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
This is a multi-part message in MIME format. --------------7507BBFF15930032111DAA7A Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit We've been told 8-10 wks for a couple of FG680 Virtex-E. It'll take that long to work out the rest of the board and what we throw into the parts anyway. These things are getting BIG inside :oO. Don McCarley wrote: > > Has anyone been able to actually obtain the larger packages for Virtex and > Virtex E devices? Like the FG456, FG676, and FG680 for the 2.5V Virtex or > the FG456, FG676, FG680, FG860, FG900, and FG1156 for the 1.8V Virtex-E. > I'm interested in the lower io-count package availability experiences also. > Thanks, DM --------------7507BBFF15930032111DAA7A Content-Type: text/x-vcard; charset=us-ascii; name="jsmith.vcf" Content-Transfer-Encoding: 7bit Content-Description: Card for John L. Smith Content-Disposition: attachment; filename="jsmith.vcf" begin:vcard n:Smith;John L. tel;work:858-320-4102 x-mozilla-html:FALSE url:http://www.visicom.com org:Visicom;Imaging Products adr:;;10052 Mesa Ridge Court;San Diego;CA;92121;USA version:2.1 email;internet:jsmith@visicom.com title:Principal Engineer x-mozilla-cpt:;30864 fn:John L. Smith end:vcard --------------7507BBFF15930032111DAA7A--Article: 21226
Tom Burgess wrote: > > Maybe > > von Neumann, J. (1963). > Probabilistic logics and the synthesis of reliable organisms from unreliable components. > In Collected works, (Taub, A. H., ed.), vol. 5, pp. 341-342, The MacMillan Company, New York. > > Mark Thorson wrote: > > > > I vaguely recall a paper with a title something like > > "On Building A Reliable System From Unreliable Nodes" > > by von Neumann. Does anyone remember that more clearly? Another good place to check out is information on computers used in space craft. My question is once you find the fault what do you do then? You can't fix the faulty part in a VLSI chip and you can't switch in a new circuit cause the switching logic may be faulty. Run 3 separate computers systems and pick the best 2 out of 3? Ben. Why does programmable logic remind me of the problems with the computer systems found on star trek? -- "We do not inherit our time on this planet from our parents... We borrow it from our children." The Lagging edge of technology: http://www.jetnet.ab.ca/users/bfranchuk/woodelf/index.htmlArticle: 21227
On Thu, 9 Mar 2000 15:36:34, Jonathan Bromley <jsebromley@brookes.ac.uk> wrote: > "Keith R. Williams" wrote: > <snip> > > I am rather surprised that P&R is so coupled to processor > > performance. > > Why? P&R is about the only thing that people do on PCs > that's nearly 100% compute bound :-) Well, I would assume that 100% CPU efficiency would be highly dependant on the memory subsystem (including caches). It may be a coincidence that these two systems are related, but it surprised the hell out of me that the numbers were so close to MHz. Remember, my office system is a PII-333 (one of the original Deschutes processors). It's running at 3.5x66MHz. The memory subsystem of this thing is crippled, at best. The other system is a PIII (Katmai, I presume) running at 6x100MHz, with a faster cache and all that goes with it. Yes, I am surprised when *anything* scales with just the processor clock. ..not a big deal, I have a room full of these things waiting for my design to actually *do* something. I can put them to work on the P&R while I do other things. Meanwhile, I have ammunition to justify a better computer. I'd really like a useful laptop. ;-) ---- KeithArticle: 21228
I have a design which is to be targeted to ASIC. This design is synthesized with 66 Mhz PCI constraint. Timings are not met with a slack of approx -4 ns I gave worst case constraints from PCI side. and best case synthesis cell selection options(timingwise) As the code changes will take a lot time again to simulate, I have another quick option I am sure that this design will work on 33 Mhz clock. Now what method I am following is 1) Synthesis with 66 Mhz constraint with worst case PCI delays. Let the logic be generated with max efforts. 2) Generate timing reports for 66 Mhz , 33 Mhz . ( Ofcource 66 mhz will fail. ) 3) Target it to ASIC. with a facility on board made for 33 Mhz, 66 Mhz. ( PCI66EN and Config space bit reset from board jumpers ). 4) Assume that PCI chipset will give me best timings ( If I put the card nearer to chipset ), I will give a try on 66 Mhz enable via board jumpers. Is this method appropriate ? What I am assuming is bus will not give me worst timings. ---Yang Li yang_li1@yahoo.co.uk * Sent from RemarQ http://www.remarq.com The Internet's Discussion Network * The fastest and easiest way to search and participate in Usenet - Free!Article: 21229
