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
Ray Andraka <ray@andraka.com> writes: > ... an all digital analog ... I like it :-) -- martin.j.thompson@trw.com TRW Conekt, Solihull, UK http://www.trw.com/conektArticle: 47626
So is there a way at all to clock this process without having to use the bufg? And is it possible to short the two pins together? (current non-global to a global one) "Dali" <dadicool@ifrance.com> wrote in message news:3D98E4DF.3010100@ifrance.com... > It doesn't work that way. In order to take advantage of the clock tree > brovided by the BUFG primitive you need to use an IBUFG instead of an > IBUF. So a normal IO pins does not do the job. > > For further details, look at the Xilinx Handbook. > > Dali > > Ho Wong wrote: > > Hello. I'm using a pin to drive a process but the thing is that it's already > > been soldered onto a normal IO pin. I've been trawling through past posts > > and newsgroups and I still haven't found a answer yet. I'm just a beginner > > to fpgas so i'm not very familiar with the low level logics. I tried doing > > PIN -> IBUF -> BUFG -> signal, but it told me the input pad net of PIN has > > illegal connection Would it be possible to short my normal IO to one of the > > dedicated I,GCK ? I'm using a Spartan 2 programming in VHDL with the ISE > > Webpack. > > > > > >Article: 47627
Hi folks I'm currently routing a design into a QuickRAM using SPDE. To set the pin locations, I use a .sc file and define the pin location of an input from my design but when SPDE is run, the pin placement is ignored. The error message that comes up is " Invalid fixed placement ioH1 for gate STM_IN_p ignored ". In the .sc file, I just written " postprop stm_in ql_placement = "ioH1"; " as shown in the SPDE manual. I use synplify to set up the pin location and it seem to have taken in the information by analysing the .qdf file that was generated. In the .qdf file, one of the lines is written as " gate STM_IN_p master Q_INPAD cell ioH1 end " which kind of suggest the constraint files was correct and that SPDE has taken in the information. Only when SPDE is run, the file seem to be ignored. The device I'm using is QuickRAM 4090 PB456. I have also changed the location to H23 which is not used when running the SPDE without the pin constraints. How can I resolve the problem? Regards PaulArticle: 47628
Hi Cisa, I have heard more than once that is not recommended to use the STARTUP block in Virtex devices. The main reason was that is very slow propagating through the device, if you are using a 10KHz clock you won't miss any clock pulses but with a 50MHz clock you will. The good thing is that it uses dedicated resources (that is why we can't see it in FPGA Editor) and therefore you get it for nothing. If I had to use an external reset I wouldn't connect it to the GSR, just writing the VHDL/Verilog properly will make the synthesis/P&R tool :o) to allocate low skew routing resources for it (this is just my opinion and it depends for aeach design) Regards. Ulises Hernandez ECS Technology Ltd. www.ecs-tech.com "Cisa" <jjyy@hotmail.com> wrote in message news:ee794a5.-1@WebX.sUN8CHnE... > I instantiate a STARTUP_VIRTEX template in my verilog design,and use the input signal GSR to reset my FFs.When I view my routed design in FPGA Editor of ISE,I can see the input signals of STARTUP,but I can't see the output signal of STARTUP to reset my FFs. > Is it a right implementation result? > Can you give me a detailed explanation of architecture about STARTUP module in virtex?And what is the right method to instantiate a STARTUP_VIRTEX in unimsims library?Article: 47629
Ray came up with some excellent advice as usual (and his company could probably do your FPGA design far quicker than others as they have the cores, knowledge etc.), however with the speeds you are talking about are you sure that a DSP-based software solution wouldn't be better, less complex to design and more than adequate for the data rates you mention. Rays designs at many 10s of megahertz definitely need FPGA but for sonabuoy processing I found that DSP was much more efficient and easily adaptable. Just an opinion. PaulArticle: 47630
