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
Hey, Eric. Seeing as you don't have the FPGA Editor, I'd try emailing Xilinx, asking for the JBits package. There's a email address you have to send your reauest to, then they ask you a question or two, and then give you access to the package. Part of the JBits package includes an explanation of th Virtex/ Spartan-II architacture. It's not as detailed as FPGA editor, but it's much better than the databook. HTH, -Kent Eric Smith <eric-no-spam-for-me@brouhaha.com> writes: > Is there any document that gives more detailed logic drawings of > the Spartan II CLBs? The description and diagram in the data sheet > are sort of, well, spartan. > > I'm trying to figure out whether I can do certain things like > simultaneously route the carry out both to the carry-in of the CLB above > *and* to the GRM or to the CLB on the right. Figure 3 in the data sheet > shows a slice, but it has big unexplained blocks like "Carray and Control > Logic". Also, the possible routing choices for signals are not obvious. > > The text talks about the F5 and F6 multipliexers, but they don't appear > in the drawing at all. > > These details aren't considered proprietary trade secrets, are they? > That would make it hard for me to figure out how to best design for > this part.Article: 29776
Kent Orthner <korthner@hotmail.nospam.com> writes: > Seeing as you don't have the FPGA Editor, I'd try emailing Xilinx, > asking for the JBits package. There's a email address you have to > send your reauest to, then they ask you a question or two, and > then give you access to the package. OK, I've done that. This brings up my next question: has anyone successfully installed and run JBits 2.5 on Red Hat Linux 7.0? Xilinx says they test JBits with the Sun JDK 1.2.2. However, JDK 1.2.2 won't run on Red Hat 7.0, and Sun recommends using JDK 1.3 instead. So I installed JDK 1.3, but when I try to install JBits, I get: % /usr/java/jdk1.3/bin/java JBits2_5.class Exception in thread "main" java.lang.NoClassDefFoundError: JBits2_5/class % I thought maybe the problem was not having the JDK bin directory in my PATH, so I added that, but I get the same results. Any idea what might be wrong? Thanks! EricArticle: 29777
The evaluation kit is available for free through Xilinx distributors. Check with your local distributor and if need further info refer http://www.xilinx.com/products/software/ise_eval.htm -Vikram Xilinx Applications. Eric Smith wrote: > I wrote: > > Has anyone managed to get (or at least order) a Foundation ISE Evaluation > > Kit? According to a Xilinx press release of 21-Feb-2001, it is available > > I should have included the URL: > > http://www.xilinx.com/prs_rls/0111iseeval.htmArticle: 29778
--------------7475764B730FED8AD0CDEDBF Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Peter, In the 4K family, this is taken into account during place and route in order to meet the constraints. You do not have to think about how we implement the route: you do have to place constraints on it so the software knows what to do. The trace report then tells you what is happening, in the worst of all possible combinations of voltage, temperature, and process. In order to make the software run faster (which it does every time we update it), we also designed the Virtex architecture to have less delay variation with load (more buffered paths). You still have to set your constraints! This has reached the complete endpoint in Virtex II where all interconnect is buffered. There is still a secondary effect of loading, but it is now much less, and easier to model. Austin peter ritchie wrote: > Hi all > > target Xilinx FPGA. > > If a signal is routed through the longline, does the > high fanout of this signal, affect its final speed, or > no ? > Say a transpose filter structure , the input is > broadcasted to all the multpliers, i know this > architecture is more optimised, at least in term of > area, than the direct form, but will this long line > broadcast not affect the final performance ? > > If we have a high fanout signal coming from the CLB > output. How far the degradation will be in function of > > of the fanout. can this output be swicthed to the > longline ? > > <~> Peter <~> > > > __________________________________________________ > Do You Yahoo!? > Get email at your own domain with Yahoo! Mail. > http://personal.mail.yahoo.com/ > > -- > Posted from web1602.mail.yahoo.com [128.11.23.202] > via Mailgate.ORG Server - http://www.Mailgate.ORG --------------7475764B730FED8AD0CDEDBF Content-Type: text/html; charset=us-ascii Content-Transfer-Encoding: 7bit <!doctype html public "-//w3c//dtd html 4.0 transitional//en"> <html> Peter, <p>In the 4K family, this is taken into account during place and route in order to meet the constraints. You do not have to think about how we implement the route: <i>you do have to place constraints</i> on it so the software knows what to do. <p>The trace report then tells you what is happening, in the worst of all possible combinations of voltage, temperature, and process. <p>In order to make the software run faster (which it does every time we update it), we also designed the Virtex architecture to have less delay variation with load (more buffered paths). <p>You still have to set your constraints! <p>This has reached the complete endpoint in Virtex II where all interconnect is buffered. There is still a secondary effect of loading, but it is now much less, and easier to model. <p>Austin <p>peter ritchie wrote: <blockquote TYPE=CITE>Hi all <p>target Xilinx FPGA. <p>If a signal is routed through the longline, does the <br>high fanout of this signal, affect its final speed, or <br>no ? <br>Say a transpose filter structure , the input is <br>broadcasted to all the multpliers, i know this <br>architecture is more optimised, at least in term of <br>area, than the direct form, but will this long line <br>broadcast not affect the final performance ? <p>If we have a high fanout signal coming from the CLB <br>output. How far the degradation will be in function of <p>of the fanout. can this output be swicthed to the <br>longline ? <p> <~> Peter <~> <br> <p>__________________________________________________ <br>Do You Yahoo!? <br>Get email at your own domain with Yahoo! Mail. <br><a href="http://personal.mail.yahoo.com/">http://personal.mail.yahoo.com/</a> <p>-- <br>Posted from web1602.mail.yahoo.com [128.11.23.202] <br>via Mailgate.ORG Server - <a href="http://www.Mailgate.ORG">http://www.Mailgate.ORG</a></blockquote> </html> --------------7475764B730FED8AD0CDEDBF--Article: 29779
"Mike H." wrote: > > Hello everybody, > > I just installed Xilinx 3.3i Service Pack 7 for Alliance-PC, > and it crashes every time I use the mapper on Virtex-E parts. > Seems to be OK on Spartan. > > I get the good ol' Windows error box and a message: > "protection fault in module LIBLMCIRCUIT.DLL". Hi I got a similar problem last week. I can't remember exactly what error I got but I traced it to the -c option of the mapper. If I didn't use it it worked fine. I haven't tried another design to see if it passed -- Nicolas MATRINGE IPricot European Headquarters Conception electronique 10-12 Avenue de Verdun Tel +33 1 46 52 53 11 F-92250 LA GARENNE-COLOMBES - FRANCE Fax +33 1 46 52 53 01 http://www.IPricot.com/Article: 29780
"Ray Andraka" <ray@andraka.com> wrote in message news:3AA8080D.EF078DAA@andraka.com... > Have you tried more than one design? Could be something in your design the > mapper doesn't like. Yes, I've tried 3 or 4 different designs, all of which were working perfectly OK with the previous software release. Plus, I've compiled the exact same design on my PC (crashes) and on my colleagues PC (works OK). I'm pretty sure it's one of these damn Microsoft DLL issues - I had the same problem a year or so ago, but I can't remember what I did to fix it! For now, I'm sticking with 3.3sp6 which works OK. MH.Article: 29781
