Site Home Archive Home FAQ Home How to search the Archive How to Navigate the Archive
Compare FPGA features and resources
Threads starting:
Authors:A B C D E F G H I J K L M N O P Q R S T U V W X Y Z
Just When You thought It Was Safe To Go On Vacation HOT CHIPS SYMPOSIUM VI A Symposium on High-Performance Chips (Advance Program) Sponsored by the IEEE Computer Society Technical Committee on Microprocessors Stanford University, Palo, Alto, California Memorial Auditorium August 14-16, 1994 Attend HOT Chips VI, a symposium on high-performance chips, which will bring together researchers and developers of chips used to construct high-performance workstations and systems. Enjoy the informal format offering interaction with speakers. The first five HOT Chips Symposiums were huge successes and prompted articles in five special issues of IEEE Micro magazine. This year's HOT Chips VI will again bring you the latest developments in chip technology. ORGANIZING COMMITTEE General Chairman: John Mashey, Silicon Graphics Vice Chairman: Nam Ling, Santa Clara University Program Co-Chairmen: Alan J. Smith, UC Berkeley Don Alpert, Intel Finance Chairman: Dennis Reinhardt, Intel Registration Chairman: Robert Stewart, Stewart Research Publication Chairman: David Gustavson, SCI Technical Consortium Publicity Chairman: S. Diane Smith, Consultant Local Arrangements Chairmen: Alan Johnson, Cary Kornfeld, Interval Research Tutorials Chairman: Qiang Li Santa Clara University At Large: Martin Freeman, Philips Research John Hennessy, Stanford University PROGRAM COMMITTEE Donald Alpert, Intel (Program Co-Chairman) Forest Baskett, Silicon Graphics Alan Baum, Apple Computer Anoop Gupta, Stanford University Norman Jouppi DEC WRL Ruby Lee, Hewlett Packard Alan J. Smith, UC Berkeley (Program Co-Chairman) PROGRAM August 14, 1994 Tutorial Schedule: 7:30-8:30 Registration & Coffee at Memorial Auditorium 8:30-12:00 Algorithms & Hardware for Video Compression 12:00-1:00 Lunch 1:00-4:30 Instruction-Set Extension for Multiprocessor Interconnects 4:30-6:00 Wine & Cheese Reception in the Dohrman Grove Just North of the Hoover Tower on Serra Street Tutorial 1: Algorithms & Hardware for Video Compression Teresa Meng, Stanford University This tutorial will give an overview of the industry standards for video compression, including compression algorithms, performance comparisons and hardware implementations. Fast algorithms to achieve real-time encoding and decoding will be one focus of this tutorial. Besides industry standards, recent developments in compression techniuqes, such as subband and wavelet filtering and vector quantization will be covered. Finally research activities in the area of low-power implementations for portable video applications will be surveyed. Tutorial 2: Instruction-Set Extensions for Multiprocessor Interconnects David James, Apple Computer Most instruction sets have been optimized for uniprocessor environments. To effectively utilize multiprocessors on scalable interconnects, processors need to support additional capabilities, including 64-bit addressing, a well-defined set of memory access capabilities (loads, stores, and locks), non-blocking interrupts and synchronized time-of-day clocks. As background, the constraints of a typical high-speed interconnect, ANSI/IEEE Std 1596-1992 Scalable Coherent Interface (SCI), are considered. We show that specialized signals (such as bus-lock and interrupt) as well as eavesdrop or broadcast-based protocols can be avoided, explicit lock instructions (such as fetch&add) scale better than LoadReserved- and StoreConditional-based instruction sequences, and mixed endian data types (big and little) require minimal hardware (but significant compiler) support. We show how these conditions impact the design of processor instruction-set extensions. August 15, 1994 9:00-9:15 Welcome and Opening Remarks John Mashey, General Chair Don Alpert, Alan Jay Smith, Program Co-Chairs 9:15-10:45 CPUs - 1 Session Chair: Norman Jouppi, DEC WRL . An Overview of the 21164 Alpha AXP Microprocessor Paul Rubinfeld, Digital Equipment Corp. . The Power2+ Processor David Shippy, IBM . A 500MHz 32b 0.4um CMOS RISC Processor (Gallop) Kazumasa Suzuki, NEC Corporation 10:45-11:15 Break 11:15-12:45 Multiprocessor and Encryption Session Chair: Howard Sachs, Sun Microsystems . The Alewife CMMU: Addressing the Multiprocessor Communications Gap John Kubiatowicz, MIT . nCube3 Integrated MPP Node Processor Robert Duzett, nCube . A 100Kbit/sec Single Chip Modular Exponentiation Processor Holger Orup, Aarhus University, Denmark 12:45-2:15 Lunch 2:15-3:45 Networks, Communications Session Chair: Forest Baskett, Silicon Graphics . UAI2110: An Universal GaAs ATM Interface Chip for High Speed Networks Premysl Vaclavik, Thomas Neuroth GmbH, Austria . A 500 MHz BiCMOS GByte/Second SCI-Link Implementation Wayne Nation, IBM . The STC104 Asynchronous Packet Switch Peter Thompson, INMOS Ltd. 3:45-4:30 Break 4:30-5:30 CPUs - 2 Session Chair: Don Alpert, Intel . The New i960 CPU that offers More for Less, the P100 Richard Brunner/Deif Atallah, Intel . SH-II- A Low Power RISC Micro for Consumer Applications Shumpei Kawasaki, Hitachi 5:30-7:30 Buffet Dinner 7:00-9:00 Evening Panel Session The Investor (Venture) Community View of What's Hot Moderator: Forest Baskett, Silicon Graphics Panelists: Cliff Friedman, Bear Stearns Stephen Shapiro, Tiger Management Peter Thomas, Institutional Venture Partners August 16, 1994: 9:00-10:00 Chipsets Session Chair: Allen Baum, Apple Computer . 