Visualization Guide
Visualization Guide
Visualization Guide
Visualization Guide
Visualization Guide
Visualization Guide
|
Where your goal is realism, at the expense of longer rendering times, you have a choice of Ray Tracing, Radiosity solving, or Particle Tracing. The latter two take into account both direct lighting and indirect lighting, such as reflection of light and refraction. Additionally, both Radiosity and Particle Tracing can calculate diffuse reflections, with Particle Tracing also able to produce caustics (such as reflected light, and refraction). For these two rendering modes, in particular, the following points should be kept in mind:
You should use real world working units for your model, and lighting values are input as lumens.
IES lighting files should be used to correctly display the lighting characteristics of different lamps, such as halogen lamps, incandescent lamps, or flourescent tubes.
More care has to be taken when defining materials, to ensure that realistic values are defined.
You should set the display gamma value for your output.
You must Ray Trace the Radiosity or Particle Traced solution to display specular highlights, reflections, and refraction.
When working with Ray Tracing (with Real World Lighting enabled), Radiosity, or Particle Tracing, the sliders in the Render tool dialog box lets you interactively adjust the brightness and contrast of the image. That is, the result is immediate, you do not have to redisplay the solution to see the updated brightness and contrast settings.
Gamma correction is used to compensate for the fact that monitors and printers don't have the same visual response as the human eye, and serves to bring out more detail in darker areas of images. Typically, the gamma correction for a standard monitor (CRT) should be set to 1.8 to 2.5, while for an LCD display, it should be left at 1.0. In MicroStation you can set the gamma correction value for your display in the View Options category of the Preferences dialog box (select Workspace > Preferences). An “exact” figure is not critical, but a good starting value to experiment with is 2.0.
Similarly, when you save an image from MicroStation, you have the option of applying Gamma Correction to it. This may be to accommodate a printer to produce hard-copy output that more closely represents what is seen on the screen. As a rule, to avoid “tying” your image to one particular display/printing device, it is better to save images with a neutral gamma value (1.0) and let other software, or printing device software, that you use add the gamma correction that they require. Each display/printing device can have different characteristics such that the correct gamma correction for one device may not be correct for another. Once gamma correction has been saved with an image, changing the gamma value will not let you return the image exactly to its original state (with gamma correction of 1.0).
There is, however, one proviso. That is, that you can perform a better gamma correction in MicroStation, while you still have floating point brightness values before they are saved in a typical integer brightness file format. What this means is, if you have a specific display or output device that you are catering for, it may be advantageous to save the image with a gamma correction value (other than 1.0). While this will tie the image to the output device, it may also give the best result.
Typically, printers tend to darken images more than display monitors, so a higher value may be required if you intend to print the finished image. Where required, gamma correction can be performed on the saved image, after rendering, using the MicroStation image display utility (select Utilities > Image > Display) or other imaging software.
Ray tracing is a photo-realistic rendering method in which an image is generated by simulating the specular reflection of light rays in a 3D scene. Ray Tracing does not display diffuse reflections. It can, however, be used in conjunction with Radiosity or Particle Tracing solution to produce the specular highlights and reflections that these processes do not display. Both Radiosity and Particle Tracing have an option for ray tracing the final display. MicroStation's Ray Tracing lets you use lighting that is more compatible with the interactive rendering modes, or you can turn on Real World Lighting, in the Ray Tracing dialog box, to work with real-world lighting values. This lets you use the Ray Tracing option to set up lighting and materials in your images prior to running a Particle Tracing or Radiosity solution, both of which can take much longer.
In the real world, light rays are emitted by one or more light sources and reflect off objects until they finally reach the eye.
On a computer, often it is more efficient to trace rays from the eye rather than from the light sources. This can save a significant amount of time by not following rays from lights to surfaces that are never seen.
 | |
Ray traced rendering | |
Ray tracing follows rays backwards from the eye into the scene, determining what can be seen. It begins by tracing (or shooting) rays of light from the viewer's eye (the camera position) through each pixel in the view.
