会员体验
专利管家(专利管理)
工作空间(专利管理)
风险监控(情报监控)
数据分析(专利分析)
侵权分析(诉讼无效)
联系我们
交流群
官方交流:
QQ群: 891211   
微信请扫码    >>>
现在联系顾问~
热词
    • 32. 发明公开
    • Method for graphic display
    • Verfahrenfüreine graphische Anzeige。
    • EP0277832A2
    • 1988-08-10
    • EP88300942.5
    • 1988-02-04
    • Hewlett-Packard Company
    • Fiasconaro, James G.
    • G06F15/72G09G1/06
    • G06T17/00G06T17/20G09G1/06
    • A graphics accelerator (91) responds to commands from a computer (86) in a graphic system by storing the definitions of non-uniform rational B-spline patches (12,13,14,15) and their associated trimming curves S₁-S₉). The graphics accelerator then produces device coordinates for trimmed polygons computed for each patch and sends these polygons to a display (94). Various improvements are incorporated to minimize the effects of roundoff error. The B-spline definitions of the trimming curves (27) in the uv parameter space of each patch are converted to approximating short straight line segments. Untrimmed polygon vertices, the end points of the straight line segments and the intersections of the straight line segments with subspan boundaries corresponding to polygon edges are kept in a data structure of linked lists of vertex tables (Fig 15). The data structure is traversed to determine new polygon vertices for trimmed polygons. The trimming mechanisms is compatible with recursive subdivision of patches to overcome practical limitations on the number of trimming curves that may be associated with each patch. The length of the straight line segments of the trimming curves is adjusted to compensate for less than ideal parameterization of the trimming curve functions. Associated with each trimming curve within a patch is information about the position of that trimming curve in the span. As each polygon for that patch is generated, those trimming curves that are clearly outside the clip limits for that polygon are excluded from consideration. This reduces the average number of trimming curves that must be processed for the patch, and increases the speed of the graphics accelerator.
    • 图形加速器(91)通过存储非均匀有理B样条片(12,13,14,15)及其相关联的修剪曲线S1-S9的定义来响应图形系统中的计算机(86)的命令, 。 然后,图形加速器产生针对每个补丁计算的修剪多边形的设备坐标,并将这些多边形发送到显示器(94)。 引入了各种改进措施以最小化舍入误差的影响。 每个贴片的uv参数空间中的修剪曲线(27)的B样条定义被转换为近似的短直线段。 未修剪的多边形顶点,直线段的端点和与多边形边缘对应的子跨边界的直线段的交点保持在顶点表的链表列表的数据结构中(图15)。 遍历数据结构以确定修剪多边形的新的多边形顶点。 修剪机制与补丁的递归细分兼容,以克服可能与每个补丁相关联的修剪曲线数量的实际限制。 调整修剪曲线的直线段的长度以补偿修剪曲线函数的理想参数化不足。 与补丁内的每个修剪曲线相关联的是有关该修剪曲线在跨度中的位置的信息。 由于生成了该贴片的每个多边形,所以明确地在该多边形的剪辑限制之外的修剪曲线被排除在考虑之外。 这减少了补丁必须处理的平均修剪曲线数,并增加了图形加速器的速度。