Site Home   Archive Home   FAQ Home   How to search the Archive   How to Navigate the Archive   
Compare FPGA features and resources   

Threads starting:
1994JulAugSepOctNovDec1994
1995JanFebMarAprMayJunJulAugSepOctNovDec1995
1996JanFebMarAprMayJunJulAugSepOctNovDec1996
1997JanFebMarAprMayJunJulAugSepOctNovDec1997
1998JanFebMarAprMayJunJulAugSepOctNovDec1998
1999JanFebMarAprMayJunJulAugSepOctNovDec1999
2000JanFebMarAprMayJunJulAugSepOctNovDec2000
2001JanFebMarAprMayJunJulAugSepOctNovDec2001
2002JanFebMarAprMayJunJulAugSepOctNovDec2002
2003JanFebMarAprMayJunJulAugSepOctNovDec2003
2004JanFebMarAprMayJunJulAugSepOctNovDec2004
2005JanFebMarAprMayJunJulAugSepOctNovDec2005
2006JanFebMarAprMayJunJulAugSepOctNovDec2006
2007JanFebMarAprMayJunJulAugSepOctNovDec2007
2008JanFebMarAprMayJunJulAugSepOctNovDec2008
2009JanFebMarAprMayJunJulAugSepOctNovDec2009
2010JanFebMarAprMayJunJulAugSepOctNovDec2010
2011JanFebMarAprMayJunJulAugSepOctNovDec2011
2012JanFebMarAprMayJunJulAugSepOctNovDec2012
2013JanFebMarAprMayJunJulAugSepOctNovDec2013
2014JanFebMarAprMayJunJulAugSepOctNovDec2014
2015JanFebMarAprMayJunJulAugSepOctNovDec2015
2016JanFebMarAprMayJunJulAugSepOctNovDec2016
2017JanFebMarAprMayJunJulAugSepOctNovDec2017
2018JanFebMarAprMayJunJulAugSepOctNovDec2018
2019JanFebMarAprMayJunJulAugSepOctNovDec2019
2020JanFebMarAprMay2020

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

Custom Search

Messages from 18725

Article: 18725
Subject: Re: where can I find fitter algorithms
From: G.S. Vigneault <z180@hotmail.DOT.com>
Date: Wed, 10 Nov 1999 00:12:24 -0500
Links: << >>  << T >>  << A >>
Many FPGA references are available at URL...
http://home.korax.net/~telic/bn/asic.htm


On Wed, 10 Nov 1999 10:42:35 +1000, anup kumar raghavan
<anup@elec.uq.edu.au> wrote:
>If there are any relevant web pages regarding this, can you please let
>me know.


{}{}{} Posted via Uncensored-News.Com, http://www.uncensored-news.com {}{}{}
 {}{}{}{} Only $8.95 A Month, - The Worlds Uncensored News Source {}{}{}{}
Article: 18726
Subject: Re: CAN tools reccomendations?
From: Don McKenzie <don@dontronics.com>
Date: Wed, 10 Nov 1999 17:17:18 +1100
Links: << >>  << T >>  << A >>


Arnold Beland wrote:
> 
> I have decided to use the CAN bus for a variety of control projects.  I
> would appreciate any suggestions.
> 
> Thanks in advance,
> 
> Arnold Beland