> -----Original Message----- > From: Paul [mailto:paul.lee@sli-institute.ac.uk] > Subject: SPDE problems > I'm currently routing a design into a QuickRAM using SPDE. > To set the pin locations, I use a .sc file and define the pin location > of an input from my design but when SPDE is run, the pin placement is > ignored. > The error message that comes up is " Invalid fixed placement ioH1 for > gate STM_IN_p ignored ". > In the .sc file, I just written " postprop stm_in ql_placement =3D > "ioH1"; " as shown in the SPDE manual. [...] > How can I resolve the problem? I don't have the data to hand, but it must be either (a) a bug in the P&R tool related to that package (very unlikely, but always possible) (b) you've chosen a pin that isn't appropriate for the input port (for example, maybe you picked a power pin). To get closer I suggest the following sanity check: - remove your pin constraint from this signal - run P&R and see what pin SpDE selects automatically - if this fails, there's something goofy in your design - set up a new constraint, specifying exactly the same pin that SpDE chose automatically in the previous run - run the job again - it should work OK If the second run fails, it's pretty sure you have a buggy tool and the nice guys at QuickLogic probably want to know about it. If my suggested procedure works, though, it's likely your choice of pin H1 for this signal is inappropriate. Possible reasons: - paper data sheet is wrong and misled you into choosing an inappropriate pin (this has happened to me before now) - take a look at the pad schematic in the chip viewer, to see the definitive version of what it really is - you chose the right pad for what you THINK the signal is doing, but your HDL code is unexpectedly specifying a driver on what you think is an input pad - or something like that - finger trouble/snafu - doubtless some more possibilities I haven't thought of Good luck -- Jonathan Bromley HDL Consultant DOULOS - Developing Design Know-how VHDL * Verilog * SystemC * Perl * Tcl/Tk * Verification * Project = Services Doulos Ltd. Church Hatch, 22 Market Place, Ringwood, Hampshire, BH24 = 1AW, UK Tel: +44 (0)1425 471223 mail: = jonathan.bromley@doulos.com Fax: +44 (0)1425 471573 Web: = http://www.doulos.com This e-mail and any attachments are confidential and Doulos Ltd. = reserves all rights of privilege in respect thereof. It is intended for the use = of the addressee only. If you are not the intended recipient please delete = it from your system, any use, disclosure, or copying of this document = is unauthorised. The contents of this message may contain personal views = which are not the views of Doulos Ltd., unless specifically stated.Article: 47632
hay iam not looking for such a solution TCP/IP in hardware is it feasible "Janusz Raniszewski" <rniski@man.koszalin.pl> wrote in message news:3D994313.F5AB54AD@man.koszalin.pl... > > What are the chances of implementing a TCP/IP stack using VHDL anybody do > > anything similar.Can the Spartan FPGA be used to hold the design > > Hello, > It is simple. You may implement some microcontroller core and ,,,,,, :-) > JanuszR >Article: 47633
Hi Karl, If your wav files were written to the SM card from a PC SmartMedia reader/writer, then your FPGA will also need to handle the file-system aspects, in addition to being able to read the raw sectors of the SM card. I did a google search, as you probably also did, and it didn't turn up a system example. But something that I came across recently may be a good starting point. June, July, August and September 2002 issues of Circuit Cellar Magazine presented articles, written by Jeff Bachiochi, on using SmartMedia in an embedded system. In this case the PIC18F452 is used. The articles deal with details of SmartMedia and a Windows/DOS-Compatible files system, which enables exchanging embedded data with PC applications. All of these articles are available in electronic form for purchase from the Circuit Cellar Website. They are US$1.50 each. http://www.circuitcellar.com/magazine/tocarchive.asp One possible way to proceed with your implementation is to put a micro core in your FPGA, which could handle the SmartMedia interface. 15K gates is enough to instantiate a micro. Or simply design a state machine to handle the interface. Of course, if you're not locked into SmartMedia, using Compact Flash cards may be an alternative for you. BurchED has an FPGA to Compact Flash plug-on module, for this kind of application http://www.burched.biz/b5compactflash.html Hope that helps:) Best regards Tony Burch http://www.BurchED.biz FPGA boards for System-On-Chip prototyping and education * * 10% off the normal price of all products, sale now on * * "Karl" <Far@East.Design> wrote in message news:3d994da5$1@news.starhub.net.sg... > Anyone knows of a example FPGA design which reads and writes a SmartMedia > card? > What are the stuff I need to read on the net, my design is simple, I want to > read the .wav > files from a SM card and feed this data into a speaker. My board will have a > 15K gate > vertex chip and a PWM digital to analog converter and amplifier, plus a > fixture which > connect up the SM card (stripped from a broken SM reader).Article: 47634
