Abstract: A graphics system configured to apply multiple layers of texture information to primitives. The graphics system receives parameters defining a primitive and performs a size test on the primitive. If the size test cannot guarantee that a fragment size of the primitive is less than or equal to a fragment capacity of a texture accumulation buffer, the primitive is divided into subprimitives, and the graphics system applies the multiple layers of texture to fragments which intersect the primitive. The graphics system switches from a current layer to the layer next when it has applied textures corresponding to the current layer to all the fragments intersecting the primitive. The graphics system stores color values associated with the primitive fragments in the texture accumulation buffer between the application of successive texture layers.

Abstract: The invention relates to a method of reproducing a gray scale image (1) in colors on a color monitor (6), in which method each shade of gray (2) is assigned, using a look-up table (4, 5), a given output value (R, G, B) for driving the electron guns (8R, 8G, 8B) of the color monitor. Assignment is performed in such a manner that a given color deviation is imparted to successive shades of gray, which deviation enables differentiation of the shades of gray by a human observer without such deviations becoming disturbing. Consequently, the information concerning the optical density of the original gray scale image (1) is preserved; this is important notably for the interpretation of medical images. Additionally, the special display properties of the monitor (6) and/or the ambient circumstances can be taken into account in the look-up table (4, 5) in order to optimize the reproduction of the gray scale image.

Abstract: A first texture is used for determining color and design of a polygon structuring an object rendered upon a two-dimensional screen. A second texture has a pattern of dense distribution of color with a predetermined slant relative to the two-dimensional screen. An rendering processing device first applies a first texture to a polygon structuring an object, and thereafter performs translucent synthesis of a second texture on an object applied with the first texture, thereby making it possible to easily render an image in a hand-drawn illustration style in, for example, home video games and computer graphics.

Abstract: The present invention provides methods and systems to facilitate pattern recognition in complex biological data using component plane presentations of clustered data.

Abstract: In a selected area of a display defined by a polygon, magnifying is simulated. The selected area may be, for example, a circle simulating a magnifying glass. Textures are represented by texel coordinates U and V, which specify the location of color components within a set of image data. Within the area selected to appear magnified, the present invention perturbs the texel location selection to simulate an angle of refraction in the selected area and offset texel coordinates.

Abstract: A cache memory for a texture mapping process which is applicable to a high performance three-dimensional graphics card for a personal computer, three-dimensional game machines and other fields requiring small and high performance three-dimensional graphics. In particular, in order to accelerate a texture mapping process based upon a hardware-used mipmapping process using a trilinear interpolation in a three-dimensional graphics system, there is provided a cache memory in which only textures by a moderate size of a working set are stored, and all eight texels needed to perform a trilinear interpolation only in one clock cycle are accessed to obtain a final texel value, and a method enabling a reduction in penalty due to a cache miss by, with hardware-based prediction, prefetching textures to be needed in the future.

Abstract: A method of processing images in images comprising three-dimensional objects, the method comprising a step of mapping a two-dimensional image on a face of a three-dimensional object, a step of interaction between a user and the two-dimensional image mapped on the three-dimensional image, intended to enable a user to displace the whole or part of the two-dimensional image on the face of the three-dimensional object. This method enables the user to map a fixed 2D image or a video image on a face of a 3D object, and to displace the image as he wishes on the face of the object he has selected. If the image is constituted by several parts, for example, the parts of a puzzle, the user can independently displace each part of this image, for example, so as to reconstitute the initial image of the puzzle.

Abstract: A system and a method for improving magnified texture-mapped pixel performance in a single-pixel pipeline. Two textured pixel addresses corresponding to two pixels may be generated. The two textured pixel addresses may then be passed to the next unit in the pipeline, where the two textured pixel addresses can be examined if the corresponding two pixels correspond to a common set of texels in texture space. The two textured pixel addresses may be merged together if the two pixels correspond to the common set of texels. Merging may operate to create a combined texel structure. Texel data may be generated in response to receiving the combined texel structure. The texel data may be filtered using one or more texture filters in order to generate texture values.

Abstract: A method is for producing the bump mapping effect for a 3D object in a computer graphic. First, a shading vector is produced at one point of the surface of the 3D object. A shift coordinate with respect to this point is determined, according to the shading vector. According to the shift coordinate, a rotational calculation matrix is set up. Then, a diffuse color value with respect to this point is computed, which is equal to the difference between an adjacent rotation color value to this point and an averaged color value, wherein the averaged color value is defined as the sum and average of the color channels for R, G, and B with respect to this point, and the adjacent rotation color value is defined as a sum of the averaged color value with respect to a number of the adjacent points and the coefficient product with respect to the adjacent points in the rotational calculation matrix. Then, the emboss image value at this point is computed.