try:
http://www.lawicel.com/candip/

Don McKenzie  mailto:don@dontronics.com http://www.dontronics.com

Don's Download Dungeon:   http://www.dontronics.com/download.html
World's Largest Range of Atmel and PICmicro Hardware and Software
Article: 18727
Subject: Re: Frequency Division in Altera AHDL ?
From: jessiek@polbox.com
Date: Wed, 10 Nov 1999 08:09:35 GMT
Links: << >>  << T >>  << A >>
<!doctype html public "-//w3c//dtd html 4.0 transitional//en">
<html>
Hi Anton,
<br>thank you for your reply. I would like to know if you have another
e-mail
<br>address than antera@mweb.co.za for future contacts, because I have
serious
<br>problems with sending any new messages via my connection (I don not
why!)
<br>Maybe you would know what it may mean and how I will be able to send
you
<br>anything without newsgroup usage (see attachment).
<p>Regards
<br>Jessie
<p>ATTACHMENT (previoulsy sent messages to you):
<p>The original message was received at Tue, 9 Nov 1999 08:15:56 GMT
<br>from [195.117.19.230]
<p>&nbsp;&nbsp; ----- The following addresses had permanent fatal errors
-----
<br>&lt;antera@mweb.co.za>
<p>&nbsp;&nbsp; ----- Transcript of session follows -----
<br>... while talking to jhb-proxy.mweb.co.za.:
<br>>>> RCPT To:&lt;antera@mweb.co.za>
<br>&lt;&lt;&lt; 550 5.7.1 Baddomain, your domain has been blocked from
sending to :
<br>antera@mweb.co.za
<br>550 &lt;antera@mweb.co.za>... User unknow
<p>Anton Erasmus wrote:
<blockquote TYPE=CITE>Hi,
<p>I am trying to get a 16MHz Frequency from a 24MHz Clock on a Altera
<br>EPM7128. The 24MHz Clock is connected to one of the global clock
<br>inputs. Can anyone help me in doing this in AHDL ?
<p>Regards
<br>&nbsp;&nbsp; Anton Erasmus</blockquote>
</html>

Article: 18728
Subject: Re: CAN tools reccomendations?
From: "Steve Lovell" <slovell@arcomcontrols.com>
Date: Wed, 10 Nov 1999 08:25:08 -0600
Links: << >>  << T >>  << A >>
Arnold,

My company, Arcom Control Systems, manufacturers a single-channel
opto-isolated CAN interface module using the Philips SJA1000 chip in PC/104
form-factor, the AIM104-CAN.  DeviceNet software is available for this board
and drivers for Dos or Windows CE operating systems.

For more information, please visit the link below for the AIM104-CAN, or
please do not hesitate to contact me.

http://www.arcomcontrols.com/products/icp/pc104/modules/AIMcan1.htm

Thanks,

Steve Lovell
Arcom Control Systems
13510 South Oak Street
Kansas City, MO 64145
Phone: 816-941-7025, x255
E-mail: slovell@arcomcontrols.com
Web: www.arcomcontrols.com

Arnold Beland wrote in message <80a9nk$127v@enews4.newsguy.com>...
>I have decided to use the CAN bus for a variety of control projects.  I
>would appreciate any suggestions.
>
>Thanks in advance,
>
>Arnold Beland
>
>


Article: 18729
Subject: Re: Simulation of FPGA design. Please Help!
From: Rick Filipkiewicz <rick@algor.co.uk>
Date: Wed, 10 Nov 1999 15:23:12 +0000
Links: << >>  << T >>  << A >>


Brian Philofsky wrote:

> The reason I suggested the Map route is because the original request
> was to simulate the logic delays but not routing delays.  Most
> netlists produced by synthesis tools do not have any delay information
> therefore only a unit delay simulation would be possible.  By
> processing the design through map, you will get true logic level
> delays but not routing delays.  Performing a post-synthesis simulation
> is fine if you are interested in only a unit delay simulation and will
> save you a few steps through our tools although in general the netlist
> translation and map stage do not take too much CPU time.
>
> True you will have to use the SIMPRIM libraries with this approach
> rather than the UNISIM libraries however I don't understand the
> problems you mention.  I successfully use both types of simulation
> models all the time for both VHDL and Verilog simulations.  The only
> major things that come to mind that you may be referring to are in the
> 1.5 release of the LUT models (for Verilog), you must drive all design
> inputs or sourced LUTs will go to a simulation value of X.  All that
> was needed to be done is to tie undriven design inputs to a known
> value.  With the 2.1 release, that requirement has been lifted and you
> should only see X's on the outputs of undriven LUTs when that input
> has a direct correlation to the result.  The only other problems I
> generally see when a global reset is not performed on the design
> before beginning simulation.  The methodology for global reset has
> been changed for Verilog in the 2.1 release so that it will work
> better with pre-compiled libraries which seems to be the trend of most
> Verilog simulators today.  This methodology is explained in our
> documentation.  Without this global reset, it is possible the design
> might come up in a unknown state.
>
> I am not trying to say the Xilinx libraries are without fault however
> they should work with most simulation/design scenarios.  There are
> always going to be corner cases which can not be easily accounted for
> and as those are encountered, we will try to fix them.
>
> If you would like to go into more detail as to the problems you
> mention, I would like to hear them.  Although I am not directly
> involved with the simulation group, I can pass on any
> comments/experiences you or anyone else might have about the
> simulation libraries.  I would like to change this perception that
> they are error-prone.
>
> Thanks,
>
>
> --  Brian Philofsky
> --  Xilinx Design Engineer
>
>

Brian, I've emailed you directly a summary of my Verilog post MAP/PAR
simulations. In general the 2.1i Verilog stuff can be made to work well
but you might need a couple of modifications. I would like to clear up a
couple of points here though.