82430NX PCIset: Companion to the Highest Performance Pentium Processor Patrick Correia, Intel Corporation . A Power PC/PCI Bridge Chip with a Cache and Memory Controller Karl Wang, Motorola, Inc. 10:00-10:30 Break 10:30-12:30 Graphics Session Chair: Ruby Lee, Hewlett Packard . An ASIC for Interactive 3D Graphics Stephanie Winner, Apple Computer Inc. . GLiNT - a 3D Graphiucs Processor Based on the OpenGL Standard Neil Trevett, DuPont Pixel . The Smart Frame Buffer Goes Hollywood: 3D and TV Joel McCormack, Digital Equipment Corporation . A Cached VRAM for 3D Graphics Michael Deering, SUN Microsystems 12:30-2:00 Lunch 2:00-3:30 Video Session Chair: Anoop Gupta, Stanford University . Video Compression Processor for H.320-to-Indeo Transcoding Bryan Martin, Integrated Information Technology . A High Performance Programmable Multi-Standard Video Compression Chip Set David Still, Array Microsystems Inc. . Multimedia Enhancements for PA-RISC Processors Ruby Lee, Hewlett-Packard 3:30-4:00 Break 4:00-6:00 CPUs - 3 Session Chair: Alan J. Smith, UC Berkeley . PowerPC 604 Marvin Denman, IBM . The Thunder SPARC Processor Bruce Lightner, Metaflow Technologies Inc. . The Superscalar Hardware Architecture of the MC68060 Joe Circello, Motorola . A High Performance, Low Power, Pentium Processor Doug Carmean, Lawrence Clark, Robert Rozploch, Intel 6:00 Closing Remarks HOUSING INFORMATION Housing is available on the Stanford University campus in student dormatories which are vacant in the summer. These have central lavatory facilities and cost about $40 per night --- single, $50 per night --- double, for the nights of the symposium. A key deposit of $50 is required that will be refunded at checkout. Housing arrangements on the Stanford campus must be made by August 4 through HOT Chips, not Stanford. Housing is also available at numerous hotels and motels on the peninsula in Palo Alto, Menlo Park, Mountain View, and Los Altos within 5 miles of the campus. You are responsible for making such reservations directly with the motel. If you would like additional housing information, please check the housing information request box on the registration form. IEEE/Computer Society Membership To join call John Gill at (202) 371-0101. With confirmation, you may register for HOT Chips VI at member rates. QUESTIONS? For more information on registration and local arrangements contact Dr. Robert Stewart at (415) 941-6699 or r.stewart@compmail.com. REGISTRATION INCLUDES: * Attendance * Sunday Evening Wine & * One Copy of the Notes Cheese Reception * Two Luncheons * Monday Evening Reception * Coffee Breaks * Parking A Stanford map, parking permit, the location of parking, and a receipt will be mailed to early registrants. ON-SITE REGISTRATION is available Sunday morning before the tutorial and each morning at the Symposium. Early advanced registration is recommended because of the large attendance. Group discounts are available. CANCELLATION OF REGISTRATION: Must be made in writing prior to Sunday, August 7, 1994. A $40 fee will be charged for cancellation. FEDERAL TAX ID NUMBER IS: 13-1656633 for the Institute of Electrical & Electronic Engineers 345 E. 47th Street New York, NY 10017 Use certified mail for registration confirmation. ============================================================================= HOT CHIPS VI REGISTRATION FORM Name________________________________________________________________________ Organization________________________________________________________________ Dept/Mail Stop______________________________________________________________ Mailing Address_____________________________________________________________ City/State/Zip______________________________________________________________ Country_____________________________________________________________________ Area Code/Phone #___________________________________________________________ Email Address_______________________________________________________________ FAX_________________________________________________________________________ Membership: IEEE/CS______ ACM_______ Student______ None______ Society Membership Number___________________________________________________ Check One: ______Check drawn on a U.S. Bank ______MasterCard Make Check Payable To: Hot Chips Symposium ______VISA Name on Credit Card_________________________________________________________ Credit Card #_______________________________________________________________ Expiration Date_____________________________________________________________ Signature___________________________________________________________________ FEES: CIRCLE APPROPRIATE VALUES IEEE/Computer Society $230 or ACM Member Non-Member $290 Student Member $100 Sunday Tutorials, Member $45 Tutorials, Non-Member $60 Tutorials, Student $30 Extra Copy of Notebook $30 (with no mailing) Stanford University Dormitory Housing ______ nights @$40 per night single; $50 double $_________ Arrival__________________ Departure___________________ TOTAL AMOUNT PAID ______________ Electronic Registration, paid via VISA or Mastercard Registration: FAX: (415) 941-5048 EMAIL: r.stewart@compmail.com Surface Mail Registration To: Dr. Robert G. Stewart Stewart Research Enterprises 1658 Belvoir Drive Los Altos, CA 94024 Do NOT put me on the Hot Chips Mailing List _____ Stanford University Housing Information Requested _____Article: 26
In article <316qf6INN5o2@sun004.cpdsc.COM> cshelor@cpdsc.com (Charles Shelor-Consultant) writes: >Welcome to the new Xilinx users: >Get ready to spend a lot more time doing place and route than you >EVER anticipated if you are using XC4010 or larger. >Best bet, skip the Xilinx PPR software and get NeoCad! Tell me more about NeoCad. We have a project on an XC4010 that takes 4 hours to place and route on a Sparc 10/51 with 96 megs of RAM. I sure would like to hear about any alternatives to PPR!!Article: 27
