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
>Hi Andreas, > >while I can't help you with the group buy, I was curious about >your experiences with the Moto FPGA. Are you using the free >software ? Which design capture method are you using (Viewlogi/ >OrCad/VHDL/Verilog)? > >Just wondering, I have tooled with the APR tool a bit and >got started on a netlist generator for one of the tools I'm >use sometime "diglog". Since diglog is free and the APR is >free, this would be a good way to get into FPGAs for people. >However, I think it's hard to get the parts. In the US, >Universities and get free parts from Moto, but this doesn't >help everyone. > >Anyway, do you think there is much interest in a MPA netlister >and libraries for free CAD tools like diglog ? I use the free software (current version is 2.3) I tried some register intensive designs with not much logic (max 5 levels deep) and got a 95% utilisation @ 40MHz with a MPA1016 and a 75% utilisation @ 40Mhz (larger but similar design) with a MPA1036 (autolayout estimations). Place&ROute prcessing time are up to 2 hours on a 90MHz Pentium/32MB. The chips are available (at least I got a quotation) from Future Electronics. Motorola donated a free sample (MPA1036HI). I use a standard EDA schematic editor as front-end tool. To write a symbol library is quite easy, just port the MPA library to your schemtaic tool. The EDIF output of my Ulticap editor needs some postprocessing but then it works OK. A free schemtic tools would be nice, but if you already have a schematic entry you can use that. Comments? Andreas -- Andreas Kugel, Karolinenstr. 4 76135 Karlsruhe, Germany Phone: (49) 721 377865, Fax (49) 721 937 49 12 E-mail: akugel@t-online.deArticle: 4976
If you are using or going to use 68HC16 microcontroller, you may be interested to look at HC16BGND, an advanced programmer’s interface that helps you to develop, test, and refine your assembly language programs for MOTOROLA’s 68HC16 microcontrollers. The key features include: - Run under MS Windows - Interface through PC’s parallel port - Motorola suggested 10 pin target connector - Download and debug HC16 assembly language program in S19 or HEX format - trace through your code by step through or step over - set up to 100 breakpoints - On-screen editing of register and data memory - On-fly assembly of instruction using mnemonics in program memory - Open infinite Register, Program, and Data windows - Watch window so that you can watch important variables - Exchange data through Windows clipboard - File I/O which can be used to automatically read a hex input file into the file and write the result to an output file. - Easy to use. Just click on the menu with left mouse or click on the windows’ area. Everything is self-explained - Convenient and easy to install due to its parallel port interface. You can use laptop to do an on-site demonstration of you products - Low cost - 30 day money back guarantee and one year product warranty For more information or need a demo program, see http://users.why.net/wctech/hc16bgnd.htm, or FTP to FTP.WHY.NET/FTP/PUB/USERS/WCTECH, or send email to WCTECH@WHY.NET Larry Chen WC TechnologyArticle: 4977
Kardos, Botond wrote: > > Hi! > > Have anyone ever realized a quartz controlled oscillator with a PLD > or an FPGA without other IC's ? I know that the simplest oscillator with > a 74hc04 won't work (the quartz is connected parralel to the inverter), > but it work's with 74hcu04 because this circuit is unbuffered > (single-stage inverter). > Thx for any ideas ! > > -- > Kardos, Botond - at Innomed Medical Co. Ltd. in Hungary > eMail: kardos@mail.matav.hu > phone/fax: (36 1) 268-0934 Hello, I did such a thing with a xilinx-fpga (xc3000 series) without any problem; these parts have special circuity on-chip for that purpose. -- Cord EliasArticle: 4978
I am replying to Eric Edwards reply, so maybe I am missing some parts of the original request.. But my suggestion would be to try to change the state machine encoding. ASYL+ is using the "compact" encoding by default. "One-hot" encoding would use more serial cells, and probably use less logic cells. Related to this is, when you use some kind of encoding within your original design, the synthesizer may not recognize this, and may produce not optimal code. I had best results when using either enumerated or integer type for state varibles (signals) and then choosing the encoding style in the synthesis process.. Thomas -- ------------------------------------------------------------------- | Thomas Hadlich hadlich@infaut.et.uni-magdeburg.de | | http://infaut.et.uni-magdeburg.de/~hadlich | -------------------------------------------------------------------Article: 4979
