Triton User Manual – version 1.0 Randall E. Hand and R. Moorhead
Contents
1. e x length y_length how large to make this layer It will extend away from the user x_length 2 in both the positive and negative X directions and y_length 2 in both the positive and negative Y directions X amp Y are the data axes not visualization axes Each number must be a positive integer power of 2 32 64 128 256 etc e data_values a data source to retrieve the scalar values used for rendering e depth_values a data source to retrieve the values for the depth of each layer Triton Technical Manual 12 of 27 e layerX on off if on then a scalar surface will be generated and rendered for this layer The entire area x_length y_ length is generated as an RGBA texture and drawn on a rectangle at the desired location The height of the quad is taken from depth_values at the user s location and the quad is centered around the user depth_values is usually an ASCII Fixed Data source to keep the data layers at constant heights independent of user position A sample scalar surface would look like this DATA SURFACE type SCALAR QUAD colormap temperature ct x length 128 y_length 128 data_values SCALAR DATA depth_values DEPTHS layer on end no Scalar Posts Triton Technical Manual 13 of 27 A scalar post renders a colormapped scalar value for a single point relative to the user s position in all layers It uses the following keys e x offset y_offset floating point numbers2. O M Smedstad Interactive Visualization of Ocean Circulation Models IEEE Visualization 96 Oct 1996 pp 429 432 B M A Chupa R J Moorhead S Nations A Johannsen K Gaither and R Vickery ISTV Interactive Structured Time varying Visualizer Proceeding of Oceans 99 Sept 1999 pp 938 944 4 A Johannsen and R J Moorhead Flow Visualization of Basin Scale Ocean Data IEEE Visualization 94 Washington D C Oct 1994 pp 355 358
3. 23 015 depth6 29 67 depth7 Sl depths 47 385 depth9 59 065 depth10 73 15 depth11 90 14 depth12 110 635 depth13 135 36 depth14 165 19 depth15 201 17 depth16 244 57 depth17 296 925 depth18 360 08 depth19 436 26 depth20 528 15 depth21 639 195 depth22 775 depth23 950 depth24 1150 depth25 1350 depth26 1550 depth27 1750 depth28 1950 depth29 2150 depth30 2350 depth31 2550 depth32 2750 depth33 2950 depth34 3150 depth35 3375 depth36 3625 depth37 3875 depth38 4125 depth39 4375 depth40 4625 depth41 4900 depth42 5200 depth43 5500 depth44 5800 depth45 6100 end Triton Technical Manual 23 of 27 Sample NCOM Configuration This configuration displays the bathymetry one scalar post 4 layers of data flow glyphs a vector post one scalar surface It also uses several data sources SYSTEM startx 900 starty 400 ComponentCount 10 AddComponent1l MAINROAM AddComponent2 DATA_POSTS AddComponent3 DATA FLOW GLYPHS AddComponent4 DATA FLOW POST AddComponent5 DATA SURFACE end DATA Databases 6 DatalName SCALAR_DATA DatalType TILED FLOAT Data2Name DEPTHS Data2Type TILED FLOAT Data3Name BATHYMETRY Data3Type TILED FLOAT Data4Name VECTOR_U_DATA Data4Type TILED FLOAT Data5Name VECTOR_V_DATA Data5Type TILED FLOAT Data6Name VECTOR MAG Data6Type MAGNITUDE FLOAT end POINTER type WAND POINTER width 0 3 length 3 0 color red 1 0 color green 0 0 color blue 0 0
4. a number to mask out of the dataset This value is optional and if is not specified no data is masked e maskop an operation to use for masking This operation is combined with landmask to determine what values to mask eg Everything greater than landmask everything less than landmask etc It can be gt or lt This value is optional and defaults to if not specified e landvalue a floating point number If masking is enabled then this value is written in place of masked values This number needs to be pre scaled This parameter is optional and if not specified masked data is simply removed from the dataset e flip_x yes no If yes the data will be inverted on the X axis of the 2 dimensional data plane effectively flipping the data top to bottom This parameter is optional and no action is taken if not specified Each y value will be replaced with height y Another way to think of the operation is that the first line is written along the top of the data plane instead of the bottom This allows Triton to handle data sources where the origin of the data is at the top left or the bottom left e scale a floating point number all data is multiplied by this scaling factor once loaded This parameter is optional and if not specified it defaults to 1 0 e minvalue a floating point number If specified then the data is moved to make this the minimum value This number needs to be pre scaled This parameter
