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
In article <1995Feb9.125827@wcsub.ctstateu.edu>, belanger002@wcsub.ctstateu.edu writes: > TI calls the product "ASSET. > From the OEM price book, the card appears to cost abour $500, > and the software, $3500. > Boy, was I ever off! I was looking at the wrong page. TI part # ASSET-AIT-PC ASSET Interconnect Test Product, $14,995.00 Double Ouch! Gerry BelangerArticle: 701
neal@ctd.comsat.com (Neal Becker) writes: >octtools is part of ptolemy. Try ftp.ptolemy.berkeley.edu. Yes, >ptolemy has been ported to Linux! No, *parts* of octtools are part of Ptolemy; the Oct database stuff and the Vem editor are shipped with Ptolemy but the CAD tools aren't. Also, the site name is ptolemy.eecs.berkeley.edu. See the WWW page http://ptolemy.eecs.berkeley.edu/ for more on Ptolemy. There's also a newsgroup: comp.soft-sys.ptolemy -- -- Joe Buck <jbuck@synopsys.com> (not speaking for Synopsys, Inc) Phone: +1 415 694 1729Article: 702
>>Does anyone know of an inexpensive method for accessing the boundary scan >>capabilities of an FPGA? I have found several devices to do this, but so >>far the boundary scan capabilities have all been add ons to extremely >>expensive test equipment. All we need is some kind of device (PC card and >>cable?) to interface with the IEEE1149.1 standard TAP, and some software to >>drive it. The Altera FLEXlogic FPGAs are programmed through a JTAG port. All they do is hook the TCK, TMS, and TDI pins to the pins of a PC port. The TDO pin hooks to one of the status lines of the printer port. Then it's a relatively simple matter of pulsing the TCK line with the appropriate values on the TMS and TDI lines to make the receiving device obey JTAG commands. I assume you could do the same with any device that has JTAG capabilities. -- || Dave Van den Bout || || Xess Corporation ||Article: 703
In article <3h5s7sINNc3r@salmon.wv.tek.com> weedk@salmon.wv.tek.com (Kirk A Weedman) writes: >Anyone know where I can get a Verilog description >for SCSI design? > > Kirk weedk@pogo.wv.tek.com Mee too! Any primer on Verilog models of SCSI protocols/components/the lot would be received with gratitude! Frank Ieromnimon, PACE Project, dept. of Computer Science, University of Essex.Article: 704
I think you're thinking about the Asset system, which TI sold off to another company, and I have their literature around here somewhere. You might want to see Proceedings of the International Test Conf. 1992, Baltimore, MD Sept. 20-24 pgs. 84-90 "Design Verification of a High Density Computer Using IEEE 1149.1" by Wayne T. Daniel, Texas Instruments 6500 Chase Oaks Blfd. M/S 8407 Plano, TX 75023. In that paper, the strategy uses Asset to scan through their design. Randy B.Article: 705
Hello. I need to Manchester decode a 5Mbit per second serial bit stream. I was wondering if it would be practical to implement a digital phase-locked loop inside an XC4008 Xilinx chip? This way, I could recover the clock signal without any external circuitry. Does this sound like a workable idea? Any idea how many CLB's/gates/etc. it would take, given that the Xilinx chip has access to a 40MHz system clock? (Ok, I'm dreaming now) Does anyone have working logic that does this? Regards, Jeff -- -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- Jeff Collins jwcollin@engr.ucdavis.edu Intelligent Manufacturing Systems/ collinsj@ece.ucdavis.edu Mechatronics Lab, 1065 Bainer Hall collinsj@cs.ucdavis.edu University of California, Davis jwcollins@ucdavis.edu -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-Article: 706
Altera just announced that it's giving out free PCI design kits (which are essentially generic templates that users can customize to their specific PCI oriented design) that target its MAX 7000, MAX 9000, FLEX 8000 and FLEXlogic device families. Note: this is not a drop-in PCI interface but something an Altera customer can use as a good start towards putting a PCI compliant interface in their FPGA design. Right now this PCI kit is written in Altera HDL (AHDL -- which is somewhat like VHDL) and it's fully supported in Altera's MAX+PLUS II proprietary synthesis tool. (I'm told that non-AHDL oriented designers can also import Verilog, EDIF & VHDL into the MAX+PLUS II design environment.) To get the free Altera PCI design kit, call (800) 525-8372 or (408) 894-7144. The PCI Local Bus spec can be purchased from the PCI Special Interest Group for $25.00 plus shipping at (800) 433-5177 or (503) 797-4207. - John Cooley part time EDA Consumer Advocate full time ASIC & FPGA Contract Designer =========================================================================== Trapped trying to figure out a Synopsys bug? Want to hear how 3196 other users dealt with it ? Then join the E-Mail Synopsys Users Group (ESNUG)! !!! "It's not a BUG, jcooley@world.std.com /o o\ / it's a FEATURE!" (508) 429-4357 ( > ) \ - / - John Cooley, EDA & ASIC Design Consultant in Synopsys, _] [_ Verilog, VHDL and numerous Design Methodologies. Holliston Poor Farm, P.O. Box 6222, Holliston, MA 01746-6222 Legal Disclaimer: "As always, anything said here is only opinion."Article: 707
In article <D3nHH4.EqD@acsu.buffalo.edu> crouse@acsu.buffalo.edu (Daniel R. Crouse) writes: >Does anyone know of an inexpensive method for accessing the boundary scan >capabilities of an FPGA? I have found several devices to do this, but so >far the boundary scan capabilities have all been add ons to extremely >expensive test equipment. All we need is some kind of device (PC card and >cable?) to interface with the IEEE1149.1 standard TAP, and some software to >drive it. Try for instance (Europe): JTAG Technologies BV tel: +31-40-785104 P.O.Box 1542 fax: +31-40-785739 5602 BM Eindhoven The Netherlands Or (USA): Corelis Inc. tel: (310) 926-6727 12607 Hidden Creek Way fax: (310) 404-6196 Cerritos, CA 90701 USA They sell a small device to plug into the Centronix port of your PC, together with software. If you need VXI-interface, they have that as well. You may also buy a getting started kit for Boundary Scan, if that is needed. We bought one of these testers this winter, but I have not yet had time to play with it. Good luck, Bjorn BL. ______________________________________________________________________ s-mail: e-mail: | | | Bjorn B. Larsen, Ph. D. bjoernb@elektro.tih.no |__ |__ | Sor-Trondelag College | \| \| Gunnerus gate 1 |__/|__/|_ N-7005 TRONDHEIM tel: +47 - 7389 6288 NORWAY fax: +47 - 7389 6286 ______________________________________________________________________Article: 708
