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
2017JanFebMarApr2017

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 42125

Article: 42125
Subject: Programming for FPGA or ASIC
From: prashantj@usa.net (Prashant)
Date: 16 Apr 2002 09:55:54 -0700
Links: << >>  << T >>  << A >>
hi,
I read in a post previously that VHDL/Verilog programming style
differs when being written for an FPGA from being written for an ASIC.
Is there any website which mentions these differences ? I'm looking to
see what should I take care of when programming for an FPGA.

Thanks,
Prashant

Article: 42126
Subject: Re: Xilinx JTAG C Source (Again)
From: Neil Glenn Jacobson <neil.jacobson@xilinx.com>
Date: Tue, 16 Apr 2002 09:55:58 -0700
Links: << >>  << T >>  << A >>
Try here:&nbsp; <a href="http://www.xilinx.com/isp/microprocessor.htm">Products
: Microprocessor Solution</a>

Kevin Hansen wrote:
<blockquote TYPE=CITE>A few months ago I downloaded some JTAG-programming
C source code, from the
<br>Xilinx site, that could be modified for my application. I have misplaced
<br>the files and am looking for them again, but I haven't been able to
find
<br>them. Can anyone point me to where they are?

Thank you,
<br>Kevin Hansen</blockquote>
</html>


Article: 42127
Subject: Re: just bought Cohen's book...Real Chip Design and Verification using Verilog and VHDL
From: "Paul Baxter" <pauljnospambaxter@hotnospammail.com>
Date: Tue, 16 Apr 2002 18:13:02 +0100
Links: << >>  << T >>  << A >>
Just thought I'd put up a short review of this book as well. I bought it on
the strength of the review (below) posted in these newsgroups

I'm not going to write about what topics it covers, I'm just aiming to
provide some overall personal opinions on the book.

I'm afraid that the ad-hoc style of presentation with poor and
inconsistently styled diagrams is the first thing to hit you.
Tables of text that look like they were converted to bitmap and then
stretched are really unacceptable in this day and age. Same applies to
certain equations (how long would that have taken to typeset
properly?)Probably an easy way of scanning source material though :(

Undeniably there is useful information here, but unfortunately, anyone who
has looked through Altera and Xilinx's app notes and followed a few links to
recommended articles posted on the news group will be in for a
disappointment. For an example Altera clocklock/clock boost (Altera App115)
is reused, as are numerous Xilinx/Altera Fifo discussions. (With permission
I hasten to add).

I felt a little cheated by the amount of information that was directly
lifted from other freely available sources.
Granted there's a CD with a lot of the files on, and the book does provide a
central repository, but I don't think that justifys the expenditure of your
effort.

There are items that everyone from newcomers to moderately well-versed HDL
engineers will learn, but it isn't presented in a coherent style, instead my
impression is that its stitching articles together in the hope of getting a
book published quickly.

The fact that its trying to be a dual language book is also intensely
annoying with page upon page of listings to skip (and providing a six page
listing of repetitive 32bit CRC equations is not my idea of a well-chosen
example when extracts from this code would have been much better). Most of
the listings are very simple and don't add too much value to the overall
content.

You may think I absolutely loathed the book. That's not the case as it does
present some useful cookbook-style topics, but its lack of a coherent
presentation style, poorly chosen example code and in the end is a missed
opportunity.

Ben is extremely helpful in these newsgroups and I did hold off before
publishing my comments in the hope that my disappointment would temper
somewhat, but at the end of the day I'd much prefer to give you a list of
seven or eight app notes/articles that are freely available that provide a
similar amount of content.

I really hope Ben doesn't take too much offence as it is books like this
that when executed well deliver a fabulous wealth of information. This one
falls short of the mark in several ways, sadly.

