Site Home Archive Home FAQ Home How to search the Archive How to Navigate the Archive
Compare FPGA features and resources
Threads starting:
Authors:A B C D E F G H I J K L M N O P Q R S T U V W X Y Z
In article <robc.801332006@appliedmicro.ns.ca> robc@appliedmicro.ns.ca "Rob Christopher" writes: >I'm trying to program an Altera/Intel Flex780 through the JTAG port and >most times get an error message from PLDShell's PENGN program. I'm >using version 3.5 on a PC. The error code is 7041 and sometimes 7045 and >my documentation (from version 3.1) only has the codes up to 7037. >Anyone have a list of error codes above 7037? The V5 book that arrived this week still only has codes listed up to 7037. -- Paul Walker 4Links for technical help +44 1908 566253 P O Box 816, Two Mile Ash paul@walker.demon.co.uk Milton Keynes MK8 8NS, UKArticle: 1276
In article <3pukf7$810@rc1.vub.ac.be>, tw38966@vub.ac.be (SH.RYU KIM HOFMANS) wrote: > A few weeks ago I've seen a posting about > 'altera vs xilinx' but i just missed that one. > Could the sender repost that one ? > (I'm not the sender but I saved a copy; I'll email it) Can anyone give us an update on the latest news? Should we expect Altera to withdraw their FLEX from sale or will it just be more expensive to cover Xilinx's royalty? Ian. -- Ian Lazarus Nuclear Physics Support Group, CCL, Daresbury Laboratory email: Lazarus@dl.ac.ukArticle: 1277
In article <3q09gb$ok2@aimnet1.aimnet.com> sjsmith@aimnet.com "Stephen J Smith" writes: >altera does have these type of macrofunctions (PCI, ATM, ...) but >currently they are only available from the bbs. >these macrofunctions will be mirrored to the ftp site in the next few weeks. ^^^^^^^^^^^^^^^^^^^^^ T'would be nice. That's what the Readme on the ftp server said, I think it was November last year. Meanwhile the sales blurbs imply it's already there. Nevertheless, many thanks for the info. -- Paul Walker 4Links for technical help +44 1908 566253 P O Box 816, Two Mile Ash paul@walker.demon.co.uk Milton Keynes MK8 8NS, UKArticle: 1278
In article <TefuvYMQzvkY083yn@io.org> Ryan Raz, morph@io.org writes: >I have been trying to use the Intel download software (PENGN) to >configure these parts without any success. The software is unable >to read the proper ID from these parts. Typically it does this if the TDI/TDO loop is not complete. Check you have not misinterpreted TDI-PORT and TDO-PORT on the cable - these are not intuitive since TDO-PORT is the input to the board and TDO board the output. > >Has anyone tried using the Intel cable and software with both >Intel and Altera parts? Yes > >_____________________________________________________________ Dr John Forrest Tel: +44-161-200-3315 Dept of Computation Fax: +44-161-200-3321 UMIST E-mail: jf@ap.co.umist.ac.uk MANCHESTER M60 1QD UKArticle: 1279
In article <3pr73f$p1m@spock.asic.sc.ti.com> clyvb@asic.sc.ti.com writes: > Does anyone know of a web site , or failing that a good book >that can cover the essentials of CUPL. For programming PLD's >Anything on ABEL (ABLE ??) would be good too. I am going to >purchase/order some books this weekend. > >Thanks >Clive B. ABEL (Advanced Boolean Expression Language) ABEL was introduced by Data I/O in 1984 and is one of the most widely used Hardware Description Languages. While Verilog and VHDL were designed for system level design and simulation, ABEL was designed specifically for describing digital circuits for PLDs and FPGAs. A book on Programmable Logic and ABEL is Digital Design Using ABEL David Pellerin and Michael Holley Prentice Hall 1994 ISBN 0-13-605874-4 Dark green hardback cover This book covers HDL-based logic design and provides many examples of how ABEL can be used for digital applications. It has a complete ABEL language reference and includes a limited version of ABEL 5.0 software. Another book on Programmable Logic is Practical Design Using Programmable Logic David Pellerin and Michael Holley Prentice Hall 1991 ISBN 0-13-723834-7 Dark blue hardback cover This book has more information on PLD and FPGA architectures and a chapter on the history of programmable logic. It also covers logic design and has many examples. ------------------------ I have no connection with the above products other then being a co-author of the books and a developer of the ABEL software. Michael Holley Data I/OArticle: 1280
