourself like it is explained in the reference section. |
|
Combining isosurface functions
We can also simulate some Constructive Solid Geometry with isosurface functions. If you do not know about CSG we suggest you have a look at What is CSG? or the corresponding part of the reference section first.
For this next group of images, consider the two functions for a cylinder and a rotated box:
#declare fn_A = function { sqrt(pow(y,2) + pow(z,2)) - 0.8 } #declare fn_B = function { abs(x)+abs(y)-1 }
- If we combine them the following way, we get a merge:
function { min(fn_A(x, y, z), fn_B(x, y, z)) }
- An intersection can be obtained by using
max()
instead ofmin()
:
function { max(fn_A(x, y, z), fn_B(x, y, z)) }
- A difference is possible, by adding a minus (-) before the second function:
function { max(fn_A(x, y, z), -fn_B(x, y, z)) }
Apart from basic CSG you can also obtain smooth transits between the different surfaces, for instance the blob object:
#declare Blob_Threshold=0.01; isosurface { function { (1+Blob_Threshold) -pow(Blob_Threshold, fn_A(x,y,z)) -pow(Blob_Threshold, fn_B(x,y,z)) } max_gradient 4 contained_by { box { -2, 2 } } }
The function{fn_A(x,y,z) + pow(Blob_Threshold,(fn_B(x,y,z) + Strength))} |
|
Noise and pigment functions
Some of the internal functions have a random or noise-like structure
Together with the pigment functions they are one of the most powerful tools for designing isosurfaces. We can add real surface displacement to the objects rather than only normal perturbation known from the normal statement.
The relevant internal functions are:
f_noise3d(x,y,z)
uses the noise generator specified inglobal_settings
and generates structures like the bozo pattern.f_noise_generator(x, y, z, noise_generator)
generates noise with a specified noise generator.f_ridged_mf(x, y, z, H, Lacunarity, Octaves, Offset, Gain, noise_generator)
generates a ridged multifractal pattern.f_ridge(x, y, z, Lambda, Octaves, Omega, Offset, Ridge, noise_generator)
generates another noise with ridges.f_hetero_mf(x, y, z, H, Lacunarity, Octaves, Offset, T, noise_generator)
generates heterogenic multifractal noise.
Using this simple noise3d function results in the image on the right. The value
|
|
In these next two images the noise function was added to a plane function. The x-parameter was set to 0 so the noise function is constant in x-direction. This way we achieve the typical heightfield structure. |
With this and the other functions you can generate objects similar to heightfields, having the advantage that a high resolution can be achieved without high memory requirements:
|
|
The noise function can of course also be subtracted which results in an inverted version:
|
|
Of course we can also add noise to any other function. If the noise function is very strong this can result in several separated surfaces.
|
|
This is a noise function applied to a sphere surface, we can influence the intensity of the noise by multiplying it with a factor and change the scale by multiplying the coordinate parameters:
|
|
As alternative to noise functions we can also use any pigment in a function:
#declare fn_Pigm=function { pigment { agate color_map { [0 color rgb 0] [1 color rgb 1] } } }
This is a vector function, it returns a color vector for use in isosurface functions. They must be pre-declared first. When using the identifier, you have to specify which component of the color vector should be used.
To do this, the dot notation is used. Refer to the above example: fn_Pigm(x,y,z).red
A color vector has five components, their supported dot types to access these components are:
fn_Pigm( ).x
|fn_Pigm( ).u
|fn_Pigm( ).red
to get the red value of the color vectorfn_Pigm( ).y
|fn_Pigm( ).v
|fn_Pigm( ).green
to get the green value of the color vectorfn_Pigm( ).z
|fn_Pigm( ).blue
to get the blue value of the color vectorfn_Pigm( ).filter
|fn_Pigm( ).f
to get the filter value of the color vectorfn_Pigm( ).transmit
|fn_Pigm( ).t
to get the transmit value of the color vector
And two special purpose operators, their supported dot types to access these operators are:
Note: The .hf
operator is experimental and will generate a warning.