Now if Ray and Ben got together with an independent typesetter... there
could be a killer cookbook :)

Paul Baxter

> strut911@hotmail.com (strut911) wrote in message
news:<4379d3e0.0203161247.eef8730@posting.google.com>...
> > its a little rough around the edges in appearance (sorry mr. cohen)
> > but its worth its weight in gold in terms of the lessons that can be
> > learned from it. i had it for one day and i already learned about five
> > new things, and i thought i had been in the industry long enough to
> > consider myself a semi-professional. there are all sorts of tips and
> > tricks that can be used to combat well-known design issues (i never
> > thought of using a flancter before) and it goes through the real way
> > to write verification suites (instead of paying lip-service to some
> > simple testbenches). overall, i am very happy with it. some issues i
> > have with it are: leans a bit more towards vhdl than verilog, although
> > he obviously tries to balance the two languages, synplify pro
> > schematics are sometimes a bit tough to follow, and in some cases it
> > might be slightly better to abbreviate the code listings instead of
> > allowing it to take up multiple pages.
> > but like i said, in the true spirit of independent publishing, its
> > rough around the edges, but definitely written by an expert. i will
> > definitely recommended the people at work to pick up a copy. the only
> > thing mr. cohen needs to worry about is that he might not get as much
> > consulting work since he already present many solutions to common
> > problems in his book.
> > i just wanted to give everyone a heads up on my experience with this
> > book. if you want to learn more about chip/fpga design and
> > verification, at least the way it is done by professionals, i
> > recommend this book. or if you want to support good independent
> > publishing and encourage more experts to publish books where they can
> > actually make money and provide quality instruction (are you listening
> > mr. andraka?), then i would also recommend this book. this is just my
> > two cents. i hope other people can present their reviews of this or
> > other industry books also.
> > strut911





Article: 42128
Subject: Re: Reconfiguring Spartan II after boot-up
From: "Falk Brunner" <Falk.Brunner@gmx.de>
Date: Tue, 16 Apr 2002 19:13:40 +0200
Links: << >>  << T >>  << A >>
"Pete Koziar" <koziar.pete@orbital.com> schrieb im Newsbeitrag
news:2951d198.0204160412.2414e128@posting.google.com...
> We have an application where we desire to remotely program a new FPGA
> image into a flash ROM then signal the system to reinitialize itself,
> including reloading the Spartan II.
>
> The Spartan II is configuring through slave parallel mode.
>
> How do I get the Spartan II to clear its internal memory and reload
> itself? I've tried just dropping the Program line, but neither Init or
> Done ever goes low.

Pulse PROGRAM LOW (>300ns), then some hundred us after PROGRAM is HIGH
again, DONE is LOW and INIT is HIGH (both are open drain).
Then you start to clock in the data, if there is a CRC error, INIT will go
LOW.
After all bytes are transfered, you need some additional clocks (4 or 6, not
sure at the moment) to start te FPGA (release global reset, Tristate etc.)
Then DONE will go HIGH.

--
MfG
Falk





Article: 42129
Subject: Re: Power supply pins
From: "Falk Brunner" <Falk.Brunner@gmx.de>
Date: Tue, 16 Apr 2002 19:18:41 +0200
Links: << >>  << T >>  << A >>
"Vladislav Vasilenko" <vlad@comsys.ntu-kpi.kiev.ua> schrieb im Newsbeitrag
news:3CBC52AF.438A6514@comsys.ntu-kpi.kiev.ua...
> Can you explain me answer or where can I read detailed information about

The problem is, that the FPGA will ALWAYS run at full speed, even if you
design has only frequencys of some kHz. That means, it will switch VERY
fast. In this moment, it requires a big current surge form the power supply
network. If you connect just 1/2 of the VCC/GND pins, this will increase the
inductance of the power supply network as seen from the FPGA and during the
small, but powerfull current spikes, the voltage will drop, causing
malfunctions.