Our client is the leading provider of highly integrated silicon solutions that enable broadband digital transmission of voice, video, and data. Using proprietary technologies and advanced design methodologies, the company designs, develops and supplies integrated circuits for a number of the most significant broadband communications markets, including the markets for cable set-top boxes, cable modems, high-speed local, metropolitan and wide area networks, home networking, Voice over Internet Protocol (VoIP), residential broadband gateways, direct broadcast satellite and terrestrial digital broadcast, optical networking, digital subscriber lines (xDSL) and wireless communications. They are currently seeking ASIC Designers with the following... Required Skills and Experiences: BSEE or equivalent; MSEE preferred, plus 5+ years' of VLSI hardware design experience. Position requires working knowledge of Verilog and Synopsys. Experience with back-end CAE, layout tools, 3D, video, and DSP is a plus. Responsibilities: Responsibilities include: definitions of the molecules specifications based on the system specification; RTL development and debugging, test bench development; design validation as part of an overall system, system integration; synthesis of the RTL at a certain clock speed; Implementation of the test strategy, scan insertion, ATPG test vectors generation. PROCOM: Established in 1978, Professional Computer Consultants Group Ltd. (Procom) is a national leader in the provision of Computer personnel on a contract and full-time basis. Our clients are comprised of the largest national and international corporations that utilize technical resources extensively across a wide range of disciplines. In the Financial Post (March 1999) Procom was ranked as the 6th largest professional Services Company in Canada. In November of 1999, Procom was named a Regional finalist in Canada' s 50 Best Managed Private companies. Our track record is proven with more than 180 consultants servicing Ottawa's high tech community and more than 1600 Procom consultants currently on assignment throughout. North America. For further information on this and other opportunities please visit our web site at www.procom.ca. Interested candidates are invited to forward their resumes or questions in confidence to: Derek Weber PROCOM 300 March Rd Suite 600 Kanata, Ontario K2K-2E2 613-270-9339 x231 613-270-9449 (FAX) derekw@procom.ca www.procom.caArticle: 29782
Hi I think that You must add RLOC to each muxcy_L AND change the XOR to XORCY with its own RLOC. Remember to annotate the design. best regards IvarArticle: 29783
Nicolas Matringe wrote: > > ERROR:MapLib:93 - Illegal LOC on symbol "SW1.PAD" (pad > > signal=SW1) or BUFGP symbol "C15" (output signal=SW1_BUFGPed), > > IPAD-IBUFG should only be LOCed to GCLKIOB site. > > ... > I think your synthesis tool (FPGA Express ?) automatically inserted the > global buffer on the clock line. There is a switch somewhere to disable > this. Yup, thanx! This was the case. In Xilinx Foundation F3.1i the switch is located in ->Versions View ->Ver<x> ->ContextMenu "Edit Synthesis Constraints" ->Tab "Ports" ->Column "Global buffer" Great thank you also goes to Michael Böhnel! StephanArticle: 29784
Hi Everybody I'm looking for an evaluation board for Xlininx Spartan-II FPGAs. Does any body know where to get a eval board that is not too expensive. Something like the XESS boards www.xess.com would be nice, but unfortunately there is no Board with larger Spartan FPGAs, I'm looking for a board that provides to least a XC2S50, better XC2S100 or even larger. Does anybody have any good pointers to manufacturers, distributors? Greetings, ChrisArticle: 29785
Ivar wrote: > I think that You must add RLOC to each muxcy_L AND change the XOR to > XORCY with its own RLOC. Remember to annotate the design. I don't understand why should I change anything in library macros. I don't think it's the way... The problem is more general: I have 3 levels of design (top, mid, bot) Bot is the compcm8 library macro (with original RLOCs). The macro is placed on mid level with RLOC attached. This level is converted to another macro within my design's library and put on the top level with RLOC_ORIGIN parameters. This gives errors - WHY ? When I move RLOC_ORIGIN to mid level everything is ok, except... this is not what I want to do. Pawel J. RajdaArticle: 29786