Tracing a ray involves testing all objects in a scene for intersection with that ray. For these initial rays, often referred to as primary rays or eye rays, the nearest intersection along each ray must be computed — the entire design must be examined to find the nearest of all the intersections along the ray. Hidden surface removal is performed by this procedure.
Once it is determined what is visible, the illumination and shading of the visible objects is computed. The illustration below shows the various rays that must be computed during ray tracing:
1 — Primary or eye ray.
2 — Reflected ray.
3 — Shadow ray, which is traced to the light source, checks for any obstructions to the light.
4 — Transmitted ray.
5 — Secondary reflected ray.
6 — Secondary transmitted ray.
 | |
Tracing a ray from the camera (eye) position. | |
Shading of the visible surface is computed for each pixel. The color of the surface is composed of three components — ambient, local, and global illumination — which are added together.
Ambient illumination is surface lighting not directly attributed to any particular light source. Ambient light brightens a scene in areas where there is little or no lighting.
Local illumination is surface shading directly attributed to light sources. Local illumination is made up of diffuse and specular components.
The Diffuse component is the light that directly strikes a surface. It is view-independent. The Specular component creates a bright highlight in the reflected direction of the light on objects with glossy surfaces. Specular highlights are view-dependent.
Local illumination of a surface depends on whether or not other objects obstruct a light from shining on the surface. To determine if any object shadows a surface, a ray must be traced from the surface in the direction of each potential light source. For these shadow rays, as they are commonly called, it is not always necessary to traverse the entire design, testing for intersections. Once an opaque object is found that intersects a shadow ray, traversal can stop, since there can be no light from that light source. If a transparent object intersects the shadow ray, the light is attenuated according to the transparency of the intersected (shadowing) object. Once a light is determined to illuminate a surface, the intensity of the light is attenuated according to the distance from the light to the surface.
Global illumination is shading on a surface due to secondary (global) effects such as reflections and transparency.
To determine the illumination a surface receives from the reflected and refracted directions, secondary rays are traced in those directions. For efficiency, rays are only traced if the surface is actually reflective or transparent. Each reflected or transmitted ray is treated as a primary ray in that the nearest intersection along the ray must be computed. Similarly, the surfaces that these rays “see” must be shaded as described above. This process is repeated recursively until a limit you define is reached or the accumulated reflectivity or transparency drops below a given threshold.
 | Ray tracing is especially useful in applications that require realistic renderings of reflective and transparent surfaces, such as metal and glass. |
 | |
Images rendered with Phong (left) and Ray Traced (right). | |
“Radiosity,” as defined in the literature of physics, is the total power leaving a point on a surface, per unit area on the surface. In the context of rendering, “power” is light energy.
Radiosity solving is a sophisticated technique that calculates the light that is reflected between diffuse surfaces. It can be used to demonstrate effects such as color bleeding (where one colored surface lends a tint to another nearby surface) and light dispersion (the reflection of indirect light onto other surfaces in a scene). Radiosity does not distribute specular light and, thus, it does not produce caustics. Also, to display specular highlights in a radiosity solution, you must set Final Display to Ray Trace (this setting is found in the Display section of the Radiosity dialog box).
Radiosity solving, unlike Ray Tracing, is not a rendering technique on its own — it merely generates a lighting solution that in turn, can be rendered. In fact, radiosity solving and ray tracing capabilities can be used together to produce realistic images with the best qualities of both methods. Radiosity solving operates as a rendering pre-process that computes the global, view-independent (diffuse) lighting solution. Ray tracing uses this radiosity solution to render a view-dependent image, adding specular highlights and reflections.