fn_Pigm( ).gray
to get the gray value of the color vector
gray value = Red*29.7% + Green*58.9% + Blue*11.4%fn_Pigm( ).hf
to get the height_field value of the color vector
hf value = (Red + Green/255)*0.996093
There are quite a lot of things possible with pigment functions. However, it should be noted that, some functions can cause longer render times:
|
|
Conditional directives and loops
Conditional directives are allowed in functions:
#declare Rough = yes; #include "functions.inc" isosurface { function { y #if(Rough=1)-f_noise3d(x/0.5,y/0.3,z/0.4)*0.8 #end } ... }
Loops can also be used in functions:
#include "functions.inc" #declare Thr = 1/1000; #declare Ang = radians(45); #declare Offset = 1.5; #declare Scale = 1.2; #declare TrSph = function { f_sphere(x-Offset,y,z,0.7*Scale) } function { (1-Thr) #declare A = 0; #while (A<8) -pow(Thr, TrSph(x*cos(A*Ang) + y*sin(A*Ang), y*cos(A*Ang) -x*sin(A*Ang), z) ) #declare A=A+1; #end }
Note: The loops and conditionals are evaluated at parse time, not at render time.
Transformations on functions
Transforming an isosurface object is done like transforming any POV-Ray object. Simply use the object modifiers, scale, translate, and rotate. However, when you want to transform functions within the contained_by
object, you have to substitute parameters in the functions.
The results seem inverted to what you would normally expect, here's why:
Remember the sphere function we created earlier in this tutorial: Sphere(x,y,z)
We know it sits at the origin because x=0
. If we want to translate it 2 units to the right to x=2
we need to write the second equation in the same form: x-2=0
. Now that both equations equal zero, we can replace the parameter x
with x-2
, call our function as: Sphere(x-2,y,z)
and it's translated two units to the right.
Let's look at how to scale our test sphere by 0.5
in the y direction. Given the default value of y=1
one unit we'd want y=0.5
. To do this we need to have the equation in the same form as the first one, so we'll multiply both sides by two: y*2 = 0.5*2
which gives y*2=1
.
Now we can replace the y
parameter in our sphere: Sphere(x,y*2,z)
. This scales the y-size of the sphere by half.
Here is an overview of some useful substitutions, we'll be using a pseudo-object designated as P(x,y,z)
in the following examples:
Scale:
To scale x
replace x
with x/scale
:
P(x/2,y,z)
Scale Infinitely:
To scale y
infinitely replace y
with 0
:
P(x,0,z)
Translate:
To translate z
replace z
with z - translation
:
P(x,y,z-3)
Shear:
To shear in xy-plane replace x
with x + y*tan(radians(Angle))
:
P(x+y*tan(radians(Angle)),y,z)
Rotate:
Note: These rotation substitutions work like normal POV-rotations, they already compensate for the inverse behavior.
To rotate around the X-axis:
replace y
with z*sin(radians(Angle)) + y*cos(radians(Angle))
replace z
with z*cos(radians(Angle)) - y*sin(radians(Angle))
To rotate around the Y-axis:
replace x
with x*cos(radians(Angle)) - z*sin(radians(Angle))
replace z
with x*sin(radians(Angle)) + z*cos(radians(Angle))
To rotate around the Z-axis:
replace x
with x*cos(radians(Angle)) + y*sin(radians(Angle))
replace y
with -x*sin(radians(Angle)) + y*cos(radians(Angle))
Flip:
To flip X - Y:
replace x
with y
AND replace y
with -x
To flip Y - Z:
replace y
with z
AND replace z
with -y
To flip X - Z:
replace x
with -z
AND replace z
with x
Twist:
To twist N turns/unit around the x
axis:
replace y
with z*sin(x*2*pi*N) + y*cos(x*2*pi*N)
replace z
with z*cos(x*2*pi*N) - y*sin(x*2*pi*N)
Improving Isosurface Speed
To optimize the approximation of the isosurface and to get maximum rendering speed it is important to adapt certain values:
accuracy
:
The accuracy value influences how accurate the surface geometry is calculated. Lower values lead to a more precise, but slower result. The default value of 0.001
is fairly low. We used this value in all the previous samples, but often you can raise this quite a lot and thereby make things faster.
max_gradient
:
For finding the actual surface it is important for POV-Ray to know the maximum gradient of the function, meaning how fast the function value changes. We can specify a value with the max_gradient
keyword. Lower max_gradient values lead to faster rendering, but if the specified value is below the actual maximum gradient of the function, there can be holes or other artefact's in the surface.
For the same reason functions with an infinite gradient should not be used. This applies for pigment functions with brick or checker patterns for example. You should also be careful when using select()
in isosurface functions because of this.