Please feel free to rebroadcast this announcement to any interested groups. One Day Xilinx Summer Workshop "Field Programmable Gate Arrays in the Classroom" Workshop 1 (Viewlogic Based) Date: Friday, August 12 Location: Painter 3.20 (Dell Lab) University of Texas at Austin Austin, Texas Workshop 2 (Mentor Based) Date: Monday, August 15 Location: University of Notre Dame Notre Dame, Indiana Workshop 3 (Mentor Based) Date: Thursday, August 18 Location: Colorado State University Fort Collins, Colorado The Xilinx University Program will be holding several regional one-day "hands-on" workshops in August to introduce university faculty, staff, and graduate students to Field Programmable Gate Arrays as a learning tool in Digital Design courses. The goal of the workshop is to provide the information and assistance necessary to develop and conduct or enhance undergraduate courses in Digital Design using FPGAs. This intensive one day workshop will allow a forum for gathering information and provide first hand experience with industry CAE and FPGA implementation tools. The hands-on labs are structured so the students will perform the entire design flow and see it working in a prototyping board. FPGAs are not just for Digital Design courses. Many universities have utilized the technology for courses in computer architecture, senior research projects, VLSI design, and ASIC design. The flexibility of the architecture, reprogrammability of the devices, and the simplicity of the design tools makes Xilinx FPGAs ideal for the university environment. An instructor can guide their students through an entire design entry, implementation, and simulation process and download results directly into silicon on the desktop. New features in the XACT 5.0 software will be covered. The workshops will be free of charge. A complimentary continental breakfast and lunch will be served on premises. An information packet will be sent to you by Stacey Pinckert of Xilinx. The packet will include directions to the university and a list of lodges in the area. Applications are due two days before the workshop date. ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Topics: Various ways to use Xilinx Xilinx Design Environment Automatic Tools XC4000 Architecture Design Entry Methods FPGA Implementation New XACT 5.0 Features BORG Prototyping Board Xilinx University Program Q&A Session Format: Begins - 9:00am Ends - 5:00pm * Labs will remain open after 5:00pm AM - Lecture (See Topics Section) PM - Hands-on * Design Entry * Implementation * Simulation ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ EMAIL APPLICATION FORM One Day Xilinx Summer Workshop "Field Programmable Gate Arrays in the Classroom" Which workshop do you wish to attend? __________ Name: _________________________________________ Title: ________________________________________ University: ____________________________________ Department: ____________________________________ Address: _______________________________________ Address: _______________________________________ Tel: ___________________________________________ Fax: ___________________________________________ Email: _________________________________________ Presently using FPGAs in course work? Yes____ No____ If yes, is it from Xilinx? Yes____ No____ Are you planning on using FPGAs in your course? Yes____ No ____ Will your University support the use of FPGAs? Yes____ No____ Which design entry tool are you familiar with? _______________ *** Please send your completed application to: Stacey Pinckert Xilinx, Inc. 2100 Logic Drive San Jose, CA 95124 Tel: 408-879-5150 Fax: 408-879-4676 Email: stacey@xilinx.com *** For more information on the Workshop Contact David Lam Xilinx University 2100 Logic Drive San Jose, CA 95030 (T) 408-879-4961 (F) 408-879-4676 Email: dlam@xilinx.com ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~Article: 28
Since some body asked, I'll respond. We are offering a new book which teaches digital logic design using the Intel/Altera FLEXlogic FPGA. The book shows how to build a FPGA prototyping board for about $100 and then uses the board to build combinatorial logic, state machines, and a 4-bit microcomputer. A completely-assembled 4'' x 2'' circuit board is also available for those who want to get started quickly. Several of these boards can be concatenated to build multi-FPGA systems. The book and board use the Intel PLDshell software for programming, and PLDshell is available free from Intel either as floppy disks or from several FTP sites. For a more complete description of these products, send e-mail to devb@vnet.net with the subject line of "fpga". -- || Dave Van den Bout || || Xess Corporation ||Article: 29
richw@lsid.hp.com (Rich Wilson) writes: >I found their support (via internet Email) to be very >rapid and helpful. We must have trained them! We begged for an internet address & we were refused! There were a few issues regarding the iFX780 which the phoneline tech support people hadn't been briefed at all. Glad your experience was a better one. After we got the problems worked out they have worked very good. Cheers, Bob -- <> Bob "Bear" Geer <> <> <> bgeer@xmission.com <> Even paranoids <> <> Salt Lake City, <> have enemies, too! <> <> Utah, USA <> <>Article: 30
I'm interested in finding descriptions of the file formats used by Xilinx and NEOCad. Are these published anywhere? Has anyone reverse-engineered them? The reason I ask is that the company I work for uses Xilinx parts a lot but the placement and routing is a major bottleneck in the design process. I'd like to experiment with some placement and routing methods of my own but I need to read these files to do it. Thanks, David Wheeler wheelerd@clark.netArticle: 31
Hi, Just how expensive is FPGA development? Is it at all possible for an individual to purchase the necessary software/hardware to use high density devices for example at a reasonable cost? I have looked at some vendors such as Lattice and Intel. The Intel SW is quite nice as it allows device independent design, something which the Lattice pLSi sw doesn't seem to do. However these vendors seem to be making a good effort to get their products into the hands of individuals at a reasonable cost. Do I really need to purchase several hundred pounds worth of software to even consider looking at the high density FPGAs? A simple functional design tool would suit me fine, it doesn't have to be fast or Windows based. Any advice appreciated. Best Regards, Roger -- ----------------------------------------------------------------------------- Roger Kinkead EMAIL: roger@rkhost.demon.co.uk "Moorcroft", 27 Antrim Road, Lisburn,County Antrim, Northern Ireland -----------------------------------------------------------------------------Article: 32