--
MfG
Falk





Article: 42130
Subject: Virtex Development Board with a 4M or more gates
From: "Cyrille de BrÚbisson" <cyrille_de-brebisson@hp.com>
Date: Tue, 16 Apr 2002 11:29:57 -0600
Links: << >>  << T >>  << A >>
Hello,

We are working on a BIG design, and we are looking for a Virtex II based
development board with a XC2V4000 or higher. Does someone has an idea of
where to find one?

Regards, Cyrille



Article: 42131
Subject: GPGA GPS Cores
From: "Kent Krumvieda" <kentkr@datafusion.com>
Date: Tue, 16 Apr 2002 11:31:33 -0600
Links: << >>  << T >>  << A >>
Does anyone know of any FPGA GPS Cores that are out on the market?

Kent




Article: 42132
Subject: Re: Command-line utility for loading Xilinx XC9572XL and Spartan II via JTAG
From: Petter Gustad <newsmailcomp2@gustad.com>
Date: Tue, 16 Apr 2002 18:05:25 GMT
Links: << >>  << T >>  << A >>
koziar.pete@orbital.com (Pete Koziar) writes:

> Does anyone know of a command-line utility for loading the Xilinx
> XC9572XL and the Spartan II via JTAG? I would like to set up a batch
> file for our medium size production run where the whole process is
> automated to the point where the tester plugs in the JTAG port and
> hits "Enter."
> 
> The operations it needs to do are:
> 
>      o Load a CPLD
>      o Load a Spartan II
> 
> After that the batch file would run our application to program the
> flash chip.
> 
> Does such a thing exist or do we need to write it?

You don't say anything about what kind of JTAG programming device you
are using.

If you are using a XILINX MultiLINX or similar then jtagprog should do
the job. I've done something like:

jtagprog -batch jtag


where the file jtag.cmd contains something like

part XC95144XL:pld
program -v pld
quit


You can also generate SVF files from jtagprog by adding the -svf
option and then copy the generated file to your other JTAG system.


A similar file for a scan chain of two 18v04's looks like:


part XC18V04:TOP_0 XC18V04:TOP_1
erase -f TOP_0
erase -f TOP_1
program -v TOP_0
program -v TOP_1
quit

Where the two MCS files TOP_0.mcs and TOP_1.mcs have been created
using promgen.

> I know that Xilinx has a big honkin app in WebPack, but the problem is
> that it takes a fair amount of resources and requires too much manual
> intervention. What we're looking for could potentially be loaded on a
> little embedded PC and run automatically.

I just start a script which does synthesis, map, par++, and then
finally jtagprog. The next day the 18v04's for the Virtex device are
ready for testing...

Petter
-- 
________________________________________________________________________
Petter Gustad   8'h2B | (~8'h2B) - Hamlet in Verilog   http://gustad.com

Article: 42133
Subject: Re: creating my own hard macro or similar
From: Theron Hicks <hicksthe@egr.msu.edu>
Date: Tue, 16 Apr 2002 14:55:58 -0400
Links: << >>  << T >>  << A >>
I found answer 10901 and tried solution #1.  I could not get it to work (I
could not get the External Macro Pin to be available.)  I looked at solution
2 (xapp123) but it seems to apply only to the tri-state buffer if I read the
readme correctly.  Any help?

Theron Hicks wrote:

> Hi,
>     My design is somewhat structural at the highest level.  I would like
> to floorplan the lower level modules and then have that floorplan (and
> perhaps even routing) be retained at the higher level.  Is that
> possible?  Is there some hidden trap that I need to be aware of (other
> than the potential problem with the blocks not being able to be placed)?
>
> If this is a reasonable and wise idea, how do I do so?  My code is VHDL
> and my target device is the Spartan2E.  I am currently using ISE4.1 but
> I have ISE4.2 available.  (I just haven't installed it yet, as I would
> rather avoid changing software in mid project.)
>
> Thanks,
> Theron Hicks


Article: 42134
Subject: Re: Need help with Insight Spartan II demo board and the counter demo.
From: Brendan Bridgford <brendanb@xilinx.com>
Date: Tue, 16 Apr 2002 11:57:54 -0700
Links: << >>  << T >>  << A >>
Jeff,

My suggestion would be to start debugging the problem by programming the
FPGA through the JTAG chain.  Don't worry about programming the part
from the PROM at this point.