Roland Welte <100070.3321@CompuServe.COM> wrote: > > - What company bought NeoCAD (Xilinx, AT&T) ? Xilinx > - Since NeoCAD was the only design system for FPGAs from > Motorola, what is happening with the support for these > devices, now that NeoCAD is not independent anymore. > - Will NeoCAD eventually disappear? There will be no NeoCAD anymore. I guess erstwhile Neocad at CO will become the focal point of Xilinx Software Development. Recently heard the AT&T has started an FPGA software development facility at Boulder,CO ?? > > Half a year ago, I decided that NeoCAD is the tool of choice > for our company's FPGA designs, mainly because of its > device independent philosophy. But now it looks quite > differently. This is purely a matter of personal opinion. But why would anyone want another company and another choice when there is Xilinx which offers all possible technologies you can think of, SRAM or Antifuse or EPLDs ... with best possible software :-) -BondArticle: 1281
I am looking for orcad 4.04 sdt library with inside sw macros of texas instruments FPGA tpc1010-tpc1460. Have anybody seen it free in internet? Does anybody know if it exist? Many thanks in advance. Marek skotnica@ugn.cas.czArticle: 1282
roger@coelacanth.com (Roger Williams) writes: >The SRAM option seems attractive, but this kind of architecture has >the potential for routing problems... It's also not clear if Altera >regards these parts as orphans--I might end up designing the whole ^^^^^^^ Intel recently stopped making the N85C508-7 programmable address decoder - I believe shortly before selling the product line to ALtera. Does anyone know where I can find some to tide me over until I can redesign my pcb to use an alternative device? While I was talking to Altera about this, they mentioned that they will integrate the Intel parts into their development software later this year. They also said that the same fab is being used to make them as before. John Walliker PS please email me if you know where I can get some 85C508s!!! j.walliker@ucl.ac.ukArticle: 1283
Thanks for posting the simple listing of Altera contacts Stephen! I hope all CPLD/FPGA vendors will take the time to post contact info. What is the chance of storing contact info in a special place in the WWW archive? Eg, somewhere up front, 1 file per vendor. btw - many thanks to whomever maintains the archive. It is most helpful. http://www.super.org:8000/FPGA/caf.html COMP.ARCH.FPGA Archive --- - Bill Wolf, Raleigh NC - My opinions, NOT my employer'sArticle: 1284
In article <D95EHF.n1M@maunakea.Data-IO.COM> holley@Data-IO.COM (Mike Holley) writes: >In article <3pr73f$p1m@spock.asic.sc.ti.com> clyvb@asic.sc.ti.com writes: >> Does anyone know of a web site , or failing that a good book >>that can cover the essentials of CUPL. For programming PLD's >>Anything on ABEL (ABLE ??) would be good too. I am going to You want to watch out which version of ABEL you'll be working with. For example, going from version 3 to version 4 was/is painful, both to figure out the differences and every time you want to migrate an old design, even when you know what to do, it's still tedious. >ABEL (Advanced Boolean Expression Language) >ABEL was introduced by Data I/O in 1984 and is one >of the most widely used Hardware Description Languages. >While Verilog and VHDL were designed for system level >design and simulation, ABEL was designed specifically >for describing digital circuits for PLDs and FPGAs. If only it supported buses well. -- Should the 1st amendment apply to the Internet? Ask Senator Feinstein.Article: 1285
