In order to realize the advanced graphics rendering algorithms which tends to become more complex and flexible, more and more graphics processor units (GPU) include a micro-processor-like core to support the programmable shading capability. However, since the number of cycles spent in the fragment shader in programmable GPU will vary with different applications, the hardware implementation of the remaining fixed function of the graphics rendering flow becomes not trivial because the suitable target throughput is hard to set. In addition, the data transfer between the shader processor and other hardware fixed-function modules will also represent a big overhead. Therefore, this thesis focuses on realizing the rasterization, which is a very important fixed rendering function, and proposes a pure-software solution that can be executed by the shader processor. The pure-software rasterization requires 98 cycles in setup-stage, and an average of 13 cycles per pixel in interpolation-stage. To further accelerate this rasterization, this thesis also proposes an hardware-software codesign which uses a embedded scan-conversion unit to cooperate with the shader processor. This unit costs about 8.5K gates, which occupies only 1.7% of the entire GPU, but can help reduce more than 30% cycles compared with the pure-software approach in the test-benches used in this thesis. The other contribution of this thesis is to implement the stereoscopic graphic rendering function. To provide stereoscopic effect, the graphic rendering system has to run the entire rendering flow for additional passes to generate the results from different views. However, this thesis will embed an additional code in the fragement shader to adjust the x-coordinate position generated by vertex shader to avoid the additional running pass of the vertex shader.

Context: Rendering 3D scenes in real-time applications is becoming more computationally heavy all the time. Applications are demanded to render high quality graphics without going below a satisfactory frame rate. A huge part of the computation of graphics goes toward complex lighting of 3D models that has to be recomputed every frame. Object Space Lighting (OSL) is a recent technique that is able to store lighting data in between frames. This thesis researches how storing all shading data can impact the performance of an application. Objective: An OSL lighting application will be tested against a standard Forward rendering application in terms of performance, and image quality and perceptual deviations. Experiments are conducted using a scene that can have either still or moving lights and produces the results for the research. Results: Analysing the images from the results indicate that OSL is capable of rendering almost identical images as Forward rendering. The images are not perceptually different either. In terms of performance the hardware used for the experiments determines which application performs better rendering a scene with non-moving. Our OSL application shows clear weaknesses when rendering a scene with moving lights however. Conclusion: Finally, saving all lighting data with OSL is an interesting technique that with further research in the field could prove to be useful in a real time application under certain conditions.

Pixel art is a popular style of digital art often found in video games. It is typically characterized by its low resolution and use of limited colour palettes. Pixel art is created manually with little automation because it requires attention to pixel-level details. Working with individual pixels is a challenging and abstract task, whereas manipulating higher-level objects in vector graphics is much more intuitive. However, it is difficult to bridge this gap because although many rasterization algorithms exist, they are not well-suited for the particular needs of pixel artists, particularly at low resolutions. In this thesis, we introduce a class of rasterization algorithms called pixelation that is tailored to pixel art needs. We describe how our algorithm suppresses artifacts when pixelating vector paths and preserves shape-level features when pixelating geometric primitives. We also developed methods inspired by pixel art for drawing lines and angles more effectively at low resolutions. We compared our results to rasterization algorithms, rasterizers used in commercial software, and human subjects---both amateurs and pixel artists. Through formal analyses of our user study studies and a close collaboration with professional pixel artists, we showed that, in general, our pixelation algorithms produce more visually appealing results than na\"{i}ve rasterization algorithms do.