Abstract: A method of performing anisotropic texture mip-mapping. The method includes determining a region of support for a set of target pixels of the image to be textured, and mapping the region of support to an area in texture map that is generally elliptical. For each axis of the ellipse the number of samples is determined and a filter function is performed on those samples to find the final color value. For four texels, the filter function is a weighted sum of the color values of each texel, where the weights are determined based on the fraction of the Level of Detail (LOD) and the fraction of the U or V coordinate.

Abstract: A system, method and computer program product are provided for computer graphics processing. Initially, a height parameter is determined. Thereafter, a depth-direction component of the height parameter is calculated. A depth-value of a pixel is then modified utilizing the computed depth-direction component of the height parameter.

Abstract: A system and method is provided for preventing the occurrence of aliasing at the edges of polygons in 3D graphics. The system may detect both polygon geometric edges and Z edges due to intersection of multiple polygons. In one embodiment, the system includes an edge anti-aliasing module configured to selectively super-sample edge portions of primitives. The system further includes a coarse memory for storing information of pixels that are not super-sampled and a fine memory for storing information of pixels that are super-sampled by the edge anti-aliasing module.

Abstract: A variable number of textures are blended together using a single texture as a mask. At least four textures are received. Masks are extracted from one of the received textures and used to blend together the remaining textures. In an embodiment, N masks are extracted from a single texture and used to blend N+1 additional textures. In this embodiment, two of the N+1 textures are initially blended together in accordance with one of the N masks to form an image. Another texture of the N+1 textures is then blended with the image in accordance with another one of the N masks. This iterative blending process continues until all of the N+1 textures have been blended together. In another embodiment, N textures are blended together by multiplying each of the N textures by one of the N masks and adding together the results of the N multiplications.

Abstract: A system, method and article of manufacture are provided for efficient storage of texture data in memory for use with a computer graphics pipeline. Provided is a data structure including at least one compressed sub-block representing a group of texels in a predetermined image plane and at predetermined locations in a first and a second dimension in a texture map. A number of the sub-blocks is based on a depth of texture data in the texture map.

Abstract: A system and method are provided for creating a vector map in a hardware graphic pipeline. Initially, one of a plurality of transforms is selected in a hardware graphic pipeline. Further, input is processed in order to generate a vector map utilizing the selected transform in the hardware graphics pipeline. Subsequently, a plurality of pixel color values is rendered utilizing the vector map.

Abstract: The present invention involves a new system and method for synthesizing textures from an input sample. A system and method according to the present invention uses a unique accelerated patch-based sampling system to synthesize high-quality textures in real-time using a small input texture sample. The patch-based sampling system of the present invention works well for a wide variety of textures ranging from regular to stochastic. Potential feature mismatches across patch boundaries are avoided by sampling patches according to a non-parametric estimation of the local conditional Markov Random Field (MRF) density function.

Abstract: A method, system, and computer program product for filtering textures applied to a surface of a computer generated object permits an application program running on a computer system to significantly increase the graphics capabilities and performance of the computer. Rendering data for a pixel of the object is received from the application program, and a first and second set of texture coordinates is generated. Next, the first and second sets of texture coordinates are used to obtain a first and second texture sample from a texture image. The first and second texture samples are then blended together to produce a texture sample having a greater degree of filtering. This produced texture sample having a higher degree of filtering is stored in a frame buffer for subsequent display.

Abstract: Methods for rendering a three-dimensional (3D) scene generated in a field of view having in-focus and out-of-focus regions on a two-dimensional (2D) screen region of pixels are described. One method includes initially rendering the scene to create color and depth texture maps and creating mip-map layers for the color texture map. The method further comprises subsequently rendering the scene by, for each pixel: creating a mip-map layer selection value as a function of a depth of the pixel from the depth texture map, generating a color value by interpolation using color values from at least one of the mip-map layers chosen according to the mip-map layer selection value, and setting a color of the pixel to the generated color texture. A graphical processing unit (GPU) for rendering a three-dimensional (3D) scene generated in a field of view having in-focus and out-of-focus regions on a two-dimensional (2D) screen region of pixels is also described.