Hello, I am looking for a way to protect intellectual property inside a Xilinx Virtex FPGA. We have developed a proprietary function and have applied for a patent on it but now we need to do a design where the customer can modify the non-proprietary parts of the FPGA. One way to deal with the problem is to separate the design into two chips with our proprietary logic in one chip and the customer specific logic in the other. We could then give them the full source code to their chip and only the compiled design for the chip that contains our IP. However, with this approach we lose some of the advantage of integration such as size, power, flexibility. It occurred to us that perhaps we could combine all the logic into one chip if we could convert our proprietary IP into a core. This way the customer could modify the circuitry around the core and we could protect our IP. Does anyone here have experience with creating a Xilinx core? We are not really interested in going through formal qualification of the core, just IP protection. Are there simpler methods to protect IP that might be sufficient? For example, could we protect the function by supplying only an edif netlist? Thanks in advance for your advice. -- Pete Dudley Arroyo Grande SystemsArticle: 21230
hi I need a tool that has a state editor and is able to produce D-type flip-flop equations: no VHDL or VERILOG code, but equations like y2+ = (y1 AND z) + (y2 EXOR z) where y2+ is the next state and y2 is present state. Does Xilinx Foundation do this job? thanks for any help EnricoArticle: 21231
This is a multi-part message in MIME format. --------------BA7EA10B1EC8EA03BFBBA6D1 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit > hi > I need a tool that has a state editor and is > able to produce D-type flip-flop equations: no VHDL > or VERILOG code, but equations like > > y2+ = (y1 AND z) + (y2 EXOR z) > > where y2+ is the next state and y2 is present state. > > Does Xilinx Foundation do this job? > > thanks for any help > Enrico I know if you use the FSM editor in Foundation and output the design as ABEL and implement it in a CPLD, then the report file will have the boolean equations in it. -- || Dr. Dave Van den Bout XESS Corp. (919) 387-0076 || || devb@xess.com 2608 Sweetgum Dr. (800) 549-9377 || || http://www.xess.com Apex, NC 27502 USA FAX:(919) 387-1302 || --------------BA7EA10B1EC8EA03BFBBA6D1 Content-Type: text/x-vcard; charset=us-ascii; name="devb.vcf" Content-Transfer-Encoding: 7bit Content-Description: Card for Dave Vanden Bout Content-Disposition: attachment; filename="devb.vcf" begin:vcard n:Vanden Bout;Dave tel;fax:(919) 387-1302 tel;work:(919) 387-0076 x-mozilla-html:FALSE url:http://www.xess.com org:XESS Corp. adr:;;2608 Sweetgum Drive;Apex;NC;27502;USA version:2.1 email;internet:devb@xess.com title:FPGA Product Manager x-mozilla-cpt:;-16464 fn:Dave Vanden Bout end:vcard --------------BA7EA10B1EC8EA03BFBBA6D1--Article: 21232
On Thu, 09 Mar 2000 13:22:09 +0000, Rick Filipkiewicz <rick@algor.co.uk> wrote: >I was thinking more of Mentor buying Synplicity as a way of reducing the >competition! Believe me Rick, from where I'm standing, it really isn't necessary! The incremental business that would result would _not_ be worth the large "valuation" that I have heard that Synplicity management would want to realise. Not by a looooong way! Cheers Stuart An employee of Saros Technology: Model Technology, Exemplar Logic, TransEDA, Renoir. www.saros.co.ukArticle: 21233