In the following example, source lighting is provided by two spotlights, pointing toward the ceiling. Only with the ray traced image that includes the radiosity solution, do you see the natural look of the scene illuminated by both direct and reflected light. This produces the subtle shadows normally present in the real world.
| Where caustics are desired (the reflection/refraction of specular light) you should consider using Particle Tracing. |
 | |
Image ray traced, with no radiosity solution. Only specular highlights are displayed. | |
 | |
Same image ray traced, with no radiosity solution, but with Global Lighting's Flashbulb turned on to illuminate the dark areas. | |
 | |
Same image ray traced with radiosity solution. The only lights in the scene are the two spotlights. Here, the darkened areas are illuminated by the reflected light from the walls and ceiling, giving a natural look with subtle shadows. | |
Since the radiosity solution is “view-independent” it can be reused to render additional images of the design from different views. Each image can be rendered using either ray tracing or smooth shading. Smooth shading can be faster than ray tracing, but does not include any of the specular effects (such as reflections, refractions, and specular highlights).
The radiosity solving process produces useful intermediate solutions in a short amount of time. It then automatically and continuously refines them into the final solution. This makes it possible to display intermediate results so that you can decide when the solution is satisfactory and stop the calculations.
Also, you can specify the stopping criteria, either as a fixed number of “shots” or as a fraction of the total global illumination to be distributed.
When viewing a radiosity solution, whether it be an intermediate solution, or a completed calculation, you can adjust the brightness and contrast of the image interactively, using the slider controls in the Render tool dialog box.
MicroStation's radiosity solving calculates the dispersion of light energy in a scene. Radiosity calculations require different lighting settings to those used for standard Phong, or Ray Traced (with Real World Lighting disabled) images. Thus, for use with a radiosity solution, the Define Light tool has a Lumens setting that acts as a multiplier of a light source's color and intensity values to simulate real world lighting values. If the Intensity value is set to 1.0 or less, the value for Lumens most closely approximates real world lumens.
Understanding the basics of the radiosity solving process can help you determine the trade-offs between solution time and solution quality that are critical to successful usage. The main control over solution quality is determined by the subdivision settings, which specify how finely each surface in the design is meshed.
 | Radiosity solving is a processing-intensive operation. As such, the practical minimum hardware requirements for radiosity solving may exceed the general minimum requirements for using MicroStation. |
During processing, surfaces are first subdivided into a set of triangles (controlled by the MicroStation Stroke Tolerance setting). These triangles are further subdivided into patches. From here, each patch is subdivided into one or more triangular elements, thus forming an element mesh. A further adaptive subdivision may occur along shadow boundaries. Settings in the Radiosity dialog box determine the sizes of the patches and elements, as well as the degree of adaptive subdivision.
The settings in the Radiosity Settings section control the size of the patches and elements. Another setting controls the way that the elements are further subdivided along shadow boundaries. Take, for example, a simple 10 Ũ 10 unit square surface.
With the following settings:
Maximum Patch Area 50
Maximum Element Area 10
the surface is first subdivided into patches having a maximum area of 50 square units.
 |
From here, the patches are further divided into elements having a maximum area of 10 square units.
 |
Light energy is received by the elements. Illumination is calculated at each vertex, and the mean value is calculated for the element. Also calculated is the amount of energy that is absorbed and reflected. This is dependent on the material definition for the surface. Finally, the amount of light energy to be “shot” (reflected) is then calculated for each patch, by gathering the values for all elements contained within it (the patch).
Each shot during radiosity solving “shoots” the light energy from a single patch to each of the elements of the other surfaces. By selecting and shooting the brightest unshot patch each time, the intermediate solution progresses as quickly as possible toward the final solution.
If a spotlight is placed above the lower right corner of the surface, the radiosity solution calculated and then ray traced, the resulting image is less than satisfactory. This is due to the element mesh being too coarse to accurately depict the circular light beam.
 | |
Coarse element mesh | |
 | |
A coarse element mesh produces inaccurate results at the boundary of the (circular) light beam in the lower right corner. | |
This can be corrected by decreasing the size of the element mesh. However, only the lower right corner of the image needs the extra resolution, not the entire surface.
