本篇論文提出一個基於砌塊式繪圖管線架構之Geometry引擎。在3D電腦繪圖中，其所描述的場景由幾何單元所表示，也就是點、線、面，而這些幾何單元再依照光源的設定繪製出受不同光源影響的顏色圖像。而本篇論文則根據OpenGL ES 1.1的規格設計一個處理幾何單元運算的Geometry引擎並對並對其硬體架構作進一步的效能評估及分析。
在Geometry引擎的設計上，頂點處理器的架構切割成兩大級管線，有效地提升硬體使用率，減少面積；另外將perspective division模組設計於頂點處理器內，與未改善前的架構相比，整體面積降低58.6%。光源的指數計算部份以內插法的方式設計查表並將其實現成硬體，與未改善前的硬體合成面積相比，降低了17倍。為了減少匯流排頻寬，除了設計Culling模組外，我們提出ATT (Abandon Tiny Triangles)演算法刪除微小三角形，減少50%至60%不會繪製在螢幕上的三角形輸出量。此Geometry引擎在0.18um的製程下總共佔389K個邏輯閘，且能在200MHz的工作頻率下正確執行，並有瞬間最高7.407M Tri/sec的三角形運算效能。此硬體在QEMU結合CoWare的完整系統模擬環境下，能完整地驗證ARM嵌入式系統經由啟動Linux並執行OpenGL ES的應用程式。

This thesis presents a Geometry engine based on a tile-based rendering architecture. In 3D computer graphics, the scene is typically constructed from geometry objects made up from vertices, lines, or triangles, and these geometry objects are displayed in different types by various drawing modes. We design a Geometry engine that deals with the geometry calculation according to OpenGL ES 1.1 specifications and evaluate the performance of the hardware architecture.

In Geometry engine designing, we split the vertex processor into a 2-stage pipeline in order to reduce the overall area cost by increasing the hardware utilization efficiently. Comparing with the traditional architecture, we propose a new architecture that puts the perspective division model in the vertex processor. This design results in overall area reduction by 58.6%. For the lighting model, we use a linear interpolation index-reduced method to design the lookup table, and the synthesis area is nearly 17 times smaller than the hardware without modification. In order to reduce the bus traffic, we design a Culling model and propose an ATT (Abandon Tiny Triangles) algorithm to discard tiny triangles. The total triangle reduction is about 50% to 60% that would not be rendered on the screen. The Geometry engine has been realized by 0.18 um technology and the total gate count is 389K. It can run up to 200 MHz, and deliver the peak fill rate of 7.407M Tri/sec. It has been simulated and verified with OpenGL ES applications executed on Linux in an ARM embedded system modelled by a full system simulation platform which combines QEMU and an ESL tool.