Does anybody have an idea on how to evaluate the time to integrate in a macrocell or fpga :a small 16bit, 65kwords, dma, i/o interface, float unit running at 18Mhz. Is it possible to do that if my company would pay 100k to 150k in r/d? Any CommentsArticle: 709
followup-to: comp,arch.fpga Hi, I'm currently working on the design of a video RAMDAC evaluation/demonstration board, and would like some comments and/or help on an idea I have. The board is like this: -------------------------------- | VRAMr | PC | _____ |- PARALLEL O ------ V |RAM| | -> RGB/sync out INTERFACE O |FPGA| R |DAC| |- ---> O ------ A ----- | | XT M | | AL b VRAMg | |______________________________| The FPGA can be programed via the pc printer port. The RAMDAC is a fast part (220MHz), so there have to be two ways to display an image, 1. A true color image downloaded to the VRAMS (works up to 30MHz dot clock officaly, but will work to 50-60MHz (most of the time) if the supply voltage is changed slightly), 2. For the fast pixel rates, via pattern generation on the FPGA. (grids, rams cross hatch etc.) The FPGA also handles the SYNC generation, and the control interface to the RAMDAC. However, I was wondering if it is possable to use the FPGA to generate a fractal image, in *real time*, to display via the RAMDAC. I know that such a task is maths intensive, but: 1. The FPGA has 5000 internal gates. 2. It can run at 250Mhz internaly 3. A low resolution image will do (say 160x100). 4. A low limit on zooming into the fractal will do. 5. A lowish video frame rate will do (<50Hz) That gives about 160x100x50=800K pixels/sec to calculate. =~300 clock cycles per pixel Can a point on the Mandel set (or any other fractal) be calculated in such a time (in hardware)? I would like to be able to pan and zoom the image, via the parallel interface, but that should be trivial once the calculation routine works. So if any good Fractal experts/ALU designers/FPGA experts find this idea interesting and/or have any ideas, please *email* me at the address below, since my NEWS access is slow, so I may not be able to read all the messages in all the groups that this was posted to. If there is interest in the topic, I will sumarise and email/post the results to interested parties/newsgroups. Thanks, Ciaran King. Ciaran King cking@accutron.ie /\ Accutron Ltd. - Product R&D / \ * Process monitoring/control / <--\- * Embedded systems design / \ * Data acquisition systems /___/\___\Article: 710
I caught some discussion about sources for JTAG S/W. Anyway, AT&T offer something they call "Tapdance" that runs on DOS & SunOS. I don't know anything about it but the phone number is: 1-800-462-8146. The datasheet I found also had a contact name of Scott Davidson sd@ohm.att.com They also produce a Boundary-Scan Master Evaluation Kit for the PC which includes H/W & S/W to turn a PC into a tester. They also produce a Boundary Scan Master I.C. part no. 497AA.Article: 711
Did a PLL and Manchester Decode in an Altera EPLD, worked out great. Altera actually had alot of it defined in one of thier applications notes. You can call Altera and they will send them to you. Good Luck. ---- J. Scott DicksonArticle: 712
Hello, I have synthecized some combinational designs with 5 input and only 1 output with FPGA Compiler from Synopsys, but they are made with more than one CLB in Xilinx. Definitly, for a 5 to 1 combinational logic, only one CLB (XC4000) is enough. What is happening? M.R.Movahedin P.S.:a sample vhdl code follows: result: 3 XC4000 CLBs without any timing constraints ! --********************************************************************* library ieee; use ieee.std_logic_1164.all; entity ctrl2 is port(a,b:in std_logic; z:out std_logic; ctrl:in std_logic_vector(2 downto 0)); end ctrl2; architecture rtl of ctrl2 is begin z<= a and b when ctrl="000" else a or b when ctrl="100" else a nand b when ctrl="001" else a nor b when ctrl="101" else a xor b when ctrl="010" else not(a xor b)when ctrl="111" else a and not b when ctrl="011" else a or not b when ctrl="110" else '-'; end rtl; --*********************************************************************Article: 713
!! To all users interested in DSP Products !! A new html catalog is available for the user community on DSPnet. You can see processors, boards and software as well as companies with DSP related products. (Over 700 Products listed) Access is free !! URL: http://www.dspnet.com or telnet to dspnet.com -- DSPnet, Inc. Voice: (617) 642-1600 49 River Street FAX: (617) 899-4449 Waltham, MA 02154 E-mail: dspadmin@dspnet.comArticle: 714
M. Movahedin (movahed@tumlis.lis.e-technik.tu-muenchen.de) wrote: : Hello, : I have synthecized some combinational designs with 5 input and only 1 output : with FPGA Compiler from Synopsys, but they are made with more than one CLB in : Xilinx. Definitly, for a 5 to 1 combinational logic, only one CLB (XC4000) is : enough. : What is happening? i This might be a stupid question, but are you sure it's reporting CLB's and not function generators? I'd expect to see three function generators for a function of five inputs. If you are in douby have a look at the XNF file, the format is pretty straightforward. Regards, Aedan Coffey. -- ================================================================= Aedan Coffey, Toucan Technology, Technology Centre, Mervue, Galway, Ireland. Phone +353-91-757223/770007 Email: coffey@toucan.ie Designers of Electronic Systems and ASICs.Article: 715