Setting Maximum Element Subdivisions to a value greater than zero meshes surfaces more finely at shadow boundaries. For example, a setting of 1 allows each element to be further subdivided into 4 triangles, and a setting of 2 allows each of these to be further subdivided into 4, and so on. Thus, the higher this value, the more accurate the boundary becomes, at a greater cost in processing time.
 | |
A finer element mesh provides a more accurate definition of the light beam on the surface. | |
 | |
Adaptive subdivision allows for greater subdivision of the surface element mesh at the shadow boundaries. Other areas are left at their initial subdivision. | |
Geometric accuracy of the shooting operations is determined by the sizes of the patches, with smaller patches giving a more accurate result at a greater cost in time. In general, each patch can shoot to the elements of every other patch, so the time needed to compute an exact solution can increase with the square of the number of patches. For example, a solution with twice as many patches takes about 4 times as long to compute; one with 10 times as many patches takes about 100 times as long to compute, assuming the ratio of the patch areas to the element areas remains the same.
When the element area remains constant, the time to compute a solution may vary linearly with the number of patches. For example, if the maximum patch size is 10.0 and the maximum element size is 1.0, reducing the patch size to 5.0 would require twice as many shots to be taken (with the time per shot remaining constant), taking twice as long to reach a similar solution (of higher quality).
 | |
Maximum Patch Area 20. | |
 | |
Maximum Patch Area 5. | |
As the Maximum Patch Area setting is reduced, the shadows from the reflected light become less distinct. In this example, the shadow cast by the lamp on the left, particularly, is much more diffuse with the smaller setting for Maximum Patch Area.
With this in mind, it is important to keep tight control over the number of patches to minimize the time needed to compute the radiosity solution. A good starting point for these settings is to use the defaults, which can be found by selecting File > Reset Default Settings in the Radiosity dialog box.
For even finer control, you can also specify, on an individual basis, which surfaces are allowed to shoot and which surfaces are allowed to receive illumination. Selection is based on whether or not Global Illumination is turned on in the material definition of the surface.
Accuracy of the visibility calculations is determined by the number of sample points on the current light source or shooting patch. You can use the Maximum Samples per Shot setting to specify a larger number of sample points to improve the visibility calculations. The radiosity solution time increases roughly one-for-one with the number of samples.
If you are using a relatively small patch size, you can generally use a smaller number of samples, and if you are using a larger patch size, you will generally need to use a larger number of samples. Similarly, if you wish to decrease the number of samples, you may need to decrease the patch size as well. If you increase the number of samples, however, you still may be unable to increase the patch size significantly without adversely affecting the accuracy of the lighting calculations.
Resolution of shadows and highlights in the solution is determined by the sizes of the elements. To improve resolution along shadow boundaries, “adaptive” subdivision of elements across the boundaries can also be performed. This is controlled by the Maximum Element Subdivisions setting.
 |
 | |
Top: Maximum Element Subdivisions = 1. Bottom: Maximum Element Subdivisions = 4. Increasing the Maximum Element Subdivisions value, results in more accurate images, particularly at shadow boundaries. | |
You can save on memory usage by using turning on Ray Trace direct illumination. With this setting on, the ray tracer is used to compute shadows from light sources, but not from reflected light. This means that if the scene is lit primarily by direct light, very high quality images can be obtained using much larger element sizes, and therefore much less memory.
 |
| For Radiosity and Particle Tracing solutions, Render All Objects, in the Ray Tracing dialog box, is assumed to be on. This is regardless of the actual setting in the Ray Tracing dialog box. |
| For radiosity solving, both Ambient and Flashbulb in the Global Lighting dialog box should be turned off, or set to zero. |
If necessary, the key-ins RADIOSITY SHOOTINTERIOR, RADIOSITY SET SHOOTINTERIOR [ON|OFF], and RADIOSITY SET EDGEDELTA <value> can be used to allow radiosity to be shot only to the interiors of polygons. This serves two purposes:
Where a light source is outside a room, it prevents light from hitting the vertices on the edge shared by a wall and a floor.
If geometry of a wall or cornice is near the edge of a surface, it may prevent that cornice from inadvertently shadowing the surface.