5. landmask 9 9999998e 30 end TEMPERATURE data NCOM t3d t3d_20000313 00 bin startx 0 starty 0 width 1024 height 640 end VECTOR_U_DATA cache 20 tilecount 1 tilel FLOW_U flip x yes layers 20 maskop lt landmask 9 9999998e 30 end FLOW U data NCOM u3d u3d_20000313 00 bin startx 0 starty 0 width 1024 height 640 end VECTOR_V_DATA cache 20 tilecount 1 flip x yes tilel FLOW V layers 20 maskop lt landmask 9 9999998e 30 end 25 of 27 Triton Technical Manual FLOW_V data NCOM v3d v3d_20000313 00 bin startx 0 starty 0 width 1024 height 640 end DEPTHS cache 20 tilecount 1 flip x yes tilel DEPTHS TIL scale 0 005 layers 20 maskop lt landmask 9 9999998e 30 end El DEPTHS TILE data NCOM model zm bin startx 0 starty 0 width 1024 height 640 end VECTOR MAG data_u VECTOR_U_DATA data_v VECTOR_V_DATA startx 0 starty 0 width 1024 height 640 layercount 20 end 26 of 27 Triton Technical Manual 27 of 27 References 1 K Gaither R Moorhead S Nations and D Fox Visualizing Ocean Circulation Models Through Virtual Environments IEEE Computer Graphics and Applications Vol 17 No 1 Jan Feb 1997 pp 16 19 2 S Nations R Moorhead K Gaither S Aukstakalnis R Vickery W C Couvillion Jr D N Fox P Flynn A Wallcraft P Hogan
6. look like this DEPTHS fixed_depths_scale 0 005 layercount 45 depth1 2 5 depth2 EE depth44 5800 depth45 6100 end NOTE This is one of the few sections that not using the end tag can be useful in rare instances If you need the same fixed data values at two different scales you could do this DEPTHS1 fixed_depths scale 0 005 DEPTHS2 fixed_depths scale 0 005 layercount 45 depthl 23 9 depth2 TERS Triton Technical Manual 7 of 27 depth44 5800 depth45 6100 end This will create two different data sources containing the same data but scaled differently In this case they are the same quantities but different signs Configuring a TILED Data Source A Tiled data source currently only supports Floating Point data so you must use TILED FLOAT A tiled data source usually only contains one tile one binary file but can support several as in GTOPO30 Tiled data sources use the following keys e cache an integer number indicating how many tiles to cache in memory at once Each layer of the data is cached independently so cache may be much larger than tilecount It should at least be as large as layers tilecount an integer number indicating how many tiles there are tileX a header found later in the file indicating what tile to load A tilex entry must exist for each X between 1 and tilecount e layers an integer number indicating how many layers exist in each file e andmask
7. Triton User Manual version 1 0 Randall E Hand and R Moorhead December 2001 ERC Technical Report MSSU COE ERC 01 15 Visualization Analysis and Imaging Lab Engineering Research Center Mississippi State University Mississippi State MS 39762 Triton Technical Manual 2 of 27 REQUIREMENTS et ageet cd steed hs VE hcdeeviiebedecaeei sedadewi tines doncdecdaea Wao avaetiniee 3 RUN IME REQUIREMENTS s asonennenenenernnrrnnrnnnre rtre krann tekar errn nren EREEREER EEEE En Enner nnne 3 COMPILE TIME REOUIBEMENTS 3 ESEGHUIONZ eege l ist Eden a eit eth eden ete cde dec Ae Een 3 USER INTERFACE ee e ee aie ei ede Sted ee ee he 3 CONFIGURATION BASICS cc ccccccccceececcecceceeecesceeceeceeeeeeeceeceueeusceceeseeeeeeeesensenees 4 CONFIGURING DATA SOURCES ccccecceecesceeceeceeceeceeceecesceseeeceeeeeseeseeeesenseeees 5 CONFIGURING THE DATA SOURCE MANAGER 5 CONFIGURING A FIXED DATA SOURCE 6 CONFIGURING A TILED DATA Soup 7 CONFIGURING A MAGNITUDE DaraGoupck 8 CONFIGURING VISUALIZATION METHODS 0 ccceeceeceeceeceeceeeeseeeceeeeeeees 10 CONFIGURING THE VISUALIZATION MANAGER 10 re TEE 11 SCA e LEE 12 Ee ele e LEE 14 Vector Flow Glyph SE gteehgeggeegere ed ed vedeteebesensece Ehe deeded 16 FR OAM FIGION THONGS E 17 SAMPLE CONFIGURATION FILES 0ccceceeceecceceeceeceeceeeeceecenseeseeceeceeseesees 20 SAMPLE PIPS CONEIGURATION 20 SAMPLE NCOM CONFIGURATION 0eceecceceeceeeeceeeceeeeceeceecaecee