Rich K. wrote: > > James W. Swonger <jws@billy.mlb.semi.harris.com> wrote in article > <5attn9$mkg@hearye.mlb.semi.harris.com>... > > If you're really interested in SEU-immune, high-rel FPGAs you should go > > to the companies which are building Class S, rad hard versions of the > > ACTEL form (Lockheed Martin, partnered with Actel) and Xilinx form > > (Honeywell). These will be able to provide you with bit error rate, > > total dose and other reliability and failure/fault analysis information > > (presuming they care to, and that's not a given; they'll probably make > > you pass through Marketing first to see if you're a customer or just a > > babbler). > > > > Also, is Honeywell building the Xilinx compatible part? Is there contact > information on this? I know Harris had a data sheet on some of these > Xililinx devices a few years ago but I believe they dropped the part. > > rk A few weeks ago I tried to find a rad hard fpga device. ACTEL can provide you with rad hard devices. Due to a previous Xilinx design I tried to find a rad hard Xilinx Device and I went through the marketing departments... As far as I could find out, there is no rad hard Xilinx Device available. The application of this was a satellite. Fritz * Friedrich Beckmann * * University of Bremen TEL: +49 421 218 4079 * * FB1 - Institute for Microelectronics FAX: +49 421 218 4434 *Article: 4980
Marcello Lajolo wrote: > > Do you have assigned device correctly? > It is necessary to specify in section "Device Options" > of the menu "Assign" that you want a configuration scheme > of type "Active Parallel Up/Down". > Thanx for the hint, you gave me the idea that I should check if the configuration EPROM contains valid data. I my design everything was OK, but for some other reasons the offset generation was wrong. So everything's fine, thanx again ! Botond -- Kardos, Botond - at Innomed Medical Co. Ltd. in Hungary eMail: kardos@mail.matav.hu phone/fax: (36 1) 268-0934Article: 4981
Hi! Have anyone ever realized a quartz controlled oscillator with a PLD or an FPGA without other IC's ? I know that the simplest oscillator with a 74hc04 won't work (the quartz is connected parralel to the inverter), but it work's with 74hcu04 because this circuit is unbuffered (single-stage inverter). Thx for any ideas ! -- Kardos, Botond - at Innomed Medical Co. Ltd. in Hungary eMail: kardos@mail.matav.hu phone/fax: (36 1) 268-0934Article: 4982
Hi folks! I'm new to this newsgroup and was wondering if some sort of a FAQ exists that I can look up ? Thanks a million! sudhirArticle: 4983
Hello Altera people, I'm trying to compile an hirarchical design into 4 deferent devices with no big success. Our design combined from 4 blocks and we tried to put each block into a deferent EPM7096LC68-12. What we did: We had 4 blocks on the top scheme, and with the clique function we attached to each block a deferent device. At the end of the compilation, the compiler shouts that it didn't use one of the devices but the project fit by putting two blocks in one device, and that aint what we wanted. Please if you can give us more info about partitioning a top design, that can be a great help. Thanx, Gabby. Digital Jerusalem. Mailto:gabby@isv.dec.comArticle: 4984
Gabby Shpirer wrote: > > Hello Altera people, > > I'm trying to compile an hirarchical design into 4 deferent devices > > with no big success. > > Our design combined from 4 blocks and we tried to put each block into a > > deferent EPM7096LC68-12. > > What we did: > > We had 4 blocks on the top scheme, and with the clique function we > > attached to each block a deferent device. > > At the end of the compilation, the compiler shouts that it didn't use > > one of the devices but the project fit by putting two blocks in one > > device, and that aint what we wanted. > > Please if you can give us more info about partitioning a top design, > > that can be a great help. > > Thanx, > > Gabby. > > Digital Jerusalem. > > Mailto:gabby@isv.dec.com The Altera Clique feature helps you acheive timing goals. If for some reason your initial compile did not meet these goals, you can list out the critical paths using the timing analyzer and then add these paths using the Clique feature to optimize timing. The Clique function tries to keep the specific nodes together (Within a LAB). Altera tools do not partition efficiently. (Actually none of FPGA tools that I know of partition very well) It is best if you partition the design yourself instead of letting the tools do the same. Depending on your design entry (VHDL, Verilog, GDF, TDF) you can describe in a modular fashion so that they are compiled into different devices. This way you can simulate as one block, but synthesize into multiple devices. Also if possible I would suggest you use one BIG device instead of using multiple devices even if it is more expensive approach. (I am assuming that you are not using those BIG devices already!!) Sundar GopalanArticle: 4985