Key-in |
Effect |
|---|---|
RADIOSITY SET SHOOTINTERIOR [ON|OFF] |
Enables or disables shooting radiosity to polygon interiors. This value should generally be set ON (the default). When enabled (on), the vertices near the edges of polygons are moved toward the polygon's interior by the amount specified with the RADIOSITY SET EDGEDELTA <value> key-in. |
RADIOSITY SHOOTINTERIOR |
Toggles the current state (on/off) for shooting radiosity to polygon interiors. |
RADIOSITY SET EDGEDELTA <value> |
Specifies the amount that the edges of polygons are moved towards the polygon's interior. This value is expressed as a fraction of a triangle. The default (0.0002) should be sufficient for most designs. |
As a radiosity solution is being computed, the status bar displays the current status — shot number, energy for this shot, unshot energy remaining, and the minimum energy threshold for stopping the solution.
If an intermediate display is specified, then intermediate solutions are rendered in the selected view at the specified frequency. When one of the stopping conditions is met, or the final solution is computed, it is rendered into the selected view (provided Final Display is not set to None), and the message “Display complete” appears in the status bar.
Rendering statistics can be viewed in the Message Details section of the Message Center.
 | |
Rendering details displayed in the Message Details section of the Message Center. | |
You can interrupt radiosity solving between shots with a Reset. After the current shot is completed, the final display is rendered. The final display can then be interrupted with another Reset.
In the case where an intermediate solution is being displayed, a single Reset interrupts the display, and the solution continues with the next shot. A second Reset then interrupts the radiosity solution, and a third Reset interrupts the final display.
You can restart the radiosity solution, from the next shot, by clicking the Add more shots to solution (to current limits) icon in the Render tool dialog box.
Radiosity settings, such as the Limit Number of Shots value, can be changed before the solution is restarted, and in most cases will take effect immediately. Other settings, such as the Maximum Patch and Element areas, will not take effect until the rendering database is cleared and a new solution commenced.
Whether Radiosity calculations have been interrupted or have been completed, you can use the Brightness and Contrast sliders in the Render settings window to fine tune an image interactively. Move the sliders to the right to brighten the image or to add contrast, or move them to the left to darken the image or to reduce the contrast. Using the sliders is similar to working with the Brightness Multiplier/Adapt to Brightness setting and the Display Contrast setting in the Display section of the Radiosity dialog box, except that the controls in the Render tool dialog box are interactive and do not require an update of the view with the Display current solution (in any view) option in the Render settings window. When you use the slider controls, the Brightness Multiplier/Adapt to Brightness, and Display Contrast settings automatically are adjusted to match the two slider control settings.
Where required, you can save the radiosity solution itself to disk. This then can be retrieved at a later time and, providing the geometry and lighting have not changed, you can add more shots to the scene or render additional images with the radiosity solution already calculated.
When saving a radiosity solution, from a model other than the default model, by default it is given a name in the form <DGN filename>_<model name>.rad. For example, if you were working in a model named “First Floor” in a DGN file named “Office.dgn”, then the radiosity solution would, by default, be given the name “Office_First Floor.rad”. If you then changed to another model, “Second Floor”, in the same DGN file, it would be given the name “Office_Second Floor.rad” and so on. Solutions created from the default model, whether it has been renamed or not, are given a default name <DGN filename>.rad.
This naming convention lets you work on different models in the same DGN file without overwriting existing solutions. You also have the option of choosing your own name for the radiosity solution.
| It is highly recommended that a radiosity solution be loaded only into the model from which it was originally saved. Doing otherwise, results are unpredictable. |
Particle tracing is not a rendering process in itself, but a global lighting solution that may be rendered. Particle tracing is an alternative to traditional radiosity, with significantly lower memory requirements. It is especially well suited to visualizing very large designs. Because the particle tracing solutions are computed directly to disk, rather than in memory, solutions may be generated for designs of virtually any size (see Disk Space Requirements for Particle Tracing).
Particle tracing solutions are view-independent. Once computed, the solution can be redisplayed from any vantage point. This feature is particularly useful for creating animations and interactive walkthroughs.