Try compiling this simple little file. ABEL's unable to optimize signals la4..la7 as 1. It does produce a working equation (cond1 & st2); however, this failure to optimize between states can lead to some lengthy equations for a complex state machine. (@DCSTATE/@DCSET doesn't change this) As a result I'd recommend avoiding ABEL's "with" statement. ---cut here--- module JUNK DECLARATIONS reset_n pin; clock pin; cond1 pin; la0 pin istype 'reg'; la1 pin istype 'reg,dc'; la2 pin istype 'reg,pos'; la3 pin istype 'reg,neg'; la4 pin istype 'reg'; la5 pin istype 'reg,dc'; la6 pin istype 'reg,pos'; la7 pin istype 'reg,neg'; STREG state_register; st1, st2 state istype 'dc'; LA = [la0..la7]; EQUATIONS LA.clk = clock; LA.oe = ^hFF; STREG.clk = clock; state_diagram STREG state st1: goto st2; state st2: if ( cond1 ) then st2 with { LA := ^h0F; } else st1; end JUNK -- Michael G. Reeves email: miker@megatek.com Megatek Corporation voice: (619) 675-4300 x2663 16868 Via Del Campo Court facsimile: (619) 675-4341 San Diego, CA 92127 -- Michael G. Reeves email: miker@megatek.com Megatek Corporation voice: (619) 675-4300 x2663 16868 Via Del Campo Court facsimile: (619) 675-4341 San Diego, CA 92127Article: 1286
pngai@mv.us.adobe.com (Phil Ngai) write:- Stuff about ABEL deleted >If only it supported buses well. Funny you should say that. MINCs PLDesigner supports busses, but it is very messy. T.H.Article: 1287
I am using the OrCAD HDL compilers with Lattice pLSI devices (1016, 2032, 1024, 1032). We bought it instead of ABEL/CUPL/MINC in order to evaluate yet another HDL and compiler. Frankly, we expected to find glitches that will make us go out and buy ABEL or CUPL. We were presently surprised by the compiler, after we got over the initial xenophobia of the new language structure and constructs. It is a nifty compiler and even though it does not support ESPRESSO or PRESTO reductions, the MACBOOLE reductions it comes with have been fairly adequate so far. I shall like to hear from others using OHDL with Lattice or other FPGAs. Som Sikdar SHOMITI SYSTEMS ssikdar@best.comArticle: 1288
William J. Wolf writes: > Thanks for posting the simple listing of Altera contacts Stephen! > > I hope all CPLD/FPGA vendors will take the time to post contact info. > > What is the chance of storing contact info in a special place in the > WWW archive? Eg, somewhere up front, 1 file per vendor. btw - many > thanks to whomever maintains the archive. It is most helpful. Thanks! I have started to collect a list of vendors' pages, and will probably add a separate page of links to the various vendors. But I think specific contact info should be provided (and maintained) by the vendors themselves, rather than by us. Please send me any links you feel should be added to the archive page. -jeff ------ Jeffrey Arnold IDA Center for Computing Sciences (formerly the Supercomputing Research Center) 17100 Science Dr. Bowie, MD 20715 email: jma@super.orgArticle: 1289
Hello Members in this news group, I'm very pleased to post my first message to news group. By the way, is there anybody who knows a company which provides FPGA-to-ASIC conversion services ? That is, I have some control logic designed by several FPGA(5-Altera EPM7064QC100). But I have to convert these chips to one-chip of ASIC, for example, LSI, Toshiba, etc. I heard some companys in U.S could provide such kind of services, but I'm not sure. Please answer me about above question. Thank you for reading. J.S. Baik -- __|__ E-mail : backfire@saturn.sst.co.kr ------oo(_)oo------ Kids : backfireArticle: 1290
Hello Members in this news group, I'm very pleased to post my first message to news group. By the way, is there anybody who knows a company which provides FPGA-to-ASIC conversion services ? That is, I have some control logic designed by several EPLD(5-Altera EPM7064QC100). But I have to convert these chips to one-chip of ASIC, for example, LSI, Toshiba, etc. The source files are written in AHDL(*.tdf - Altera HDL), so I could not retarget this design into other ASIC tech- nology in Synopsys Design Compiler. I heard some companys in U.S could provide such kind of services, but I'm not sure. If you know their name, please let me know. Thank you for reading. J.S. Baik -- __|__ E-mail : backfire@saturn.sst.co.kr ------oo(_)oo------ Kids : backfireArticle: 1291
