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
"VhdlCohen" <vhdlcohen@aol.com> wrote in message news:20010723114205.22055.00000763@ng-fx1.aol.com... > What kind of Flip-flop would yield a 'Z' onto its output? > What you need to do is use a regular FF that yields a '0' or '1' as output. > Follow that with a concurrent signal assignmet to infer a tri-state: > D_Output <= Qvector_FF when tri_condition = '0' else > (others => 'Z'); I've written code before which synchronously assigned 'Z' to a flip-flop output. Synplify was smart enough to figure out to create another flip-flop whose output goes to the tri-state control. Sounds like the synthesis tool isn't being as intelligent as it could be. I agree with your suggestion to try a different coding style, though. Cheers, JamieArticle: 33326
Falk, Recently we had the opportunity to get a specific customer design, do the simulation, and calculate the power from the node analysis using Xpower., and compare with their actual power in their system. Since the simulation uses WORST CASE numbers ( and the transition file), the actual power in the customer's system was less than that predicted. The conculsion was "this lousy tool!!??!!..." but that is the wrong conclusion! In the worst case, the power might be more, and more by the amount predicted. I seldom get in trouble for having engineers design a power supply bigger than the actual load, and a heat removal system more capable than what is required. Good engineering practice uses safety factors of 2 or 3 or 4. Of concern is when each person in the chain applies their X2, leading to a over engineering of the system (X2 X2, X2 ... X8???). Of course, the simulation file (transitions) may be inaccurate, and as the saying goes, "garbage in, garbage out" may be exactly what was happening. I recommend a convenient 0 ohm shunt in line with the power supplies so that current may be conveniently measured in the prototype system in extreme cases. The ultimate solution is the thermal one, as watts in, will result in watts out, as heat. Austin Falk Brunner wrote: > Hello everybody, > > I (and I suppose many other people in this group ;-) have the problem, > that it is really difficult to estimate AND measure the power > consumption of a FPGA design. I know that power consumption strongly > depends on size of the design, clock frequency and, one of the most > difficult to estimate parameters, average toggle rate. OK, we have the > virtex power estimator, XPower and some nice application notes, but > after all its just a very raw guess, then we add 100% security plus a > layout to change the power supply easy on the prototype ;-) > But what is VERY sad is the fact, that it is mostly impossible, to > measure the power consumption on the real, working board :-((( > There inst just one VCC pin, there are many, all connectet to the power > plane. Cutting the power plane and feeding the whole power over a zero > Ohm resistor is no good idea. Socket?? With BGAs?? Signal integrity?? > Damit!!! > > Whats your expirience with this? > Any hints? > > -- > MFG > FalkArticle: 33327
Michael Wichmann schrieb: > > Hey Everyone, > > Does anyone know how to interface from the Altera software package > programming output to jtag. > Altera tries to sell its serial programming cable BitBlaster for around > $500, which IMHO is a little overpriced ;) (since we go from serial data to > ... yep serial data, right??) The is AFAIK a low cost programmer (bitblaster), schematics available on the Altera website. -- MFG FalkArticle: 33328
Back in the day, when I had to do a lot of soldering, I soldered BGAs without a lot of specialized equipment. I would set the BGA down where it was supposed to go, and then superglue a "corral" of SMT resistors around it so it wouldn't be able to move during reflow. Then I would blast it with one of those big red hot air guns that look like hair dryers until the solder melted. There was no subtle technique and no need to pre-heat. "Anthony Ellis" <xxxa.ellis@logicworks.co.za> wrote in message news:3b594878.0@news1.mweb.co.za... > This is not an FPGA question but I guess many of you may have designed > FPGA's using similar technology and work for companies that have some advice > to part with. > > We have a PCB designed for a 255 pin Ceramic BGA and are struggling to > solder this correctly. Plastic BGA's are not an issue however. > Obviously the ceramic absorbs the reflow energy but we haven't been able to > find the correct technique to pre-heat the part. I guess we could measure > the pre-heated component temperature if we had some idea about it's required > value. > > Any help please. > > Thanks Anthony > > > > > > >Article: 33329