Gabby Shpirer wrote: > > Hello Altera people, > > I'm trying to compile an hirarchical design into 4 deferent devices > > with no big success. > > Our design combined from 4 blocks and we tried to put each block into a > > deferent EPM7096LC68-12. > > What we did: > > We had 4 blocks on the top scheme, and with the clique function we > > attached to each block a deferent device. > > At the end of the compilation, the compiler shouts that it didn't use > > one of the devices but the project fit by putting two blocks in one > > device, and that aint what we wanted. > > Please if you can give us more info about partitioning a top design, > > that can be a great help. > > Thanx, > > Gabby. > > Digital Jerusalem. > > Mailto:gabby@isv.dec.com The Altera Clique feature helps you acheive timing goals. If for some reason your initial compile did not meet these goals, you can list out the critical paths using the timing analyzer and then add these paths using the Clique feature to optimize timing. The Clique function tries to keep the specific nodes together (Within a LAB). Altera tools do not partition efficiently. (Actually none of FPGA tools that I know of partition very well) It is best if you partition the design yourself instead of letting the tools do the same. Depending on your design entry (VHDL, Verilog, GDF, TDF) you can describe in a modular fashion so that they are compiled into different devices. This way you can simulate as one block, but synthesize into multiple devices. Also if possible I would suggest you use one BIG device instead of using multiple devices even if it is more expensive approach. (I am assuming that you are not using those BIG devices already!!) Sundar GopalanArticle: 4986
Xilinx XC3000 and XC31000 have an on-chip single-stage inverter, intended for being wrapped around a 1 MHz to 40 MHz crystal, and there are thousands of designs using this feature. More recent Xilinx architectures don't have this on-chip circuitry for the following reasons: You can buy a complete oscillator can for essentially the same price as a crystal ( around one dollar in single quantities ). IThe complete oscillator is only slightly larger than a crystal, and the dedicated oscillator consumes significantly less power ( Icc ) than the invertyer on the FPGA. A canned oscillator keeps the digital designer away from the high-frequency analog issues of crystal oscillator design, including the start-up conditions. Eliminating that inverter avoids exotic analog testing at Xilinx. It avoids fingerpointing between FPGA and crystal manufacturer when something does not work reliably. ( "You don't provide enough gain" vs. "you specified the load capacitance incorrectly". There have been no complaints from our customer base when we deleted the crystal oscillator option from the newer architectures, but the option remains and will remain on the XC3000 and XC3100 family devices, which will be available for many years to come. Peter Alfke, Xilinx ApplicationsArticle: 4987
,, , wrote: > > Hi folks! I'm new to this newsgroup and was wondering if > some sort of a FAQ exists that I can look up ? > > Thanks a million! > > sudhir AFAIK, No FAQ, but check out COMP.ARCH.FPGA Archive at http://www.super.org:8000/FPGA/thread_old.html for old posts. -- John L. Smith, Pr. Engr. | Sometimes we are inclined to class Univision Technologies, Inc. | those who are once-and-a-half witted 6 Fortune Dr. | with the half-witted, because we Billerica, MA 01821-3917 | appreciate only a third part of their wit. jsmith@univision.com | - Henry David ThoreauArticle: 4988
eswar@mega.megamed.com (Subramnyeswar Saladi) wrote: >John Cooley (jcooley@world.std.com) wrote: >: simonson@skopen.dseg.ti.com (Kevin M Simonson) writes: > >: > I'm feeling really ignorant right now. Is there anyone who can tell >: >me what ASIC stands for? Thanks. >: > >: > ---Kevin Simonson > >: Aren't "ASICs" a brand of tennis shoe? :^) > >Why were the shoes named ASIC's ? > We're on a tangent here, but I think it's Application-Specific Integrated Cross-training Shoe. At least that's what my pointy-haired manager told me. ;)Article: 4989
