Harvard Graphics 3.0

HarvardGraphics3. Robert Gaskins Home Page. Hackers Conference. Sausalito, California. In 1. 98. 4 I was one of about 1. November 91. 1. at the worlds most beautiful repurposed 1. Fort Cronkhite. on the Marin headlands just north of the Golden Gate, attending the original. Hackers Conference. This meet up was initiated by Kevin Kelly and Stewart Brand, and designed by Lee Felsenstein. Bill Budge, Andy Hertzfeld, and Doug Carlston, timed to coincide with the publication of Steven Levys book. Strategy and entrepreneurship are often viewed as polar opposites. Strategy is seen as the pursuit of a clearly defined pathone systematically identified in. According to a study published last week in the EPJ Data Science journal, researchers at Harvard University and the University of Vermont questioned whether markers. BICCs arent dead far from it. The need is greater than ever. But a BICC needs to continually evolve to meet the changing needs of the business. Harvard Graphics 3.0' title='Harvard Graphics 3.0' />AMD, which purchased Nvidias previous competitor, ATI Graphics, has been losing the GPU war for a while. Nvidia is currently producing the majority of discrete. Connecting decision makers to a dynamic network of information, people and ideas, Bloomberg quickly and accurately delivers business and financial information, news. Hackers Heroes of the Computer Revolution. Stewart Brand claimed that the invitees were the most interesting and effective body of. U. S. Constitution, a claim that escaped criticism from those attending. Invitees paid a flat 9. Steve Wozniak donated 5,0. DVD. The t shirt design was by Don Knuths student. Scott Kim. Western Institute of Computer Science. Santa Cruz, California. In 1. 97. 9 after I had joined Bell Northern Research I attended the International Course in. Programming Methodology, an advanced course in programming taught by Edsger W. Dijkstra. followed by a course of forty nine lectures from Dijkstra, C. A. R. Hoare, Ole Johan Dahl. John Backus, David Gries, and over twenty additional members of. IFIP WG 2. 3. This was a four week residential course, held on the campus of the University of California at Santa Cruz. August, 1. 97. 9. Classes were held in a steeply raked auditorium with a raised aisle behind every curved row of. Dijkstra assigned in class programming problems, and would then walk along the aisles, looking. It was unnerving when he stopped and lingered behind. Dijkstra recorded his own thoughts about the course in his contemporaneous typescript. EWD 7. 14 now archived online at the. University of Texas I found the UCSC UC Santa Cruz campus not an inspiring place. I stayed there, the more depressing it. We. had to share the foodand what was worse. American. football In. The audience was of a higher calibre than we had been led to. Bell. Laboratories 3. WG2. 3 meeting is the very lousy impression I got from Xerox. PARC a. place where research in computing science is primarily viewed as gadget development. Twenty two years after this, Dijkstra was still using a pen to write his own overhead foils transparencies. Power. Point. From another trip report EWD 1. Preparing to receive an honorary doctorate in Athens, May 2. Fortunately I had discovered. I had left my prepared foils in Nuenen and the University had provided me with. I used Wednesdays free moments to make new ones. And a month later, June 2. Software Pioneers conference in Bonn that featured 1. I mention another. Power. Point, for the less of it you use, the better your lecture. The bloody electronics only encourage the next steps of the replacement of content by. University of California, Berkeley. Computer Scientist. Berkeley, California. July 1. 97. 3May 1. While a graduate student, I co authored with Laura Gould a textbook on programming for linguistic and humanities research. Berkeley and Stanford, and in summer sessions for college teachers in the humanities. American Council of Learned Societies. I did extensive consulting with Berkeley faculty members. I was graphics consultant for the. Berkeley Campus Computer Center. I spent some years as chief programmer for Berkeley machine translation research. Chinese to English. I did the programming of ancient Egyptian hieroglyphic fonts and typesetting for the. Berkeley Late Egyptian Dictionary. I wrote a program to generate haiku, which was embedded in the idle loop. CDC6. 40. 0 and became the most prolific poet up till that date, with a selection published in an. Richard W. Bailey Computer Poems, 1. One of my motivations for. Berkeley was Professor Bertrand Bronson in the English Department, who had pioneered the study of the. English and Scottish popular ballads by coding the music and transcribing it to punched cards. I was able to help him with computer analyses and concordances. There were many more projects in graphics and music and natural language. University of California, Berkeley. Berkeley, California. M. A. 1. 97. 