PRELIMINARY ADVANCED PROGRAM ************************************************** * * * FIFTH GREAT LAKES SYMPOSIUM ON VLSI * * MARCH 16 - 18, 1995, * * BUFFALO MARRIOTT, BUFFALO, NEW YORK, USA * ************************************************** CO-SPONSORED BY: IEEE COMPUTER SOCIETY TECHNICAL COMMITTEE ON VLSI IEEE CIRCUITS AND SYSTEMS SOCIETY STATE UNIVERSITY OF NEW YORK AT BUFFALO IN COOPERATION WITH ACM SIGDA GENERAL CHAIR : S. Chakravarty, University at Buffalo PROGRAM CHAIRS: R. Sridhar & S. J. Upadhyaya, University at Buffalo PROGRAM COMMITTEE: Jacob A. Abraham University of Texas at Austin Vishwani D. Agrawal AT&T Bell Laboratories C. Anagnostopoulos Eastman Kodak Co Jason Cong University of California at Los Angeles Warren Debany Griffiss AFB M.I. Elmasry University of Waterloo Eby Friedman University of Rochester Dwight Hill Synopsys Rajiv Jain University of Wisconsin-Madison Bozena Kaminska University of Montreal S. M. Kang University of Illinois at Urbana-Champaign John Oldfield Syracuse University C. A. Papachristou Case Western Reserve University C.A.T. Salama University of Toronto Edwin Hsing-Mean Sha University of Notre Dame Naveed Sherwani Western Michigan University S. Verdonckt-Vandebroek Xerox Corporation LOCAL ARRANGEMENTS: Victor Demjanenko, University at Buffalo STEERING COMMITTEE: Naveed Sherwani, Western Michigan University PROGRAM SCHEDULE: MARCH 16, 1995 THURSDAY ----------------------- ------------------------------------------------------------------ 8:00 - 9:00 AM Continental Breakfast ------------------------------------------------------------------ 9:00 - 9:30 AM OPENING REMARKS ************************************************************************ 9:30 - 10:30 AM ** KEYNOTE ADDRESS ** 21ST CENTURY GIGASCALE INTEGRATION JAMES D. MEINDL, Joseph M. Pettit Chair Professor of Microelectronics Georgia Institute of Technology ************************************************************************ ----------------------------------------------------------------- 10:30 - 11:00 AM Coffee Break ----------------------------------------------------------------- 1A: 11:00 - 12:30 PM SESSION 1A: SYNTHESIS I Timing-Driven Circuit Implementation D. Karayiannis and S. Tragoudas Southern Illinois University Optimization Using Implicit Techniques for Industrial Designs F. Poirot, R. Roane and G. Tarroux Compass Design Automation, France Optimal Technology Mapping for Single Output Cells U. Hinsberger and R. Kolla Universitat Wurzburg, Germany 1B: 11:00 - 12:30 PM SESSION 1B: ANALOG VLSI A Time-Differential Equation Approach to Analog Design Automation M. L. Manwaring and D. J. Klein Washington State University A New Approach for Modeling and Optimization of Analog Systems E. Penn and L. Schelovanov University of New Haven A Scalable Analog Architecture for Neural Networks with On-Chip Learning and Refreshing B. A. Alhalabi and M. Bayoumi University of Southwestern Louisiana ----------------------------------------------------------------- 12:30 - 2:00 PM Lunch Break ----------------------------------------------------------------- 2A: 2:00 - 3:30 PM SESSION 2A: PHYSICAL DESIGN I Bus Minimization and Scheduling of Multi-Chip Systems M. Sheliga and E. Sha University of Notre Dame Thumbnail Rectilinear Steiner Trees J. L. Ganley and J. P. Cohoon University of Virginia A Two-Stage Simulated Annealing Methodology J. M. Varanelli and J. P. Cohoon University of Virginia Wiring Space Optimization in Slicing Floorplans J. T. Mowchenko and Y. Yang University of Alberta 2B: 2:00 - 3:30 PM SESSION 2B: LOW POWER DESIGN Estimating Worst-Case Power Consumption of CMOS Circuits Modeled as Symbolic Neural Networks E. Macii and M. Poncino Politecnico di Torino, Italy Design and Analysis of a Low Power Energy-Recovery Adder N. Tzartzanis and W. C. Athas University of Southern California - ISI Coding a Terminated Bus for Low Power M. R. Stan and W. P. Burleson University of Massachusetts Circuit/Architecture for Low-Power High-Performance 32bit Adder I. S. Abu-Khater, A. Bellaouar, M. I. Elmasry and R. H. Yan University of Waterloo ----------------------------------------------------------------- 3:30 - 4:00 PM Coffee Break ----------------------------------------------------------------- 3A: 4:00 - 5:30 PM SESSION 3A: SYNTHESIS II Symbolic Execution of Data Paths C. Monahan and F. Brewer University of California, Santa Barbara Specification and Synthesis of Bounded Indirection K. Rath, M. E. Tuna and S. D. Johnson Indiana University Synthesis of SEU - Tolerant Sequential Circuits using Concurrent Error Correction H. Hollander, B. S. Carlson and T. D. Bennett SUNY at Stony Brook Scheduling Conditional Data-Flow Graphs with Resource Sharing J. Siddhiwala and L.-F. Chao Iowa State University 3B: 4:00 - 5:30 PM SESSION 3B: VERIFICATION Automated Verification of Temporal Properties Specified as State Machines in VHDL Y. V. Hoskote, J. A. Abraham and D. S. Fussell University of Texas at Austin Partitioning Transition Relations Effeciently and Automatically Z. Zhou, X. Song, F. Corella, E. Cerny and M. Langevin University of Montreal Using Symbolic Rademacher-Walsh Spectral Transforms to Evaluate the Correlation between Boolean Functions E. Macii and M. Poncino Politecnico di Torino, Italy PANEL 5:45 - 7:00 PM PANEL DISCUSSION: Low Power vs High Speed: Can you have both? Moderator: Eby G. Friedman, University of Rochester Panelists: James D. Meindl, Georgia Institute of Technology William C. Athas, University of Southern California/ISI Tom Wik, AT\&T Bell Laboratories, Allentown, PA. ----------------------------------------------------------------- 7:30 - 9:00 PM Conference Banquet ----------------------------------------------------------------- MARCH 17, 1995, FRIDAY ---------------------- ------------------------------------------------------------------ 8:00 - 9:00 AM Continental Breakfast ------------------------------------------------------------------ 4A: 9:00 - 10:30 AM SESSION 4A: PHYSICAL DESIGN II An Efficient Building Block Layout Methodology for Compaction Placement N. G. Bourbakis and M. Mortazavi SUNY Binghamton Performance Driven Standard-cell Placement Using the Genetic Algorithm S. M. Sait, H. Youssef, K. Nassar and M. S. T. Benten King Fahd Univ. of Petroleum & Minerals, Saudi Arabia An Efficient Heuristic Approach on Minimizing the Number of Feedthrough Cells in Standard Cell Placement J.-T. Yan National Chiao Tung University, Taiwan Priority Driven Channel Pin Assignment I. Peters Humboldt-University of Berlin, Germany 4B: 9:00 - 10:30 AM SESSION 4B: ARCHITECTURE AND DESIGN I A Systolic Algorithm and Architecture for Image Thinning N. Ranganathan and K. Doreswamy University of South Florida Analysis and Verification of Locally Clocked Circuits with the Concurrency Workbench G. Baulch, D. Hemmendinger and C. Traver Union College Automatic Rapid Prototyping of Semi-Custom VLSI Circuits using FPGAs J.