The edif netlist is about all the security you'd get with a core too. A determined person can reverse engineer the edif netlist fairly easily. You can make his job harder by obfuscating the names of signals and functions, and by making the edif netlist flat. The fixed cores, at least in coregen 1.5 are delivered as an edif netlist. What makes them a core is the extra stuff...documentation, simulation models, instantiation templates etc. peter dudley wrote: > Hello, > > I am looking for a way to protect intellectual property inside a Xilinx > Virtex FPGA. We have developed a proprietary function and have applied for a > patent on it but now we need to do a design where the customer can modify > the non-proprietary parts of the FPGA. > > One way to deal with the problem is to separate the design into two chips > with our proprietary logic in one chip and the customer specific logic in > the other. We could then give them the full source code to their chip and > only the compiled design for the chip that contains our IP. However, with > this approach we lose some of the advantage of integration such as size, > power, flexibility. > > It occurred to us that perhaps we could combine all the logic into one chip > if we could convert our proprietary IP into a core. This way the customer > could modify the circuitry around the core and we could protect our IP. > > Does anyone here have experience with creating a Xilinx core? We are not > really interested in going through formal qualification of the core, just IP > protection. > > Are there simpler methods to protect IP that might be sufficient? For > example, could we protect the function by supplying only an edif netlist? > > Thanks in advance for your advice. > > -- > Pete Dudley > > Arroyo Grande Systems -- -Ray Andraka, P.E. President, the Andraka Consulting Group, Inc. 401/884-7930 Fax 401/884-7950 email randraka@ids.net http://users.ids.net/~randrakaArticle: 21234
Freeware Newsreader Newsreader freeware at http://www.webmasterfree.com/xnews.html is worth looking into. Multiple ID and server support. It's full version freeware. ICQ 67129994Article: 21235
I have a design which is to be targeted to ASIC. This design has been synthesized with 66 Mhz PCI constraint. Timings are not met with a slack of approx -4 ns I gave worst case constraints from PCI side. As the code changes will take a lot time again to simulate, I have another quick option I am sure that this design will work on 33 Mhz clock. Now what method I am following is 1) Synthesis with 66 Mhz constraint with worst case PCI delays. Let the logic be generated with max efforts. 2) Generate timing reports for 66 Mhz , 33 Mhz . ( Ofcource 66 mhz will fail. ) 3) Target it to ASIC. with a facility on board made for 33 Mhz, 66 Mhz. ( M66EN and Config space bit to be tied from board jumpers ). 4) Assume that PCI chipset will give me best timings ( If I put the card nearer to chipset ), I will give a try on 66 Mhz enable via board jumpers. Is this method appropriate ? What I am assuming is bus will not give me worst timings. Thanks in advance Yang Li * Sent from RemarQ http://www.remarq.com The Internet's Discussion Network * The fastest and easiest way to search and participate in Usenet - Free!Article: 21236
I'm looking for a quick example of how to use an LPM module from VHDL in Altera.. thanks, Tobin.Article: 21237
Call for Participation: SIGDA Ph.D. Forum at DAC'2000 ******************************************************************** * EXTENDED Submission Deadline: Friday, March 17, 2000 * ******************************************************************** http://www.eng.uci.edu/~daforum/ Tuesday, June 6, 2000, 7-9pm Los Angeles Convention Center The Ph.D. Forum at the Design Automation Conference is a poster session hosted by SIGDA for Ph.D. students to present and discuss their thesis work with people in the DA community. It is a great chance for the Ph.D. students to get feedback on their work, and for the industry to preview academic work-in-progress. Travel grants will be available to select students. Eligible students are * those within 1-2 years from completion of thesis, with - either a university-approved thesis proposal, - or at least one published conference paper. * those who have completed their theses in the 1999-2000 academic year What to Submit * A one-page abstract of the thesis in PDF, not including figures or references, and not to exceed 750 words. * A university-approved thesis proposal, or a published paper. this is required of ALL students. * Names of five reviewers whom the student would like to review the abstract Dates * Submission deadline (extended): Friday, March 17, 2000 * Travel Grants Notification: April 30, 2000 * Forum Presentation: Tuesday, June 6, 2000, 7-9pmArticle: 21238