[1] T. Akenine-Moller and E. Haines, Real-time rendering, 2nd edition. A. K. Peters, Ltd, Natick, MA, 2002.
[2] H. Fuchs et al., “Pixel-Planes 5: A Heterogeneous Multiprocessor Graphics System Using Processor-Enhanced Memories,” In Proc. ACM SIGGRAPH, pp. 79-88, 1989.
[3] M. Chen, G. Stoll, H. Igehy, K. Proudfoot, and P. Hanmhan, “Simple Models of the Impact of Overlap in Bucket Rendering,” In Proc. ACM SIGGMPH/EURDGRAPHICS Workshop on Graphics Hardware, pp. 105-112, 1998.
[4] M. Cox and N. Bhandari, “Architectural Implications of Hardware-Accelerated Bucket Rendering on the PC,” In Proc. ACM SIGGMPH/EURDGRAPHICS Workshop on Graphics Hardware, pp. 25 - 34, 1997.
[5] I. Antochi, B. Juurlink, S. Vassiliadis, and P. Liuha, “Memory Bandwidth Requirements of Tile-Based Rendering,” In Proceedings of the Third and Fourth International Workshops SAMOS 2003 and SAMOS 2004 (LNCS 3133), Samos, Greece, pp. 323-332, Jul. 2004.
[6] I. Antochi, “Suitability of Tile-Based Rendering for Low-Power 3D Graphics Accelerators,” PhD Thesis, Delft University of Technology, Oct. 2007.
[7] G. Humphreys, M. Houston, R. Ng, R. Frank, S. Ahern, P. Kirchner, and J. Klosowski, “Chromium: A Stream Processing Framework for Interactive Rendering on Clusters,” In Proc. 29th Annual Conf. on Computer Graphics and Interactive Techniques (SIGGRAPH 2002), pp. 693–702, 2002.
[8] S. Molnar, M. Cox, D. Ellsworth, and H. Fuchs, “A Sorting Classification of Parallel Rendering,” In IEEE Comput. Graph. Appl., vol 14(4), pp. 23–32, 1994.
[9] C. Mueller, “The Sort-First Rendering Architecture for High-Performance Graphics,” In Proc. 1995 Symp. on Interactive 3D Graphics, pp. 75–84, 1995.
[10] D. Harris, “An Exponentiation Unit for an OpenGL Lighting Engines,” In IEEE Trans. Computers, vol. 53, no. 3, pp. 251-258 , Mar. 2004.
[11] B.-G. Nam, H.-J. Kim, and H.-J. Yoo, “Power and Area-Efficient Unified Computation of Vector and Elementary Functions for Handheld 3D Graphics Systems,” In IEEE Trans. Computers, vol. 57, no. 4, Apr. 2008.
[12] B. Mochocki, K. Lahiri, and S. Cadambi, “Power analysis of mobile 3D graphics,” In Design, Automation, and Test in Europe (DATE), 2006.
[13] B.-G. Nam, M.-W. Lee, and H.-J. Yoo, ”Development of a 3-D Graphics Rendering Engine with Lighting Acceleration for Handheld Multimedia Systems,” In IEEE Trans. Consumer Electronics, vol. 51, no. 3, pp.1020–1027, Aug. 2005.
[14] D.-H. Kim et al., “An SoC with 1.3Gtexels/s 3D Graphics Full Pipeline Engine for Consumer Applications,” In IEEE J. Solid-State Circuits, vol. 41, pp. 71, Jan. 2006.
[15] R.-C. Woo et al., “A 210 mW graphics LSI implementing full 3-D pipeline with 264 Mtexels/s texturing for mobile multimedia applications,” In IEEE J. Solid-State Circuits, vol. 39, pp. 358, Feb. 2004.
[16] C.-H. Chen and C.-Y. Lee, “A Cost Effective Lighting Processor for 3D Graphics Application,” In Proc. IEEE Int’l Conf. Image Processing, vol. 2, pp. 792-796, 1999.
[17] H.-J. Kim and B.-G. Nam,“A 231-MHz, 2.18-mW 32-bit Logarithmic Arithmetic Unit for Fixed-Point 3-D Graphics System,” In IEEE Journal of Solid-State Circuits, vol. 41, no. 11, pp. 2373-2381, Nov. 2006.
[18] T. Aila, V. Miettinen, and P. Nordlund, “Delay streams for graphics hardware,” In ACM Trans. Graphics, vol. 22, no. 3, pp. 792-800, Jul. 2003.
[19] CoWare, ConvergenSC Training Manual rev: 1Mar06, CoWare, Inc., 2006.
[20] J.-W. Bae, D.-H. Kim, and L.-S. Kim, ”An 11M-Triangles/sec 3D Graphics Clipping Engine for Triangle Primitives,” In Proc. IEEE International Symposium on Circuits and Systems, vol. 5, pp. 4570 - 4573, May. 2005.
[21] R. S. Wright, Jr., and B. Lipchark, OpenGL SuperBible, 3rd edition. Sams Publishing, 2005.
[22] The Mesa 3D Graphics Library website. Available: http://www.mesa3d.org.
[23] Vincent Mobile 3D Rendering Library. Available:
http://sourceforge.net/projects/ogl-es/.
[24] D. Blythe and A. Munshi, “OpenGL ES Common/Common-Lite Profile Specification Version 1.1.10,” Khronos Group, Inc. Available: http://www.khronos.org/registry/gles/specs/1.1/es_full_spec.1.1.10.pdf.
[25] Microsoft Corp., DirectX Resource Center,
http://msdn2.microsoft.com/zh-tw/xna/aa937781.aspx.
[26] A. Stevens, White paper: ARM Mali 3D Graphics System Solution, December 2006, http://www.arm.com/miscPDFs/16514.pdf.
[27] Imagination Technologies Ltd., PowerVR MBX 2D/3D Graphics, http://www.imgtec.com/Downloads/index.asp?download=true&f=PowerVRMBX.pdf&t=PowerVR%20PDP%20Documents.
[28] PowerVR (Imagination Technologies). Imagination Technologies’ PowerVRTM in STMicroelectronics’ KYROTM PC Graphics Accelerator Unveiled, http://www.imgtec.com/News/Release/index.asp?ID=548.
[29] QEMU Processor Emulator, official site : http://bellard.org/qemu/.
[30] Synopsys, DesignWare IP Family Reference Guide,
https://www.synopsys.com/dw/doc.php/doc/dwf/datasheets/.
[31] C. Lomont, “Fast inverse square root,” Feb. 2003.[Online], http://www.lomont.org/Math/Papers/2003/InvSqrt.pdf.
[32] M. Segal and K. Akeley, “The OpenGL Graphics System: A Specification Version 2.1,” Silicon Graphics, Inc,
http://www.opengl.org/registry/doc/glspec21.20061201.pdf.
[33] Free 3D Model, http://fantasyartdesign.com/.
[34] R.-T. Gu et al., “An 8.69 Mvertices/s 278 Mpixels/s Tile-based 3D Graphics Full Pipeline with Embedded Performance Counting Module, Real-Time Bus Tracer and Protocol Checker for Consumer Electronics,” In International Symposium on VLSI Design, Automation & Test, 2008.
[35] Design Compiler, http://www.synopsys.com/home.aspx.