-Generate the .bit file with startup clock set for JTAG Clock and
readback/reconfiguration enabled, as you mentioned.
-Set the Mode pins for Boundary Scan mode
-Launch iMPACT and initialize the chain.  The Spartan-II will appear as
a Virtex because they use the same IDCODEs.
-You can assign the $xilinx\xc18v00\data\xc18v01.bsd file (BSDL file) to
the PROM instead of a .mcs file for the purpose of bypassing the PROM.
-Assign the .bit file to the Spartan-II as you did before
-Right-click on the Spartan-II and program it.

At this point, the critical question is whether or not the DONE pin went
high.  If it did, configuration was successful, and the problem is
somehow related to the design or your board setup.  Make sure that your
oscillator is hooked up and oscillating, and make sure that any enable
signals are correctly driven.  Check your design.pad file (the PAD
report) and make sure that design clock input is locked to the correct
pin, and that the LED outputs are locked to the correct pins.

Hope this helps!

--Brendan


jeff wrote:

> I've been trying to make the counter demo work with
> Insight's $125 Spartan II demo board, but it fails.
> All I get is a blank LCD.  It should have incrementing
> numbers.
>
> I got the Verilog source for the counter demo here:
> http://208.129.228.206/solutions/kits/xilinx/spartan-ii.html
>
> I created a project in Webpack 4.2, and it compiled OK.
>
> In the process called "Generate Programming File", I
> opened the property box, and selected CCLK as the startup
> clock.  I ran the process, and then went to Generate
> Prom File, and made the counter.mcs file.
>
> Then I went back again, and set the startup to JTAG clock.
> I ran the process, which made a counter.bit file with
> the JTAG clock.  I didn't make a prom file, because this
> file is for the Spartan II chip.
>
> (I'm sure I selected the clocks correctly, because if I
> don't the iMPACT program tells me).
>
> After that, I clicked on Configure Device.  I have the board
> jumpered for JTAG mode (M1 jmpr on, M0 & M2 off).  I clicked
> on Initialize chain, and it shows a 18V01 and a XCV100.
> I assign counter.mcs to the 18V01 and counter.bit to the
> next chip.   After I do that, the XCV100 changes to a
> XC2S100.  Now I right-click on the 18V01 and tell it
> to program. It succeeds.
>
> Now I unplug power from the board, and re-jumper it for
> Master Serial mode, which is M0,M1,M2 jumpers all on.
> I plug in power, fully expecting the LCD to start incrementing
> but it doesn't.  Nothing happens.  Can anyone tell me
> what I'm doing wrong ?
>
> -----= Posted via Newsfeeds.Com, Uncensored Usenet News =-----
> http://www.newsfeeds.com - The #1 Newsgroup Service in the World!
> -----==  Over 80,000 Newsgroups - 16 Different Servers! =-----


Article: 42135
Subject: Re: creating my own hard macro or similar
From: Bret Wade <bret.wade@xilinx.com>
Date: Tue, 16 Apr 2002 13:25:58 -0600
Links: << >>  << T >>  << A >>
Hi Theorn,

You need to be in macro editing mode in FPGA Editor, look for an .nmc extension
in the banner. If it's .ncd instead, you need to first do a "save as macro".
Once you're in macro editing mode, an external pin is added by selecting a
component pin (not just a site pin) and running the command Edit-->Add Macro
External Pin. Fill in the resulting form. You need to give the external pin the
same name as the corresponding port name for the black box in your logical
design.