Enrico Migliore wrote: > > hi > I need a tool that has a state editor and is > able to produce D-type flip-flop equations: no VHDL > or VERILOG code, but equations like > > y2+ = (y1 AND z) + (y2 EXOR z) > > where y2+ is the next state and y2 is present state. > > Does Xilinx Foundation do this job? > > thanks for any help > Enrico I don't know of any tools that will output a generic equation that you are asking for. But I will offer my 2 cents worth on how to design FSMs in FPGAs. I assume that you are working with Xilinx FPGAs or something similar and using schematic capture. These devices have 4 input LUTs and lots of registers. I find that FSMs are best done one hot encoded. This allows each state to be represented by one signal. The state diagrams can be easily translated with a one to one correspondence between the trasition lines and logic elements. Each trasistion is an AND gate with the from state as one input and the transistion condition as another input. The outputs of all the AND gates for a given state are ORed together. This signal must be negated on the exit conditions (another AND gate). This signal then defines the state. If you change the transitions into a state, you only have to add or delete the AND gates feeding the OR gate. Changing the output from a state affects the negation of the OR output. This is actually not hard to do in schematic form. I would bet that Ray A. can describe it better. I believe he even has some functional blocks to make these elements. -- Rick Collins rick.collins@XYarius.com remove the XY to email me. Arius - A Signal Processing Solutions Company Specializing in DSP and FPGA design Arius 4 King Ave Frederick, MD 21701-3110 301-682-7772 Voice 301-682-7666 FAX Internet URL http://www.arius.comArticle: 21239
I do. I've got the basic elements as hierarchical schematics. The symbols for those components look like flow chart items. For example, the decision diamond is drawn as you'd expect with an input a yes branch output and a no branch output. Addtionally, there is a condition input. Underneath the symbol, the schematic is a pair of and gates with the condition and input wired to both. The no output and has its condition input inverted. I have the OR gates separate, and these have a symbol that looks like a wire junction. This works out nicely because you wind up with a schematic that looks for all the world like a flowchart. It is easy to grok what it does and almost as easy to make changes. Even complicated state machines are quick to design and easy to enter. Rickman wrote: > Each trasistion is an AND gate with the from state as one input and the > transistion condition as another input. The outputs of all the AND gates > for a given state are ORed together. This signal must be negated on the > exit conditions (another AND gate). This signal then defines the state. > If you change the transitions into a state, you only have to add or > delete the AND gates feeding the OR gate. Changing the output from a > state affects the negation of the OR output. > > This is actually not hard to do in schematic form. I would bet that Ray > A. can describe it better. I believe he even has some functional blocks > to make these elements. The credit for this technique goes to the guys at highgate though. I just adopted it very early on. -- -Ray Andraka, P.E. President, the Andraka Consulting Group, Inc. 401/884-7930 Fax 401/884-7950 email randraka@ids.net http://users.ids.net/~randrakaArticle: 21240
Hi Folks, Just a quick note that the list of FPGA prototype boards and systems has been updated on The Programmable Logic Jump Station (http://www.optimagic.com/boards.html). If you know of some boards that are missing, please let me know. Ideally, we're trying to limit the list to those that are commercially available. Thanks. -- ----------------------------------------------------------- Steven K. Knapp OptiMagic, Inc. -- "Great Designs Happen 'OptiMagic'-ally" E-mail: sknapp@optimagic.com Web: http://www.optimagic.com -----------------------------------------------------------Article: 21241