Particle tracing calculates the paths of light particles as they are emitted from light sources and are reflected and transmitted through the scene. When ray traced, these lighting solutions provide photorealistic images, including reflections, refraction and other caustic lighting effects such as light reflected by mirrors or focused through lenses. To view specular highlights, reflections and refraction, in a particle traced image, you must turn on Ray Trace Specular Effects, in the Display section of the Particle Tracing dialog box.
 | |
Image ray traced with no particle tracing solution present, displays only specular highlights | |
 | |
Same image ray traced with a particle tracing solution present. Notice that the walls are more naturally illuminated by reflected as well as direct light. Notice in particular, the circle of light on the left wall above the table. This is caused by light reflecting from the silver tray (with decanter and glasses) on the table. | |
Particle tracing computes a view-independent, global lighting solution that includes all diffuse lighting effects, such as color bleeding. Additionally, particle tracing accounts for all specular light effects including reflections, refractions and caustics.
Since the particle trace solution is view-independent, it can be viewed from any vantage point in the design. This makes it very useful for animation applications such as walk-throughs. Because it also has the low memory overhead of ray tracing, particle tracing can be used to render very large designs.
To further speed up the process for an image, you can turn on Visible Surfaces Only so that only those surfaces visible in the view are meshed. If you then select another view that contains surfaces not visible in the first view then only the new surfaces will need be meshed.
Particle tracing works by calculating the paths of light particles as they are emitted from light sources and are reflected and transmitted throughout the scene. This process occurs in two distinct phases — Particle Shooting and Meshing.
During this phase, particles are emitted or shot, into the scene, from each of the light sources. You can specify the total number of particles to be shot. From this total figure, the relative value of the Lumens setting for each light source then determines the number of particles shot by each source.
Particle paths are traced through the scene, interacting with the surfaces encountered along the way. These interactions can include caustics, which are the lighting effects caused by light reflected off surfaces or refracted through transparent objects. Examples of these effects are light reflecting off a mirror or being focused through a lens. A caustic reflection differs from a ray-traced reflection in that the caustic reflection adds light to the surface receiving the caustic, while a ray-traced reflection just shows what the viewer would see.
When a particle strikes a surface, it is either absorbed or bounced. A bounce can be a diffuse reflection, a specular (mirror) reflection, or a specular transmission (with refraction). Each time that a particle is diffusely reflected or absorbed, a hit point is recorded for that surface. All hits are recorded to disk in a hit point file, having the extension “.shp.”
Each surface's material properties determine the relative probabilities of these interactions. Take, for example, a diffuse white surface having a diffuse setting of 0.7 and a specular setting of 0.0. This surface will diffusely reflect 70% of the particles that hit it and absorb the remaining 30%. If the material also has a specular value greater than zero then a percentage of the particles will be specularly reflected. Similarly, if the material is partially transparent then some particles will be transmitted as well.
Once the shooting phase has finished, the display phase begins, meshing surfaces as needed.
When Visible Surfaces Only is enabled, the progress bar and time estimates the amount of time required to complete the view, not the complete solution.
When Visible Surfaces Only is disabled, first the view is rendered, meshing surfaces as needed. When the view is complete, all remaining surfaces are meshed. For these situations the progress bar and time estimates reflect the amount of time required to complete the entire solution. Note that the view is likely to be fully rendered long before the progress bar reaches 100%. After meshing is completed with Visible Surfaces Only disabled, the particle-traced solution can be redisplayed quickly and easily, from any viewpoint. Because the solution automatically is saved to disk, it is not lost even if you close the design or exit MicroStation.
| Particle tracing stores most of its data in temporary files. These files are located in the directory specified by the environment variable MS_PTDIR. If this variable is not defined, then the files are stored in the directory pointed to by the environment variable MS_TMP. |
 |
Section |
Description |
|---|---|
Defines the number of particles to be shot, and how they interact with materials in the design. |
|
Controls the level of detail and smoothness of the rendering mesh |
|
Defines the display of the particle traced solution. |
Because Particle Tracing computations are written to disk, you need to ensure that you have enough free space for the image being rendered. Specifically, 8 bytes is required for each hit point initially, in the unsorted hit point file (.uhp). Later, this reduces to 4 bytes / hit point when that file is converted into a sorted hit point file (.shp).