8. d e ComponentCount an integer number greater than or equal to the number of visual components e AddComponentX a header name indicating where to get information for this component X must be between 1 and ComponentCount inclusive Each header listed in an AddComponentX key must appear later in the configuration script or the program will exit with an error Each corresponding configuration block must also contain the key type and it must be one of the following e WAND_POINTER a simple line indicating the direction of the wand Not very useful except for VRJuggler configuration of the wand and screens SCALAR_POST a scalar post SCALAR_QUAD layers of scalar values VECTOR_POST a vector post VECTOR_GLYPHS layers of vector flow glyphs ROAM_FLOAT a ROAM heightfield The last five types correspond to the 5 currently supported visualization methods listed below A sample SYSTEM block would look like this SYSTEM startx 900 starty 400 Triton Technical Manual ComponentCount 10 AddComponent1 MAINROAM AddComponent2 DAT AddComponent3 DAT AddComponent4 DAT AddComponent5 DAT end Scalar Surfaces TA POSTS TA FLOW GLYPHS TA FLOW POST TA SURFACE 11 of 27 A scalar surface renders a colormapped scalar value for all data points in a rectangular subsection of a single layer centered around the user following keys e colormap an ISTV colormap file to map to this layer It uses the
9. d the transition points will change to match the current position This is typically used to indicate the flow at a single lat long position in all layers When the user is between data points the data from the nearest point is used A sample vector post block would look like DATA_FLOW POST type VECT colormap flowu_va flowv_va depth_va OR_POST temperature ct ues VECTOR U DATA ues VECTOR_V_DATA ues DEPTHS color va x_ offset v offset ues VECTOR_MAG 3 scaling 5 0 end Triton Technical Manual 16 of 27 Vector Flow Glyph Layers A vector flow glyph layer renders a colormapped scalar value onto a set of oriented lines indicating a vector value for all points in a rectangle around the user in a single layer It uses the following keys e colormap An ISTV colormap to use for rendering e x length y_length integers indicating the size of the area to render a rectangle centered around the user e color_values a data source containing the data values to which to map the colors depth_values a data source containing the depths of each data layer flowu_values a data source containing u flow values flowv_values a data source containing v flow values layerX on off if on then this layer is rendered X must be between 1 and the number of layers in the above data sources For each data point in a rectangle x_length y_length around the user in
10. ded through two Raw Binary data loaders and the magnitude of the flow can be calculated through an On The Fly Magnitude Data loader to create colormapped flow magnitude surfaces Data is also cached in memory to minimize memory consumption This also helps to minimize network traffic and hard drive accesses thereby improving performance Data is loaded on its first access and least recently data is replaced later by new data on an as needed basis Also supported is multi file tiled data like GTOPO30 where each tile is cached individually improving memory usage Each data loader supports scaling amp masking operations meaning that none of the visualization methods need to know about them Each loader also supports multiple layers but does not support multiple timesteps at this time Configuring the Data Source Manager The DATA section configures the Data Source Manager The following keys are supported e Databases an integer indicating the exact number of data sources your program uses e DataXName a header name indicating from which section to read this data source s configuration information X is a number between 1 and Databases e DataXType the type of data source to use for data source X This must be one of o TILED _FLOAT data that is read from a series of raw binary files o MAGNITUDE_FLOAT data that is calculated on the fly from two data files o FIXED data that is constant across an area and is s
11. e o O filled and textured o 1 filled and lit no texture o 2 filled No lights no texture o 3 wireframe This parameter is optional and defaults to 0 filled and textured e map_size an Integer indicating the size of the map used to adjust the resolution e gDesiredTris the Desired number of triangles per frame e FrameVariance the starting allowable variance before splitting another level e MinVariance The lowest value FrameVariance can be This parameter is optional and defaults to 0 e AutoLOD 1 0 if 1 then the FrameVariance is automatically changed each frame to optimally hit gDesiredTris ROAM renders the data as a surface beneath the user Each area is generated to a full detail texture then the texture is mapped on the lower detail geometry to ensure the color data is accurate even if the geometry is not These textures are also used to render a map in front of the user with a small cone indicating the current view This method also implements View Frustrum Culling on a per wall basis This means each wall draws only the parts of the scene visible from that side This vastly increases the framerate because each side wall can only see A of the scene and the floor can only see a tiny area when the user is bound close to the ground Several articles can be found on ROAM but this implementation has been heavily modified to support the following e Dynamic loading of data As the user moves a