I am pleased to post a copy of my snug 2001 presentation on "Component Verification by Example". This paper presents, by example, some of the key features of the front-end processes for specifying the planning of both the implementation and verification of a design to ensure that the implemented design meets its intended requirements and costs. Presentation and code is available at my site: http://www.vhdlcohen.com under "Models and Papers" File: verifxmpl.zip. SNUG 2001 Presentation + VHDL approaches + Verilog Code on Verification by example using transaction based methods -------------------------------------------------------------------------- ------------------------------------------ Ben Cohen Publisher, Trainer, Consultant (310) 721-4830 http://www.vhdlcohen.com/ vhdlcohen@aol.com Author of following textbooks: * Component Design by Example ... a Step-by-Step Process Using VHDL with UART as Vehicle", 2001 isbn 0-9705394-0-1 * VHDL Coding Styles and Methodologies, 2nd Edition, 1999 isbn 0-7923-8474-1 * VHDL Answers to Frequently Asked Questions, 2nd Edition, isbn 0-7923-8115 -------------------------------------------------------------------------- ------------------------------------------Article: 29787
Christian Plessl <cplessl@gmx.net> writes: > I'm looking for an evaluation board for Xlininx Spartan-II FPGAs. Does > any body know where to get a eval board that is not too expensive. > > Something like the XESS boards www.xess.com would be nice, but > unfortunately there is no Board with larger Spartan FPGAs, I'm looking > for a board that provides to least a XC2S50, better XC2S100 or even > larger. Burch sells a very inexpensive board with an XC2S200, the BED-SPARTAN2+. The cost is AU $220, which is approximately US $120. They are out of stock of the current version, but will have a new version available on March 19th. I've used the old version and am very happy with it.Article: 29788
"Pawe³ J. Rajda" <pjrajda@uci.agh.edu.pl> wrote in message news:3AA91E60.498810E4@uci.agh.edu.pl... > The problem is more general: I have 3 levels of design (top, mid, bot) > Bot is the compcm8 library macro (with original RLOCs). The macro > is placed on mid level with RLOC attached. This level is converted to > another macro within my design's library and put on the top level with > RLOC_ORIGIN parameters. This gives errors - WHY ? > When I move RLOC_ORIGIN to mid level everything is ok, except... > this is not what I want to do. Try looking in the EDIF file generated by your netlist software. This can be a lot of work, but you may be able to spot something wierd in the EDIF which correlates with the problem.Article: 29789
I just read that Handspring, Xilinx, and Portable Design magazine are jointly sponsoring a Springboard module design contest. Details are at www.xilinx.com/contest. I don't work for any of the above but can't help but wonder what goodies will come of it. There are a lot of sharp CPLD whizzes on this group and the Springboard design looks straightforward. Biggest question for me is - what to design? Best of luck to all! Jim Horn, WB9SYN/6 (just enjoy cranking telecom via programmable logic)Article: 29790
Anyone using LVDS I/O on Virtex-II ? We're having trouble getting Edif2NGD to correctly expand the lower level primitives (IOBUF_LVDS). We're using Synplicity V6.13, Alliance 3.3i (w/ Service Pack 7). Not sure if we have all the files we're supposed to. When Edif2NGD fails, we get an error 432. It appears that it's looking for a .NGO file for the IOBUF_LVDS primitive. Any ideas ??Article: 29791
"PROCOM" <derekw@procom.ca> wrote in message news:98amoj$db1$1@news.igs.net... > Our client is the leading provider of highly integrated silicon solutions > that enable broadband digital transmission of voice, video, and data. Using > proprietary technologies and advanced design methodologies, the company > designs, develops and supplies integrated circuits for a number of the most > significant broadband communications markets, including the markets for > cable set-top boxes, cable modems, high-speed local, metropolitan and wide > area networks, home networking, Voice over Internet Protocol (VoIP), > residential broadband gateways, direct broadcast satellite and terrestrial > digital broadcast, optical networking, digital subscriber lines (xDSL) and > wireless communications. > > They are currently seeking ASIC Designers with the following... > > Required Skills and Experiences: > > BSEE or equivalent; MSEE preferred, plus 5+ years' of VLSI hardware design > experience. > Position requires working knowledge of Verilog and Synopsys. > Experience with back-end CAE, layout tools, 3D, video, and DSP is a plus. > > > Responsibilities: > > Responsibilities include: definitions of the molecules specifications based > on the system specification; RTL development and debugging, test bench > development; design validation as part of an overall system, system > integration; synthesis of the RTL at a certain clock speed; Implementation > of the test strategy, scan insertion, ATPG test vectors generation. > > PROCOM: Established in 1978, Professional Computer Consultants Group Ltd. > (Procom) is a national leader in the provision of Computer personnel on a > contract > and full-time basis. Our clients are comprised of the largest national and > international corporations that utilize technical resources extensively > across a wide range of disciplines. In the Financial Post (March 1999) > Procom was ranked as the 6th largest professional Services Company in > Canada. In November of 1999, Procom was named a Regional finalist in Canada' > s 50 Best Managed Private companies. Our track record is proven with more > than 180 consultants servicing Ottawa's high tech community and more than > 1600 Procom consultants currently on assignment throughout. North America. > For further information on this and other opportunities please visit our web > site at www.procom.ca. > > Interested candidates are invited to forward their resumes or questions in > confidence to: > > Derek Weber > PROCOM > > 300 March Rd Suite 600 > Kanata, Ontario > K2K-2E2 > 613-270-9339 x231 > 613-270-9449 (FAX) > > derekw@procom.ca > www.procom.ca ************************************************************* Derek, This is an outstanding ad, espcially compared to some I've seen on comp.arch.fpga and comp.lang.verilog. Keep up the good work and good luck finding the right candidates. Simon Ramirez, Consultant Synchronous Design, Inc. Oviedo, FL USAArticle: 29792
Rick Collins <spamgoeshere4@yahoo.com> writes: > Muzaffer Kal wrote: > > > > On 04 Mar 2001 00:25:08 +0100, Magnus Homann > > <d0asta@mis.dtek.chalmers.se> wrote: > > > > >Ray Andraka <ray@andraka.com> writes: > > > > > >[Simons wisdom deleted] > > > > > >> Simon failed to point out that using the state machine as the second > > >> flip flop should only be done if the possibly metastable signal > > >> affects only one flip-flop in the state machine. If if affects more > > >> than one flip-flop, there is a potential for a metastable event to > > >> be detected as different levels by the multiple flip-flops that > > >> depend on its condition. Simon, I'm sure this was your intention, I > > >> just wanted to make sure your short response didn't mislead anyone. > > > > > >Really? Hmm, the idea is that FF1 would settle within the cycle > > >time. Now, if it doesn't, does it really matter if one or several of > > >the statemachines FFs wil go into metastability? > > > > > >The idea is that the cycle time will be enough for all metastability > > >to vanish. If it hasn't, is there anything you can do to save your FSM > > >from running amok? > > > > > >Your advice might be god, but probably mostly due to the fact that having > > >only one FF after FF1 usaully reduces the routing delay, and > > >hence increases the available settling time (iff you have the same > > >cycle time in both cases). > > > > > >Comments invited. > > > > > >Homann > > > > I think that "two FFs" was just a rule of thumb which ensured a > > certain MTBF under certain conditions (which are probably not valid in > > < .25u processes anymore). You can further reduce the probability of > > metastability by adding more FFs. The metastability never vanishes > > completely. There is always a non-zero probability that the metastable > > event will pass through all the FFs. It can be made arbitrarily small > > though. That's where the MTBF comes from. > > > > Muzaffer > > Homann and Muzaffer, > > The factor that settles metastability is not FFs, but rather time. The > need for the second FF is a function of the speed of the clock. As > Homann says, using the second FF separates the metastability settling > time from the time used for routing and logic prop delay. So you get the > entire clock cycle for settling. My point was that I don't think it matters if the second FF is just one or if its several copies. Only the time matters between FF1 and FF2(1..n). At least, it's the most important issue.. Thanks, Homann -- Magnus Homann, M.Sc. CS & E d0asta@dtek.chalmers.seArticle: 29793