-T. Yoo, K. F. Smith and E. Brunvand University of Utah A Local Clocking Approach for Self-timed Datapath Designs S.-J. Kim and R. Sridhar SUNY at Buffalo ----------------------------------------------------------------- 10:30 - 11:00 AM Coffee Break ----------------------------------------------------------------- 5A: 11:00 - 12:30 PM SESSION 5A: SYNTHESIS III A Soft Computing Approach to Hardware Software Codesign V. Catania, N. Fiorito, M. Malgeri and M. Russo Universita di Catania, Catania Technology Mapping Algorithms for Sequential Circuits Using LUT Based FPGAs S. Habib and Q. Xu City University of New York Modeling of Communication Protocols in VHDL A. Assi and B. Kaminska Ecole Polytechnique, Montreal Using EDIF for Software Generation M. J. van der Westhuizen, R. G. Harley and D. C. Levy University of Natal, South Africa 5B: 11:00 - 12:30 PM SESSION 5B: TESTING A Protocol Extraction Strategy for Control Point Insertion in Design for Test of Transition Signaling Circuits H. F. Li and P. N. Lam Concordia University, Montreal Statistical Estimation of Delay Fault Detectabilities and Fault Grading Z. Zhang, B. McLeod and G. E. Bridges University of Manitoba, Winnipeg Test Application Time Reduction for Scan Based Sequential Circuits H. Zheng, K. K. Saluja and R. Jain University of Wisconsin-Madison Pseudo-Random Behavioral ATPG A-L. Courbis and J. F. Santucci Parc Scientifique G. Besse, France ----------------------------------------------------------------- 12:30 - 2:00 PM Luncheon ----------------------------------------------------------------- 6A: 2:00 - 3:30 PM SESSION 6A: PHYSICAL DESIGN III Fast Algorithm for Performance-oriented Steiner Routing M. Borah, R. M. Owens and M. J. Irwin Pennsylvania State University On Locally Optimal Breaking of Nondisjoint Cyclic Vertical Constraints in VLSI Channel Routing A. D. Johnson University of Toledo OPRON: A New Approach to Planar OTC Routing S. Danda, S. Madhwapathy, N. A. Sherwani and A. Sureka Western Michigan University Parallel Hierarchical Global Routing for General Cell Layout S. Khanna, S. Gao and K. Thulasiraman Concordia University, Montreal 6B: 2:00 - 3:30 PM SESSION 6B: ASYNCHRONOUS CIRCUITS Improving Self-Timed Pipeline Ring Performance Through the Addition of Buffer Loops H. Zhao, N. M. Sabine and E. Sha University of Notre Dame Scan Testing of Asynchronous Sequential Circuits O. A. Petlin and S. B. Furber University of Manchester, UK A New Look at the Conditions for the Synthesis of Speed-Independent Circuits E. Pastor, J. Cortadella and O. Roig Universitat Politecnica de Catalunya, Spain ----------------------------------------------------------------- 3:30 - 4:00 PM Coffee Break ----------------------------------------------------------------- 7A: 4:00 - 5:30 PM SESSION 7A: VLSI EDUCATION Design and Manufacture of a 2K Transistor Pwell CMOS Gate Array in a student run factory at RIT C. Kraaijenvanger, H. Enschede and L. Fuller Rochester Institute of Technology Using a Reconfigurable Field Programmable Gate Array to Demonstrate Boundary Scan with Built in Self Test H. J. Pottinger and C.-Y. Lin University of Missouri - Rolla Linking Fabrication and Parametric Testing to VLSI Design Courses R. Pearson Rochester Institute of Technology A Personal Computer Based VLSI Design Curriculum W. B. Leigh Alfred University 7B: 4:00 - 5:30 PM SESSION 7B: ARCHITECTURE AND DESIGN II A Scalable Shared Buffer ATM Switch Architecture A. Agrawal, A. Raju, S. Varadarajan and M. Bayoumi University of Southwestern Louisiana ATM Burst Traffic Generator P. P. Chu Cleveland State University A Universal Formalization of the Effects of Threshold Voltages for Discrete Switch-Level Circuit Models W. Korver University of Surrey, UK MARCH 18, 1995 SATURDAY ----------------------- T1: 9:00 AM - 11:00 AM TUTORIAL 1: Issues in Multichip Modules P.R. MUKUND, Rochester Institute of Technology This tutorial will address the following issues pertaining to MCMs. Electronic packaging constraints, Multichip module technology, Choices and trade-offs, Electrical/system considerations, Thermal considerations and Testing considerations. T2: 9:00 AM - 11:00 AM TUTORIAL 2: Issues in IDDQ Testing J. H. PATEL, University of Illinois at Urbana-Champaigne & S. Chakravarty, SUNY at Buffalo This tutorial will address the basic concept of Iddq testing, test generation, fault simulation and diagnosis. ----------------------------------------------------------------- 11:00 - 11:30 PM Break ----------------------------------------------------------------- 11:30 AM - 12:30 PM Facilities Tour Registration Table Open: 8:00 AM - 5:00 PM Thursday 8:00 AM - 5:00 PM Friday IEEE 1995 FIFTH GREAT LAKES SYMPOSIUM ON VLSI LOCATION The IEEE Fifth Great Lakes Symposium will be held at the Buffalo Marriott, which is adjacent to the State University of New York, North Campus. This site is central to many other conveniences such as restaurants, malls and hotels. Buffalo is the hub of a metropolitan area with a population exceeding one million individuals. A wealth of scenic, recreational, and cultural opportunities enrich University life. Closer attractions include Niagara-on-the Lake's Shaw Festival and New York State's Artpark, noted for its spring and summer festivals of art, theatre, and music. People from all over the world visit Niagara Falls (20 minutes from the University). Sunsets over Lake Erie are truly breathtaking. An inch of snow may create havoc in some places, but Buffalo moves during almost anything. Skiers always make it to the many slopes south of the city; cross-country skiers, sleders and skaters make enthusiastic use of the many parks. Spring and summer are usually cool and clear; autumn, crisp and colorful. Average temperature for the month of March is around 45 degree F and it snows in March only occasionally. Warm jackets are advised. Buffalo offers the advantages of a major city but is situated in one of the most scenic areas in New York State -- the Niagara Gorge, the Boston Hills, Letchworth State Park, and the Finger Lakes are all nearby. Buffalo airport is directly accessible from other major cities such as New York City, Washington DC, Detroit, Pittsburgh and Chicago. HOTEL INFORMATION A block of rooms has been reserved at the Buffalo