In article <775585746snz@rkhost.demon.co.uk>, Roger Kinkead <roger@rkhost.demon.co.uk> wrote: >Hi, > >Just how expensive is FPGA development? Is it at all possible for an individual >to purchase the necessary software/hardware to use high density devices for >example at a reasonable cost? > Hmm... I remember about ~12 years ago I would get Fortran source code for Palasm from MMI; thats right, joe user had to compile the stuff himself - I even added a few routines to create nroff data sheets that would become 'unix-like' man pages. Those were the days... Now we have companies like Xilinx, Actel, etc; these guys are really software companies - Xilinx and Actel will even tell you so if you have the chance to talk to some of there high level engineering folks. --- I wonder why the PC based tools are cheaper than the Unix tools. In the past the argument was that Unix was multi-tasking, therefore you could 'do more' than with a PC running DOS. Five years ago with the home brew Unix license mechanisms of these vendors this was true; today with Highland Digital you can restrict all the way down to a single task per node. What gives? -- /* Mark A. Indovina, mark_indovina@pts.mot.com */ /* Motorola, Inc., Applied Research, Advanced IC Technology Laboratory */ /* Mail Stop 63, 1500 Gateway Avenue, Boynton Beach, FL 33436-8292 USA */ /* phone, 1-407-364-2379, fax, 1-407-364-3904 */Article: 33
Arnim Littek (arnim@actrix.gen.nz) wrote: : the situation. Xilinx seems to be moving away from ASCII outputs, : and NeoCAD has never been that open, as far as I know. However, : that's another issue altogether. I don't think that this is true. All my intermediate files are ASCII. I can't even find a "convert to binary" converter. I'm using XACT 5.0. Roger. -- * As a protest against the recent bunch proposed anti-cryptography * * laws, this message has been doubly encrypted using the rot13 algorithm. * EMail: wolff@dutecai.et.tudelft.nl ** Tel +31-15-783643 or +31-15-142371Article: 34
In article 3809@krabat.marco.de, frank@krabat.marco.de (frank) writes: >We were thinking of using the flexlogic chips here, but >now that Intel is apparently selling all its PLD and flexlogic business >to Altera, who knows what the future for these chips is. Anybody know >and like to comment? > I have included the press release about the sale at the end of this article. You can read it for yourself. Altera will acquire this business on October 1, pending government approval. Until then, we are definitely here doing our jobs. The PLDshell software is available, for free, at ftp.intel.com:/pub/pld_fpga/software/dos/pldshell.zip (about 2 Megabytes to download) As for the book "FPGA Workout", the designs are available in ftp.intel.com:/pub/pld_fpga/designs/fpga-workbook/fpga-wor.zip -or- ftp.intel.com:/pub/pld_fpga/designs/fpga-workbook/fpga-workout.tar An order form for the book and board is a Postscript file in ftp.intel.com:/pub/pld_fpga/designs/fpga-workbook/orderform.ps You can call Xess Corp. at 800-549-XESS (USA) or email to dev@vnet.net for more information. --Richard Vireday Intel PLD & FPGA Business Unit ================= ftp.intel.com:/pub/pld_fpga/press/sold ===================== OTC 07/12 1622 Altera to acquire Intel programmable logic business SAN JOSE, CALIF. (JULY 12) BUSINESS WIRE - July 12, 1994--Altera Corp. (NASDAQ:ALTR) announced Tuesday that it will acquire Intel Corp.'s programmable logic business for a price of approximately $50 million. The purchase price consists of $25 million in cash and 779,277 shares of Altera common stock, subject to certain post-closing adjustments. The transaction is expected to be completed on Oct. 1, subject to governmental approval. Immediately after the closing, Robert Reed, senior vice president and general manager of Intel's Semiconductor Products Group, is expected to join Altera's board of directors. The relationship between the two companies dates back to 1984 when Altera licensed Intel to Altera's first family of products in exchange for process technology and wafer manufacturing services. "Intel is the premier semiconductor company in the world. Our partnership with Intel played a significant role in Altera's early development and we look forward to continuing our relationship with Bob Reed on our board of directors, " said Rodney Smith, president, chief executive officer, and chairman of Altera Corp. As part of the sale, Altera will receive Intel PLD products and licenses to use associated process technologies and intellectual property to build PLD products. In addition, Intel agrees to supply wafers to Altera in support of the business. "Altera is a recognized leader in the programmable logic market. Intel's investment in Altera allows Intel to share in the continued growth and development of the PLD market and to realize the best possible return on its prior investment in this business. I will work actively with Altera to ensure a smooth transition for Intel customers," according to Reed of Intel. The two companies are each planning to prepare customers for the transition of the products to Altera by Oct. 1. Employee transfers are not included in the transaction. However, Altera, at its discretion, may make employment offers to any of Intel's current PLD employees. Altera pioneered the high density PLD market which remains one of today's fastest growing semiconductor areas. The addition of the Intel products complements Altera's broad range of programmable logic solutions. Software support will continue to be provided by Altera through Intel's PLDshell product and will be added into Altera's MAX+PLUS II software in the future. The Intel product line consists of 15 devices ranging from the