If the real maximum gradient differs too much from the specified value POV-Ray issues a warning together with the found maximum gradient. It is usually sufficient to use this number for the max_gradient
parameter to get fast and correct results.
POV-Ray can also dynamically change the max_gradient
when you specify evaluate
with 3 parameters in the isosurface definition. Concerning the details on this and other things see the evaluate keyword in the reference section.
contained_by
:
Make sure your contained_by
object fits as tightly as possible. An oversized container can sky-rocket the render time. When the container has a lot of empty space around the actual isosurface, POV-Ray has to do a lot of superfluous sampling: especially with complex functions this can become very time consuming. On top of this, the max_gradient
needed to get a proper surface will also increase rapidly, almost proportional to the oversize! You could use a transparent copy of the container (using exactly the same transformations) to check how it fits. Getting the min_extent
and max_extent
of the isosurface is not useful because it only gives the extent of the container and not of the actual isosurface.
Poly Object
The polynomial object (and its shortcut versions: cubic
, quartic
and quadric
)
of POV-Ray is one of the most complex and mathematical primitives of the program. One could think that it is seldom
used and more or less obsolete, but we have to remember that for example the torus primitive is just a shortcut for the equivalent quartic
, which is just a shortcut for the equivalent poly
object. Polys are, however, seldom used in scenes due to the fact that they are so difficult to define and it is far from trivial to get the desired shape with just a polynomial equation. It is mostly used by the most mathematically oriented POV-Ray users.
This tutorial explains the process of making a polynomial object in POV-Ray.
Note: Since version 3.5, POV-Ray includes the new isosurface
object
which makes the polynomial object more or less obsolete. The isosurface is more versatile (you can specify any mathematical function, not
just polynomials) and easier to use. You can write the function as is, without needing to put values in a gigantic vector. Isosurfaces also often (although not always) render considerably faster than equivalent polys.
However, the most mathematically oriented still like polys because isosurfaces are calculated just by approximating the right value, while the poly is calculated in a mathematically exact way. Usually isosurfaces are more than good enough for most applications, though.
Note: A maximum of 35th degree polynomial can be represented with the poly object. If a higher degree polynomial or other non-polynomial function has to be represented, then it is necessary to use the isosurface object.
Creating the polynomial function
The first step is to create the polynomial function to be represented. You will need some (high-school level) mathematical knowledge for this.
1) Let's start with an easy example, a sphere:
The sphere function is:
Now we have to convert this to polynomial form, we will need a polynomial of the 2nd degree to represent this:
2) A more elaborated example:
Let's take the function:
Converting this to polynomial form we get:
Although the highest power is 4 we will need a 5th order polynomial to represent this function (because we cannot represent y4z with a 4th order polynomial).
3) And since we talked about the torus, let's also take it as an example.
A torus can be represented with the function:
where r1 is the major radius and r2 is the minor radius.
Now, this is tougher to convert to polynomial form, but finally we get:
A 4th order polynomial is enough to represent this.
Note: Not every function can be represented in polynomial form. Only functions that use addition (and substraction), multiplication (and division) and scalar powers (including rational powers, eg. the square root) can be represented. Also, the poly primitive supports only polynomials of the 35th degree at max.
Converting a function to polynomial form may be a very laborious task for certain functions. Some mathematical programs are very helpful in this matter.
Writing the polynomial vector
Now that we have the function in polynomial form, we have to write it in POV-Ray syntax. The syntax is specified in the sections on polynomial and quadric of the reference section. There is also a table in this chapter which we will be using to make the polynomial vector. It is easier to have this table printed on paper.
Note: It is also possible to make a little program with your favorite programming language which will print the poly vector from the polynomial function, but making a program like this is up to you.
1) Let's start with the easy one, ie. the sphere.
Since the sphere can be represented with a polynomial of 2nd degree, we look at the column titled 2nd in the table. We see that it has 10 items, ie. we need a vector of size 10. Each item of the vector will be the factor of the term listed in the table.
The polynomial was:
Writing the poly in this way we get:
#declare Radius=1; poly { 2, <1,0,0,0,1, 0,0,1,0,-Radius*Radius> }
Put each group of factors (separated with lines in the table) in their own lines.
In the table we see that the first item is the factor for x2, which is 1 in the function. The next item is xy. Since it is not in the function, its factor is 0. Likewise the next item, which is xz. And so on. The last item is the scalar term, which is in this case -r2.