"Andrew Gray" <andrewgray@iafrica.com> wrote in message news:<3b5aa232.0@news1.mweb.co.za>... > Hi > > Does anyone know if there is an alternate site to download Baseline 10.1 and > Maxplus II from? > The Altera FTP server does not seem to support resuming, and the files are > too big to download in one go. Get GoZilla from http://www.gozilla.com. It automatically resumes if you have a problem downloading large files. LeonArticle: 33330
>"VhdlCohen" <vhdlcohen@aol.com> wrote in message >news:20010723114205.22055.00000763@ng-fx1.aol.com... >> What kind of Flip-flop would yield a 'Z' onto its output? >> What you need to do is use a regular FF that yields a '0' or '1' as >output. >> Follow that with a concurrent signal assignmet to infer a tri-state: >> D_Output <= Qvector_FF when tri_condition = '0' else >> (others => 'Z'); > >I've written code before which synchronously assigned 'Z' to a flip-flop >output. Synplify was smart enough to figure out to create another flip-flop >whose output goes to the tri-state control. > >Sounds like the synthesis tool isn't being as intelligent as it could be. I >agree with your suggestion to try a different coding style, though. > You're correct about Synplify on automatically inferring a FF for the condition, with the output of the FF condition driving a tri-state buffer. Our IEEE P1076.6 spec is not clear on this point. Thanks for brining it out. I'll bring this point to our VHDL and Verilog synthesis groups. -------------------------------------------------------------------------- ---------------------------------------- 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: 33332
Hey, I have a hotii-xl board from VCC and am trying to get the user interrupt working, ie hardware generate an interrupt using the "userint" signal and a user defined function in software handle the generated interrupt. So far I have had no luck in getting this to work. Does anyone else have any experiences with this? Any examples of working code, or realization of underlying problem(s)? Thanks, KenArticle: 33333
Falk Brunner wrote: > > Ray Andraka schrieb: > > > > Measure the differential power on the board between a configured and > > running FPGA and one that is either unconfigured or loaded with a > > placeholder static design. > > Hmmm, this may be a first attempt. But I am afraid this won't work at all . . . It works very well for comparing alternate designs instances. It can also tell you quickly when something is very wrong. Better to tie up an ammeter than burn your thumb. --Mike TreselerArticle: 33334
> > Does anyone know how to interface from the Altera software package > programming output to jtag. > Altera tries to sell its serial programming cable BitBlaster for around > $500, which IMHO is a little overpriced ;) (since we go from serial data to > ... yep serial data, right??) It's got a microcontroller in it. The ByteBlaster is a lot cheaper if you have a printer port available, and it's easy to make your own. LeonArticle: 33335
Michael Boehnel (boehnel@iti.tu-graz.ac.at) wrote: : I am looking for an online description of the EDN netlist format. Does : anybody know such a site? : Is there free source code for the analysis of EDN-Files available? : Michael Do you mean EDIF?Article: 33336
This is a multi-part message in MIME format. --------------CB1D2D4A90962111D83802A7 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit The Xilinx WebPACK software tools can be found here: http://www.xilinx.com/products/software/webpowered.htm Dave Vanden Bout wrote: > Philipp Krause wrote: > > > Are there any free tools around? I'd like to learn VHDL, but don't want > > to spend money on commercial software since I don't know which chip > > family I'll use when it comes to implementing something. > > > > Philipp Krause > > Download the free WebPACK tools from Xilinx or the free Maxplus2 tools from > Altera. Both contain VHDL and Verilog synthesizers. > > -- > || 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 || --------------CB1D2D4A90962111D83802A7 Content-Type: text/x-vcard; charset=us-ascii; name="jennifer.jenkins.vcf" Content-Transfer-Encoding: 7bit Content-Description: Card for Jennifer Jenkins Content-Disposition: attachment; filename="jennifer.jenkins.vcf" begin:vcard n:Jenkins;Jennifer tel;fax:(505) 858-3106 tel;work:(505) 798-4813 x-mozilla-html:FALSE url:http://www.xilinx.com org:Xilinx CoolRunner CPLDs;<br><img src="http://www.xilinx.com/images/xlogoc.gif" alt="Xilinx"> adr:;;7801 Jefferson St. NE;Albuquerque;New Mexico;87109; version:2.1 email;internet:Jennifer.Jenkins@xilinx.com title:Applications Engineer fn:Jennifer Jenkins end:vcard --------------CB1D2D4A90962111D83802A7--Article: 33337