Does anyone have any recommendations or experiences of functionally debugging pcb's containing Xilinx extra fine pitch (XFP) parts. Due to the physical and functional constraints placed on our latest design, we chose to use XFP parts. We performed extensive simulations prior to manufacture to remove the majority of functional errors and used a variety of probes, wires and clips when the board was returned to complete the debug. The main problem we had was with the test clips which we found to be expensive, have a very short life and could only be used reliably when the board was flat on the bench. As we have a double sided board, this proved a bit of a pain. Has anyone seen similar problems with the XFP clips? If so, how did you get round the problems? If not, what clip manufacturers would you recommend? What other experiences did you have debugging with XFP parts? What recommendations do you have if you have to use these parts? Any input greatly appreciated. Jim Banks. -- *************** __ ************************************************** ************* / / ************ Jim Banks * *********** / / ********* HP Queensferry Telecom Operation * ********* / /___ ______ ******* email: jbanks@hpsqf.sqf.hp.com * ******** / __ // __ / ****** Mail: Station Road * ******** / / / // /_/ / ****** South Queensferry * ********* /_/ /_// ____/ ******* West Lothian, EH30 9TG * ********** / / ********* Scotland * ************ / / *********** Telnet: 313-2949 (+44-131-331-7949)* ************* /_/ ****************************************************Article: 1292
In article <3qbot9$oge@noc.sait.samsung.co.kr> backfire@saturn.sst.co.kr (baik jong sung <4224>) writes: > By the way, is there anybody who knows a company which > provides FPGA-to-ASIC conversion services ? > That is, I have some control logic designed by several > EPLD(5-Altera EPM7064QC100). But I have to convert these > chips to one-chip of ASIC, for example, LSI, Toshiba, etc. Orbit Semiconductor, ATS (Advanced Technical Solutions) & AMI all do turnkey FPGA conversions. I would guess all three could convert the chips you mention into a single ASIC. In the case of the first two, you need to supply full production test vectors. In the case of AMI, you can opt to just give them functional test vectors and they will (for a fee) insert scan elements and generate production test vectors themselves. AMI will also (for a fee) automatically add full JTAG boundary scan. -Jeff.Article: 1293
Hi, We are performing research into memory architectures on FPGAs. More specifically, we are looking at how large (several Kbit) configurable SRAM arrays can be included on an FPGA along with logic resources. To evaluate some of our ideas, we need benchmark circuits. Unfortunately, all the "standard" benchmark suites (as far as we know) contain circuits without memory. If anyone has circuit descriptions (with memory) that they are willing to part with, we would be grateful. We are looking at circuits with less than (about) 30,000 gates of logic, and less than (about) 64Kbits of memory (significantly smaller circuits are OK too). The netlists can be in any format (as long as we can understand it). If you can help us out, please send email to: wilton@eecg.toronto.edu Thanks!Article: 1294
Dear Srs, We are using an FPGA 3090 from XILINX and we stated some problems with programming it. Could someone please help us with useful information ? Next we describe our design: - The board we use consists on a transputer T425, memory and LCA3090 devices. - The transputer programs the LCA in Peripheral Mode. The schematic used during configuration is the following: i) notCS1=GND, CS2=Vcc, notCS0 is activated on each byte we send to the LCA; ii) notWR (from T425) is connected to notWS (LCA) and is activated on each byte we send to the LCA; iii) INIT (LCA) - is not used; iv) RDY/notBUSY (LCA) is read by T425, via bit 0 of Data Bus, using a Tri-state buffer; v) DONE/notPROGRAM (LCA) is activated by T425 via an Open collector buffer. - The transputer sends each byte of the configuration file (.BIT) to the LCA and waits by the LcaReady signal (which arise successfully). - At the end of the configuration file, LCA device should deactivate the DONE/notPROGRAM signal, which stays at LOW LEVEL. Why the I/O pins of the LCA do not become active, when the signal RDY/notBUSY seems to behave correctly on each configuration writing cycle? Is normal the DONE/notPROGRAM signal to be at LOW level during all the configuration process ? - The