Marriott, Hampton Inn, and Red Roof Inn for your convenience. All three hotels are located directly across from the University at Buffalo, North Campus. For Hotel Reservations Contact: BUFFALO MARRIOTT HAMPTON INN RED ROOF INN, Amherst 1340 Millersport Highway 10 Flint Road I-290 & Millersport Hwy Amherst, New York 14221 Amherst, New York 14226 Williamsville, NY 14221 Single: $74.00 + 13% tax Single: $59.00 + 13% tax Single: $42.99 + 13% tax Double: $74.00 + 13% tax Double: $66.00 + 13% tax Double: $49.99 + 13% tax Phone: (716) 689-6900 Phone: (716) 689-4414 Phone: (716) 689-7474 or 1-800-228-9290 or 1-800-426-7866 or 1-800-843-7663 Fax: (716) 689-0483 Fax: (716) 689-4382 Fax: (716) 689-2051 Please state that you are attending the University at Buffalo/GLSVLSI Conference when making your reservation. Other hotels adjacent to the University at Buffalo North Campus are: SUPER 8 Marriott's Residence Inn University Inn 1 Flint Road 100 Maple Road 2401 North Forest Amherst, New York 14226 Williamsville, NY 14221 P.O. Box 823 (716) 688-0811 (716) 632-6622 Amherst, NY 14226 or 1-800-800-8000 or 1-800-331-3131 (716) 636-7500 Fax: (716) 688-2365 Fax: (716) 632-5247 Fax: (716) 636-8296 REGISTRATION FORM Fifth Great Lakes Symposium on VLSI, March 16-18, 1995, Buffalo, New York, USA Name _______________________________________________________________________ First Middle Initial Last/Family Institution ________________________________________________________________ Address ____________________________________________________________________ City _____________________________ State ___________ ZIP ___________________ Country ________________________ Email Address _____________________________ Phone Number ___________________________ Fax _______________________________ Membership: IEEE ________________________ ACM __________________________ REGISTRATION FEES For Preregistration BEFORE February 22, 1995 IEEE/ACM Member Non-Member Student Conference $225 $280 $85 Tutorial 1 $85 $105 $85 Tutorial 2 $85 $105 $85 If Registration is received AFTER Febrary 22, 1995 or on-site: IEEE/ACM Non-Member Student Conference $270 $335 $100 Tutorial 1 $100 $125 $85 Tutorial 2 $100 $125 $85 Registration Total $ _____________________ Please circle appropriate fees. NOTES: 1) Registration by AT LEAST ONE PERSON AT REGULAR RATE is required for a paper to be presented and included in the proceedings. If a paper is to be presented by a student and no other coauthor is registered at the regular rate, the student rate does not apply. 2) Registration includes a copy of the proceedings, two continental breakfasts, two luncheons and a conference banquet on Thursday evening. Student registration includes all the above except the banquet. 3) We reserve the right to cancel tutorials due to insufficient participation. Form of Payment: ____________ Check Enclosed (in U.S. dollars only) ____________ Credit Card (Visa, Mastercard ) Make check payable to: UB Foundation/GLSVLSI Credit Card Number _____________________________ Expiration Date ______________ Name that appears on Card ______________________ Signature ______________________________________ MAIL REGISTRATION AND PAYMENT TO: University at Buffalo Office of Conference Operations 314 Crofts Hall Buffalo, New York 14260 Fax: (716) 645-3869 Phone: (716) 645-2018 Cancellation Refunds Honored Until March 3, 1995 (Less 10% Service Charge) TRANSPORTATION 1) From/To Buffalo Airport: Complimenrary shuttle service from the Buffalo Airport to the Marriott is available for those staying at the Marriott. There are courtesy phones located in the airport to call the Marriott for pick-up. They will also shuttle clients to the airport. 2) From/To Toronto Airport: International visitors can use Toronto, Canada route to reach Buffalo. From Toronto, the best way to reach Buffalo is through ground transportation such as rental cars or shuttle service. Contact Niagara Airbus shuttle at (905) 374-8111 (24 hours). The round-trip cost is $66. Three to four days advance booking is needed. The Niagara Airbus phone number at Buffalo is (716) 835-8111. The toll-free Fax line from North America is 1-800-206-7222 and from Europe is 0-800-89-7629. International visitors traveling via Toronto must have valid visa to enter Canada. FURTHER INFORMATION For additional information on the technical contents of the Conference, please contact any one of the following: Dr. Sreejit Chakravarty, General Chair (716) 645-3180 Dr. Ramalingam Sridhar, Program Co-Chair (716) 645-2422 Dr. Shambhu Upadhyaya, Program Co-Chair (716) 645-2422 Dr. Victor Demjanenko, Local Arrangements Chair (716) 645-2422 Fax: (716) 645-3656 Email: glsvlsi@eng.buffalo.edu For registration and accommodation related enquiries, please contact the Office of Conference Operations or phone (716) 645-2018.Article: 716
SAVE TAX DOLLARS!!! =================== Did you know that your PC, home office, and equipment may be tax deductible under current tax laws? This is true even if you use your PC in very limited business applications! Federal tax laws are often confusing and most tax preparation services do not understand the significant savings that your investment in office and computing equipment may qualify you for. Based on a legal brief by a prominent Los Angeles tax attorney for his celebrity clients, this is an easy-to-read, step-by-step guide to home office tax deductions. This write-up is written in plain English and covers all current tax law. It is an invaluable aid to legal tax deductions. Others have paid hundreds and even thousands of dollars for this expensive service. Now you can have this valuable money-saving information for only $7.95! With April 15 coming soon, you have no time to lose. If you pay U.S. taxes and use a portion of your home for business purposes, you need this write-up! E-mail for further information.Article: 717