Background. 3D sound propagation is important for immersion and realism in interactive and dynamic virtual environments. However, this is difficult to model in a physically accurate manner under real-time constraints. Computer graphics techniques are used in acoustics research to increase performance, yet there is little utilization of the especially efficient rasterization techniques, possibly due to concerns of physical accuracy. Fortunately, psychoacoustics have shown that perceived realism does not equate physical accuracy. This indicates that perceptually realistic and high-performance 3D sound propagation may be achievable with rasterization techniques. Objectives. This thesis investigates whether 3D sound propagation can be modelled with high performance and perceived realism using rasterization-based techniques. Methods. A rasterization-based solution for 3D sound propagation is implemented. Its perceived realism is measured using psychoacoustic evaluations. Its performance is analyzed through computation time measurements with varying sound source and triangle count, and theoretical calculations of memory consumption. The performance and perceived realism of the rasterization-based solution is compared with an existing solution. Results. The rasterization-based solution shows both higher performance and perceived realism than the existing solution. Conclusions. 3D sound propagation can be modelled with high performance and perceived realism using rasterization-based techniques. Thus, rasterized 3D sound propagation may provide efficient, low-cost, perceptually realistic 3D audio for areas where immersion and perceptual realism are important, such as video games, serious games, live entertainment events, architectural design, art production and training simulations. / Bakgrund. 3D-ljudpropagering är viktig för inlevelse och realism i interaktiva och dynamiska virtuella miljöer. Dock är detta svårt att modellera på fysiskt träffsäkert sätt med realtidsbegränsningar. Tekniker inom datorgrafik används inom akustikforskning för att öka prestanda, ändock används knappt de synnerligen effektiva rasteriseringsteknikerna, möjligtvis på grund av osäkerhet kring fysisk träffsäkerhet. Lyckligtvis har psykoakustiken visat att uppfattad realism inte är detsamma som fysisk träffsäkerhet. Detta är en indikation på att högpresterande och perceptuellt realistisk 3D-ljudpropagering kan åstadkommas med rasteriseringstekniker. Syfte. Denna avhandling undersöker huruvida 3D-ljudpropagering kan modelleras med hög prestanda och perceptuell realism med rasteriseringstekniker. Metod. En rasteriseringsbaserad lösning för 3D-ljudpropagering implementeras. Dess perceptuella realism mäts genom psykoakustiska utvärderingar. Dess prestanda analyseras genom körtidsmätningar vid varierande antal ljudkällor och trianglar, och teoretiska uträkningar över minnesanvändning. Den perceptuella realismen och prestandan hos den rasteriseringsbaserade lösningen jämförs med en existerande lösning. Resultat. Den rasteriseringsbaserade lösningen påvisar både högre perceptuell realism och prestanda än den existerande lösningen. Slutsatser. 3D-ljudpropagering kan modelleras med hög prestanda och perceptuell realism med rasteriseringsbaserade tekniker. Alltså kan rasteriserad 3D-ljudpropagering bistå med effektivt, billigt och perceptuellt realistiskt 3D-ljud för områden där inlevelse och perceptuell realism är viktiga, såsom videospel, seriösa spel, live underhållningsevents, arkitekturdesign, konstproduktion och träningssimulationer.

This term project is aimed on analysis of graphic pipeline which can rasterize required picture. Document is specialized to drawing algorithms that are used in rasterization block. Major aim of this project is describing of rasterization algorithms that can be implemented on hardware. Type of aimed hardware is field-programmable gate array FPGA.

<p>Implement a high-level model of computationally intensive part of 3D graphics pipe-line. Increasing popularity of handheld devices along with developments in hardware technology, 3D graphics on mobile devices is fast becoming a reality. Graphics processing is essentially complex and computationally demanding. In order to achieve scene realism and perception of motion, identifying and accelerating bottle necks is crucial. This thesis is about Open-GL graphics pipe-line in general. Software which implements computationally intensive part of graphics pipe-line is built. In essence a rasterization unit that gets triangles with 2D screen, texture co-ordinates and color. Triangles go through scan conversion, texturing and a set of other per-fragment operations before getting displayed on screen.</p>

Implement a high-level model of computationally intensive part of 3D graphics pipe-line. Increasing popularity of handheld devices along with developments in hardware technology, 3D graphics on mobile devices is fast becoming a reality. Graphics processing is essentially complex and computationally demanding. In order to achieve scene realism and perception of motion, identifying and accelerating bottle necks is crucial. This thesis is about Open-GL graphics pipe-line in general. Software which implements computationally intensive part of graphics pipe-line is built. In essence a rasterization unit that gets triangles with 2D screen, texture co-ordinates and color. Triangles go through scan conversion, texturing and a set of other per-fragment operations before getting displayed on screen.

This report documents implementation details, results, benchmarks and technical discussions for the work carried out within a master’s thesis at Linköping University. Within the master’s thesis, the field of software rendering is explored in the age of parallel computing. Using the Open Computing Language, a complete graphics pipeline was implemented for use on general processing units from different vendors. The pipeline is tile-based, fully-configurable and provides means of rendering visually compelling images in real-time. Yet, further optimizations for parallel architectures are needed as uneven work loads drastically decrease the overall performance of the pipeline.

Context. Many occlusion culling algorithms have to cope with the task of balancing performance and accuracy. While it is desirable to accurately identify all occluded scene objects, settling with a rough estimate is often more beneficial for the overall performance. Algorithms that rely on a depth buffer can often gain a lot of performance by performing the occlusion culling at a lower resolution than the resolution of the screen. This calls for more advanced methods to render the depth buffer as the standard rasterizer will not guarantee inner coverage. Objectives. The goal of this thesis is to find a solution to generate a depth buffer where all rasterized pixels are fully covered by overlapping occluders. An algorithm is proposed that is based on previous work on inner conservative rasterization. The algorithm addresses some of the problems existing methods are suffering from, but also has some flaws of its own. Methods. The proposed algorithm is tested by comparing it to two methods that also produce conservative results. A GPU-based occlusion culling system is developed to conduct an experiment. The experiment is performed by measuring performance and culling efficiency in two different views of a scene. The scene is set up to represent an average setting in a game. Results. The results from the experiment show that the proposed algorithm outperforms its competitors in many cases. In the first scene view, the total frame time is 5% faster at a full screen resolution of 1366x768 pixels and 8% faster at a full screen resolution of 1920x1080 pixels. The depth buffer generated by the proposed algorithm is culling atleast as many occludees as its competitors and often surpasses them. In the second scene view, the total frame time is 2% faster at a full screen resolution of 1366x768 pixels and 3% faster at a full screen resolution of 1920x1080 pixels. The depth buffer generated by the proposed algorithm is often culling more occludees than its competitors, but is at lower resolutions less efficient, up to 3%. Conclusions. The conclusions show that the goal has been reached. The proposed algorithm lacks flexibility, but provides good performance and accuracy. Future work to improve the proposed algorithm is outlined.