popular 22V10 to the recently introduced FLEXlogic line. FLEXlogic contains several unique features compared to other high density programmable logic devices on the market and complements Altera's industry leading MAX 7000 and FLEX 8000 families. About Altera Altera Corp., founded in 1983, is a worldwide leader in high-performance, high-density programmable logic devices and associated computer aided engineering (CAE) logic development tools. Programmable logic devices are semiconductor chips that offer on-site programmability to customers. The chips are programmed using tools that run on personal computers or engineering workstations. User benefits include ease of use, lower risk, and fast time-to-market. The company offers the broadest line of CMOS programmable logic devices that address high-speed, high-density and low-power applications. Altera products serve a broad range of market areas, including telecommunications, data communications, computers, and industrial applications. About Intel Intel, the world's largest chip maker, is also a leading manufacturer of personal computer, networking and communications products. -0- NOTE TO EDITORS: Altera, MAX, and FLEX are registered trademarks, and MAX 7000, FLEX 8000, and specific device designations are trademarks of Altera Corp. --30--ahc/sf cs/sf CONTACT: Altera Corp., San Jose Erik Cleage or Robert K. Beachler, 408/894-7000 Intel Corp. Larry Palley or Janet Woodworth, 916/356-6653 Franson, Hagerty and Associates Susan Cain, 415/462-1605 KEYWORD: CALIFORNIA INDUSTRY KEYWORD: COMPUTERS/ELECTRONICS COMED TELECOMMUNICATIONS MANUFACTURING MERGERS/ACQ MANAGEMENT CHANGES Z REPEATS: New York 212-575-8822 or 800-221-2462; Boston 617-330-5311 or 800-225-2030; SF 415-986-4422 or 800-227-0845; LA 310-820-9473Article: 35
> I wonder why the PC based tools are cheaper than the Unix tools. I've always thought it was the perception that those people/groups running UNIX workstations had more money and thus would be able to pay more. -GeorgeArticle: 36
Charles F. Shelor writes: >Welcome to the new Xilinx users: >Get ready to spend a lot more time doing place and route than you >EVER anticipated if you are using XC4010 or larger. >Best bet, skip the Xilinx PPR software and get NeoCad! Yeah. I heard through the grapevine that it may take up to a day to coplete a compilation of Xilinx designs before you can program silicon. But of course this is unconfirmed hearsay and I have not worked w/ the Xilinx design tools. I have heard about NeoCad. I thought it was only a design capture and sim tool. Does it compile designs for target FPGAs as well? John _____________________ All opinions are my own and not those of TI.Article: 37
Hi Everybody, The following is a list of articles and references for FPGA based designs. I would like to thank all of you who replied to my "by-now-forgotten" posting on request for references on FPGA based designs. I have found most of the references. I havent particulary paid any attention to the periodicals so i maniar@charlie.ece.scarolina.edu and i will post a continuation of this posting. Prashant. ~~~~~~~~~ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ~THE FUTURE BELONGS TO THOSE WHO BELIEVE IN THE BEAUTY OF THEIR DREAMS !!!~ ~ ~ ~ \\|// PRASHANT B. MANIAR ~ ~ (o|o) email : maniar@charlie.ece.scarolina.edu ~ ~-----ooO-~-Ooo------- ~ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Literature Survey for FPGA based Design's. ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 1> Electronic design ( periodical ) 2> Computer Design 3> ASIC and EDA ( magazine ) 4> ISSCC proceedings ( what does ISSCC stand for...never heard of it ) ( has details on latest FPGA semiconductor processes and some information on architecture trends.) 5> % DPGA Related Reference Collection % i> http://www.ai.mit.edu/projects/transit/tn95/tn95.html ftp://transit.ai.mit.edu/transit-notes/tn95.ps.Z Transit Note #95 Unifying FPGAs and SIMD Arrays Michael Bolotski, Andre DeHon, and Thomas F. Knight, Jr. Original Issue: September, 1993 ii> http://www.ai.mit.edu/projects/transit/tn100/tn100.html ftp://transit.ai.mit.edu/transit-notes/tn100.ps.Z Transit Note #100 DPGA-Coupled Microprocessors: Commodity ICs for the Early 21st Century Andre DeHon Original Issue: January, 1994 @INPROCEEDINGS{asic_microprocessors, AUTHOR = {M. J. Flynn and R. I. Winner}, TITLE = {ASIC Microprocessors}, BOOKTITLE = {Proceedings of the 22nd Annual Workshop on Microprogramming and Microarchitecture}, YEAR = {1989}, PAGES = {237-243}, ORGANIZATION = {ACM}, PUBLISHER = {ACM Press}, MONTH = {August} } @PROCEEDINGS{fccm93, TITLE = {Proceedings of the IEEE Workshop on FPGAs for Custom Computing Machines}, YEAR = {1993}, EDITOR = {Duncan A. Buell and Kenneth L. Pocek}, ORGANIZATION = {IEEE Computer Society}, PUBLISHER = {IEEE Computer Society Press}, ADDRESS = {Los Alamitos, California}, MONTH = {April} } @ARTICLE{prism_computer93, AUTHOR = {Harvey F. Silverman}, TITLE = {Processor Reconfiguration Through Instruction-Set Metamorphosis}, JOURNAL = {IEEE Computer}, YEAR = {1993}, VOLUME = {26}, NUMBER = {3}, PAGES = {11--18}, MONTH = {March} } @INPROCEEDINGS{prism2_fccm93, AUTHOR = {M. Wazlowski and L. Agarwal and T. Lee and A. Smith and E. Lam and P. Athanas and H. SIlverman and S. Ghosh}, TITLE = {PRISM-II Compiler and Architecture}, BOOKTITLE = {Proceedings of the IEEE Workshop on FPGAs for Custom Computing Machines}, YEAR = {1993}, EDITOR = {Duncan A. Buell and Kenneth L. Pocek}, PAGES = {9--16}, ORGANIZATION = {IEEE Computer Society}, PUBLISHER = {IEEE Computer Society Press}, ADDRESS = {Los Alamitos, California}, MONTH = {April} } @INPROCEEDINGS{cm2x_fccm93, AUTHOR = {Steven A. Cuccaro and Craig F. Reese}, TITLE = {The CM-2X: A Hybrid CM-2/Xilinx Prototype}, BOOKTITLE = {Proceedings of the IEEE Workshop on FPGAs for Custom Computing Machines}, YEAR = {1993}, EDITOR = {Duncan A. Buell and Kenneth L. Pocek}, PAGES = {121--130}, ORGANIZATION = {IEEE Computer Society}, PUBLISHER = {IEEE Computer Society Press}, ADDRESS = {Los Alamitos, California}, MONTH = {April} } @INPROCEEDINGS{splash, AUTHOR = {M. Gokhale, et. al.