In article 4vv@barnacle.iol.ie, coffey@iol.ie (Aedan Coffey) writes: > M. Movahedin (movahed@tumlis.lis.e-technik.tu-muenchen.de) wrote: > : Hello, > > : I have synthecized some combinational designs with 5 input and only 1 output > : with FPGA Compiler from Synopsys, but they are made with more than one CLB in > : Xilinx. Definitly, for a 5 to 1 combinational logic, only one CLB (XC4000) is > : enough. > : What is happening? > i > This might be a stupid question, but are you sure it's reporting CLB's > and not function generators? I'd expect to see three function generators for > a function of five inputs. YES, I am sure. definitly, for a 5 in/1 out combinational logic we need only 2 F or G function generators and one H function generator , which can be implemented in only one CLB. for more information, you can see XC4000 data books at page 2-9. Movahedin > > If you are in douby have a look at the XNF file, the format is pretty > straightforward. > > Regards, > > Aedan Coffey. > > -- > ================================================================= > Aedan Coffey, Toucan Technology, Technology Centre, Mervue, Galway, > Ireland. Phone +353-91-757223/770007 > Email: coffey@toucan.ie > Designers of Electronic Systems and ASICs.Article: 718
I've been looking for technical info about fpgas (electrical characteristics, timing, pwr diss., packaging, etc.).Does anybody can suggest me where to find the right databooks? Or at least a way to get info from manufacturers? I'd appreciate any reply. Manuel JimenezArticle: 719
In <3htibu$4vv@barnacle.iol.ie> coffey@iol.ie (Aedan Coffey) writes: >M. Movahedin (movahed@tumlis.lis.e-technik.tu-muenchen.de) wrote: >: Hello, >: I have synthecized some combinational designs with 5 input and only 1 output >: with FPGA Compiler from Synopsys, but they are made with more than one CLB in >: Xilinx. Definitly, for a 5 to 1 combinational logic, only one CLB (XC4000) is >: enough. >: What is happening? >i If your using Xilinx PPR tool and the FPGA is no full, the router may partition the design into multiple CLBs although it could fit into one. Check the report file for occupied CLBs and packed CLBs. Occupied CLBs are those containing some logic. Packed CLBs are the number of CLBs that the design could be sqeezed into. David LanzaArticle: 720
!!! "It's not a BUG, jcooley@world.std.com /o o\ / it's a FEATURE!" (508) 429-4357 ( > ) \ - / _] [_ An Independant Critique of Two New Synopsys Tools by John Cooley, the ESNUG guy Holliston Poor Farm, P.O. Box 6222, Holliston, MA 01746-6222 [ Tuesday, Feb. 7th, Synopsys held a big press event where it announced that they were entering into the Verilog/VHDL source level tool market. For what it's worth, I founded and run from my sheep farm the E-mail Synopsys Users Group (ESNUG), a completely independant grassroots user's group with 3200 members. (So independent, in fact, that Synopsys's lawyers tried to shut me down when I started ESNUG some 3 1/2 years ago -- but that's another story.) I make my living as an autonomous ASIC & FPGA design consultant helping EDA users with their tools. - John Cooley, the ESNUG guy ] A quickie summary of what Synopsys announced at the press event is that they're trying to push designers out of the gate tweaking stage and into higher level approaches. That is, they're interested moving designers into doing things like consider the effects of test, power, and physical design characteristics at the source level *prior* to synthesis. The two major products that are the foundation of this new approach are "DesignSource" and "HDL Advisor." Synopsys's DesignSource ----------------------- Essentially, DesignSource is a block diagram schematic editor with the design database management idea thrown in of having the ability to associate HDL source code with synthesis scripts, Verilog/VHDL simulation test beds and their resulting outputs in one "database". It has bells & whistles thrown in like a hierarchy browser, a rather simple Verilog/VHDL template generator, a block-level connectivity checker, a VHDL-specific "attribute" speadsheet editor and the ability to fire up simulation & synthesis scripts as a front end to VSS & Design Compiler. The Block Editor automatically propagates VHDL information like names, types & attributes to Synopsys's VSS simulator. In reviewing DesignSource I couldn't help but feel like I was some sort of art critic at a cocktail party who was given the challenge to identify where every part of a certain new montage came from. That is, with DesignSource, I wasn't looking at anything new but something more like a potpourri of ideas and parts of other tools that had been kicking around for some time in the EDA community. It just sang out Cadence Composer!, Mentor Design Architect!, ViewLogic ViewDraw!, Summit Design!, Escalade!, parts of Intergraph's tool offerings! and even the old Racal-Redac's Visula schematic editor! The Synopsys Verilog/VHDL template generator seemed rather primitive at best. Based on what was entered in the graphical Block Editor, for VHDL it threw down a simple piece of empty boiler plate ENTITY/ARCHITECTURE; for Verilog it spits out an empty module and its associated input/output statements -- the user is then left on his own with the "vi" editor to fill in these empty pieces of boiler plate with his sophisticated codings of state machines, data paths, controllers, etc. (My question is: "If our user can't create this initial no-brainer type of boiler plate in ten minutes on his own, how can he be trusted with coding the rest of the design?") For designers interested in something a little more involved in writing VHDL, I'd suggest you take a look at Summit Design's "Insight" -- it's best described as interacting with a living, breathing VHDL Language Reference Manual -- a hog wild VHDL syntax thingy. For Verilog, Cadence reports having a similar language sensitive editor in its Verilog-XL Design Environment with an additional waveform viewer and Verilog debugger thrown in. For the super-cheap user you can even look into the free GNU-EMACS editor with Verilog/VHDL specific settings. The carefully-manage-all-your-ASIC-design's-associated-files concept is as old as UNIX software is. For many a year, ASIC designers have been using *free* UNIX utilities like SCCS and RCS to keep track of various revisions of their source Verilog/VHDL files, their synthesis scripts and the changes in their Verilog/VHDL testbenches for their designs. (SCCS and RCS design database management is characterized by designers checking in and checking out various associated design files. Synopsys's DesignSource doesn't have check in/check out; it just associates various design files together.) I freely admit that