Tri simulating with Modelsim. In a simple 32 bit configuration (512 registers) it takes around 75 clk cycles to switch. Matjaz "Peter Sommerfeld" <peter@vtecna.com> wrote in message news:amvoa0$566$1@avvidasystems.com... > Does anyone have the ISR switch and release times for Nios? Ideally for Nios > 2.1? > > -- Pete > >Article: 47635
Indeed,the reason I use STARTUP in my verilog desing is that I can't get my virtex-e device to work after configuration.As for the phenomenon,the DONE pin and INIT pin of the device both goes high. And I suspect that the STARTUP sequence after the configuration is not completed.So I add a STRATUP module in my verilog design. Who can give me any advice about the configuration process and the corresponding code desing? Another question,What is the function of clk declared in STARTUP_VIRTEX module?3x!!!Article: 47636
Thank you Tony. I am a electronics hobbyist. I visited your website, it's pretty cool though. But I wonder where you have found out a good PCB software? I used CadStar from Zuken, but it didn't seem very user-friendly. "Tony Burch" <tony@burched.com.au> wrote in message news:3d9973a3$0$30861$afc38c87@news.optusnet.com.au... > Hi Karl, > > If your wav files were written to the SM card from > a PC SmartMedia reader/writer, then your FPGA > will also need to handle the file-system aspects, > in addition to being able to read the raw sectors of > the SM card. > > I did a google search, as you probably also did, > and it didn't turn up a system example. But something > that I came across recently may be a good starting > point. June, July, August and September 2002 issues > of Circuit Cellar Magazine presented articles, written > by Jeff Bachiochi, on using SmartMedia in an embedded > system. In this case the PIC18F452 is used. > > The articles deal with details of SmartMedia > and a Windows/DOS-Compatible > files system, which enables exchanging embedded data > with PC applications. All of these articles are available > in electronic form for purchase from the Circuit Cellar > Website. They are US$1.50 each. > http://www.circuitcellar.com/magazine/tocarchive.asp > > One possible way to proceed with your implementation > is to put a micro core in your FPGA, which could > handle the SmartMedia interface. 15K gates is > enough to instantiate a micro. Or simply design > a state machine to handle the interface. > > Of course, if you're not locked into SmartMedia, > using Compact Flash cards may be an > alternative for you. BurchED has an FPGA to Compact > Flash plug-on module, for this kind of application > http://www.burched.biz/b5compactflash.html > > Hope that helps:) > > Best regards > Tony Burch > http://www.BurchED.biz > FPGA boards for System-On-Chip prototyping and education > * * 10% off the normal price of all products, sale now on * * > > "Karl" <Far@East.Design> wrote in message > news:3d994da5$1@news.starhub.net.sg... > > Anyone knows of a example FPGA design which reads and writes a SmartMedia > > card? > > What are the stuff I need to read on the net, my design is simple, I want > to > > read the .wav > > files from a SM card and feed this data into a speaker. My board will have > a > > 15K gate > > vertex chip and a PWM digital to analog converter and amplifier, plus a > > fixture which > > connect up the SM card (stripped from a broken SM reader). > > >Article: 47637
"Blackie Beard" <BlackBeard@FearlessImmortalWretch.com> wrote in message news:<YHRl9.7422$1F3.6383@nwrddc02.gnilink.net>... > For your own board manufacture, there was some Iron-on > stuff that only works for traces over 20 mil., or so. Then > there's them cool machines that use a router, my last > company had that. But you can't do fine traces with that > either. Better to just bite the bullet. There are prototype > shops you can just email your gerbers to, 2-3 day turn, > low quantity. > > For MIPS core, you aren't going to get that for free, me Oh yeah ?! Well, think again, and then kindly follow this link: www.OpenCores.org :^P You shall find TWO (!) MIPS cores (Plasma & Yellow Star) and many, many other FREE IP Cores ! Cheers, rudi ---------------------------------------------- www.asics.ws - Solutions for your ASIC needs - > thinks! But you can sometimes buy a test chip for around > $10-20, and build the SoC interconnects using an FPGA. > You'll probably need at least 400K equiv gates. That > won't be cheap either. Good luck, and reply if you find > something interesting. > > > "RCU" <nemesis@icequake.no_spam.net> wrote in message > news:pan.2002.09.30.02.18.12.141498@icequake.no_spam.net... > > > > > For Layout, I found that I liked the