Abstract: Terrain is rendered in a three-dimensional computing environment by attaching a number of fixed terrain layers to the viewer, in a configuration such that each layer resembles a “magic carpet” on which the viewer flies. Each concentric layer covers a successively larger area of the visual database, and the vertex density and texture resolution of each layer is independent of other layers. Each layer may feature a fading band around its circumference, which can facilitate fading between consecutive layers. This approach achieves optimal vertex density/range distribution, which enables both very far horizons (exceeding 100 nmi) and very high elevation detail in the foreground, while maintaining strict performance requirements. High texture resolution and sparsely populated terrain texture are also achieved without the use of specialized hardware because each layer is textured independently.

Abstract: A texture picture with a highest resolution is selected among a plurality of pictures taken from different points of view for each patch surface, and texture mapping is carried out by pasting a texture picture selected for a patch surface on the patch surface. Then, processing is carried out to correct differences in concentration, which are caused by differences in beam-source condition, between pictures or between patch surfaces. As the correction processing, inter-frame picture-concentration-value correction, inter-patch texture-picture-shift correction and inter-patch boundary concentration smoothing correction are performed to allow texture mapping with a high picture quality to be carried out to produce a three-dimensional picture having a high quality. It is thus possible to provide a texture mapping technique capable of producing a three-dimensional picture having a high picture quality.

Abstract: A method and system are disclosed for creating three-dimensional models in a graphics texture mapping system and are especially useful for creating three-dimensional urban models. A computer includes a graphics application program and an application programming interface. A single rectangular, composite building texture map is formed for each of a plurality of three-dimensional objects corresponding to individual buildings and have a reduced area of pixels within each composite building texture map as caused by rotating individual images. A site model clip map image contains each composite building texture map in a spatially correct position that has been expanded such that no overlap remains.

Abstract: A method and an apparatus for retrieving a mipmap from memory. The method and apparatus provide an efficient method of determining the location of the desired mipmap in memory by storing the address of each row of mipmaps and calculating the offset from the start of the row to the desired mipmap. The mipmap is retrieved from memory at the location corresponding to the sum of the start address and the offset.

Abstract: Methods and apparatus are provided for allocating correlated data sets, such as texture data, among first and second areas of memory in a computer graphics system. Each texture map in a series of texture maps is divided into a set of blocks of data. Each texture map that has a width greater than one block is divided into first and second map areas. Typically, the first and second map areas are the left and right halves of each texture map. Blocks of data from the first map areas of odd level texture maps are stored in the first memory area, blocks of data from the second map areas of even level texture maps are stored in the first memory area, blocks of data from the second map areas of odd level texture maps are stored in the second memory area and blocks of data from the first map areas of even level texture maps are stored in the second memory area. The blocks of data representing each texture map in the series of texture maps are stored in consecutive blocks of memory.

Abstract: A system and method for selecting a cartographically-preferred label position from a plurality of potential label positions for a feature on a computer-generated map. The mechanism analyzes each pixel within each potential label position to determine the presence of a colliding label or a colliding feature, and calculates a total penalty for each potential label position. Then, based on the several total penalties, the mechanism determines which of the potential label positions is the cartographically-preferable label position and places the label in that label position.

Abstract: An input two-dimensional image is divided into equal-size blocks. MIP maps with different resolutions are generated in response to each of the blocks. A memory stores data representative of the MIP maps for each of the blocks. Polygon data represent conditions of a polygon and also a correspondence between the polygon and positions of pixels of a two-dimensional image to be applied to the polygon. On the basis of the polygon data, at least one is selected from the MIP maps represented by the data in the memory as a desired MIP map on a pixel-by-pixel basis. Data representative of the desired MIP map are read from the memory. Calculation is made as to a color intensity of each pixel in a final image in which the desired MIP map is applied to the polygon on a pixel-by-pixel basis. Data representative of each calculated pixel color intensity are outputted.

Abstract: The present invention provides a method and apparatus for color adjustment of a display screen which can provide an interactive user interface which allows a user to perform color adjustment efficiently and effectively. The color adjustment of the display screen in a color display device, such as LCD, CRT or the like, is performed by displaying a color image without any color adjustment at a certain portion of the display screen, displaying a color image with a color adjustment at another portion of the display screen, and referring to the color image without any color adjustment. And to implement this method for color adjustment, the apparatus is configured so that the color images which have passed the color adjustment block and have their colors adjusted, and the color images which have not passed the color adjustment block and are not subject to the color adjustment can be both displayed on the same display screen.