First the description of the LUT problem is not complete. As well as the
issues you mention above the 1.5 models themselves were wrong and
couldn't handle _legitimate_ `x' conditions. If anybody wants to try
this all they have to do is configure a 1.5 LUT2 model as a simple ``and
gate'' [INIT = 4'h8], set one input to 0 and the other to `x' and see
what happens. This was very depressing since at the time I badly needed
at least post-MAP simulation in my hunt for the Virtex MAPper
``disappearing inverter'' bug. The 2.1i LUT models have been completely
re-written.

Secondly, and directly relevant to the original posting, is a problem
with post-MAP Virtex simulation where the design contains BlockRAMs with
several inputs e.g. addresses tied to GND. The Verilog netlist generator
may sometimes leave these inconnected. This doesn't seem to happen with
PAR simulation.

Finally the 2.1i UNISIMS Verilog models seem to be a step backwards from
the 1.5 ones. They have been rewritten as behavioural and seem to have
lost all attempts to e.g. handle `x' conditions which the 1.5 versions
for the most part modeled correctly. This means that anyone using them
for post-synth simulation is likely to get differences when they come to
post-route simulation. Of course you can always push on through NGDBUILD
& then use the SIMPRIMS lib.

If anybody else is interested in my findings please mail me & I'll try
to make a summary available on our FTP.

Anybody had any experience with the VHDL simulation ?

Article: 18730
Subject: Re: CAN tools reccomendations?
From: Ted Wood <ted.wood@sortex.com>
Date: Wed, 10 Nov 1999 16:33:23 GMT
Links: << >>  << T >>  << A >>
Armold:

The NSI Canpocket analyser is a handy little tool and is pretty useful.
I liked it a lot, although the version I used had only French language
support, I believe it now supports English. About 1000 UKP last time I
looked...

Vector Software's Canalyzer is an expensive but heavyweight tool which
needs a PC with a PCMCIA slot for their own CAN card. Comprehensive but
only for those with about 4000 UKP to spend...

Both available from Hitex, who are in England, Germany and the USA.

On the other hand, if you buy a cheapo CAN card for your PC you can
knock up an analyser program using your preferred PC development tool.
We did one using a National Instruments CAN card and LabWindows CVI.

There are much cheaper CAN cards available than the NI ones. Check the
CAN In Automation journal for details.



Cheers
TW

Any views expressed in this message are those of
the individual  sender, except  where  the  sender
specifically  states them to be the views of Sortex Ltd


Sent via Deja.com http://www.deja.com/
Before you buy.
Article: 18731
Subject: Re: CAN tools reccomendations?
From: "Steve Letkeman" <zanthic@zanthic.com>
Date: Wed, 10 Nov 1999 17:15:53 GMT
Links: << >>  << T >>  << A >>

Hi Arnold,

   Be sure to check through the CAN mailing list archives,
you can find them in the "News"-section at http://www.vector-informatik.de/.
  Please check out my web page for both PC interfaces, HC11
with CAN and Infineon  C505C boards. www.zanthic.com
  Also check out Lar's page for a really neat, small and low cost
CAN node  http://www.lawicel.com/candip/ (and other neat CAN stuff)

Good luck,
Steve

Zanthic Technologies Inc. located at    http://www.zanthic.com
Controller Area Network boards for HC11, ISA/PC104 and Parallel
I/O boards for ISA and Parallel, Win95/98 and NT drivers included

Arnold Beland wrote in message <80a9nk$127v@enews4.newsguy.com>...
>I have decided to use the CAN bus for a variety of control projects.  I
>would appreciate any suggestions.
>
>Thanks in advance,
>
>Arnold Beland
>
>


Article: 18732
Subject: Re: CAN tools reccomendations?
From: Tom Burgess <tom.burgess@hia.nrc.ca>
Date: Wed, 10 Nov 1999 10:25:07 -0800
Links: << >>  << T >>  << A >>
You might want to check out http://www.ibutton.com/TINI/index.html . A SIMM
JAVA controller with Ethernet, 4 Mb RAM, RTC, CAN interface and other goodies.
$50 Qty 1. Very cool.