3, in Computer Science, Linguistics, and English. I was admitted to the Ph. D. program in the English Department at the University of California, Berkeley in 1. My wife was admitted to the Linguistics Department. Ph. D. at the same time, and 1. Berkeley. I entered with a Special Career Fellowship. Ford Foundation for five years of complete support. My intention was to specialize in Shakespeare and follow an academic career teaching literature and linguistics. I registered for my first classes, I read in the catalogue and discovered classes in the Computer Science Department. College of Letters and Science, spun off from the Math Department. My advisor, Josephine Miles, the poet, agreed that some exposure to computers would be broadening, so I enrolled. I was immediately enthralled, took more classes CDC 6. Butler Lampson was memorable, and soon I formally broadened my program this was possible because the Special Career Fellowship. I was approved to undertake an individual interdisciplinary Ph. D. program in the College of Letters and Science. Computer Science, Linguistics, and English Departments. This was a wise step for the purpose of getting an education, but not necessarily wise for the purpose of ever completing. Pursuing a Ph. D. The computer scientists thought I was reasonably smart, at least for a student whose interests were in such nebulous areas. The literary people thought I was a tolerable critic, at least for a student. The linguists thought I had some. Over the next ten years, I passed. Ph. D. language examinations in Latin and in French Computer Science Department Ph. D. comprehensive written exam Linguistics Department Ph. D. comprehensive written exam English Department Ph. D. comprehensive written exam Ph. D. oral examination, with examiners from all three departments. Ph. D. dissertation, with examiners from all three departments at this period, the dissertation defense was conducted after preliminary work and before writing, with the topic Use of Two Level van Wijngaarden Grammars for Natural Language Analysis. Advisor Charles J. Fillmore. This completed the requirements for the Cand. Phil., which at the time was the Berkeley official status corresponding. ABD all but dissertation degree status for Ph. D. students, with an M. A. degree awarded along the way. But by 1. 97. 8, after ten years at Berkeley, I had decided I needed to write that dissertation far less than I needed. Silicon Valley, where I could get the experience to do a software startup for the new single user personal computers. These were very early days at the time I felt I had to leave Berkeley, Bill Gates had left Harvard less than three years earlier. Albuquerque, writing software for the MITS Altair Bill only relocated Microsoft to the Seattle area. I had moved to Silicon Valley. The first Apple had been. And just as 1. 96. Berkeley, 1. 97. 8 was a great year to move. Silicon Valley. University of Southern California. Los Angeles, California. A. B. 1. 96. 8, in English Literature. I transferred to the University of Southern California as a junior. English Department. There, one of my major advisors was Professor Virginia Tufte, who was. Her son Edward Tufte, who was about my same age, would later attract a great deal of attention for. Power. Point. I was elected to Phi Beta Kappa. Order of the Palm award at graduation. Married. Los Angeles, California. Leanna Jean Koehn. General purpose computing on graphics processing units. General purpose computing on graphics processing units GPGPU, rarely GPGP is the use of a graphics processing unit GPU, which typically handles computation only for computer graphics, to perform computation in applications traditionally handled by the central processing unit CPU. The use of multiple video cards in one computer, or large numbers of graphics chips, further parallelizes the already parallel nature of graphics processing. In addition, even a single GPU CPU framework provides advantages that multiple CPUs on their own do not offer due to the specialization in each chip. Essentially, a GPGPU pipeline is a kind of parallel processing between one or more GPUs and CPUs that analyzes data as if it were in image or other graphic form. While GPUs operate at lower frequencies, they typically have many times the number of cores. Thus, GPUs can process far more pictures and graphical data per second than a traditional CPU. Migrating data into graphical form and then using the GPU to scan and analyze it can create a large speedup. GPGPU pipelines were developed at the beginning of the 2. These pipelines were found to fit scientific computing needs well, and have since been developed in this direction. HistoryeditGeneral purpose computing on GPUs only became practical and popular after about 2. Notably, problems involving matrices andor vectors  especially two, three, or four dimensional vectors  were easy to translate to a GPU, which acts with native speed and support on those types. The scientific computing communitys experiments with the new hardware began with a matrix multiplication routine 2. GPUs than CPUs was an implementation of LU factorization 2. These early efforts to use GPUs as general purpose processors required reformulating computational problems in terms of graphics primitives, as supported by the two major APIs for graphics processors, Open. GL and Direct. X. This cumbersome translation was obviated by the advent of general purpose programming languages and APIs such as ShRapid. Mind, Brook and Accelerator. These were followed by Nvidias CUDA, which allowed programmers to ignore the underlying graphical concepts in favor of more common high performance computing concepts. Newer, hardware vendor independent offerings include Microsofts Direct. Compute and AppleKhronos Groups Open. CL. 6 This means that modern GPGPU pipelines can leverage the speed of a GPU without requiring full and explicit conversion of the data to a graphical form. ImplementationseditAny language that allows the code running on the CPU to poll a GPU shader for return values, can create a GPGPU framework. As of 2. 