12. e the same size This could cause problems for certain visual components The entire dataset is generated on the first access so there is no noticeable delay during run time for static data A sample Magnitude data source would look like this VECTOR MAG data_u VECTOR_U_DATA data_v VECTOR_V_DATA startx 0 starty 0 width 1024 height 640 layercount 20 end Triton Technical Manual 10 of 27 Configuring Visualization Methods Triton currently supports the following visualization methods e Scalar Surfaces e Scalar Posts e Vector Posts e Vector Flow Glyph Layer e Heightfields Rendered with ROAM Each of these methods will load data from any of the available data loaders and supports ISTV style colormaps 3 4 Any combination of these visualization methods can be operating at the same time and any one can be duplicated as often as wished For example the user could have a Scalar Surface at layer 7 One Scalar Surface Vector Flow Glyph layers for layers 1 through 5 5 Vector Flow Glyphs and several Scalar Posts scattered around all while rendering the context bathymetry using a ROAM Heightfield Configuring the Visualization Manager The visualization manager is responsible for creating and maintaining all of the visual components It is setup in the other required block SYSTEM It uses the following keys e startx starty integer numbers indicating the starting x and y location of the user on loa
13. eeenecuscessueateeeeass 23 RERERERNE Eeer 27 Triton Technical Manual 3 of 27 Requirements Run Time Requirements Triton v1 0 requires e An SGI with on board Texture memory OpenGL Irix 6 5 Read access to all data Configuration files for VRJuggler specific to your system Triton was designed to run on SGI Irix systems VRJuggler is compatible with other platforms such as Sun PC etc but Triton has never been tested on these systems and may not run properly Compile Time Requirements For compilation you also need e VRJuggler 1 0 refer to VRJuggler documentation for additional requirements and instructions e GNU Compiler utilities gcc g and gmake specifically e MIPSpro Compiler version 7 3 1 1m or greater Execution To run Triton after it has been compiled simply use the following syntax roamer vrjuggler config files c config Where e vrjuggler config files all of the files built with vjcontrol necessary for your VRJuggler setup e config the configuration script for your system If the configuration file is not specified then by default config tile will be loaded User Interface Once the system has started you can use the following simple user interface similar to CTHRU 1 2 To Rotate simply turn the wheel or analog axis left or right to rotate To Move Use the trigger or other analog axis to move forward and backward Triton Technical Manual 4 of 27 Config
14. end DATA_FLOW GLYPHS type VECTOR_GLYPHS colormap temperature ct x length 64 y_length 64 fFlowu_values VECTOR U DATA Clown values VECTOR V DATA depth_values DEPTHS color_values VECTOR MAG layerl on layer2 on layer3 on layer5 on end Triton Technical Manual DATA_FLOW POST type VECTOR POST colormap temperature ct flowu values VECTOR U DATA flowv_values VECTOR V DATA depth_values DEPTHS color values VECTOR MAG x offset 3 y_ offset I scaling 5 0 end DATA SURFACE type SCALAR QUAD colormap temperature ct x length 128 y length 128 data values SCALAR DATA depth_values DEPTHS layer7 on end DATA_POSTS type SCALAR POST x_offset 3 0 y_offset 1 0 colormap temperature ct data_values SCALAR DATA data_depths DEPTHS end MAINROAM type ROAM FLOAT colormap terrain ct data BATHYMETRY data_layer 1 color shared patchsize 32 viewsquare 9 variance depth 9 max nodes 50000 render 0 map size 65536 gDesiredTris 8000 FrameVariance 50 MinVariance 0 0 AutoLOD 1 end BATHYMETRY scale 0 005 landmask 1 tilecount 1 cache 1 tilel PIPS layers 1 24 of 27 Triton Technical Manual flip x yes end PIPS data NCOM model bathymetry bin startx 0 starty 0 width 1024 height 640 end SCALAR DATA cache 20 tilecount 1 flip x yes tilel TEMPERATURE layers 20 maskop lt