conditions used in XACT to generate the configuration file (.BIT) for a simple system, were: TLL inputs, Pullup resistor in DONE/notPROGRAM, no Readback, XtalOsc disabled, DONE/notPROGRAM and RESET signals activated 1 cicle after the end of the configuration. We didn't use Tie of unused logic, and the logic outside the LCA is TTL. Thanks a lot in advance. Best regards. -- +-----------------------+-----------------------------------------------------+ | Antonio J. A. Esteves | | | Dep. Informatica | esteves@di.uminho.pt | | Universidade do Minho | Tel: +351 53 604479 | | Largo do Pac,o | Fax: +351 53 612954 | | 4719 Braga Codex | PGP: finger -l esteves@shiva.di.uminho.pt | | Portugal | WWW: http://www.di.uminho.pt/~esteves/index.html | +-----------------------+-----------------------------------------------------+Article: 1295
Hi all Does anyone have any info on a low cost programmer that I can use on Flex780 devices. I've heard of a product called the PROTAG programmer made by a company called Products in Motion but the number I have for them has been disconnected (probably a bad sign). I'm interested in hearing from anyone with a lead for me. Thanks Rob Christopher robc@appliedmicro.ns.ca Hardware Designer Applied Microelectronics, Inc. phone: (902)421-1250 1046 Barrington St. fax: (902)429-9983 Halifax, N.S. CanadaArticle: 1296
In article <3qctns$2b2@icaro.uminho.pt> esteves@di.uminho.pt writes: > >Dear Srs, and Madams, > >We are using an FPGA 3090 from XILINX and we stated some problems with >programming it. >Could someone please help us with useful information ? > >Next we describe our design: > >- The board we use consists on a transputer T425, memory and LCA3090 devices. > >- The transputer programs the LCA in Peripheral Mode. The schematic used > during configuration is the following: > > i) notCS1=GND, CS2=Vcc, notCS0 is activated on each byte we send to the LCA; > ii) notWR (from T425) is connected to notWS (LCA) and is activated on each > byte we send to the LCA; > iii) INIT (LCA) - is not used; > iv) RDY/notBUSY (LCA) is read by T425, via bit 0 of Data Bus, using > a Tri-state buffer; > v) DONE/notPROGRAM (LCA) is activated by T425 via an Open collector buffer. > >- The transputer sends each byte of the configuration file (.BIT) to the LCA and >waits by the LcaReady signal (which arise successfully). The .BIT file has a header that you DO NOT want to send to the chip. About 64 to 70 bytes into the file you will find the sequence 'FF' '20'. the 'FF' is where you need to start. Personally, I prefer using the .RBT file and reformatting it to suit the application. That way I know exactly what the data looks like. use 'makebits -b design.lca' to create design.rbt If you look at page 2-120 of the 1994 data book, the 8 bits labeled dummybits are the 'FF', and the '0010' is the '2' of the '20'. the '0' of the '20' is the first 4 bits of the 24 bit length count. > >- At the end of the configuration file, LCA device should deactivate the >DONE/notPROGRAM signal, which stays at LOW LEVEL. Do you have a Pull up resistor on this signal? if you dont, then it can't go high. > Why the I/O pins of the LCA do not become active, when the signal RDY/notBUSY >seems to behave correctly on each configuration writing cycle? The I/O pins will only go active if the chip is done. > Is normal the DONE/notPROGRAM signal to be at LOW level during all the >configuration process ? Yes, and so is LDC. HDC is high during configuration. > >- The conditions used in XACT to generate the configuration file (.BIT) for a > simple system, were: > TLL inputs, Pullup resistor in DONE/notPROGRAM, no Readback, You still need an external pull-up resistor, I use 4.7k ohms. You also need one on INIT. > XtalOsc disabled, DONE/notPROGRAM and RESET signals activated 1 cicle after > the end of the configuration. We didn't use Tie of unused logic, and > the logic outside the LCA is TTL. > After the end of your data, you should probably send one extra byte of data, and it should be an 'FF'. this is sometimes needed for the startup sequence to complete, as the sequence takes 3 clocks, and if your data happens to have its last byte end within this sequence, then