Arnold Beland wrote:
> 
> I have decided to use the CAN bus for a variety of control projects.  I
> would appreciate any suggestions.
> 
> Thanks in advance,
> 
> Arnold Beland

-- 
Tom Burgess
-- 
Digital Engineer
National Research Council of Canada
Herzberg Institute of Astrophysics
Dominion Radio Astrophysical Observatory
P.O. Box 248, Penticton, B.C.
Canada V2A 6K3
Article: 18733
Subject: Re: orcad synthesis for simplepld
From: wq998@my-deja.com
Date: Wed, 10 Nov 1999 20:01:06 GMT
Links: << >>  << T >>  << A >>
Do not waste your time and money on OrCAD Express.

In article <38285ffc@ant.wmin.ac.uk>,
  seamang@westminster.ac.uk (Graham Seaman) wrote:
> I'm new to using ORCAD (I have tried mailing them/using their notice
board
> with no response so far) - does anyone have any experience of using
> the Express synthesis for simpleplds (actually GAL22V10)?
> I have a little behavioural VHDL FSM which analyzes ok; when I try to
build
> it Express generates an error message telling me I have too many
> product terms for a couple of rows. On inspecting the vhdl netlist
> its generated, there are actually not many PTs in the expressions
> its complaining about - well within the powers of a 22V10!
> Is there a common user error which could generate this? Or is this
> a known problem for which a patch is available?  (I have v. 7.1)
>
> Thanks for any advice,
>
> Graham Seaman
>


Sent via Deja.com http://www.deja.com/
Before you buy.
Article: 18734
Subject: Re: orcad synthesis for simplepld
From: Ray Andraka <randraka@ids.net>
Date: Wed, 10 Nov 1999 16:34:43 -0500
Links: << >>  << T >>  << A >>
wq998@my-deja.com wrote:

> Do not waste your time and money on OrCAD Express.
>

Now there's a helpful answer!

Are the pins constrained?  The 22v10 has 10 product terms on the 'outside'
outputs and 16 on the center ones.  If you have, say 12 product terms,
there are four outputs that the equation can't be assigned to.

> In article <38285ffc@ant.wmin.ac.uk>,
>   seamang@westminster.ac.uk (Graham Seaman) wrote:
> > I'm new to using ORCAD (I have tried mailing them/using their notice
> board
> > with no response so far) - does anyone have any experience of using
> > the Express synthesis for simpleplds (actually GAL22V10)?
> > I have a little behavioural VHDL FSM which analyzes ok; when I try to
> build
> > it Express generates an error message telling me I have too many
> > product terms for a couple of rows. On inspecting the vhdl netlist
> > its generated, there are actually not many PTs in the expressions
> > its complaining about - well within the powers of a 22V10!
> > Is there a common user error which could generate this? Or is this
> > a known problem for which a patch is available?  (I have v. 7.1)
> >
> > Thanks for any advice,
> >
> > Graham Seaman
> >
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.



--
-Ray Andraka, P.E.
President, the Andraka Consulting Group, Inc.
401/884-7930     Fax 401/884-7950
email randraka@ids.net
http://users.ids.net/~randraka


Article: 18735
Subject: ## Information Auction - Sell What You Know -Buy What You Dont ## 3845
From: ozfbgs@hotmail.com
Date: 10 Nov 1999 17:26:31 -0500
Links: << >>  << T >>  << A >>
Hi All,		
Check this site out - http://www.KnowToday.com

I recently registered on an interesting new site called - KnowToday.com that allows its members to buy and sell information! They are currently recruiting new members and will launch the site in several months. For
a limited time, they are offering an incentive plan, where they pay you
for successful referrals.

It sounds really interesting. You can go on their site and ask detailed questions or make money by answering questions. They are designing
the site with many different categories, so almost anyone will be able to ask and answer questions. Also, you can make money if any of your referrals use the site. 

It only takes a minute to join -  http://www.knowtoday.com/ -  they
explain everything on the web site. It's totally free and you have no 
further obligations after registering 

Hope you like it! 