15. is optional and if not specified no action is taken e spread a floating point number indicating the desired range between the minimum and maximum value If specified it automatically calculates a scaling factor to force the minimum and maximum to be this far apart This parameter is optional and if not specified no action is taken Triton Technical Manual 8 of 27 The data is multiplied by scale before the minimum and maximum is calculated Masked values are ignored during this calculation If andvalue is specified a separate pass is made next to write this value in so the andvalue needs to be scaled in the configuration block Another pass is made through the scaled and masked data to set the minvalue and spread if specified The data keys minvalue and spread are useful when you do not know the range of the data that you are loading and want to force it within a certain range ie force a minimum value of minvalue and a maximum value of minvalue spread For each tileX a configuration block needs to exist detailing the exact specifications of the file The keys for each of those blocks looks like this e data the actual path and filename of the data file no spaces e startx starty integer numbers the x and y location of the data in object space e width height integer numbers indicating the width and height of the data in object space In this context object space refers to the plane below the user Usually
16. nd El DATA Databases 3 DatalName SCALAR_DATA DatalType TILED FLOAT Data2Name DEPTHS Data2Type FIXED Data3Name BATHYMETRY Data3Type TILED FLOAT end POINTER type WAND POINTER width 0 3 length 3 0 color red 1 0 color green 0 0 color blue 0 0 end DATA SURFACE type SCALAR QUAD colormap temperature ct x length 128 y_length 128 data_values SCALAR DATA depth_values DEPTHS layerl2 on end DATA_POSTS type SCALAR POST x offset 2 0 v offset 1 0 colormap temperature ct data_values SCALAR DATA data_depths DEPTHS end MATNROAM Triton Technical Manual type ROAM FLOAT colormap terrain ct data BATHYMETRY data_layer 1 color shared patchsize 32 viewsquare 9 variance depth 9 max nodes 50000 render 0 map size 65536 gDesiredTris 8000 FrameVariance 50 MinVariance 0 0 AutoLOD 1 end BATHYMETRY scale 0 005 landmask 1 tilecount 1 cache 1 tilel PIPS layers 1 end PIPS data PIPS bathymetry bin startx 0 starty 0 width 1280 height 720 end SCALAR DATA cache 45 tilecount 1 tilel TEMPERATURE layers 45 landmask 1 e 30f end TEMPERATURE data PIPS flat potential_temp bin startx 0 starty 0 width 1280 height 720 end DEPTHS fixed_depths scale 0 005 layercount 45 depth1 25 depth2 125 depth3 E depth4 Las 21 of 27 Triton Technical Manual 22 of 27 depth5
17. pecified in the configuration script A sample DATA section would look like this DATA Databases 5 DatalName BATHYMETRY DatalType TILED FLOAT Data2Name DEPTHS Data2Type FIXED Triton Technical Manual 6 of 27 Data3Name VECTOR U DATA Data3Type TILED FLOAT Data4Name VECTOR_V_DATA Data4Type TILED FLOAT DataSName VECTOR MAG Data5Type MAGNITUDE FLOAT end Each DataName must correspond to a Header somewhere in the configuration script If it does not the program will quit in an error Data sources that you do not use may be defined as they will only be loaded when accessed by a visualization method Configuring a FIXED Data Source A fixed data source is defined inline in the configuration script It creates a data source that returns a fixed value for any location within a layer of data This is used most often to list depths for use by Vector Posts and Scalar Posts Only a few keys are used here e fixed_depths_scale a floating point number indicating a scale to multiply all following numbers by Useful if you frequently scale your data differently and don t want to recalculate all of the depths This key is optional defaults to 1 0 e layercount an integer number indicating the number of data values e depthx a floating point number indicating the value to return for layer X A depthxX key must exist for each X between 1 and ayercount A sample Fixed Data source would
18. position in all layers Instead of being a flat fixed width post though the surface bends around the striped post to reflect two other scalar fields defined by the user It has the following keys e x offset y_offset floating point numbers that indicate relative to the user s position where to place the post These coordinates are in the data s object space so the post will move around the user as the user rotates e colormap an ISTV colormap to use for rendering e color_values a data source containing the data values to which to map the colors e depth_values a data source containing the depths of each data layer e scaling a floating point number by which to multiply the widths to