the chip wont go done. The extra byte will always supply sufficient clocks to do this, and the FF code just looks like postamble (see 2-120 again) Hope this gets you going, All the best Philip Freidin. Fliptron@netcom.com >Thanks a lot in advance. >Best regards. >-- >+-----------------------+-----------------------------------------------------+ >| Antonio J. A. Esteves | | >| Dep. Informatica | esteves@di.uminho.pt | >| Universidade do Minho | Tel: +351 53 604479 | >| Largo do Pac,o | Fax: +351 53 612954 | >| 4719 Braga Codex | PGP: finger -l esteves@shiva.di.uminho.pt | >| Portugal | WWW: http://www.di.uminho.pt/~esteves/index.html | >+-----------------------+-----------------------------------------------------+ > > > >Article: 1297
Hi, we are doing some FPGA/ASIC design in VHDL. To see the output of the design it would be a nice thing to get the EDIF netlist displayed.Does anyone know about a EDIF display program, or a converter for ORCAD? Any help would be appreciated. Horst Eyermann -- ------------------------------------------------------------------- | Evolution Electronics Ltd EMail horste@evolution.co.uk | -------------------------------------------------------------------Article: 1298
I might be missing something here, since I don't have immediate access to ABEL right now, but I'll comment anyway (in true Internet tradition). I would not expect the reduced equation to be LA<4:7> = 1 but rather LA<4:7> = (STREG==st2) & cond1. Is this what ABEL did? A big difference between ABEL and Verilog is that ABEL implies "else 0" in state machines while Verilog implies "else keep the previous state." miker@megatek.com (Mike Reeves) writes: > > Try compiling this simple little file. ABEL's unable to optimize > signals la4..la7 as 1. It does produce a working equation (cond1 & st2); > however, this failure to optimize between states can lead to some lengthy > equations for a complex state machine. (@DCSTATE/@DCSET doesn't change this) > As a result I'd recommend avoiding ABEL's "with" statement. > > ---cut here--- > module JUNK > > DECLARATIONS > > reset_n pin; > clock pin; > cond1 pin; > > la0 pin istype 'reg'; > la1 pin istype 'reg,dc'; > la2 pin istype 'reg,pos'; > la3 pin istype 'reg,neg'; > la4 pin istype 'reg'; > la5 pin istype 'reg,dc'; > la6 pin istype 'reg,pos'; > la7 pin istype 'reg,neg'; > > STREG state_register; > st1, > st2 state istype 'dc'; > > > LA = [la0..la7]; > > EQUATIONS > > LA.clk = clock; > LA.oe = ^hFF; > > STREG.clk = clock; > > state_diagram STREG > > state st1: goto st2; > > state st2: if ( cond1 ) then > st2 with { > LA := ^h0F; } > else > st1; > > end JUNKArticle: 1299
Hi you FPGA designers!!! I am designing a data communication add-in card for the PCI bus. We have ALTERA Max+Plus II software at your company so I have decided to use FLEX 8K device(s) for the design. The add-in card has to operate in a master and target configuration. However I have experienced that to drive the PCI interface signals as master you need to use 6 I/O cell Output Enable signals to keep witin the input-setup time and clock-to-output time requirements in the spec. Unfortunately the Flex 8K devices only have 4 (except EPF81500, it has 10). What is the solution to this problem? Multiple devices can be a solution where the master driver logic with my back-end device specific logic could go in one device and PCI synchronous inputs and target logic in another. But what about load problems??? Is it a better idea to use PCI compliant MAX devices. Please note folks that I have only worked with the FLEX devices and unfortunately I am new in digital design. Please give me some ideas or solutions (if any) ASAP. My time schedule is very tight. ---------------------------------------------------------------------------- ----------------------- Finn M. Johansen, M.Sc.E.E. Hardware Design Engineer, R&D ------------------------------ PURUP PREPRESS A/S 5 Soenderskovvej DK-8520 Lystrup Denmark Phone: +45 87 434 343 Fax: +45 87 434 445 E-mail: fmj@pe.dk ------------------------------ ---------------------------------------------------------------------------- ---------------------
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