If we make a proper scene and render it, we get:
camera { location y*4-z*5 look_at 0 angle 35 } light_source { <100,200,-50> 1 } background { rgb <0,.25,.5> } #declare Radius=1; poly { 2, <1,0,0,0,1, 0,0,1,0,-Radius*Radius> pigment { rgb <1,.7,.3> } finish { specular .5 } }
Note: There is a shortcut for 2nd degree polynomials: The quadric
primitive. Using a shortcut version, whenever possible, can lead to faster
renderings. We can write the sphere code described above in the following way:
quadric { <1,1,1>, <0,0,0>, <0,0,0>, -Radius*Radius pigment { rgb <1,.7,.3> } finish { specular .5 } }
2) Now lets try the second one. We do it similarly, but this time we need to look at the column titled 5th in the table.
The polynomial was:
Writing the poly primitive we get:
poly { 5, <0,0,0,0,0, 0,0,0,0,0, 0,0,0,0,0, 0,0,0,1,0, 0,0,0,0,0, -2,0,0,0,0, 0,0,0,0,0, 0,1,0,0,0, 0,0,0,0,0, 0,0,0,0,0, 0,0,0,0,0,0> }
With the proper scene we get:
camera { location <8,20,-10>*.7 look_at x*.01 angle 35 } light_source { <100,200,20> 1 } background { rgb <0,.25,.5> } poly { 5, <0,0,0,0,0, 0,0,0,0,0, 0,0,0,0,0, 0,0,0,1,0, 0,0,0,0,0, -2,0,0,0,0, 0,0,0,0,0, 0,1,0,0,0, 0,0,0,0,0, 0,0,0,0,0, 0,0,0,0,0,0> clipped_by { box { <-4,-4,-1><4,4,1> } } bounded_by { clipped_by } pigment { rgb <1,.7,.3> } finish { specular .5 } rotate <0,90,-90> }
3) And finally the torus:
The polynomial was:
And we get the proper 4th degree poly primitive:
camera { location y*4-z*5 look_at 0 angle 35 } light_source { <100,200,-50> 1 } background { rgb <0,.25,.5> } #declare r1=1; #declare r2=.5; poly { 4, <1,0,0,0,2, 0,0,2,0,-2*(r1*r1+r2*r2), 0,0,0,0,0, 0,0,0,0,0, 1,0,0,2,0, 2*(r1*r1-r2*r2),0,0,0,0, 1,0,-2*(r1*r1+r2*r2),0,pow(r1,4)+pow(r2,4)-2*r1*r1*r2*r2> pigment { rgb <1,.7,.3> } finish { specular .5 } }
When rendered we get:
There is a shortcut for 4th order polynomials: The quartic
primitive. We can write the torus like this:
quartic { <1,0,0,0,2, 0,0,2,0,-2*(r1*r1+r2*r2), 0,0,0,0,0, 0,0,0,0,0, 1,0,0,2,0, 2*(r1*r1-r2*r2),0,0,0,0, 1,0,-2*(r1*r1+r2*r2),0,pow(r1,4)+pow(r2,4)-2*r1*r1*r2*r2> pigment { rgb <1,.7,.3> } finish { specular .5 } }
Polynomial made easy
Since consulting the table in the section Polynomial or writing a program to get the right poly vector can be a bit cumbersome, especially when the poly vector is not a write-once-only expression and that you want to get it back, so let's examine how those equations would be rewritten using the simplified syntax.
You should refer to the images in the previous section, as these examples produce exactly the same results.
1) The sphere example can be rewritten as:
#declare Radius=1; polynomial { 2, xyz(2,0,0):1, xyz(0,2,0):1, xyz(0,0,2):1, xyz(0,0,0):-Radius*Radius }
2) Let's now see the second one:
polynomial { 5, xyz(2,0,1):1, xyz(0,4,1):1, xyz(1,2,0):-2 }
3) And finally the torus example:
polynomial { 4, xyz(4,0,0):1, xyz(2,2,0):2, xyz(2,0,2):2, xyz(2,0,0):-2*(r1*r1+r2*r2), xyz(0,4,0):1, xyz(0,2,2):2, xyz(0,2,0):2*(r1*r1-r2*r2), xyz(0,0,4):1, xyz(0,0,2):-2*(r1*r1+r2*r2), xyz(0,0,0):pow((r1*r1-r2*r2),2) }
Height Field Object | Superquadric Ellipsoid Object |
This document is protected, so submissions, corrections and discussions should be held on this documents talk page. |