block editors in themselves, regardless of who's selling them, just don't impress me. I personally just don't see beaucoup value-add here for what little they do and how much they cost. A quick comparison of pricing on UNIX tools that are somehow similar: Tool Price Comments ------------ ------- --------------------------------------- GNU EMACS, "vi" & No user hand holding. Clever people SCCS or RCS *free* save lots $$$ here; idiots lose $$$ here. Mentor Graphics Only works with Mentor specific VHDL Design Architect $9,900 and no Verilog (yet) Cadence Verilog-XL Waveform viewer, results analyzer, Design Environment $7,500 language sensitive editor & debugger Cadence Leapfrog VHDL Same as above. "Free" when you purchase Design Environment "free" their Leapfrog VHDL at $20,000. Cadence Composer $14,500 Overpriced schematic block editing w/dbase Concept $12,500 mngmnt. -- very similar to DesignSource Synopsys Overpriced (like Cadence Composer & Concept) DesignSource $17,000 for what it does. Should be $4K to $7K. For those EDA users interested in shelling out $12,500 to $17,000 per engineer at your site for cheesy Verilog/VHDL templates and revision control software, please immediately phone me, John Cooley, at (508) 429-4357 and I'll easily beat their price by 50%! (I really *am* an independent ASIC & FPGA design consultant, you know!) Mind you, it is a REAL and BIG problem for designers if they're not up to working with file management & revision control -- then I'd STRONGLY recommend getting something like Synopsys's DesignSource as a fairly light-weight full-source control environment because you'll lose significant time & energy & money & sleep without it... but I'd still haggle like crazy with the salesman to not pay $17,000 for it! :^) (I know I'll catch hell for this but it's true!) Synopsys's HDL Advisor ---------------------- Inasmuch as I reserve the right to publicaly yawn at DesignSource because it seemed to be an overpriced Cadence Composer copycat product for Synopsys -- I also reserve the right to publicaly applaud Synopsys for creating a unique, new, interesting & useful type of tool with its HDL Advisor. It's HOT! One of the early questions that I had when first taking an in-depth look at HDL Advisor was: "Is this just warmed over, productized Source-To-Gates?" (Source-To-Gates was a feature that offered Synopsys customers a very crude correlation between a group of gates from synthesis and where in the source Verilog/VHDL it sort of thought it came from. Source-To-Gates didn't work very well because synthesis would add & remove gates all the time to fit some timing or area constraint.) HDL Advisor turned out to much more of a true analysis tool focused on the *before* synthesis problem of coding one's Verilog/VHDL design. Here's what I saw in Synopsys's HDL Advisor: * - HDL Browser: - the "home" window that lets you examine your source Verilog/VHDL and how various specific lines of source code relates to graphics/reports you generate in other windows. The Hot Cursor & Selection Inspector (described later) play a significant role here, too. At the pre-synthesis point you can keep track of timing via the number of logic levels your source Verilog/VHDL would create, a rough area estimate under the component count heading and your source Verilog/VHDL related connectivity issues like fan-in and fan-out. Note: it's *before* synthesis. At the post-synthesis point you get to keep track of area, power, timing violations in relation to the constraints you gave the design, and connectivity issues like fan-in, fan-out & capacitance. * - Histogram & Profiler: - two graphics packages that display the pre- and post- synthesis data decribed above in an intelligent & useful manner for the ASIC designer. Let's say I'm interested in timing issues for my pre-synthesis Verilog source code. (Since there's no real timing info at the pre-synthesis phase, the next best thing that correlates to timing is the number of logic levels my piece of Verilog would have if mapped to gates.) I'd fire up the Histogram to show me the distribution of pre-synthesis logic levels in my design. Click on a specific Histogram bin and, voila!, the list of specific design data points in that bin are listed with the HDL Browser showing the associated source Verilog code highlighted. (Profiler works in a similar manner but as a single verticle bar broken into segments showing percentages. It's useful, for example, for tracking how much area various parts of your source Verilog/VHDL results in.) * - Path Browser & Logic Inspector: Once I've identified an interesting issue with my Verilog source (like losts of high fan-out points discovered from looking at the Histogram for fan-out) I can chase these problem children through my Verilog *source* by using the Path Browser which explores pure connectivity. If I was curious down to the gate level, I fire up Logic Inspector to see a Boolean gate level representation of what I'm interested in. (All the time, every point I touch in these explorations is cross referenced and highlighted in the *source* in HDL Browser.) Also, If I'm interested in chasing something like post-synthesis capacitance, I can use Path Browser & Logic Inspector in a similar manner. * - Hot Cursor & Selection Inspector: These two properties run throughout all of the windows in HDL Advisor. Essentially, a hot cursor tells the user quickie information without having to "click" on anything. Just wander the cursor over something and it'll tell you a useful snippet like a module name in a particular part displayed in Profiler or the Boolean expression at a specific point seen in Logic Inspector. Selection Inspector is a follow-up window that spews out as much info as is known about a point if it's "clicked" by the cursor -- stuff like: name, fan-in, fan-out, logic levels, area, hierarchy path, timing slack, power -- whatever it knows at the time! With its focus on getting feedback to the designer while he's writing his Verilog/VHDL source code, HDL Advisor helps circumvent the dreaded Synthesis Iteration Hell. That is, there are a lot of ASIC design teams that first write their Verilog/VHDL, spend months verifying its functionality with reams of regressions and then take this golden source code to synthesis. If the