It does matter if you've used up most of your clock cycle in routing & logic delays. In that case, you leave yourself very little time to resolve the metastabiltiy, and unequal delays to multiple flip-flops makes it fairly likely to have the two FF's resolve different levels at the next clock edge. Time is critical for metastability, and time propagating to the next FF's doesn't count. Depending on the implementation, the metastability in the first flop, which is guaranteed to happen with an async input, can manifest itself as an oscillation, a runt pulse, a slow rise/fall time or a delayed edge. I've even seen a case (not in an FPGA) where a metastable event triggered a self-sustaining oscillation that would continue until the register was clocked again. It all comes down to clock rate, or more specifically, the slack time before the data is clocked into the next flip-flop. If you have plenty of slack, one flip-flop will usually do it. If not, then be careful! Magnus Homann wrote: > > > My point was that I don't think it matters if the second FF is just > one or if its several copies. Only the time matters between FF1 and FF2(1..n). > At least, it's the most important issue.. > > Thanks, > Homann > -- > Magnus Homann, M.Sc. CS & E > d0asta@dtek.chalmers.se -- -Ray Andraka, P.E. President, the Andraka Consulting Group, Inc. 401/884-7930 Fax 401/884-7950 email ray@andraka.com http://www.andraka.com or http://www.fpga-guru.comArticle: 29794
Magnus Homann wrote: > > Rick Collins <spamgoeshere4@yahoo.com> writes: > > > Muzaffer Kal wrote: > > > > > > On 04 Mar 2001 00:25:08 +0100, Magnus Homann > > > <d0asta@mis.dtek.chalmers.se> wrote: > > > > > > >Ray Andraka <ray@andraka.com> writes: > > > > > > > >[Simons wisdom deleted] > > > > > > > >> Simon failed to point out that using the state machine as the second > > > >> flip flop should only be done if the possibly metastable signal > > > >> affects only one flip-flop in the state machine. If if affects more > > > >> than one flip-flop, there is a potential for a metastable event to > > > >> be detected as different levels by the multiple flip-flops that > > > >> depend on its condition. Simon, I'm sure this was your intention, I > > > >> just wanted to make sure your short response didn't mislead anyone. > > > > > > > >Really? Hmm, the idea is that FF1 would settle within the cycle > > > >time. Now, if it doesn't, does it really matter if one or several of > > > >the statemachines FFs wil go into metastability? > > > > > > > >The idea is that the cycle time will be enough for all metastability > > > >to vanish. If it hasn't, is there anything you can do to save your FSM > > > >from running amok? > > > > > > > >Your advice might be god, but probably mostly due to the fact that having > > > >only one FF after FF1 usaully reduces the routing delay, and > > > >hence increases the available settling time (iff you have the same > > > >cycle time in both cases). > > > > > > > >Comments invited. > > > > > > > >Homann > > > > > > I think that "two FFs" was just a rule of thumb which ensured a > > > certain MTBF under certain conditions (which are probably not valid in > > > < .25u processes anymore). You can further reduce the probability of > > > metastability by adding more FFs. The metastability never vanishes > > > completely. There is always a non-zero probability that the metastable > > > event will pass through all the FFs. It can be made arbitrarily small > > > though. That's where the MTBF comes from. > > > > > > Muzaffer > > > > Homann and Muzaffer, > > > > The factor that settles metastability is not FFs, but rather time. The > > need for the second FF is a function of the speed of the clock. As > > Homann says, using the second FF separates the metastability settling > > time from the time used for routing and logic prop delay. So you get the > > entire clock cycle for settling. > > My point was that I don't think it matters if the second FF is just > one or if its several copies. Only the time matters between FF1 and FF2(1..n). > At least, it's the most important issue.. > > Thanks, > Homann > -- > Magnus Homann, M.Sc. CS & E > d0asta@dtek.chalmers.se As Ray pointed out, metastability is resolved not in the cycle time, but in the slack time. So the more places you have to route to, the less slack time you are left with. So what you say is basicially true, you do have to watch your cycle time vs the routing time. If the routing eats up too much of the cycle time you will degrade your metastability resolution. -- Rick "rickman" Collins rick.collins@XYarius.com Ignore the reply address. To email me use the above address with the XY removed. Arius - A Signal Processing Solutions Company Specializing in DSP and FPGA design URL http://www.arius.com 4 King Ave 301-682-7772 Voice Frederick, MD 21701-3110 301-682-7666 FAXArticle: 29795
I have just received the Xilinx Web Pack ISE v. 3.2. I tried to synthesize the code released by Xilinx itself associated with an Application Note describing how to implement an asynchronous FIFO using Spartan II devices. I have some questions: 1) how to locate the pins? I think I have to use the "attribute" reserved word, but which is the precise syntax? 2) The code uses some device-dependent components (i.e. BUFGB). The synthesizer messages say that it generates black-box for these components. In other words, it seems it can not understant what component to instantiate. How to solve it (missing libraries???)? Thanks a lot in advance. -- llandre e-mail : andmars@tin.it web : http://www.dei.unipd.it/~patchArticle: 29796
> I'm looking for an evaluation board for Xlininx Spartan-II FPGAs. Does > any body know where to get a eval board that is not too expensive. > > Something like the XESS boards www.xess.com would be nice, but > unfortunately there is no Board with larger Spartan FPGAs, I'm looking > for a board that provides to least a XC2S50, better XC2S100 or even > larger. > > Does anybody have any good pointers to manufacturers, distributors? > Have a look here: www.insight-electronics.com. I think they have a board for $130 or something like that, equipped with XC2S100. -- llandre e-mail : andmarsNOSPAM@tin.it web : http://www.dei.unipd.it/~patchArticle: 29797
llandre wrote: > > > I'm looking for an evaluation board for Xlininx Spartan-II FPGAs. Does > > any body know where to get a eval board that is not too expensive. > > > > Something like the XESS boards www.xess.com would be nice, but > > unfortunately there is no Board with larger Spartan FPGAs, I'm looking > > for a board that provides to least a XC2S50, better XC2S100 or even > > larger. > > > > Does anybody have any good pointers to manufacturers, distributors? > > > Have a look here: www.insight-electronics.com. > I think they have a board for $130 or something like that, equipped > with XC2S100. The best I've found is the BurchEd board (XCS200 for $120 US, works with WebPack). Found no problems :-)) ATB, Simon. -- Freedom ? What's that ? (see http://www.domesday.co.uk/ )Article: 29798
Rick Collins wrote: > As Ray pointed out, metastability is resolved not in the cycle time, but > in the slack time. So the more places you have to route to, the less > slack time you are left with. So what you say is basicially true, you do > have to watch your cycle time vs the routing time. If the routing eats > up too much of the cycle time you will degrade your metastability > resolution. And by a lot. MTBF is not linear with slack time but exponential. ----------------------------------------------------------------------- rk Micromanaging leads to macro f'ups. stellar engineering, ltd. -- Richard Marcinko, Seal Team Six stellare@erols.com.NOSPAM Hi-Rel Digital Systems DesignArticle: 29799
glen herrmannsfeldt wrote: > The old story about Huygens hanging multiple pendulum clocks on the > same wall and noticing that they would phase lock. Interesting and a bit ironic. I note that on the probe to Saturn named after him, there is a problem since the probe's transmitter and receiver on the spacecraft can't lock, as the doppler shift wasn't accounted for properly. ----------------------------------------------------------------------- rk Micromanaging leads to macro f'ups. stellar engineering, ltd. -- Richard Marcinko, Seal Team Six stellare@erols.com.NOSPAM Hi-Rel Digital Systems Design
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