The ratio of hit points to particles can vary greatly, of course, depending on the geometry and materials used in your design. This ratio, however, remains relatively constant as you add more particles. Thus, you can get an estimate of the disk space required as follows:
First run a small (1 million particle) solution, and check the size of your hit point file (.shp).
You can use this figure then to estimate how much free space you'll need.
For example, if a 1 million particle solution produces a 4 MB hit point file (.shp), then for a 400 million particle solution, the final hit point file would be 1600 MB (400 x 4 MB). The disk space required, however would be double that, 3200 MB, to allow for the initial “.uhp” file.
Where you donīt have enough free space, there is a “trick” that you can use to overcome the shortage. What if, in the above example, you had only 3000 MB free? Still you could get a 400 million particle solution, as follows:
Shoot 200 million particles, which would require 1600MB for the initial “.uhp” file, then reducing to 800MB for the final “.shp” file.
Add 200 million particles, which would require only 2400 MB (800 MB for the first “.shp” file, plus 1600 MB for the second “.uhp” file).
In this example, the final “.shp” file still will be 1600 MB, exactly the same as if you did it all in one step.
| In order to create files greater than 4 GB, you must be using an NTFS file system. FAT file systems are limited to 4GB for a single file. |
Particle tracing calculates the paths of light particles as they “bounce” around a model and then calculates the effect of the light particle “hit points” on each surface. The more particles that you use, the more hit points that the particle tracer has to work with, to produce a better image. The Smoothness setting controls the size of the local area over which hit points are spread out. The Smoothness setting involves a trade-off between “noisiness” and “blurriness”. In brief, setting the Smoothness setting:
Too low — your details (such as shadows and caustics) will be sharp, but you will also see noise (more commonly known as “splotches”) in your image.
 | |
With Smoothness set too low, details are sharp, but noise (splotches) appear in the image. | |
Too high — everything normally will appear smooth, but you will lose detail in your shadows and caustics.
 | |
With Smoothness set too high, details are indistinct, but surfaces are smooth. | |
Using the default setting (3.0) normally should provide a good balance. If there is noise in the image, adding more particles will reduce the noise without loss of detail.
 | |
Adding more particles to the solution reduces noise, while retaining detail. | |
You can interrupt the processing of a particle traced view by entering a reset. When you do this, the Continue After Reset icon becomes active. Clicking this icon lets you continue the process from where it was stopped. This applies even after you have exited and restarted MicroStation.
If Visible surfaces only is enabled and processing is aborted after the shooting phase was complete, Display current solution is enabled and can be used to continue the solution. As the view is rendered, any unmeshed surfaces are meshed as they are displayed.
If Visible surfaces only is not enabled or processing is aborted during the shooting phase, the solution can be continued by clicking the Continue after reset icon.
| Often it is advisable to save the partial solution as a PTD file, just in case any of the design changes. |
When Particle Tracing calculations have been completed, you can use the brightness and contrast sliders in the Render tool dialog box to fine tune an image interactively. Move the sliders to the right to brighten the image and/or to add contrast, or to the left to darken it and/or reduce the contrast. Using the sliders is similar to working with the Brightness Multiplier/Adapt to Brightness and Display Contrast settings in the Display area of the Particle Tracing dialog box (Advanced Settings), except that it is interactive and does not require an update of the view with the Display current solution (in any view) option in the Render tool dialog box. When you use the Slider controls in the Render tool dialog box, the Brightness Multiplier/Adapt to Brightness and Display Contrast settings are adjusted automatically to match the two slider control settings.