make them more visible The screenshot above used 5 0 This parameter is optional and defaults to 1 0 e flowu_values a data source containing u flow values e flowv_values a data source containing v flow values Triton Technical Manual The data in color_values at x y each depth_values is rendered as a ribbon spanning all layers of the data with the width and direction determined by scaling flowu_values scaling flowv_values Smooth shading is used to make the colors smoothly transition So that the user can tell the exact extent of the layers of depth_values a striped post is drawn next to it with the transitions indicating the exact layers depth_values is usually a Fixed Data source but can be a varying data source an
19. round data moves off one edge of the meshable area and new data must be loaded This implementation only loads new data when required where most load all data at startup e Separate tessellation and rendering Most methods tessellate into triangles in a binary tree and render directly from the tree The implementation herein achieves an order of magnitude speedup from converting the tree into a simple array list before rendering and rendering from that Triton Technical Manual 19 of 27 e Per Context View Frustrum Culling Textures are only generated and loaded for areas that can be seen from the current wall and surfaces are only rendered if it is possible they could be seen from there e Several other optimizations through the use of lookup tables ROAM is usually used to render the bathymetry for contextual information A sample ROAM block would look like this MAINROAM type ROAM FLOAT colormap terrain ct data BATHYMETRY data_layer 1 color shared patchsize 64 viewsquare 9 variance depth 9 max nodes 50000 render 0 map size 65536 gDesiredTris 8000 FrameVariance 50 MinVariance 0 0 AutoLOD 1 end Triton Technical Manual 20 of 27 Sample Configuration Files Sample PIPS configuration This configuration file loads up bathymetry one scalar post and one scalar surface SYSTEM startx 900 starty 400 ComponentCount 10 AddComponent1l MAINROAM AddComponent2 DATA POSTS AddComponent3 DATA _SURFAC e
20. startx amp starty are 0 However in a system with multiple adjacent data tiles sfartx A starty will indicate each tile s starting point within the whole data set All following examples use a single tile data system A sample tiled data source block would look like BATHYMETRY scale 0 005 landmask 1 tilecount 1 cache 1 tilel PIPS layers 1 flip x yes end PIPS data NCOM model bathymetry bin startx 0 starty 0 width 1024 height 640 end Configuring a MAGNITUDE Data Source A Magnitude data source takes two data sources and uses them to calculate a scalar magnitude point by point to generate another data source This is useful if your using vector data and want to colormap by magnitude but don t have a separate magnitude dataset Triton Technical Manual 9 of 27 A magnitude data source uses the following keys data_u a data source listed BEFORE this one containing scalar u values data_v a data source listed BEFORE this one containing scalar v values startx starty integer numbers indicating the starting points within data_u amp data_v width height integer numbers indicating the area to read from data_u amp data_v layercount an integer number indicating the number of layers to generate from data_u amp data_v If startx and starty are not 0 or width and height are not equal to the width and height of data_u and data_v then the resulting dataset will not b
21. that indicate where to place the post relative to the user s position These coordinates are in the data s object space so the post will move around the user as the user rotates e colormap an ISTV colormap to use for rendering e data_values a data source containing the actual data values to render e depth_values a data source containing the depths of each data layer The data in data_values at x y each depth_values is rendered as a post penetrating all layers of the data Smooth shading is used to make the colors smoothly transition So that the user can tell the exact extent of the layers of depth_values a striped post is drawn next to it with the alternating color transitions indicating the exact layer extents data_heights is usually a Fixed Data source but can be a varying data source and the transition points will change to match the current position This visualization method is typically used to visualize a variable such as temperature that can vary significantly from layer to layer in value When the user is between data points the data is clamped to the nearest data point A sample scalar post block would look like this DATA_POSTS type SCALAR POST x offset 3 0 y_ offset 1 0 colormap temperature ct data_values SCALAR DATA data_depths DEPTHS end Triton Technical Manual 14 of 27 Vector Posts A vector post renders a colormapped scalar value for a single x y position relative to the user s