Our company is preparing to introduce ASIC synthesis-related EDA tools for the Linux (x86) platform, in the near future. To help our planning, please answer the 2 questions below: 1) Which of the following release(s) of Linux do you currently run (or plan on running in the near future) ? [ ] 1.1.?? [ ] 1.2.?? [ ] 2.0.?? [ ] other: _______________ (please specify) 2) If you use Sun machines; which OS release(s) do you currently run (or plan on running in the near future) ? [ ] SunOS 4.1.?? [ ] SunOS 5.4.?? (Solaris 2.4) [ ] SunOS 5.5 (Solaris 2.5) You can check your OS version by typing "uname -sr" at the Unix prompt. PLEASE EMAIL US DIRECTLY and we will post the results to these newsgroups, in a short while. Thank you. -- Synthesis Solutions, Inc. http://www.ez-synthesis.com vmail: (415) 431-6429Article: 4990
In article <E3oII7.K8A@nonexistent.com> kardos@mail.matav.hu writes: >Hi! > > Have anyone ever realized a quartz controlled oscillator with a PLD >or an FPGA without other IC's ? I know that the simplest oscillator with >a 74hc04 won't work (the quartz is connected parralel to the inverter), >but it work's with 74hcu04 because this circuit is unbuffered >(single-stage inverter). > Thx for any ideas ! Sure it works (with both). You may just need to put a resistor in parallel with the crystal (20,000,000 ohms), and maybe some capacitors to ground on both sides (10 to 20 pF). An easy way to get this right is to put the circuit together, and leave out the crystal. It should oscillate anyway. Adjust the resistor an capacitor to give an RC oscillator at about the frequency of the xtal. then put the xtal back into the circuit. As for FPGAs, all the Xilinx XC3000, XC3100, XC3000A, and XC3100A devices have specifically got an XTAL in and out pin to support an XTAL oscillator with no extra ICs. > >-- >Kardos, Botond - at Innomed Medical Co. Ltd. in Hungary >eMail: kardos@mail.matav.hu >phone/fax: (36 1) 268-0934 Philip Freidin. Self starting oscillators that dont, and amplifiers that do :-)Article: 4991
The first announcement went out just before Christmas. I'm not sure everyone was able to read it before it expired, hence the repost. Announcement of Trianus/Hades tools for Xilinx XC6200 FPGA ========================================================== The Institute for Computer Systems of ETH Zurich is proud to announce the first publicly available CAD tool suite for the new Xilinx XC6200 FPGA architecture. The Trianus/Hades package was developed using the Oberon System over the past three years by two graduate students, Stephan Gehring and Stefan Ludwig. It uses and runs under the Oberon System V4 from the Institute for Computer Systems, ETH Zurich, Switzerland and the Institute for System Software, University of Linz, Austria. See the text below for a detailed description of the capabilities of the CAD software. Go to http://www-cs.inf.ethz.ch/Wirth/CADTools.html#Trianus for more information and for downloading instructions or use anonymous ftp to ftp-cs.inf.ethz.ch and go to directory /pub/Trianus The development of this software was possible due to the open architecture of the XC6200. It was not developed by Xilinx, Inc. and is different from their Camelot software for the XC6200. We would like to thank Xilinx Scotland and Xilinx San Jose for their continuing support over the past years, without which the development of this software would not have been possible. Comments and criticism is welcomed. Happy New Year! Stefan Ludwig, Stephan Gehring ------------------------------------------------------------------------ Stefan H-M Ludwig Institute for Computer Systems mailto:ludwig@inf.ethz.ch ETH Zentrum http://www.inf.ethz.ch/personal/ludwig CH-8092 Zurich, Switzerland Phone: +41-1-632 7301 Fax: +41-1-632 1307 Trianus/Hades by Stephan Gehring/Stefan Ludwig The Trianus/Hades software system consists of a suite of tightly integrated tools for the efficient design and implementation of algorithms for a custom computing machine. The software is built upon a generic framework for FPGA circuit design and comprises a compiler for the Lola hardware description language, a layout editor, a circuit checker, a technology mapper, a placer, a router, and a bit-stream generator and loader for the Xilinx XC6200 architecture. We argue that a tight coupling of design tools provides a base for fast iterative and interactive circuit design, a feature which current systems provide only in a very limited form. The Hades synthesis back-end uses simple, deterministic algorithms, which give predictable results at high speeds. Typically, compile, place and route times are under 1 minute on a Pentium-class PC. Hence, it is possible to develop hardware with the same turnaround times as is common for software development. Hardware Description Language Lola by Niklaus Wirth Lola was designed as a simple, easily learned hardware description language for describing synchronous, digital circuits. In addition to its use in a digital design course for second year computer science students at ETH Zurich, the Institute for Computer Systems uses it as a hardware description language