John
mzjhqujmzfwylkwlwmsykbermgdctgyjrlxdqjxbjthkjvdcxptmzwleudgizvhvxbfckwvwgxf

Article: 18736
Subject: Re: CAN tools reccomendations?
From: Bruce L <bruceandjoan@ultranet.com>
Date: Wed, 10 Nov 1999 18:19:25 -0500
Links: << >>  << T >>  << A >>
Precise has a CAN implementation with their RTOS.  www.psti.com.

Arnold Beland wrote:

> I have decided to use the CAN bus for a variety of control projects.  I
> would appreciate any suggestions.
>
> Thanks in advance,
>
> Arnold Beland

Article: 18737
Subject: FS: New Altera Max+Plus II full VHDL $1K
From: fidonews2@my-deja.com
Date: Thu, 11 Nov 1999 02:43:57 GMT
Links: << >>  << T >>  << A >>
Hi,

I'm selling a full Altera MAX+Plus II package, V9.01. It's new and still
sealed (I never used it).  This package is their "Magnum" product which
supports full VHDL.  Great for DSP design.  Includes manual, CD-ROM and
dongle.  The standard 3-month maintainance has expired.  I can transfer
registration to you upon purchase.  Originally $7000, will sacrifice for
$1000 including shipping.

- Chris
fidonews2@my-deja.com



Sent via Deja.com http://www.deja.com/
Before you buy.
Article: 18738
Subject: Re: Altera Files vho and sdo too big
From: Rick Filipkiewicz <rick@algor.co.uk>
Date: Thu, 11 Nov 1999 11:28:37 +0000
Links: << >>  << T >>  << A >>


giuseppe giachella wrote:

> After placing and routing my design on an Altera Flex 10KA250, Maxplus2
> created two output files .vho and .sdo in order to start
> a postlayout simulation. These two files have a dimension of 16 MB
> each. It is almost impossible for me to compile such a vhdl and load
> an sdf file so big (I have tried using Modelsim and VSS).
>
> Is there anyone who succeeded in simulating files so big ?
>
> What should I expect if I migrate my design an a APEX or VIRTEX
> (they contain much more gates than a Flex 10KA250) ?
>

One approach would be to use a perl script to split the postroute netlist
into separate pieces and compile them seprately into a library.
Alternatively if VHDL makes that difficult and you have the co-simulation
version of ModelSim you could output the netlist in the more compact & easy
to split Verilog form.

For my postlayout Virtex XCV300 simulation I get an 8MB Verilog netlist+8MB
SDF and ModelSim [5.3aPE] handles this with no problems, takes just under 2
minutes to compile - 450MHz P-II + 512MBytes.


Article: 18739
Subject: looking for Xilinx/Actel Board
From: Andreas Barthel <barthel@htwm.de>
Date: Thu, 11 Nov 1999 14:11:36 +0100
Links: << >>  << T >>  << A >>
Does anyone know a FPGA evaluation (demonstration) board for Xilinx
FPGA's
larger PC84 packages or reprogrammable Actel FPGA's ?
Where can I get it ?

Thanks

Andreas Barthel
Article: 18740
Subject: Re: orcad synthesis for simplepld
From: seamang@westminster.ac.uk (Graham Seaman)
Date: 11 Nov 1999 15:03:28 GMT
Links: << >>  << T >>  << A >>
Ray Andraka (randraka@ids.net) wrote:
: wq998@my-deja.com wrote:
: 
: > Do not waste your time and money on OrCAD Express.
: >
: 
: Now there's a helpful answer!
And I don't have any choice... but it sounds like there are well
known problems with it. What are they?

: 
: Are the pins constrained?  The 22v10 has 10 product terms on the 'outside'
: outputs and 16 on the center ones.  If you have, say 12 product terms,
: there are four outputs that the equation can't be assigned to.
: 
No, no pins constrained at all. 6 outputs are used, 2 registered with 3 PTs 
each (none of the pts have more than 4 variables). These are the ones I
get the error messages for. 
The rest are combinational and only need 1 pt. So its
clearly not a 'real' fitting problem. My guess is that maybe there's a
problem with the simplepld library - perhaps something I haven't included,
the documentation is very patchy. I guess I have to track down an Orcad user...


