mslater@cup.portal.com (Michael Z Slater) (07/08/90)
MICROPROCESSOR FORUM - ADVANCE PROGRAM Hyatt Regency Hotel, San Francisco Airport October 10-11, 1990 Sponsored by Microprocessor Report ------------------------------------------------------------------------- Day One: Wednesday, October 11 Keynote The Effect of Free Silicon on Microprocessor Design Andy Rappaport, The Technology Research Group General-Purpose Microprocessors A High-Performance Superscalar Processor that is Binary-Compatible with the 386/486 Thampy Thomas, Nexgen Microsystems The First Microprocessor Designed for Notebook Computers Dave Vannier, Intel The Lightning Superscalar SPARC Implementation Bruce Lightner, Metaflow Technologies The Next-Generation 88000-Family Processor Keith Diefendorff, Motorola A Superscalar Clipper Implementation Howard Sachs, Intergraph Panel Discussion All speakers above Lunch: The Second Annual Microprocessor Report Awards Nick Tredennick, Master of Ceremonies Embedded Microprocessors The 68340: A 68020-Based Controller Brad Cohen, Motorola 32000-Family Microprocessors with Signal Processing Support Gideon Intrater, National The 29050: Architecture, Implementation, and Performance Tim Olson, AMD Upward and Downward Mobility in the Next-Generation 960-Family Processors Steve McGeady, Intel A Next-Generation R3000-Based Embedded Control Processor Phil Bourekas, IDT An 88000 Implementation for Embedded Control John Sansing, Motorola The Next Generation of Transputers: Microprocessors for Multiprocessing Ian Pearson, Inmos Panel Discussion All speakers above Reception ------------------------------------------------------------------------- DAY TWO: Thursday, October 11 Digital Signal Processors The 96002 Floating-Point DSP Mike Collins, Motorola The 340C40 Floating-Point DSP Ray Simar Jr., Texas Instruments A Low-Cost Floating-Point DSP for PC/Workstation Multimedia Applications Craig Garen, AT&T Applications Multifunction Imaging Peripherals Moshe Doron, National A RISC-based X-Windows Terminal Peter Weyman, Samsung Software America Multiprocessor Systems using Intel's Next-Generation i860 Benny Maytal, Intel A Multiprocessor Chip Set for 386/486 Systems Dado Banatao, S3 Lunch: The Zen of Change Paul Saffo, Institute for the Future Multichip Modules An Overview of Multichip Technologies Jay Block, G & B Technical Services A High-Performance Multichip SPARC Module Bruce McWilliams, nCHIP, Inc. Panel Discussion: The Future of MCM Speakers above, joined by John Mick (IDT), Clyde Loftal (Raychem), John Reche (Polycon), Howard Davidson (Sun) Architectural Issues for the 1990s Superscalar and Superpipelined Microprocessors Mike Johnson, AMD Dataflow Processors Greg Papadopoulos, MIT VLIW and Compiler-Directed Parallelism Monica Lam, Stanford University Panel Discussion: Personal Views on The Next Decade of Microprocessors Dave Patterson (UC Berkeley), Greg Papadopoulos (MIT), Norm Jouppi (DEC), Monica Lam (Stanford), John Mashey (MIPS), Mike Johnson (AMD), Pat Gelsinger (Intel), Keith Diefendorff (Motorola) ========================================================================= Registration Information Until August 15, 1990, the early registration price of $645 for Microprocessor Report subscribers ($795 for non-subscribers) is in effect. Orders must be prepaid to qualify for this discounted rate. After August 15, the prices go up $100 to $745 for subscribers and $895 for non-subscribers. Multiple registration discounts are available. Two seminars on "Understanding High-Performance Microprocessors" will be offered on October 9 in conjunction with the conference. One version is designed for those without engineering backgrounds, while another is intended for engineers. The seminars will be presented by Michael Slater, Editor and Publisher of Microprocessor Report, and by John Wharton and Brian Case, independent consultants and contributing editors to Microprocessor Report. For a complete brochure and registration form, send an email request (with your U.S. mail address) to deena@cup.portal.com, call (800) 527-0288 or (707) 823-4004, or fax your request to (707) 823-0504.
wallach@motcid.UUCP (Cliff H. Wallach) (07/14/90)
In article <40052@mips.mips.COM> Mashey writes: >Since this discussion pops up here every 6-9 months, maybe it's worth >a shot at some generalizations, to put some structure around the >flood of anecdotal examples and counter-examples. >First MY anecdote: I wrote all of the C string library in assembler for >the R2000 a long while back, because you ALWAYS do that when you do a new >UNIX. THEN, after looking at the generated code from the standard C functions, >we threw out most of the assmebly code, because the compiler was good >enough that it wasn't worth QAing the assembly code. Typically there is a learning curve... Has anyone immersed themselves in R2000 assembly and then written the C string library? >Generalizations: the following factors will encourage one to use >assembly code (left) or compiled code (right). I think most of the >anecdotes so far fit at least one of these. The interesting questions are: >1) Are there more good genralizations? >2) What kind of experiments can be done to quantify these more specifically? > >ASSEMBLER COMPILER >1. Old, non-optimizing compiler Modern, global optimizer, at least > In scientific world, vectorizer/parallizer > if appropriate. >2. Architecture/language mismatch Architecture/language well matched > > 2a. Architecture weird > 2b. Language not expressive enough > > >3. One architecture target, forever Portability > > (as in some embedded) (as in UNIX commands) > >4. Absolute speed requirement Premium on development cost, maintenance > >5. Absolute space requirement Space costs, but no hard, low limit > >6. Programming time is cheap Programming time is not cheap 7. A competitive environment 8. Easier to debug. Improved algorithms are independent of the language used. Assembly language provides a constant improvement over hll. It is commonly claimed that compiler code is almost as fast as assembler. It is also common for vendors to claim that this years compiler generates faster code than last years compiler. Cliff Wallach ...uunet!motcid!wallach filler more filler and more filler and still more filler