01. 6update, Open. CL is the dominant open general purpose GPU computing language, and is an open standard defined by the Khronos Group. Open. CL provides a cross platform GPGPU platform that additionally supports data parallel compute on CPUs. Open. CL is actively supported on Intel, AMD, Nvidia, and ARM platforms. The Khronos Group is currently involved in the development of SYCL, which has its implementations with Compute. CPP and SYCL STL, the first being developed by Codeplay, and currently only supported in Linux Operating Systems. The second one, being hosted by Khronos Group on Git. Hub, and possible to be compiled for any modern operating system. The dominant proprietary framework is Nvidia. CUDA. 1. 0 Nvidia launched CUDA in 2. SDK and application programming interface API that allows using the programming language C to code algorithms for execution on Ge. Force 8 series GPUs. Programming standards for parallel computing include Open. CL vendor independent, Open. ACC, and Open. HMPP. Mark Harris, the founder of GPGPU. GPGPU. Open. VIDIA was developed at University of Toronto during 2. Nvidia. Altimesh Hybridizer1. Altimesh1. 3 compiles Common Intermediate Language to CUDA binaries. It supports generics and virtual functions. Debugging and profiling is integrated to visual studio and Nsight. Its available as a Visual Studio Extension on Visual Studio Marketplace. Microsoft introduced the Direct. Compute GPU computing API, released with the Direct. X 1. 1 API. Alea GPU1. Quant. Alea1. 7 introduces native GPU computing capabilities for the Microsoft. NET language F1. C. Alea GPU also provides a simplified GPU programming model based on GPU parallel for and parallel aggregate using delegates and automatic memory management. MATLAB supports GPGPU acceleration using the Parallel Computing Toolbox and MATLAB Distributed Computing Server,2. Jacket. GPGPU processing is also used to simulate Newtonian physics by Physics engines, and commercial implementations include Havok Physics, FX and Phys. X, both of which are typically used for computer and video games. Ibm T43 Serial Port. Close to Metal, now called Stream, is AMDs GPGPU technology for ATI Radeon based GPUs. C Accelerated Massive Parallelism C AMP is a library that accelerates execution of C code by exploiting the data parallel hardware on GPUs. Mobile computerseditDue to a trend of increasing power of mobile GPUs, general purpose programming became available also on the mobile devices running major mobile operating systems. Google. Android 4. Render. Script code on the mobile device GPU. Apple introduced a proprietary Metal API for i. OS applications, able to execute arbitrary code through Apples GPU compute shaders. Hardware supporteditComputer video cards are produced by various vendors, such as Nvidia, and AMD and ATI. Cards from such vendors differ on implementing data format support, such as integer and floating point formats 3. Microsoft introduced a Shader Model standard, to help rank the various features of graphic cards into a simple Shader Model version number 1. Integer numberseditPre Direct. X 9 video cards only supported paletted or integer color types. Various formats are available, each containing a red element, a green element, and a blue element. Sometimes another alpha value is added, to be used for transparency. Common formats are 8 bits per pixel Sometimes palette mode, where each value is an index in a table with the real color value specified in one of the other formats. Sometimes three bits for red, three bits for green, and two bits for blue. Usually the bits are allocated as five bits for red, six bits for green, and five bits for blue. There are eight bits for each of red, green, and blue. There are eight bits for each of red, green, blue, and alpha. Floating point numberseditFor early fixed function or limited programmability graphics i. Direct. X 8. 1 compliant GPUs this was sufficient because this is also the representation used in displays. This representation does have certain limitations, however. Given sufficient graphics processing power even graphics programmers would like to use better formats, such as floating point data formats, to obtain effects such as high dynamic range imaging. Many GPGPU applications require floating point accuracy, which came with video cards conforming to the Direct. X 9 specification. Direct. X 9 Shader Model 2. Full precision support could either be FP3. FP2. 4 floating point 3. FP1. 6. ATIs. Radeon R3. GPUs supported FP2. FP3. 2 was supported in the vertex processors while Nvidias NV3. FP1. 6 and FP3. 2 other vendors such as S3 Graphics and XGI supported a mixture of formats up to FP2.