Graham 
Article: 18741
Subject: fast programmable divider using xilinx xc4002xl
From: Theron Hicks <hicksthe@egr.msu.edu>
Date: Thu, 11 Nov 1999 10:12:14 -0500
Links: << >>  << T >>  << A >>
Can any one point me to a design for a programmable divide-by-n using an
XC4002xl device?  I plan on using the fastest speed grade available in
this part.  I need to divide by n, where n can be any integer from 2 to
32.  I don't care if the counter is up or down or whatever but it needs
to count with a 50% clock at 102.4 MHz.  Also I am looking for as
detailed a schematic as I can find on how to connect the 4002 to an
EPROM, etc.  I have hired a grad student to design the actual FPGA code
and he seems to be doing a good job but the actual hardware connection
between the FPGA and the EPROM is still unclear to me.  If anyone has
any comments on how to decouple the FPGA, etc. I would be thankful.  I
currently have a discrete version of this system in the prototyping
stage but the switching noise in the digital system is killing my analog
performance.  By the way I guess I should explain that this system
involves a relatively slow ramp driving a comparator.  The output of the
comparator is coupled to the digital section.  The basic nature of the
system precludes using hysteresis to clean up the signal so I need to
clean up the digital noise before the comparator.

Thanks,
Theron Hicks


Article: 18742
Subject: read back Altera
From: Euripides Sotiriades <esot@mhl.tuc.gr>
Date: Thu, 11 Nov 1999 17:58:15 +0200
Links: << >>  << T >>  << A >>
I would like to know if  i can ReadBack an Altera FPGA so I can get the
file that it is programmed. I have an Altera Programmer (the card, mpu
unit and the serial cable)

Article: 18743
Subject: Re: fast programmable divider using xilinx xc4002xl
From: Ray Andraka <randraka@ids.net>
Date: Thu, 11 Nov 1999 11:03:22 -0500
Links: << >>  << T >>  << A >>
Use a loadable down-counter that uses the carry chain.   Invert and connect
the tc output to the load input so that the counter automatically reloads on
the next rising edge after the counter reaches zero.  The parallel load
input is one less than the count length.  You can use the Xilinx logiblox
downcounter, but you are better off rolling your own if you use the count
enable because the xilinx macro puts an OR gate on the CE input, which will
degrade maximum performance.  In an XLA-07 this circuit will run at around
140 MHz for a 16 bit counter.  That curcuit should give you 103 MHz
performance in a Spartan XSC05XL as well.  I'd send a picture, but I'm
having trouble with my viewlogic license at the moment.

The EEPROM connections for the FPGA are outlined pretty well in the data
book.  Also look on http://www.xilinx.com in the support answers for
additional information on configuration and eeproms.  For bypassing, you
want to put a cap at each voltage pin.  Keep the caps as close as possible,
and connect them so that the trace from the via from the vcc plane to the
device goes through one of the cap pads.  The connection to the ground plane
from the other cap pad should be kept as short as possible as well.  The
idea here is to keep the inductance in series with the cap as low as
possible, and to use the trace inductance in the supply lead to your
advantage.  Also, keep the traces from the FPGA ground pins to the ground
plane vias as short a possible.  Use at least a 4 layer board with solid VCC
and ground planes.  You might want to separate the digital and analog
voltage and ground planes as well.

Theron Hicks wrote:

> Can any one point me to a design for a programmable divide-by-n using an
> XC4002xl device?  I plan on using the fastest speed grade available in
> this part.  I need to divide by n, where n can be any integer from 2 to
> 32.  I don't care if the counter is up or down or whatever but it needs
> to count with a 50% clock at 102.4 MHz.  Also I am looking for as
> detailed a schematic as I can find on how to connect the 4002 to an
> EPROM, etc.  I have hired a grad student to design the actual FPGA code
> and he seems to be doing a good job but the actual hardware connection
> between the FPGA and the EPROM is still unclear to me.  If anyone has
> any comments on how to decouple the FPGA, etc. I would be thankful.  I
> currently have a discrete version of this system in the prototyping
> stage but the switching noise in the digital system is killing my analog
> performance.  By the way I guess I should explain that this system
> involves a relatively slow ramp driving a comparator.  The output of the
> comparator is coupled to the digital section.  The basic nature of the
> system precludes using hysteresis to clean up the signal so I need to
> clean up the digital noise before the comparator.
>
> Thanks,
> Theron Hicks