}, TITLE = {SPLASH: A Reconfigurable Linear Logic Array}, BOOKTITLE = {Proceedings of the International Conference on Parallel Processing}, YEAR = {1990}, PAGES = {526--532}, MONTH = {August} } @TECHREPORT{splash2, AUTHOR = {D. A. Buell}, TITLE = {A Splash~2 Tutorial}, INSTITUTION = {Supercomputing Research Center}, YEAR = {1992}, TYPE = {Technical Report}, NUMBER = {SRC-TR-92-087}, ADDRESS = {Bowie, Maryland}, MONTH = {December} } @TECHREPORT{pam_intro, AUTHOR = {Patrice Bertin and Didier Roncin and Jean Vuillemin}, TITLE = {Introduction to Programmable Active Memories}, INSTITUTION = {DEC Paris Reserch Laboratory}, YEAR = {1989}, TYPE = {PRL Report}, NUMBER = {3}, ADDRESS = {85, Av. Victor Hugo, 92563 Rueil-Malmaison Cedex, France}, MONTH = {June} } @MANUAL{algotronix_cal1024, TITLE = {The Configurable Logic Data Book}, ORGANIZATION = {Algotronix Ltd.}, ADDRESS = {Edinburgh, UK}, YEAR = {1990} } @MANUAL{xilinx, TITLE = {The Programmable Gate Array Databook}, ORGANIZATION = {Xilinx, Inc.}, ADDRESS = {2100 Logic Drive, San Jose, CA 95124}, YEAR = {1989} } @INPROCEEDINGS{furtek_fccm93, AUTHOR = {Frederick Furtek}, TITLE = {Arithmetic Benchmarks for the CLi6000}, BOOKTITLE = {Proceedings of the IEEE Workshop on FPGAs for Custom Computing Machines}, YEAR = {1993}, EDITOR = {Duncan A. Buell and Kenneth L. Pocek}, ORGANIZATION = {IEEE Computer Society}, PUBLISHER = {IEEE Computer Society Press}, ADDRESS = {Los Alamitos, California}, MONTH = {April}, Note = {Paper does not appear in printed proceedings; contact author {\tt fred@clogic.com}} } %% get full names and issue number for rsa_cacm %% NUMBER = {}, @ARTICLE{rsa_cacm, AUTHOR = {R. Rivest and A. Shamir and L. Adleman}, TITLE = {Public Key Cryptography}, JOURNAL = {Communications of the ACM}, YEAR = {1979}, VOLUME = {21}, PAGES = {120-126}, NOTE = {} } @INPROCEEDINGS{pam_rsa, AUTHOR = {M. Shand and J. Vuillemin}, TITLE = {Fast Implementations of RSA Cryptography}, BOOKTITLE = {Proceedings of the 11th Symposium on Computer Arithmetic}, YEAR = {1993}, EDITOR = {Earl {Swartzlander Jr.} and Mary Jane Irwin and Graham Julien}, PAGES = {252--259}, ORGANIZATION = {IEEE Computer Society}, PUBLISHER = {IEEE Computer Society Press}, ADDRESS = {Los Alamitos, California}, MONTH = {June} } @TECHREPORT{pam_performance_assessment, AUTHOR = {P. Bertin and D. Roncin and J. Vuillemin}, TITLE = {Programmable Active Memories: A Performance Assessment}, INSTITUTION = {DEC Paris Reserch Laboratory}, YEAR = {1992}, TYPE = {PRL Report}, ADDRESS = {85, Av. Victor Hugo, 92563 Rueil-Malmaison Cedex, France}, MONTH = {June} } @INPROCEEDINGS{dzung_splash2_fccm93, AUTHOR = {Dzung T. Hoang}, TITLE = {Searching Genetic Databases on Splash~2}, BOOKTITLE = {Proceedings of the IEEE Workshop on FPGAs for Custom Computing Machines}, YEAR = {1993}, EDITOR = {Duncan A. Buell and Kenneth L. Pocek}, PAGES = {185--191}, ORGANIZATION = {IEEE Computer Society}, PUBLISHER = {IEEE Computer Society Press}, ADDRESS = {Los Alamitos, California}, MONTH = {April} } @ARTICLE{splash_computer91, AUTHOR = {Maya Gokhale and William Holmes and Andrew Kopser and Sara Lucas and Ronald Minnich and Douglas Sweely and Daniel Lopresti}, TITLE = {Building and Using a Highly Programmable Logic Array}, JOURNAL = {IEEE Computer}, YEAR = {1991}, VOLUME = {24}, NUMBER = {1}, PAGES = {81--89}, MONTH = {January} } @INPROCEEDINGS{lopresti_advvlsi91, AUTHOR = {Daniel Lopresti}, TITLE = {Rapid Implementation of a Genetic Sequence Comparator Using Field-Programmable Logic Arrays}, BOOKTITLE = {Advanced Research in VLSI}, YEAR = {1991}, EDITOR = {Carlo H. S\'equin}, PAGES = {138--152}, PUBLISHER = {MIT Press}, ADDRESS = {Cambridge, MA}, MONTH = {April} } @INPROCEEDINGS{sorting_on_algotronix_fccm93, AUTHOR = {Wayne Luk and Vincent Lok and Ian Page}, TITLE = {Hardware Acceleration of Divide-and-Conquer Paradigms: a Case Study}, BOOKTITLE = {Proceedings of the IEEE Workshop on FPGAs for Custom Computing Machines}, YEAR = {1993}, EDITOR = {Duncan A. Buell and Kenneth L. Pocek}, PAGES = {192--1201}, ORGANIZATION = {IEEE Computer Society}, PUBLISHER = {IEEE Computer Society Press}, ADDRESS = {Los Alamitos, California}, MONTH = {April} } @INPROCEEDINGS{dbc_fccm93, AUTHOR = {Maya Gokhale and Ron Minnich}, TITLE = {FPGA Computing in a Data Parallel C}, BOOKTITLE = {Proceedings of the IEEE Workshop on FPGAs for Custom Computing Machines}, YEAR = {1993}, EDITOR = {Duncan A. Buell and Kenneth L. Pocek}, PAGES = {94--101}, ORGANIZATION = {IEEE Computer Society}, PUBLISHER = {IEEE Computer Society Press}, ADDRESS = {Los Alamitos, California}, MONTH = {April} } @MANUAL{xilinx_product_guide_93, TITLE = {Product Description and Selection Guide}, ORGANIZATION = {Xilinx, Inc.}, ADDRESS = {2100 Logic Drive, San Jose, CA 95124}, MONTH = {April}, YEAR = {1993} } @INPROCEEDINGS{luk-fccm94, AUTHOR = {Wayne Luk and Teddy Wu and Ian Page}, TITLE = {Hardware-Software Codesign of Multidimensional Programs}, BOOKTITLE = {Proceedings of the IEEE Workshop on FPGAs for Custom Computing Machines}, YEAR = {1994}, EDITOR = {Duncan A. Buell and Kenneth L. Pocek}, ORGANIZATION = {IEEE Computer Society}, PUBLISHER = {IEEE Computer Society Press}, ADDRESS = {Los Alamitos, California}, MONTH = {April} } @INPROCEEDINGS{siu-fccm94, AUTHOR = {Greg J. Gent and Scott R. Smith and Regina L. Haviland}, TITLE = {An FPGA-based Custom Coprocessor for Automatic Image Segmentation Applications}, BOOKTITLE = {Proceedings of the IEEE Workshop on FPGAs for Custom Computing Machines}, YEAR = {1994}, EDITOR = {Duncan A. Buell and Kenneth L. Pocek}, ORGANIZATION = {IEEE Computer Society}, PUBLISHER = {IEEE Computer Society Press}, ADDRESS = {Los Alamitos, California}, MONTH = {April} } 6> %% stuff to potentially include %% ftp.eecg.toronto.edu %% pub/tech_reports/SteveBrown %% cicc94.fm.ps.Z %% phd.ps.tar.Z %% stochastic.ps.Z 7> August '93 edition of "Journal of VLSI Signal Processing", Special Issue on FPGAs (Kluwer Academic Publishers). 