Leon Heller wrote: > > I've just built up a prototype of a simple prototyping system for the > Altera Flex 10K10. If there is any interest, I'll get some more boards > made. I'll be placing the design in the public domain, and making the > Gerbers available for anyone who wants to get their own PCB made. > > I've put a picture on my web site: > > http://www.geocities.com/leon_heller/flex.gif > > Leon Heller Leon, this sounds interesting, but the page is not found. There doesn't appear to be a link from your home page either. regards... --GaryArticle: 33338
Ben Franchuk <bfranchuk@jetnet.ab.ca> wrote in message news:<3B5BC9A0.43FA581C@jetnet.ab.ca>... > Leon Heller wrote: > > > > I've just built up a prototype of a simple prototyping system for the > > Altera Flex 10K10. If there is any interest, I'll get some more boards > > made. I'll be placing the design in the public domain, and making the > > Gerbers available for anyone who wants to get their own PCB made. > > > > The PCB is four-layers, 80mm X 100mm (half Eurocard). Unregulated DC > > (8-12V) needs to be provided from a suitable PS, like a wall-wart. > > There is an 'idiot' diode to prevent damage if the supply is round the > > wrong way. A socket is provided for a 14-pin crystal oscillator > > module. The Flex chip is configured via a standard Byteblaster header, > > there is no provision for a configuration device. Most of the Flex I/O > > and +5V is available via a 64-way DIN41612 plug, to which a suitable > > Eurocard prototyping board may be attached. > > > > I've put a picture on my web site: > > > > http://www.geocities.com/leon_heller/flex.gif > > > > Leon Heller > Burch use to make a nice 10k10 FPGA board but they moved to bigger chips. > Having the this board has me thinking about the few problems I have using this > board - I can't interface it to to a large solderless breadboard. Rather than > making another full sized PC board is make a FPGA adapter PCB's to fit a > solderless breadboards. Ben. > PS. Your URL gives me a not found error. Sorry, I should have uploaded it as a JPEG. Try http://www.geocities.com/leon_heller/flex.jpg You could plug my board into a Eurocard board with the solderless breadboard mounted on it. The chips are a bit big to put onto some sort of DIL PCB, which could be plugged into a solderless breadboard and there might be problems with decoupling LeonArticle: 33339
Good Luck is all I can say - these can be a real pain in the... But find a board builder with experance to place your part (recommendation - so you can kick them when it don't work) However I found the folloing help full. Keep the part towards the center of the board, Don't flood fill too think or unless you realy need to (this will all pull heat away). Now this I cant remember - somthing to do with the pad, you get copper or pre solderd ones if I remember. I think the copper ones were better? Get your stuff X-rayed and JTAGed afterwords - At this point its realy worth rembering that complex designs can be a lot easyer to fault find with lots of boudary scan devices on your board. We found corner pins the biggest problem, would not flow or cracked during the cool down. I the end we payed someone that knowns what they are doing - cost a lot less that our own labour and time. Cyber_Spook_Man Anthony Ellis wrote: > This is not an FPGA question but I guess many of you may have designed > FPGA's using similar technology and work for companies that have some advice > to part with. > > We have a PCB designed for a 255 pin Ceramic BGA and are struggling to > solder this correctly. Plastic BGA's are not an issue however. > Obviously the ceramic absorbs the reflow energy but we haven't been able to > find the correct technique to pre-heat the part. I guess we could measure > the pre-heated component temperature if we had some idea about it's required > value. > > Any help please. > > Thanks AnthonyArticle: 33340