golden code runs into problems designers will literally spend weeks/months chasing down all sorts of settings and switches for Design Compiler to avoid having to change their errant source Verilog/VHDL. HDL Advisor promotes better coding styles for synthesis *while* the source code is being written. A few of the old time Synopsys customer may say: "Sounds like a good tool for beginners. I don't need it because I already know what does & doesn't get you in trouble for Synopsys." -- but they're missing a crucial point. It's one thing to know that "nets with high fan-outs are bad news"; it's another to know exactly which nets fit in that category in *your* specific design before synthesizing. That is, HDL Advisor lets you ask meaningful questions on large & complex hunks of Verilog/VHDL source code like: - where are the longest paths? what modules do they go through? - what are the fan-ins and fan-outs at this point? where do they lead? - how does the timing slack or capacitance relate back to the source? - what's the boolean equation here in the design? Instead of burning up brain cells trying to figure out how your code (or worst yet, some source code you "inherited") will be digested by the Synthesis Monster, this tool lets you quickly know where the hot spots are and where hot spots aren't in your code. I *strongly* recommend checking out Synopsys's HDL Advisor because it lets you check out your design even before simulate (much less synthesize) it! Summary ------- DesignSource is a Synopsys version of Cadence's Composer -- neither product I'd personally buy with their hefty asking prices ($17K & $14K) for what little these block-schematic-editors-with-revision-control-thrown-in tools do. You can get free rev control using UNIX's SCCS/RCS and the wimpy coding templates created by the block schematic editors can be made by hand in ten minutes with "vi". Mind you, if you're not disciplined enough to handle file management & revision control on your own, IMMEDIATELY BUY DesignSource because you'll lose significant time & energy & money & sleep without it. On the other hand, Synopsys HDL Advisor is definitely an interesting new type of tool altogether. I'd recommend giving it a look-see if you're a newbie designer or an old pro. HDL Advisor lets you quickly know where the hot spots are and where hot spots aren't in your Verilog/VHDL source code -- even before you simulate (much less synthesize) the design! Check it out! Overall, I think most Synopsys customers enjoyed the big press announcement get together because it told them through acts (not lip service) that Synopsys really is serious about getting open customer input in front of the trade press, industry analysts and financial people. Although there weren't hundreds of customers there cheering & shouting "Encore! Bravissimo!" when the event was done, I do think they *liked* the idea of being invited to the *big* press announcements. - John Cooley part-time EDA Consumer Advocate full-time contract ASIC & FPGA designer P.S. I'd like to hear from readers what they think about getting in-depth critical reviews like this. They're a lot of work. Are they worth it? =========================================================================== Trapped trying to figure out a Synopsys bug? Want to hear how 3196 other users dealt with it ? Then join the E-Mail Synopsys Users Group (ESNUG)! !!! "It's not a BUG, jcooley@world.std.com /o o\ / it's a FEATURE!" (508) 429-4357 ( > ) \ - / - John Cooley, EDA & ASIC Design Consultant in Synopsys, _] [_ Verilog, VHDL and numerous Design Methodologies. Holliston Poor Farm, P.O. Box 6222, Holliston, MA 01746-6222 Legal Disclaimer: "As always, anything said here is only opinion."Article: 721
In article <3hfv31$8p@seralph9.essex.ac.uk> ierof@csc2.essex.ac.uk (Ieromnimon F) writes: >From: ierof@csc2.essex.ac.uk (Ieromnimon F) >Subject: Re: VERILOG >Date: 10 Feb 1995 15:00:48 GMT >In article <3h5s7sINNc3r@salmon.wv.tek.com> weedk@salmon.wv.tek.com (Kirk A Weedman) writes: >>Anyone know where I can get a Verilog description >>for SCSI design? >> >> Kirk weedk@pogo.wv.tek.com >Mee too! Any primer on Verilog models of SCSI protocols/components/the lot would >be received with gratitude! >Frank Ieromnimon, >PACE Project, >dept. of Computer Science, >University of Essex. Try talking to the guys at NCR in colorado springs ,CO, they have the 53C80 etc implementable as a standard cell, so my guess would be they have some version of verilog / VHDL perhaps available for it Wassail, KenArticle: 722
cking@accutron.ie writes :- >Hi, > I'm currently working on the design of a video RAMDAC >evaluation/demonstration board, and would like some comments >and/or help on an idea I have. The board is like this: etc. etc... > 1. The FPGA has 5000 internal gates. > 2. It can run at 250Mhz internaly ^^^^^^ 250MHz ! Are you sure about this, sounds like a toggle frequency to me, in which case would I not expect your design to run at this speed. Please correct me if I am wrong and let me know who is making such a device. Cheers, T.H. (trev@wg.icl.co.uk)Article: 723
Hello, I would like to know something diffrent among them? I was always cunfused by them all. In my previous impression, they are: PAL: programmable AND, fixed OR PLD: programmable AND, programmable OR PLA: ???????????? AND, ???????????? OR GAL=PLD ?? Please correct the above, Thanks in advance! JasonArticle: 724
In article <3i02ch$d85@msunews.cl.msu.edu> jimenez2@oscar.egr.msu.edu (Manuel Alejandro Jimenez-Cede) writes: >I've been looking for technical info about >fpgas (electrical characteristics, timing, pwr diss., packaging, etc.).Does anybody can suggest me where to find the right databooks? >Or at least a way to get info from manufacturers? > Yeah, CALL them. Go to your University library and look in any of the technical magazines: EDN, Electronic Design, Computer Design, EE Times, ASIC & EDA, etc. You'll find dozens of addresses and phone numbers for Xilinx, Altera, Atmel, AT&T, Cypress, Quicklogic, etc. Hope this helps, Mircea >I'd appreciate any reply. > >Manuel Jimenez > -- Mircea R. Stan | "Without immortality the whole world would UMass, ECE Dept. | be nonsense, all of creation an absurdity." Amherst, MA 01003 | Karl F. Gauss
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