There is a list of VHDL books on The Programmable Logic Jump Station at http://www.optimagic.com/books.html. -- ----------------------------------------------------------- Steven K. Knapp OptiMagic, Inc. -- "Great Designs Happen 'OptiMagic'-ally" E-mail: sknapp@optimagic.com Web: http://www.optimagic.com ----------------------------------------------------------- <stang99@my-deja.com> wrote in message news:89gro4$vbo$1@nnrp1.deja.com... > I am in need of a good book covering VHDL techniques for synthesis in > PLD's, especially FPGA's. A more general book covering VHDL for > synthesis (not PLD's specifically) would be just as good. If anyone has > any suggestions I would greatly appreciate it. Thanks in advance. > > John M. > > > Sent via Deja.com http://www.deja.com/ > Before you buy.Article: 21242
Hi. I'm new to this field. I have a good idea of what we can do with FPGA but I saw something that I can understand. I found a companie (in the Web) that do DSP with FPGA. If I'm not wrong DSP are like MCU, but specifically designed to handle analog data. So how can we handle analog data in FPGA pins. What's worst, I saw (in the same compagny) "Modulation and demodulation with FPGA". How can we do this?Article: 21243
Does anyone know of a prewired standalone EEPROM memory that can be used for storing data logging information in a hostile environment, unpowered, for long periods of time? We are after about 1MByte of storage area. If not all of those features, than what would you suggest? * Sent from RemarQ http://www.remarq.com The Internet's Discussion Network * The fastest and easiest way to search and participate in Usenet - Free!Article: 21244
Hi! DSP is *Digital* signal processor or processing. You can not handle analog data in with an FPGA, you first have to convert the analog signal to a digital signal and then process it. / Jonas On Mon, 13 Mar 2000 00:15:14 -0500, "Jean-Réginald Louis" <louis.reginaldjean@teccart.qc.ca> wrote: >Hi. I'm new to this field. I have a good idea of what we can do with FPGA >but I saw something that I can understand. I found a companie (in the Web) >that do DSP with FPGA. If I'm not wrong DSP are like MCU, but specifically >designed to handle analog data. So how can we handle analog data in FPGA >pins. What's worst, I saw (in the same compagny) "Modulation and >demodulation with FPGA". How can we do this? > > >Article: 21245
Hello, I'm looking for data concerning the various Xilinx FPGA devices (from XC3000 to Virte-E). Specifically i'm looking for their year of introductions (i'd like to see how FPGA density evolve with time). I have been looking on Xilinx web site but did not find what I was looking for. Can someone help me with this ? StevenArticle: 21246
Jean, DSP is DIGITAL signal processing. An Analog to Digital converter (ADC) is used to convert the analog signal to a discrete time series of digital values representing the analog signal. The digital values are typically 8-16 bits each. Some modern ADC's can sample at 200MHz or more. The basic math theory is pretty much the same with the exception of dealing with discrete time signals instead of continuous time. With that in mind, the FPGA is simply doing the math needed to demodulate the signal into a discrete digital base-band signal. By performing the signal processing digitally, we gain advantages in the flexibility of the system, and at the same time eliminate many of the analog calibration headaches. "Jean-Réginald Louis" wrote: > Hi. I'm new to this field. I have a good idea of what we can do with FPGA > but I saw something that I can understand. I found a companie (in the Web) > that do DSP with FPGA. If I'm not wrong DSP are like MCU, but specifically > designed to handle analog data. So how can we handle analog data in FPGA > pins. What's worst, I saw (in the same compagny) "Modulation and > demodulation with FPGA". How can we do this? -- -Ray Andraka, P.E. President, the Andraka Consulting Group, Inc. 401/884-7930 Fax 401/884-7950 email randraka@ids.net http://users.ids.net/~randrakaArticle: 21247