22. the layerX a line is drawn from the grid point in the direction of flowu_values flowv_values The line is colormapped by the values in color_values and is the length Triton Technical Manual 17 of 27 sqrt flowu_values flowu_values flowv_values flowv_values color_values is usually a magnitude data source calculating the magnitude of the vector at each point A sample vector flow glyph block would look like DATA FLOW GLYPHS type VECTOR GLYPHS colormap temperature ct x_length 64 y_length 64 Clown values VECTOR _ U DATA flowv values VECTOR V DATA depth_values DEPTHS color_values VECTOR_MAG layerl on layer3 on layer5 on end ROAM Heightfields ROAM was used to render the bathymetry in the previous images Probably the most complex of all the visualization methods ROAM Heightfields let you render a heightfield of any user defined quality at interactive framerates ROAM is a run time level of detail algorithm that stands for Real Time Optimally Adapting Meshes It recursively subdivides a surface into triangles until either a certain detail is achieved or a certain number of triangles is obtained By changing this upper bound one can ensure that the framerate stays high no matter how complex the geometry is It recognizes the following keys e colormap an ISTV colormap mapped onto the surface e data the data source from which to read the heightfield e data_layer
23. the layer of data to use This parameter is optional and defaults to 1 e color the data source to use for generating colors NOTE if shared is specified here then co or will be the same as data This is much faster than actually listing color to be the same as data e color_layer the layer of color to use This parameter is optional and defaults to 1 e patchsize the patch size to use Must be a positive integer power of 2 16 32 64 etc This sets the minimum resolution the surface will always render these points Large patch sizes will degrade visual quality while small patch sizes will increase computation time e viewsquare the area to visualize It must be an odd integer A square of viewsquare viewsquare patches will be rendered centered around the user Triton Technical Manual 18 of 27 e variance_depth an integer indicating the depth of the variance tree This should be sqrt patchsize 1 Setting it lower will reduce the number of trianges tessellated correctly Setting it higher simply wastes memory but does not impact performance If this number is set lower than recommended then when a surface is split variance_depth levels it will automatically split it the remaining levels regardless of requirements e max_nodes an integer sets the maximum number of splits This should be much higher than the number of triangles that you want e render an integer specifying the rendering method to us
24. uration Basics Configuration takes place through text files called configuration scripts A configuration script is formed from several blocks each looking something like this HEADER key1 value1 key2 value2 key3 value3 end The Header key and value fields can contain any printable characters except white spaces The Header is enclosed in square braces and for readability are usually all caps Keys vary depending on the type of section this Header represents Headers keys and default values are listed below in their corresponding sections The end key is critical as it indicates the end of a configuration block Without it blocks will run together Sometimes this may be useful to avoid duplicating sections of the file but usually it is not Only two Headers are required for Triton SYSTEM and DATA These sections are discussed in more detail later Keys and headers listed below are case specific Keys are actively searched for not parsed This means that if you insert unrecognized keys there will be no error Triton Technical Manual 5 of 27 Configuring Data Sources Triton supports the following data formats e Raw Binary rectilinear data e ASCII Fixed Data e On The Fly Magnitude Data calculated from other data sources These data formats are specified at run time in the configuration files and can be mixed and matched to form a very flexible data set For example U flow and V Flow files can be loa
Download Pdf Manuals
Related Search