8> Altera's data books. 9> Xilinx data books. 10> A. Wolfe, J.P.Shen, " Flexible Processors : A promising Application Specific Processor Design Approach," Proceedings of MICRO '21. 11> Christian Iseli & Eduardo Sanchez, "Beyond Superscalar Using FPGAs", IEEE International Conference on Computer Design, Cambridge Mass., October 1993. 12> Christian Iseli & Eduardo Sanchez, "Spyder: A SURE (SUperscalar and REconfigurable) Processor", The Journal of Supercomputing, to be published in autumn 1994. 13> anon ftp from ftp.inf.ethz.ch i> Author: N. Wirth, S. Ludwig Title: An Extension-Board with an FPGA for Experimental Circuit Design CL - An Editor for the CLi6000 Field Programmable Gate Array and its Implementation CL-Editor User Manual Dokument: 198 / 1993 ii> Author: H. Eberle, S. Gehring, S. Ludwig, N. Wirth Title: Tools for Digital Circuit Design using FPGAs Document: 215 / 1994Article: 38
In article <Ctv06F.671@pts.mot.com> ep520mi@pts.mot.com (MARK INDOVINA Xxxxx Ppppp) writes: >Now we have companies like Xilinx, Actel, etc; these guys are really >software companies - Xilinx and Actel will even tell you so if you have >the chance to talk to some of there high level engineering folks. I don't agree. These software products look great on the marketing man's papers, but when you really start doing something you'll notice that the tools do just about the minimum, what they have to. Well, some people like to say that a SW product is never ready. And these companies do get their bread from the HW... qru -- lkuru@snakemail.hut.fiArticle: 39
In article <m363Yc.gdZ@asic.sc.ti.com>, jmassoth@asic.sc.ti.com (John Massoth) writes: > > Yeah. I heard through the grapevine that it may take up to > a day to complete a compilation of Xilinx designs before you > can program silicon. But of course this is unconfirmed hearsay > and I have not worked w/ the Xilinx design tools. I have been using Xilinx and ppr for about two years. I have not used any other competing programs, so have no other point of reference. The compilation time of ppr depends very much on how far you are pushing the technology, both in how much of the Xilinx device is being used, and in your speed requirements. I have used devices ranging from a 4005 to a 4010. I run ppr on a SparcStation IPX with clock-doubled CPU. For any of the devices, with up to about 70% utilization of function generators (I tend to use them up first) and about 60% utilization of flip flops, I see from 10 to 15 minutes of ppr time for one run. If you use the path timing capability of ppr (where you can place an upper limit on the time from a flop's output to another flop's input, the time can go up by a factor of three or four. I've tried path timing only once and found it very lacking compared to my usual approach of flagging critical nodes and doing hand placement of flops for time critical portions of the circuit. I have found that if I go over about 75% utilization of function generators, ppr has real problems. I suspect that it is because my circuits don't always fit cleanly into the Xilinx model of four inputs feeding each flop, so the actual resource utilization is much more than 70%. Occasionally ppr will take a long time (30 minutes instead of 10 minutes) when it does a really poor placement. My really time-critical circuits meet the four-input constraint for all time critical flops, and I have completely hand-placed a couple of 4006's to get things to run fast enough. ppr didn't have a chance of meeting my timing requirements without this. All of this is to say that "up to a day" is, in my experience, not close to the truth. Of course, if you need 100 runs to get one with good enough timing, it can easily take a day. Also, getting the first one through the tool set will take a while, just to learn how the tools work. David Eby Tektronix, Inc.Article: 40
Arnim Littek wrote in article <Cto8vK.FzJ@actrix.gen.nz> : >> >>cshelor@cpdsc.com writes >>>Get ready to spend a lot more time doing place and route than you >>>EVER anticipated if you are using XC4010 or larger. >>> >>>Best bet, skip the Xilinx PPR software and get NeoCad! >> >>I don't suppose you could be a bit more specific in this case? Have >>you got some side-by-side examples of the same data being submitted >>to PPR and to the NeoCAD tools? >> >>We are still using the 3K stuff principally, and the APR tool, so I >>am not willing to comment on the PPR stuff, but in our comparisons, >>from 2018 devices right through to 3195, there are some patterns. >> >>Firstly, the NeoCAD tools are a lot faster than the Xilinx factory >>tools, by a factor of 3-4 times, particularly on the larger designs. >>This is based on the APR V3.30 which was in the last pre-XACT5.0 >>release. (Haven't tried with the APR in XACT 5.0) The NeoCAD tools >>I've tried on are V4.1. >> >>However, in defence of APR, it is seems more flexible, allowing >>us to tailor the placement and routing in ways that the NeoCAD >>isolates the user from. For that matter, APR also seems to have that >>all over PPR too. Too bad its being put on the shelf. >> >>The trend seems to be to isolate the user from any understanding of >>what's happening. We much prefer the use of intermediate files in >>ASCII, rather than hiding things in proprietary binary files. Nobody >>is perfect at error messages, and the more information we have to go >>on when the going gets tough, the more likely we are to get out of >>the situation. Xilinx seems to be moving away from ASCII outputs, >>and NeoCAD has never been that open, as far as I know. However, >>that's another issue altogether. >> >>Arnim Littek. arnim@digitech.co.nz >> >> >>-- >> arnim@actrix.gen.nz >>Article: 41