Excuse me! I didn't have a clear idea of what it's a DSP. In my course, I'm learning microcontroller. I have friends studying telecom (college level) and they using DSP to manipulate signal (like sound) and I automatically made an association between telecom and DSP. Can someone tell me the difference between a MCU (microcontroller), a DSP and a CPU. What type of processing unit are use in what type of project/environement/situation/etc ? Why a graphics card use a DSP (like the nVidia GeForce (if I'm not wrong)) ? > Hi! > > DSP is *Digital* signal processor or processing. > > You can not handle analog data in with an FPGA, you first have to > convert the analog signal to a digital signal and then process it. > > / Jonas > > On Mon, 13 Mar 2000 00:15:14 -0500, "Jean-Réginald Louis" > <louis.reginaldjean@teccart.qc.ca> wrote: > > >Hi. I'm new to this field. I have a good idea of what we can do with FPGA > >but I saw something that I can understand. I found a companie (in the Web) > >that do DSP with FPGA. If I'm not wrong DSP are like MCU, but specifically > >designed to handle analog data. So how can we handle analog data in FPGA > >pins. What's worst, I saw (in the same compagny) "Modulation and > >demodulation with FPGA". How can we do this? > > > > > > > > Hi! > > DSP is *Digital* signal processor or processing. > > You can not handle analog data in with an FPGA, you first have to > convert the analog signal to a digital signal and then process it. > > / Jonas > > On Mon, 13 Mar 2000 00:15:14 -0500, "Jean-Réginald Louis" > <louis.reginaldjean@teccart.qc.ca> wrote: > > >Hi. I'm new to this field. I have a good idea of what we can do with FPGA > >but I saw something that I can understand. I found a companie (in the Web) > >that do DSP with FPGA. If I'm not wrong DSP are like MCU, but specifically > >designed to handle analog data. So how can we handle analog data in FPGA > >pins. What's worst, I saw (in the same compagny) "Modulation and > >demodulation with FPGA". How can we do this? > > > > > > >Article: 21248
Hi I write two VHDL programs, one for a modulator and one for a demodulator (for sound transmission). I know that the modulator works good. And now I want to test the demodulator, so I have to write a testbench for it. I have never written a testbench and I don't know if my idea is the best. If I use structural VHDL and use the modulator as component no 1, then the modulator as component no 2 and then look at the output of the demodulator. Is this a "simple" way to check the demodulator or does anyone else know a better way? Björn LindegrenArticle: 21249
This is a multi-part message in MIME format. --------------CBCE2188E38A687915F5A540 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit Hi, To my knowledge , you can write testbench for any type of description . if you want to verify the demodulator , you can seperately simulate that . otherwise you can interface the modulator and demodulator in the structural description . Well , the book " the designers guide for Vhdl " by peter ashenden explains clearly about testbench with lot of examples or you can refer Vhdl primer by jayaram bhaskar . if i am wrong , please correct me hope this helps kamal "Björn Lindegren" wrote: > Hi > > I write two VHDL programs, one for a modulator and one for a demodulator > (for sound transmission). > > I know that the modulator works good. > > And now I want to test the demodulator, so I have to write a testbench for > it. > > I have never written a testbench and I don't know if my idea is the best. > > If I use structural VHDL and use the modulator as component no 1, then the > modulator as component no 2 and then look at the output of the demodulator. > > Is this a "simple" way to check the demodulator or does anyone else know a > better way? > > Björn Lindegren --------------CBCE2188E38A687915F5A540 Content-Type: text/x-vcard; charset=us-ascii; name="raja.vcf" Content-Transfer-Encoding: 7bit Content-Description: Card for raja Content-Disposition: attachment; filename="raja.vcf" begin:vcard n:kamalanathan;Raja tel;home:07 38761962 tel;work:07 33658849 x-mozilla-html:FALSE adr:;;;;;; version:2.1 email;internet:raja@elec.uq.edu.au fn:Raja kamalanathan end:vcard --------------CBCE2188E38A687915F5A540--
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