for describing hardware designs in general and coprocessor applications in particular. The purpose of Lola is to statically describe the structure and functionality of hardware components and of the connections between them. A Lola text is composed of declarations and statements. It describes the hardware on the gate level in the form of signal assignments. Signals are combined using operators and assigned to other signals. Signals and the respective assignments can be grouped together into types. An instance of a type is a hardware component. Types can be composed of instances of other types, thereby supporting a hierarchical design style and they can be generic (e.g. parameterizable with the word-width of a circuit).Article: 4992
In message <5b0igk$h2r@Bayou.UH.EDU> st7jp@Bayou.UH.EDU (,, ,) writes: > Hi folks! I'm new to this newsgroup and was wondering if > some sort of a FAQ exists that I can look up ? > Thanks a million! > sudhir If you want a grounding in FPGAs and Programmable Logic you could try my book:- FPGAs and Programmable LSI - A Designer's Handbook Published by Newnes (Butterworth Heinemann) - ISBN 0 7506 2883 9 It is not too high-powered but gives (I think!) a good introduction to the subject. Geoff BostockArticle: 4993
Peter Collins (peter@exemplar.com) wrote: : There will be a Birds of a Feather Session on Linux and EDA at : Usenix 97 in Anaheim on Jan 7 at 7 pm. We'll share : information on EDA software currently available on Linux, ideas : If you can't make it, but would like a summary of the : BOF and further Linux & EDA developments, you can also send email : to info@linuxeda.com Any chance you could post a summary here? -- Al --------------------------------------------------------- I plan to die like my grandfather, peacefully in his sleep. Not in screaming terror, like his passengers.Article: 4994
In comp.lang.verilog Peter Collins <peter@exemplar.com> wrote: : There will be a Birds of a Feather Session on Linux and EDA at : Usenix 97 in Anaheim on Jan 7 at 7 pm. We'll share : information on EDA software currently available on Linux, ideas : for promoting more support and the current state of LinuxEDA. : For developers there will be information on how to support Linux : from the practical side of third party tools like license servers, : memory managers and online documentation tools. Since about one year I try to convince Oliver Bartels (http://www.bartels.de) to risk a Linux port of his PCB Layout package BAE. Mr. Bartels argues that he had only about 5 requests for a Linux version at all, so he doubts real interesst and market chance. If you think other and have need for a Linux PCB package, please let him know (obartels@bartels.de). F.y.i: Many commercial PCB packages have Mr. Bartels Autorouter Kernel licenced. -- Uwe Bonnes bon@elektron.ikp.physik.th-darmstadt.de Institut fuer Kernphysik Schlossgartenstrasse 9 64289 Darmstadt --------- Tel. 06151 162516 -------- Fax. 06151 164321 ----------Article: 4995
RSA Data Security will soon sponsor a 'DES Challenge', in which participants are asked to attempt to decrypt a message supplied by RSA. The first person or group to email RSA the correct key and the decrypted message will receive a $10,000 prize. I believe that a reconfigurable logic system is a near-ideal candidate to attack this problem. The winner, aside from the money, will also gain a great deal of positive publicity. This challenge is acheivable with a plausible level of time and effort. This is a 'known plaintext challenge': The first few bytes of the encrypted message are known, but the key used for encryption, and the rest of the message, are not. If the key is found, the rest of the message can be easily decrypted. There are no restrictions on the mechanisms which can be used to recover the DES key. The fastest method is probably brute force; testing each of the 2^56 keys until a match is found. DES was originally designed for hardware implementation, and is thus easier to run on a reconfigurable logic system than many later encryption algorithms. In 1993, Michael J. Wiener of Bell-Northern Research, published a paper 'Efficient DES Keysearch', in which he gave a detailed description of such a keysearch engine, based on custom chips, which could search the entire keyspace in about 7 hours. The paper includes a gate-level description of the custom chip, which can test one key per clock cycle due to a clever pipelined design. The chip he describes is constructed using 0.8 um CMOS, and contains 26086 gates, spread over 76637 sites. It runs at 50 MHz, testing 50 million keys/sec. By contrast, the best known key search algorithm capable of running on a Pentium tests about 100,000 keys/sec at that speed. Applying