Eagle for Windows free > > > layout program. You can make boards up to a certain size, > > > which is OK for a lot of stuff. Something like 3x4 inches? > > > > What about the actual physical manufacturing of a board? I see lots of > > homebrew-available kits (Press N Peel, et al), and a few homebrewn > > methods: > > http://www.netcomuk.co.uk/~wwl/pcbs.html > > > > for example. > > > > > For FPGA, it depends on size. What is your first design > > > going to be? > > > > I have no idea. :) I thought it would be fun to implement a MIPS core or > > something similar. If that turns out to be too complex for the bits I can > > afford, then I might play with implementing old hardware anew, such as a > > portable Super NES, or something along those lines. > > > > I don't think anything I would want to do is anywhere near, say, a CPU > > produced even in the last 5 years. I just want flexibility for > > experimentation. If that makes any sense. > > > > Thanks! > >Article: 47639
"news.terra.es" <securez@teleline.es> wrote in message news:<QRVl9.310521$sI1.2734123@telenews.teleline.es>... > I have a MC68VZ328 system interfacet to a Acex 1K30 device, i have a doubt, > the processor are a 32bit but the system bus are 16bit, i want to include > the a core int the FPGA for ide interface, but the whisbone slave core for > ide have 32bit registers, and SEL_i need to be 1111 32bit data, it's posible > to cahnge the master controler to drive this 32 bit data from he 16 bit bus > of the system¿? > > Any help will be apreciated. I assume you are using the OCIDEC from www.opencores.org. You can easily modify the 32 bit wishbone interface to be 16 bits only. It's such a simple core should be very easy. Cheers, rudi ---------------------------------------------- www.asics.ws - Solutions for your ASIC needs -Article: 47640
I agree with Paul in principle. As I've stated before, if a process can be handled by a single DSP we usually recommend using that over an FPGA simply because the design talent is cheaper and easier to find. Paul Baxter wrote: > Ray came up with some excellent advice as usual (and his company could > probably do your FPGA design far quicker than others as they have the cores, > knowledge etc.), however with the speeds you are talking about are you sure > that a DSP-based software solution wouldn't be better, less complex to > design and more than adequate for the data rates you mention. > > Rays designs at many 10s of megahertz definitely need FPGA but for sonabuoy > processing I found that DSP was much more efficient and easily adaptable. > > Just an opinion. > > Paul -- --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 "They that give up essential liberty to obtain a little temporary safety deserve neither liberty nor safety." -Benjamin Franklin, 1759Article: 47641
No, it is not. Are you giving it enough cclks after programming completes? If you are using JTAG, make sure you set the configuration to use the JTAG clock or it will do this. Cisa wrote: > I have configured a virtex-e device, > and DONE goes high,INIT also goes high .That indicates that configuration has finished,but the device can't start to work. > I suspect the virtex-e does not finish the STARTUP sequence. > Is it necessary to instantiate a STARTUP_VIRTEX module in my verilog design for STARTUP sequence? -- --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 "They that give up essential liberty to obtain a little temporary safety deserve neither liberty nor safety." -Benjamin Franklin, 1759Article: 47642
X-Post without folowup isn't the best choice to spread your lack of wisdom :) X-Post to caf und clv without followup "Djohn" <deepucjohn@yahoo.com> wrote: > I am now designing a Uart chip for study purposes. Now I am almost over > with my design and VHDL Coding . I would like to integrate the DFT strategy > in the chip . Interesting excercise. But I have problem with _the_ DFT strategy. > 2) I have a fair good knowledge of JTAG . So If I integrate a JTAG circuitry > in the chip (the tap controller , data registers and instruction > registers), will it become DFT compatible? Answer your self: will the design be testable with this cirquit? If you just forgot to mention boundary scan, it might be the first step to DFT. DFT=Design For Testability, not design to fake test. > 3) What exactly is BIST ? How is it implemented ? Any online tutorials or > good links ... BIST=Build in Self Test, any logic on chip, that tests your cirquit without external hardware is BIST. > 4) What is ATPG ? how is it implemented? Any online tutorials ....... Automatic Test Pattern Gerneration. Typical a Programm that generates testpattern for your cirquit. > 6) A comparison between these strategies? any links also welcome www.google.com is your best friend. Buzzwords: - atpg - bist - dft - marchtest (for RAM) - boundary scan - scanchains - scantest - LFSR For ATPG you could start at the homepage of synopsys or cadence I think you should start with some books concerning DFT, because it might be too complex reading all the papers concerning these themes without knowing where to start and where to go along. If you have more detailed questions, you might get more answers here. bye ThomasArticle: 47645