I need large net lists for a current study of cpld/fpga/asic design. 10K gates/1000clb/1000le/1000modules would be fine, 100K gates even better. Edif/xnf/verilog/vhdl is ok. Does anyone know of such a database (or have any to donate). I don't need the functional specification, I am only interested in the geometric structure. I already know about mcnc, so no need to reference them. Thanks for your help Stephen -------------------------------------------------------- home: sjsmith@netcom.com work: sjsmith@altera.com --------------------------------------------------------Article: 42
--- We have developed some (in house) layout tools that are timing driven and do great job when compared with Xilinx tools for some of the MCNC and other benchmark examples. I am interested in getting some design examples with timing constraints that are desired. Currently we do routing using our own fast multi-terminal router but that is going thru changes. The placement part is fully automated. Correct lca files are generated to perform routing using Xilinx router(s). Anybody willing to share design examples (nda no problem) would help us a lot!! --dinesh Dinesh Bhatia University of Cincinnati --Article: 43
Hi all, How does XACT5.0 compare to XACT 4.03? I have used xact 4.03 and don't think much of it, espcially the xblox library. Has the routing become more predictable? have they allowed the designer more access to routing resources? Are there any *TRUE* incremental routing utils in XACT 5.0? Can we generate "hard macros" in xact 5.0? can we route a functional-block of arbitrary fix it in that position in subsequent routing ? your views would be appreciated Ahmed... ------------------- Attention To Detail | Tel: +44 926 843 444 Unit 3, Nunhold Business Centre | FAX: +44 926 843 363 Dark Lane, Hatton | Modem: +44 926 843 587 Warwick, U.K. CV35 8XB | Email: ahmed@atdetail.demon.co.uk -------------------Article: 44
In article <LKURU.94Aug2200742@alpha.hut.fi> lkuru@snakemail.hut.fi writes: >>Now we have companies like Xilinx, Actel, etc; these guys are really >>software companies - Xilinx and Actel will even tell you so if you have >>the chance to talk to some of there high level engineering folks. > >I don't agree. These software products look great on the marketing >man's papers... >And these companies do get their bread from the HW... Not to be grumpy or anything :-), but if they got their bread from their hardware, one would expect them to be enthusiastically handing out free copies of the software to promote hardware sales. Nothing could be farther from the truth... -- "We must choose: the stars or the dust.| Henry Spencer @ U of Toronto Zoology Which shall it be?" -H.G.Wells | henry@zoo.toronto.edu utzoo!henryArticle: 45
Just to add another data point, I have been using Xilinx ppr tool for about a year. My target devices are 4010 parts, with a utilization of 50-60%. However, my place and route times are on the order of 2-3 hours using a sparcstation 10. Some of the things which are no doubt causing added processing are: 1) The abundant use of path timing constraints, some of which are tight. 2) The use of several hard counter macros (which obstruct routing). 3) The use of several 16 and 32-bit busses (which require lots of routing). 4) Setting place and route effort levels fairly high to get good timing. --Charles flaig@3do.comArticle: 46
Baiju Jacob (jake@astro.ocis.temple.edu) wrote: Ok, A design I was part of used 14 Altera MAX devices to build a superclone of the RCA 1802. We called it the 18020. This design was then exported to workstations and then built into custom ASICs. All in all a great experience! bob proffittArticle: 47
Has anyone tried ever to create a pierce x-tal oscillator using I/O pins from a FPGA (I'm thinking here about the intel 780)? I would need this for a slowish (<10MHz) oscillator on a single-chip solution using the FPGA. Any luck, or should I use an external oscillator? Frank -- ------------------------------------------------------------------------ Frank A. Vorstenbosch +31-70-355 5241 P.O. Box 85800 Kingswood Software Fax/Modem +31-70-355 8674 2508 CM The Hague prompt@hacktic.nl The NetherlandsArticle: 48
The key to getting the s/w free is to dangle a large enough quantity of parts to be sold in front of the Rep/Disti/Dude. Most of these companies want to sell 10,000 or 100,000 parts and are willing to hose the smaller companies/individuals and only hand out freebies where they think they can get business. Otherwise software is just another profit center :-( In a sense this is similar to NRE costs on ASICs - you can pay now, or make it up in volumes later. I would love to see all of the FPGA guys come out with a bare-bones product for free - no auto place & route, no support, just a disk - take netlist in, spit timing out, and maybe an interface to a simulator and a bitstream for the part. That would keep the 'freebie' people happy, and save the real effort for the paying customers.Article: 49
Hello, I'm having problems with my mouse when I start MAKEBITS from XDE. Somehow the program recognizes my mouse as a XCHECKER cable. Thus I can't use my mouse. I already tryed the 'port' command, and to save a proper configu- ration in the profile, but that doesn't help. Somehow the problem can be fought by moving the mouse when starting MAKEBITS. But that works only in 2 of 3 cases. And I feel really stupid jerking around while starting a program.. :) I'm not quite sure if this is the right place to ask, but I know no better place.. If you can help me please answer by pm. Thank you Thomas Hadlich -- -------------------------------------------------------------------------- | Thomas Hadlich hadlich@infaut.e-technik.tu-magdeburg.DE | | hadlich@csmd.cs.tu-magdeburg.de | | | | snail: Joh.-Kirsch-Strasse 19; D-39106 Magdeburg; Germany | | Tel. : (0391) 5611021 | --------------------------------------------------------------------------
Site Home Archive Home FAQ Home How to search the Archive How to Navigate the Archive
Compare FPGA features and resources
Threads starting:
Authors:A B C D E F G H I J K L M N O P Q R S T U V W X Y Z