Abstract: Methods, systems, and computer program products for blending textures used to render computer generated images are provided. In an embodiment of the invention, a MIP-mapped mask texture is constructed. Each MIP-level of the MIP-mapped mask texture includes texels representative of different mask information. The MIP-mapped mask texture is sampled during rendering to obtain mask information. The obtained mask information is used to blend between textures. The invention is used to blend, for example, between multiple textures wherein, zero, one, or more of the textures are MIP-mapped and/or between different levels of one or more three-dimensional textures. In an embodiment, the most appropriate texture amongst multiple textures, each providing coverage at different resolutions, is selected for a fragment being rendered, thereby avoiding texture scintillation.

Abstract: A system and method for generating multiple potential label positions for a polygon at run time, rather than selecting from pre-configured label positions. The present invention analyzes the visible portion of the polygon to generate potential label positions dynamically based on the visible portion of the polygon, rather than using pre-configured label positions. As many rectangles are created as can fit within the visible portion of the polygon, and which can support the placement of a label. The rectangles are sized in accordance with the size of the label to be displayed. Once the rectangles are created, the present invention allows for the selection of one of the rectangles as a preferred label position for the polygon. The present invention also allows for the generation of a form following baseline created by connecting the centerpoints of horizontally disposed rectangles, and then smoothing the resulting baseline until a desirable baseline is obtained.

Abstract: A graphics system including a custom graphics and audio processor produces exciting 2D and 3D graphics and surround sound. The system includes a graphics and audio processor including a 3D graphics pipeline and an audio digital signal processor. The same texture mapping hardware used for color texturing provides resampled z texturing for sprites with depth or other applications. A z blender performs a z blending operation in screen space to blend surface z values with z texel values to provide per-pixel mapping of resampled z textures onto sampled 3D surface locations. Z texels can represent absolute depths or depth displacements relative to primitive surface depth. The z texel values may add to or replace primitive surface z values, and a constant bias may be added if desired. The resulting depth values are used for occlusion testing. Z textures can be generated by copying out portions of an embedded z buffer and providing the copied depth values to the texture mapping hardware.

Abstract: A system, method and computer program product are provided for performing shader calculations in a graphics pipeline. Initially, a shading calculation is performed in order to generate output. Thereafter, an additional shading calculation is carried out. Such additional shading calculation includes converting the output of the shading calculation into a floating point format. Further, a dot product is calculated utilizing the converted output and texture coordinates. The dot product is then clamped. Next, the clamped dot product is stored in a plurality of color components.

Abstract: A first copy of an object is rendered using a texture sample selected from a texture image. This texture sample is selected from the texture image according to a first set of texture coordinates. The rendered object is stored in a frame buffer. Next, a second copy of the object is rendered using a second texture sample selected from the texture image. The second texture sample is selected from the texture image according to a second set of texture coordinates calculated in accordance with the first set of texture coordinates and one or more Jitter factors. The second set of calculated texture coordinates is displaced from the first set of texture coordinates along an axis of anisotropy. This second rendered copy of the object is then blended with the first rendered copy of the object to produce an object with anisotropic filtering. In embodiments of the invention, more than two copies of the object are rendered and blended together to form an object with anisotropic filtering.

Abstract: A method and apparatus for utilizing mip maps in a video graphics system begins by setting a dynamically configurable level of detail bias that is used to select between potential mip maps. The level of detail bias is set based on the screen resolution. The selection of the mip map, or mip maps, utilized for texturing operations with respect to a particular pixel is performed based on the configurable level of detail bias and the texel-per-pixel ratio between the potential mip maps and the particular pixel to be textured. The dynamic configuration of the level of detail bias allows texture detail to be maintained across multiple display resolutions.

Abstract: A graphics accelerator which includes a dedicated virtual memory manager which manages at least some host memory, as well as dedicated graphics memory, and which manages memory during mipmapping using at least two separate pools of memory.