some of the speedups and simplifications learned from the work done on software search algorithms, I beleive it should be simple to reduce the site count by about 3600, getting it down to about 73,000 sites. It should also be possible to simplify some of the wiring. In this mode, the chip would identify 'candidate keys', roughly one in 2^32, which would then be tested on a general purpose computer. I am not conversant with the conventions and constraints of reconfigurable hardware design. I give this detailed information in hope that it will enable readers to decide whether their systems are capable of emulating this chip. Other designs are possible. Anyone interested? RSA has issued a press release and given out some example data for testing systems, but still has nothing on their web site. The actual challenge will be issued during the RSA Data Security Conference later this month. Peter Trei (not affiliated with RSA) trei@process.com ptrei@acm.org I append RSA's press release. ------------------------------ RSA Announces New "DES Challenge" Tens of thousands of dollars in cash prizes offered; contest should improve overall Internet security by illustrating relative strength of different crypto algorithms and keysizes. Business Editors and Computer Writers REDWOOD CITY, Calif.-Jan 2, 1997--RSA Data Security, Inc., a wholly owned subsidiary of Security Dynamics Technology, Inc. (NASDAQ: SDTI), today announced an Internet-based contest with cash prizes. The contest, known as the "RSA DES Challenge", challenges mathematicians, hackers and computer experts around the world to decipher encrypted messages. The goal of the contest is to quantify the security offered by the government endorsed DES encryption standard and other secret-key ciphers at various key sizes. The challenge proper will be launched during the RSA Data Security Conference to be held in San Francisco, January 28-31, with the target ciphertexts for the different contests being simultaneously posted on the company web-site, at http://www.rsa.com/ RSA Data Security pioneered the Internet-based "cracking" contest, when it launched the original "RSA Factoring Challenge" back in 1991. Since then, the company has paid out over $100,000 in prize money to mathematicians and hackers around the world, and the data gained from that Challenge (which is ongoing) has greatly increased mathematicians' understanding of the strength of encryption techniques based on the "factoring problem", such as the RSA Public Key Cryptosystem T. Background It's widely agreed that 56-bit keys, such as those offered by the government's DES standard, offer marginal protection against the committed adversary. By inertia as much as anything else, DES is still used for many applications, and the 20-year-old algorithm is proposed to be exportable under the latest incarnation of Clipper. It is the perfect time to demonstrate to the world that better systems are both required - and available - thus improving the world's security. There have been theoretical studies done showing that a specialized computer "DES cracker" could be built for a modest sum, which could crack keys in mere hours by exhaustive search. However, no one is known to have built such a machine in the private sector - and nobody knows if one has been built in any government, either. The successes of the RSA Factoring Challenge show that for some types of problems, it's possible to recruit spare "cycles" on a large number of machines distributed around the Internet. Therefore, by offering a suitable incentive, it should be possible to recruit sufficient CPU power across the Internet to exhaustively search the DES keyspace in a matter of weeks. Computer scientists have already developed software that will allow even the novice computer user to participate in the cracking effort. By incorporating the key search software in a "screen saver", a simple PC anywhere on the Internet can devote its spare time to working on the problem - remotely and completely unattended. Even people with limited computer skills will be able to participate. In the RSA DES Challenge, the motto will definitely be "The More, The Merrier". The Contest Full details of the RSA DES Challenge will be posted on the RSA home page (http://www.rsa.com/) during the first weeks of January. Complete rules for the competition will be provided as well as example challenges and solutions against which computer scientists and hackers can test their software. In conjunction with the RSA DES Challenge, RSA will simultaneously launch a series of other contests based around the RC5 Symmetric Block Cipher (another encryption algorithm). Since RC5 is a variable key length block cipher, targets that offer increasing