Ray tracing comes together with a tremendous computational cost [1]. Therefore, Keller et al., expressed that possible cost reduction appears when a hybrid rendering pipeline is implemented by combining rasterization and ray tracing, which have already been introduced to the film and game industries. Such a rendering method within Grand Strategy Games (GSG) has been an unexplored task. The standard rendering method of GSG has been rasterization. Implementing hybrid rendering for GSG would allow this niche to follow the continuously developing rendering techniques. Therefore, this thesis examined the advantages and disadvantages of hybrid rendering compared to a path traced pipeline. The study measured different camera angles applied to three GSG-inspired scenes by rendering time and quality according to pixelby-pixel comparison focusing on effects like shadows and reflections. Closeup images have been taken on the rendered scenes to evaluate interesting pieces in the scenes. Steady time performance for all angles was the significant advantage of the hybrid pipeline. The angles at lower grades resulted in an increased difference in shadows and reflections for two out of three scenes. Additionally, the entire pixel-by-pixel comparison did not generate more than ten percent difference for any scene and not more than twelve percent difference on closeup images. Still, differences were noticeable to the eye since the path tracer was superior for developing sharp shadows. The hybrid pipeline generated a massive reflection compared to the path tracer. Since the path tracer was defined as the ground truth, this quantity of reflections was not considered positive. The thesis concludes that a simple hybrid rendering pipeline could be an exciting future for GSG, especially for angles above 67.25◦ . Additionally, improving the sharpness of the shadows for the hybrid rendering pipeline could increase the interest in hybrid rendering for GSG even at angles below 67.25◦ . Some interesting future work is rendering advanced 3D map-based GSG scenes, including more shadows and reflections. Another suggestion is a qualitative analysis of users playing a game with the two rendering pipelines before attending a user study about their possible improved graphical experience and how the game experience has been affected. / Strålspårning kommer tillsammans med en stor beräkningskostnad [1]. Därför har Keller m.fl. uttryckt att kostnaderna kan reduceras genom att implementera en hybridrenderingsmetod baserad på en kombination av rastrering och strålspårning, vilket redan har introducerats till film- och spelindustrin. En sådan renderingsmetod inom Grand Strategy Games (GSG) har dock varit ett outforskat område. Standard renderingsmetoden för GSG har varit rastrering. Implementering av hybridrendering för GSG skulle tillåta denna nisch att följa de ständigt utvecklande renderingsteknikerna. Därför undersöker denna avhandling fördelarna och nackdelarna med hybridrendering jämfört med en renderingspipeline baserad på strålspårning. Studien har mätt olika kameravinklar applicerade på GSG-inspirerade scener mätt med renderingstid och kvalitet enligt pixel-för-pixel-jämförelse och med fokus på effekter som skuggor och reflektioner. Närbilder har tagits på de renderade scenerna för att utvärdera intressanta delar i scenerna. Stabil tidprestanda av samtliga vinklar var den betydande fördelen med hybridpipelinen. Vinklarna vid lägre grader resulterade i ökad differens av skuggor och reflektioner för två av tre scener. Dessutom resulterade hela pixeljämförelsen inte mer än tio procents skillnad för någon av scenerna och inte mer än tolv procents skillnad på närbilderna. Ändå var skillnaderna märkbara för ögat eftersom strålspårningen var överlägsen för att generera skarpa skuggor. Hybridlösningen genererade en stor andel reflektion jämfört med strålspårningen. Eftersom strålspårningen definierades som målbilden var denna mängd reflektioner inte positiva. Avhandlingen drar slutsatsen att en enkel hybridmetod kan vara en spännande framtid för GSG, speciellt för vinklar över 67,25◦ . Dessutom kan en förbättring av skärpan på skuggorna för hybridrenderingen öka intresset för hybridrendering för GSG även vid vinklar under 67,25◦ . Intressanta framtida arbeten är rendering av avancerade GSG scener, som inkluderar fler skuggor och reflektioner. Ett till förslag är kvalitativ analys av användare som spelar ett spel med de två renderingsmetoder följt av användarstudie om deras möjliga förbättrade grafiska upplevelse och om spelupplevelsen har drabbats.