Abstract: Method and apparatus for rendering texture to an object to be displayed on a pixel screen display. This technique makes use of linear interpolation between perspectively correct texture address to calculate rates of change of individual texture addresses components to determine a selection of the correct LOD map to use and intermediate texture addresses for pixels of the object between the perspectively correct addresses. The method first determines perspectively correct texture address values associated with four corners of a predefined span or grid of pixels. Then, a linear interpolation technique is implemented to calculate a rate of change of texture address components in the screen x and y directions for pixels between the perspectively bound span corners. This linear interpolation technique is performed in both screen directions to thereby create a potentially unique level of detail value for each pixel, which is then used as an index to select the correct pre-filtered LOD texture map.

Abstract: A method and apparatus for managing texture mapping data in a computer graphics system, the computer graphics system including a host computer, primitive rendering hardware and a primitive data path extending between the host computer and the primitive rendering hardware. The host computer passes primitives to be rendered by the system to the primitive rendering hardware over the primitive data path. The host computer has a main memory that stores texture mapping data corresponding to the primitives to be rendered. The primitive rendering hardware includes a local texture memory that locally stores the texture mapping data corresponding to at least one of the primitives to be rendered. When a primitive passed to the primitive rendering hardware is to be rendered, a determination is made as to whether its corresponding texture mapping data is in the local texture memory.

Abstract: A graphic processing apparatus which can provide a high quality picture with a small hardware configuration, comprising a DRAM for storing a plurality of texture data corresponding to different reducing rates, a representative point determining circuit for determining a pixel as a representative point from among pixels positioned inside a unit graphic to be processed among a plurality of pixels simultaneously processed, a reducing rate calculation circuit for determining a reducing rate from the homogeneous coordinates and a homogeneous term corresponding to the determined representative point, and a texture data reading circuit for reading from the DRAM the texture data specified by the determined reducing rate, homogeneous coordinates, and homogeneous term for a plurality of the pixels simultaneously processed.

Abstract: A circuit arrangement and display apparatus are described for use in 3-D graphics, where 2-D texture mass stored at different resolutions in a pyramidal array are indexed by a pair of texture coordinates and an associated level coordinate (L). In order to introduce depth cues to the 2-D transformed image of the 3-D environment, a focus depth (F) for the image is specified and those image components having a depth (z) other than the focus depth have their texture blurred to direct the viewer's eye to the depth of interest. The texture blurring is introduced by applying an offset (FS) to the level coordinate (L) indexing the pyramidal array in an amount determined by the distance to the focus depth.

Abstract: In order to store texture images in a smaller data amount, there is provided a method for recording a texture recording image 28 which contains a number of texture images in a memory to thereby record the images of the number of texture in the memory. In this method, the texture images are arranged on the texture recording image 28 based on the shape information of each texture such that a region not being occupied by the images can be reduced on the texture recording image 28.

Abstract: A graphics system comprises a texture memory, a rendering engine, a sample buffer and a filtering engine. The rendering engine renders received primitives based on a render pixel array whose vertical and horizontal resolutions are dynamically programmable. The rendering engine determines render pixels that geometrically intersect a primitive. For each intersecting render pixel, a texture access may be required (if texture processing is turned on) to determine texture values. The texture values may be used to compute sample values at sample positions interior to the sample render pixel and the primitive. A controlling agent may decrease the vertical and horizontal resolutions of the render pixel array to control frame render time. The filtering engine may programmably generate virtual pixel centers covering the render pixel array. Any change in the render pixel resolutions may require an accommodating change in the virtual pixel array parameters.

Abstract: A 3D-graphics engine has several texture maps with different levels of detail (LOD). The largest of the four derivatives of the u,v texture-map coordinates with respect to the x,y screen coordinates determines which LOD texture map to select. Using bi-linear interpolation, the four nearest texture pixels or texels are fetched from the texture map in a texture memory and a weighted-average texel generated. Distortion in space and time can be visible when a triangle transitions from one LOD texture map to the next LOD map. Tri-linear interpolation eliminates this LOD-transitioning distortion by generating weighted-average texels for both the LOD map and for four texels from a next LOD map. Unfortunately the calculational complexity is more than doubled for tri-linear rather than bi-linear interpolation. Tri-linear interpolation is employed only near a transition to a next LOD map. When the derivatives are not near an LOD-map transition, only bi-linear interpolation is performed.

Abstract: A method and system for providing a graphical image on a display is disclosed. The image is provided from data describing at least one object. The display includes a plurality of pixels. Each of the plurality of pixels has a size and a plurality of display elements. Each of the plurality of display elements has a color. The data includes a plurality of fragments for the at least one object. The plurality of fragments intersects a portion of the plurality of pixels. Each of the plurality of fragments includes a texture and at least one color. The method and system include ensuring that a texture area corresponds to the size of the pixel for the plurality of fragments and taking a plurality of samples of the at least one color for each of the plurality of fragments. The plurality of samples corresponds to the plurality of display elements. The method and system also include processing the texture for each of the plurality of fragments using the texture area.