resistance against so-called "exhaustive search attacks" will be posted in the hope of assessing the full impact of a widely-distributed exhaustive search. There will be 12 challenges based on the use of RC5. Prizes will be awarded for the recovery of each of 12 keys which are chosen to be of lengths varying from 40 bits all the way up to 128 bits, with the length increasing in steps of eight bits. The email sender of the first correctly formatted submissions to each contest will receive a cash prize. For the RSA DES Challenge the first sender of the secret DES key will receive $10,000. For the other contests the prize money awarded will vary with the difficulty of the RC5 key attacked. For more information about the ongoing RSA Factoring Challenge send email to challenge-administrator@rsa.com and for the latest news and developments send email to challenge-news@rsa.com. About RSA Data Security, Inc. RSA Data Security, Inc., a wholly owned subsidiary of Security Dynamics Technologies, Inc., is the world's brand name for cryptography, with more than 75 million copies of RSA encryption and authentication technologies installed and in use worldwide. RSA technologies are part of existing and proposed standards for the Internet and World Wide Web, IT4, ISO, ANSI, IEEE, and business, financial and electronic commerce networks around the globe. The company develops and markets platform-independent developer's kits and end-user products, and provides comprehensive cryptographic consulting services. For more information on any of RSA's encryption technologies, please call RSA directly at 415/595-8782 or send electronic mail to sales@rsa.com. RSA also provides information on its Web site at http://www.rsa.com. **************************************************************** Kurt R. Stammberger Director, Technology Marketing RSA Data Security, Inc. (A Security Dynamics Company) 415-595-8782 vox 415-595-1873 fax kurt@rsa.com www.rsa.comArticle: 4996
Don't be fooled by imitations... I have been keeping the BEST listing anywhere on the internet for the past 1 1/2 years and it gets updated frequently. Right now, there are 327 CHIP MANUFACTURER websites listed! Check it out (brief or verbose versions available) and bookmark it. http://www.scruznet.com/~gcreager/hello5.htm -- Gray Creager Applications Engineer Xicor, Inc. http://www.xicor.comArticle: 4997
Hello, I'm 14 years old and I think I may be a gay. I'm looking for some support and friendship with a older male age 18-40. Please email if you can help.Article: 4998
Mark Adam's suggestion for using the Linked SNF extractor is a good one, but you can also do multi-chip designs with one project. The key is not to "chips" instead of "cliques." A clique is a strong hint to the compiler that logic should be kept together, but the compiler can ignore it. A "chip" forces the compiler to put certain specified logic in a certain device. Since you want four devices, create four chips and assign pins and as much logic as you want to specify to each chip. If you leave some logic unassigned, Maxplus2 will automatically divide that logic among the chips. Also "reserve resources" can be used to keep utilization within each device down. I think that Maxplus2 does a pretty good job of paritioning, but pins created by to tranfer "internal signals" between chips can be a hard to work with if the design changes. If you know exactly which logic goes in which chip, the Linked SNF extractor is probably best. If you want Maxplus2 to do some of the paritioning work, use chips. benArticle: 4999
Gabby Shpirer wrote: > > Hello Altera people, > > I'm trying to compile an hirarchical design into 4 deferent devices > > with no big success. > > Our design combined from 4 blocks and we tried to put each block into a > > deferent EPM7096LC68-12. > > What we did: > > We had 4 blocks on the top scheme, and with the clique function we > attached to each block a deferent device. > At the end of the compilation, the compiler shouts that it didn't use > one of the devices but the project fit by putting two blocks in one > device, and that aint what we wanted. > Please if you can give us more info about partitioning a top design, > that can be a great help. > > Thanx, > > Gabby. > You said you used the Clique function. Did you try the Assign Chip menu option? Grab each block (one at a time) and assign them to a different chip (under the Assign -> Pin/Location/Chip menu). You can assign any block of logic to a particular chip this way (the chip you can call what you want; c1, c2, etc...). Just make sure you make the appropriate interconnections on the board to connect them together, if needed. Wayne
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