Paul, I am not surprised to here your response. I wondered myself about whether the FPGA was overkill. The reason for the FPGA is that the particular student is realy interested in dsp in FPGA. His area of interest is not signal processing algorythms but rather hardware based DSP. He is, therefore, more interested in the FPGA based approach. As you can imagine, working in an academic environment is a little different than working in industry. I suspect that this system is somewhere between radar and sonobouys. By the way, who do you do sonobouys for? In the late 1970's to early 1980's, I used to work for Spartan Corporation (in Jackson, MI) doing bouys and other similar things. Thanks, Theron Hicks "Paul Baxter" <pauljnospambaxter@hotnospammail.com> wrote in message news:3d996a27$0$8510$cc9e4d1f@news.dial.pipex.com... > > Ray came up with some excellent advice as usual (and his company could > probably do your FPGA design far quicker than others as they have the cores, > knowledge etc.), however with the speeds you are talking about are you sure > that a DSP-based software solution wouldn't be better, less complex to > design and more than adequate for the data rates you mention. > > Rays designs at many 10s of megahertz definitely need FPGA but for sonabuoy > processing I found that DSP was much more efficient and easily adaptable. > > Just an opinion. > > Paul > > > >Article: 47646
"Blackie Beard" <BlackBeard@FearlessImmortalWretch.com> wrote: >One of the stupidest things about Altera Quartus S/W (at least the >first versions, I don't know about later ones) was that hidden about >4 submenus down, was the setting for what to do with unused pins, >the default choice; "As Outputs Driving Ground". What if you have >the board made first, with the I/O in place, and want to add the >various interface blocks of the design in incrementally? You can >get some real hot chips. You must know about this setting, and >find it and change it to "As Tri-State Inputs", and remember to do >so with any design revisions, or your prototype is toast. I tried to >tell Altera, but to no avail. > The problem with "As Tri-State Inputs" is that this makes every unused pin a floating input and subject to noise injection and excessive poser consumption unless you connect them on the board to a valid signal level. A better solution in my experience is to add at the beginning all of the pin definitions even if that module is not used (after all you know what pin they are going to since the board is built.) -- richard_damon@iname.com (Redirector to my current best Mailbox) rdamon@beltronicsInspection.com (Work Adddress) Richad_Damon@msn.com (Just for Fun)Article: 47647
Hi all, After installing ISE WebPACK 5.1 I can configure devices only when I run the iMPACT as an administrator in win2k. I'm using parallel cable and I haven't found parallel port driver as in all previous versions of WebPACK. Is it possible to configure devices from iMPACT using parallel cable when I run it as normal - non privileged - user under windows 2000? Thanks a lot, Jan _________________________ j.pech _AT_ ieee.org http://fpga.f2g.netArticle: 47648
[suggest software] > iam not looking for such a solution TCP/IP in hardware is it feasible Look at the source code for a TCP stack. It's a huge state machine. Just take a page or two of code and try to convert that to hardware. I haven't seen any reports of doing TCP in hardware, but I could easily have missed something. (But somebody would probably mention it here.) As an example, consider just routing, a small part of the TCP/IP stack. High end routers do some of the table lookup in hardware, but they normally drop back to software for the hard/rare cases. -- The suespammers.org mail server is located in California. So are all my other mailboxes. Please do not send unsolicited bulk e-mail or unsolicited commercial e-mail to my suespammers.org address or any of my other addresses. These are my opinions, not necessarily my employer's. I hate spam.Article: 47649
Where can i buy xilinx fpga online? Tanks Henrique.
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