--
-Ray Andraka, P.E.
President, the Andraka Consulting Group, Inc.
401/884-7930     Fax 401/884-7950
email randraka@ids.net
http://users.ids.net/~randraka


Article: 18744
Subject: Altera Files vho and sdo too big
From: "giuseppe giachella" <il_templare@hotmail.com>
Date: Thu, 11 Nov 1999 09:09:07 CET
Links: << >>  << T >>  << A >>
After placing and routing my design on an Altera Flex 10KA250, Maxplus2 
created two output files .vho and .sdo in order to start
a postlayout simulation. These two files have a dimension of 16 MB
each. It is almost impossible for me to compile such a vhdl and load
an sdf file so big (I have tried using Modelsim and VSS).

Is there anyone who succeeded in simulating files so big ?

What should I expect if I migrate my design an a APEX or VIRTEX
(they contain much more gates than a Flex 10KA250) ?




______________________________________________________
Get Your Private, Free Email at http://www.hotmail.com


 Sent via Deja.com http://www.deja.com/
 Before you buy.
Article: 18745
Subject: Re: looking for Xilinx/Actel Board
From: xiaocong9313@my-deja.com
Date: Thu, 11 Nov 1999 17:51:32 GMT
Links: << >>  << T >>  << A >>
Try www.xess.com. Good luck!
In article <382AC088.B56986B7@htwm.de>,
  barthel@htwm.de wrote:
> Does anyone know a FPGA evaluation (demonstration) board for Xilinx
> FPGA's
> larger PC84 packages or reprogrammable Actel FPGA's ?
> Where can I get it ?
>
> Thanks
>
> Andreas Barthel
>


Sent via Deja.com http://www.deja.com/
Before you buy.
Article: 18746
Subject: Re: looking for Xilinx/Actel Board
From: "Steven K. Knapp" <sknapp@optimagic.com>
Date: Thu, 11 Nov 1999 09:51:38 -0800
Links: << >>  << T >>  << A >>
I haven't seen a board for Actel's reprogrammable FPGAs but there are a
number of Xilinx boards, supporting the larger devices, listed on The
Programmable Logic Jump Station at http://www.optimagic.com/boards.html.



--
-----------------------------------------------------------
Steven K. Knapp
OptiMagic, Inc. -- "Great Designs Happen 'OptiMagic'-ally"
E-mail:  sknapp@optimagic.com
   Web:  http://www.optimagic.com
-----------------------------------------------------------

Andreas Barthel <barthel@htwm.de> wrote in message
news:382AC088.B56986B7@htwm.de...
> Does anyone know a FPGA evaluation (demonstration) board for Xilinx
> FPGA's
> larger PC84 packages or reprogrammable Actel FPGA's ?
> Where can I get it ?
>
> Thanks
>
> Andreas Barthel


Article: 18747
Subject: Re: orcad synthesis for simplepld
From: Greg Neff <gregneff@my-deja.com>
Date: Thu, 11 Nov 1999 17:53:15 GMT
Links: << >>  << T >>  << A >>
In article <38285ffc@ant.wmin.ac.uk>,
  seamang@westminster.ac.uk (Graham Seaman) wrote:
> I'm new to using ORCAD (I have tried mailing them/using their notice
board
> with no response so far) - does anyone have any experience of using
> the Express synthesis for simpleplds (actually GAL22V10)?
> I have a little behavioural VHDL FSM which analyzes ok; when I try to
build
> it Express generates an error message telling me I have too many
> product terms for a couple of rows. On inspecting the vhdl netlist
> its generated, there are actually not many PTs in the expressions
> its complaining about - well within the powers of a 22V10!
> Is there a common user error which could generate this? Or is this
> a known problem for which a patch is available?  (I have v. 7.1)
>
> Thanks for any advice,
>
> Graham Seaman
>

We have encountered a number of problems using OrCAD Express.  OrCAD
themselves admit that the synthesis engine they wrote was less than
perfect.  That's why they bolted in the Exemplar synthesis engine in
release 9.  They still use the homegrown synthesizer for simple PLDs
though.

We are working with Xilinx CPLDs and FPGAs.  The problems that we are
experiencing are mostly related to the interface between Capture and
Express, and the interface between Express and the Xilinx A2.1i tools.
We have problems passing constraints through, and we have found one bad
macro (XIL_M1\XC4000E\CB2CLE.EDN).