Abstract: The present invention provides a picture processing apparatus and method comprising a displacement computing unit which computes displacements in a UV coordinate system of a 2-dimensional picture texture to be pasted on a multidimensional figure from displacements in a texture coordinate system of the figure and an XY coordinate system onto which the figure is mapped and supplies the computed displacements to a displacement adjusting unit for adjusting the displacements by multiplying them by adjustment coefficients to produce adjusted displacements which are then supplied to a shrinkage-factor computing unit for computing a shrinkage factor to be used as a shrinkage factor of MIPMAP filtering.

Abstract: A method for MIP mapping index texture. This invention has index texture with texel index values which refer to physical material properties. First, a lookup table is stored and the lookup table has table entries defining material types. Each table entry has an index, and material property type values. Then at least two texels are selected from the index texture, where each texel has index values corresponding to table entries. Next, the material property type values are averaged for each separate property type from the table entries for the selected texels. This produces an average material property value for each material property type. Another step is selecting a new material index based on the material which most closely matches the average material property values. The final step is generating the next lowest MIP level by using the new material indexes to form a new index texture with fewer texels.

Abstract: A trilinear MIPmap filtering technique wherein the LOD bias is derived from both major axis and minor axis minification.

Abstract: A method and apparatus for rendering lightpoints is provided. For the method of the present invention, a programmer creates a series of texture maps. Each texture map approximates the lobe of a lightpoint at a respective distance from the lightpoint. Each texture map includes transparency texture information. This allows the lightpoint to correctly model fog and other atmospheric conditions. The series of texture maps are encoded in a mipmap associated with the lightpoint. During use, a simulation environment renders the lightpoint using a billboarding technique. The billboarding technique keeps the lobe of the lightpoint oriented towards the eye point. The simulation environment dynamically tracks the distance from the lightpoint to the eye point. Each time the distance changes, the simulation environment selects an appropriate texture map from the mipmap. The appropriate texture map is the texture map that correctly depicts the lightpoint at the distance between the eye point and the lightpoint.

Abstract: A system and method for supporting texture patterns larger than natively supported by a graphics processor divides the texture pattern into quadrants and then decomposes object triangles into clipped triangles that fit completely within the quadrants.

Abstract: A method and circuit for determining the address of texture maps in memory, when only the base address of the primary texture map is known. The various maps associated with a given texture are sized and stored in a manner that allows any texel in any of the maps to be located based on the map number and the base address of the primary map. A circuit is provided that determines the necessary addresses with minimal calculations.

Abstract: An apparatus and method for quickly and efficiently providing texel data relevant for displaying a textured-image. A large amount of texture source data, such as photographic terrain texture, is stored as a two-dimensional or three-dimensional texture MIP-map on one or more mass storage devices. Only a relatively small clip-map representing selected portions of the complete texture MIP-map is loaded into faster, more expensive memory. These selected texture MIP-map portions forming the clip-map consist of tiles which contain those texel values at each respective level of detail that are most likely to be mapped to pixels being rendered for display based upon the viewer's eyepoint and field of view. To efficiently update the clip-map in real-time, texel data is loaded and discarded from the edges of tiles.

Abstract: In order to store texture images in a smaller data amount, there is provided a method for recording a texture recording image 28 which contains a number of texture images in a memory to thereby record the images of the number of texture in the memory. In this method, the texture images are arranged on the texture recording image 28 based on the shape information of each texture such that a region not being occupied by the images can be reduced on the texture recording image 28.

Abstract: A method and apparatus for managing texture mapping data in a computer graphics system, the computer graphics system including a host computer, primitive rendering hardware and a primitive data path extending between the host computer and the primitive rendering hardware. The host computer passes primitives to be rendered by the system to the primitive rendering hardware over the primitive data path. The host computer has a main memory that stores texture mapping data corresponding to the primitives to be rendered. The primitive rendering hardware includes a local texture memory that locally stores the texture mapping data corresponding to at least one of the primitives to be rendered. When a primitive passed to the primitive rendering hardware is to be rendered, a determination is made as to whether its corresponding texture mapping data is in the local texture memory.