You can save the particle traced solution to disk. This then can be retrieved at a later time and, providing the geometry and the lighting has not changed, you can add more particles or render additional images with the particle traced solution already calculated.
When saving a particle tracing solution, from a model other than the default model, by default it is given a name in the form <DGN filename>_<model name>.ptd. For example, if you were working in a model named “First Floor” in a DGN file named “Office.dgn”, then the particle traced solution would, by default, be given the name “Office_First Floor.ptd”. If you then changed to another model, “Second Floor”, in the same DGN file, it would be given the name “Office_Second Floor.ptd” and so on. Solutions created from the default model, whether it has been renamed or not, are given a default name <DGN filename>.ptd.
This naming convention lets you work on different models in the same DGN file without overwriting existing solutions. You also have the option of choosing your own name for the particle traced solution.
Associating solutions with different models makes it convenient to have various setups, such as different models with different lighting setups, or with different levels of detail, or even different rooms of a house. By default, each of these models will use different work files for particle tracing solutions.
| If Use Alternate Workfile is enabled, the alternate file name is used for particle tracing work files without modification. |
| To use particle tracing solutions that were computed with earlier versions of the software, for non-default models, turn on Use Alternate Workfile (in the Advanced Settings interface of the Particle Tracing dialog box) and key in the name of the DGN file. Alternatively, you can rename the files in the Particle Tracing working directory (MS_PTDIR) from the relevant DGNfilename.* to DGNfilename_modelname.*. |
| It is highly recommended that a particle traced solution be loaded only into the model from which it was originally saved. Otherwise, the results are unpredictable. |
Once you have saved a particle traced solution to disk, you only need this file, the design, and the pattern/bump map files to re-display the solution at any point in the future. Once this “.ptd” file is saved, and you are sure that you will not require more particles to be added, you can safely delete any remaining hit point files, mesh files, or other temporary files associated with particle tracing.
When creating particle traced images, you can use multiple machines across a network. The requirements and procedures are as follows:
MicroStation must be installed on each machine.
All machines must use the same directory for the solution files. In other words, all machines must use a common network path for the Configuration Variable MS_PTDIR, the Particle Tracing Work Dir configuration variable in the Rendering/Images category of the Configuration dialog box.
Start the solution on one machine and let it finish shooting particles (the first stage).
Once the view starts to render on the first machine, use Display current solution for another view on a different machine. Each machine will mesh any unmeshed surfaces as needed, and will in turn, finish the complete solution faster.
| Often it is necessary to save a PTD file on the first machine and then load that PTD file on the others. This ensures that the rendering database is the same on all machines. If necessary, you can Reset to interrupt the process on the first machine, save the PTD file, then continue. |
During processing, before a machine creates a new mesh for a surface, it first checks to see if another machine has already meshed that surface and, if so, it will use the existing mesh for display. Thus, multiple views can be meshed simultaneously from the same hit point file.
During processing, each machine writes its meshes into a separate mesh (UM) file. For example, when creating a solution for gallery.dgn, the first machine writes meshes into gallery.um, and the others write meshes into gallery$1.um, gallery$2.um, and so on. Index files (with extensions of MIN and PIN) are used to keep track of where all the meshes are located. These mesh and index files, along with the hit point (SHP) file, must be in a single directory that is shared among all of the participating machines.
When processing is complete, it is recommended that the solution be saved to a PTD file, which will combine the individual mesh files into the single PTD. In addition to providing a permanent solution, redisplaying from the PTD file is somewhat more efficient than redisplaying from the individual mesh files. Solutions are saved by selecting File > Save Solution in the Particle Tracing dialog box, or by keying in PARTICLETRACE SAVE.
In a similar manner, Display current solution can be used to extend a PTD file that contains a partial solution. Again, it is recommended that the completed solution be saved. This will combine the original PTD file with any new mesh files into a single solution file.
After creating a partial solution, a full solution can be completed as follows:
At this point, only after saving the full solution, it is safe to delete the mesh, index and hit point files. If, however, you might want to re-mesh later, then the hit point file (SHP) should NOT be deleted.