--
Greg Neff
VP Engineering
*Microsym* Computers Inc.
greg@guesswhichwordgoeshere.com


Sent via Deja.com http://www.deja.com/
Before you buy.
Article: 18748
Subject: FPGA Expess vs. Synplify vs. Leonardo Spectrum
From: Greg Neff <gregneff@my-deja.com>
Date: Thu, 11 Nov 1999 17:59:58 GMT
Links: << >>  << T >>  << A >>
We are evaluating these tools.  I would appreciate any comments from
those of you that have experience with at least two of the following
tools:

Synopsys FPGA Express
Synplicity Synplify
Exemplar Leonardo Spectrum

--
Greg Neff
VP Engineering
*Microsym* Computers Inc.
greg@guesswhichwordgoeshere.com


Sent via Deja.com http://www.deja.com/
Before you buy.
Article: 18749
Subject: Re: fast programmable divider using xilinx xc4002xl
From: Peter Alfke <peter@xilinx.com>
Date: Thu, 11 Nov 1999 10:14:40 -0800
Links: << >>  << T >>  << A >>


Theron Hicks wrote:

> Can any one point me to a design for a programmable divide-by-n using an
> XC4002xl device?  I plan on using the fastest speed grade available in
> this part.  I need to divide by n, where n can be any integer from 2 to
> 32.  I don't care if the counter is up or down or whatever but it needs
> to count with a 50% clock at 102.4 MHz.

In the 1992 Xilinx Data Book ( page 6-28) I published a design called:
"Fully Synchronous 5-Bit Counter Uses Only Three CLBs."

This design would easily run at >150 MHz in XC4002, and even faster in the
newer 'XL devices.There is no carry involved, it's a Linear-Feedback Shift
Register Counter. Output wave-shape is also programmable.

This particular design defines the modulus during configuration, but it can
easily be made programmable from a 5-bit input at the expense of a few more
CLBs.

If you want, I can fax you the design, it's very simple, but non-obvious,
something a grad student might appreciate.

Today I would use a 3.3 V device (XL), either XC4000XL or Spartan XL.
But that may depend on your power supply situation. For purely digital
designs, 5 V is history ( albeit a long history!)

Peter Alfke, Xilinx Applications




Site Home   Archive Home   FAQ Home   How to search the Archive   How to Navigate the Archive   
Compare FPGA features and resources   

Threads starting:
1994JulAugSepOctNovDec1994
1995JanFebMarAprMayJunJulAugSepOctNovDec1995
1996JanFebMarAprMayJunJulAugSepOctNovDec1996
1997JanFebMarAprMayJunJulAugSepOctNovDec1997
1998JanFebMarAprMayJunJulAugSepOctNovDec1998
1999JanFebMarAprMayJunJulAugSepOctNovDec1999
2000JanFebMarAprMayJunJulAugSepOctNovDec2000
2001JanFebMarAprMayJunJulAugSepOctNovDec2001
2002JanFebMarAprMayJunJulAugSepOctNovDec2002
2003JanFebMarAprMayJunJulAugSepOctNovDec2003
2004JanFebMarAprMayJunJulAugSepOctNovDec2004
2005JanFebMarAprMayJunJulAugSepOctNovDec2005
2006JanFebMarAprMayJunJulAugSepOctNovDec2006
2007JanFebMarAprMayJunJulAugSepOctNovDec2007
2008JanFebMarAprMayJunJulAugSepOctNovDec2008
2009JanFebMarAprMayJunJulAugSepOctNovDec2009
2010JanFebMarAprMayJunJulAugSepOctNovDec2010
2011JanFebMarAprMayJunJulAugSepOctNovDec2011
2012JanFebMarAprMayJunJulAugSepOctNovDec2012
2013JanFebMarAprMayJunJulAugSepOctNovDec2013
2014JanFebMarAprMayJunJulAugSepOctNovDec2014
2015JanFebMarAprMayJunJulAugSepOctNovDec2015
2016JanFebMarAprMayJunJulAugSepOctNovDec2016
2017JanFebMarAprMayJunJulAugSepOctNovDec2017
2018JanFebMarAprMayJunJulAugSepOctNovDec2018
2019JanFebMarAprMayJunJulAugSepOctNovDec2019
2020JanFebMarAprMay2020

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

Custom Search