what are the limitations on the free webpack software? which vhdl syntax are supported or not supported? where can i find technical specifications? thanks.Article: 33342
Daniel Hañczewski wrote: > Hi, > > I have also made this cable and the only problem I met was cable length. > Xilinx advices wire connection between JTAG header and Parallel III > electronics to be "as short as possible" and between electronics and LPT > port no longer than 2m. I have placed electronics inside standard DSUB25 > connector and according to Xilinx's schematic JTAG-electronics cable length > musn't exceed 0,5m. Finally I got rid of all 100pF capacitors connected to > JTAG signal lines and now my cable has 2m and works without reservation. I > haven't tried to make it longer... > > Regards > Daniel > No real criticism Daniel but I've heard this sort of thing a lot but I reckon the length thing is basically voodoo. I've run 3x 3 meter Centronics extension cables + the Parallel-3 & configured CPLDs with not problem (back in the early days of the JTAG programmer when it didn't run on Win95 I had to do this to reach the NT box on the far side of the office). Now our boards have the Xilinx P-III h/w on them & it only takes 2 extension cables.Article: 33343
I am wondering what exactly is meant by "full vhdl and verilog support", because i keep getting "...not supported" kind of error messages with the webpack software.Article: 33344
I had this problem with the demo version that came with the book. The problem was the design I was compiling was too big for the demo version. "Dave Feustel" <dfeustel@mindspring.com> wrote in message news:9jceos$guu$1@nntp9.atl.mindspring.net... > I have the Silo-3 Demo from "Verilog HDL_ by Samir Palnitkar > installed on two machines running Windows 2000 Pro. > One is a Dell Dimension XPS Pro200n, the other is a Dell 4100. > Silo runs on the XPS, but crashes with a memory access violation > as soon as I attempt to run a simulation on the 4100. This behavior > occurs with exam1a.s. (While Silo does run on the XPS, it crashed > while I was adjusting a window size. How stable is the demo version > of Silo?) > > Does anyone have any idea what might be causing this? > > Thanks, > > Dave Feustel > >Article: 33345
Hi Dave, "Dave Feustel" <dfeustel@mindspring.com> wrote in message news:9jhc90$g0p$1@nntp9.atl.mindspring.net... > > I downloaded the Silos demo program but it reports fatal errors when I try > to run any of the sample projects. I'd be willing to try the evaluation version > What kind of error does it report? I had a similar situation sometime ago and it said: " Said that I must select a device. I was using Silos III demo version." If this is the error that you have encountered I can sugegst a workaround, do let me know. SriniArticle: 33346
> > > Sorry, I should have uploaded it as a JPEG. Try > > http://www.geocities.com/leon_heller/flex.jpg For some reason, clicking on the above doesn't work. If you type the address in it does work!!! LeonArticle: 33347
Good Morning, Always a question on my QPSK classical modulator schema containing two branches each one with a polyphase SRRC filter having it's output going to a multiplier together with sine or cosine produced by a DDS working at 165MHz to produce 40MHz , I've these two questions about : 1)I've arranged the interpolator rate so to always have at the multiplier the same rate of the DDS, 165MHz , so if for example I've at the modulator input a data rate of 55Mbps corresponding to a symbol rate of 27.5MSpS , I interpolate 6 to have the same rate , is this correct ??? There are other methods to do the same ??? 2) What do you think about my choise to use 165MHz instead of 160MHz to produce 40MHz , there is any consideration to do in terms of SFDR ??? Thank you all again for your help ... Antonio D'OttavioArticle: 33348
> Sounds like the synthesis tool isn't being as intelligent as it could be. I > agree with your suggestion to try a different coding style, though. That's what I thought, because it synthesized fine with FPGA Express but the problem came up with XST VHDL. Thanks for the replies, ClemensArticle: 33349
ts1@its.com.sg wrote: > I am new in vb design, anybody can email some sample code or show me some site to download, I just want to design a vb program to load an MS access database(phonebook directory), I can search by name , by postcode, by tel no, to select some company. > > thanks > ts Wrong news group. You probably need to look around in comp.lang.XXX.
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