Hope this helps,
Bret

Theron Hicks wrote:

> I found answer 10901 and tried solution #1.  I could not get it to work (I
> could not get the External Macro Pin to be available.)  I looked at solution
> 2 (xapp123) but it seems to apply only to the tri-state buffer if I read the
> readme correctly.  Any help?
>
> Theron Hicks wrote:
>
> > Hi,
> >     My design is somewhat structural at the highest level.  I would like
> > to floorplan the lower level modules and then have that floorplan (and
> > perhaps even routing) be retained at the higher level.  Is that
> > possible?  Is there some hidden trap that I need to be aware of (other
> > than the potential problem with the blocks not being able to be placed)?
> >
> > If this is a reasonable and wise idea, how do I do so?  My code is VHDL
> > and my target device is the Spartan2E.  I am currently using ISE4.1 but
> > I have ISE4.2 available.  (I just haven't installed it yet, as I would
> > rather avoid changing software in mid project.)
> >
> > Thanks,
> > Theron Hicks


Article: 42136
Subject: IO Standards supported in Spartan-II devices
From: creon100@yahoo.com (Sean)
Date: 16 Apr 2002 13:08:20 -0700
Links: << >>  << T >>  << A >>
Different documentation seems to say different things about what IO
standards are supported in Spartan-II.  One table says it doesn't
support PCI33_5, but another document about the Virtex (the Xilinx
site takes you to that doc since Spartan-II and Virtex have the same
IO) says it does support PCI33_5.  Does anyone know if it actually
does or not?

Article: 42137
Subject: Re: Need Help to Implement Div Operation
From: "Speedy Zero Two" <david@manorsway.NOSPAMPLEASE.freeserve.co.uk>
Date: Tue, 16 Apr 2002 21:13:48 +0100
Links: << >>  << T >>  << A >>
All,

Think first principles,
    remember long division? now think binary!
    think async or sync

    That's all folks :-)

"kn" <kn@cetic.be> wrote in message
news:1b1252c2.0204160700.379a8571@posting.google.com...
> Hi all
>
> I am trying to implement a module in VHDL which performs an 8 bit
> disvison operation.
> I am looking for an efficient algorithm suitable for hw
> implementation.
>
> Kind Regards
> kn



Article: 42138
Subject: Re: creating my own hard macro or similar
From: Theron Hicks <hicksthe@egr.msu.edu>
Date: Tue, 16 Apr 2002 16:23:13 -0400
Links: << >>  << T >>  << A >>
That was the problem, Thanks!

Bret Wade wrote:

> Hi Theorn,
>
> You need to be in macro editing mode in FPGA Editor, look for an .nmc extension
> in the banner. If it's .ncd instead, you need to first do a "save as macro".
> Once you're in macro editing mode, an external pin is added by selecting a
> component pin (not just a site pin) and running the command Edit-->Add Macro
> External Pin. Fill in the resulting form. You need to give the external pin the
> same name as the corresponding port name for the black box in your logical
> design.
>
> Hope this helps,
> Bret
>
> Theron Hicks wrote:
>
> > I found answer 10901 and tried solution #1.  I could not get it to work (I
> > could not get the External Macro Pin to be available.)  I looked at solution
> > 2 (xapp123) but it seems to apply only to the tri-state buffer if I read the
> > readme correctly.  Any help?
> >
> > Theron Hicks wrote:
> >
> > > Hi,
> > >     My design is somewhat structural at the highest level.  I would like
> > > to floorplan the lower level modules and then have that floorplan (and
> > > perhaps even routing) be retained at the higher level.  Is that
> > > possible?  Is there some hidden trap that I need to be aware of (other
> > > than the potential problem with the blocks not being able to be placed)?
> > >
> > > If this is a reasonable and wise idea, how do I do so?  My code is VHDL
> > > and my target device is the Spartan2E.  I am currently using ISE4.1 but
> > > I have ISE4.2 available.  (I just haven't installed it yet, as I would
> > > rather avoid changing software in mid project.)
> > >
> > > Thanks,
> > > Theron Hicks


Article: 42139
Subject: Re: IO Standards supported in Spartan-II devices
From: Kevin Brace <ihatespam99kevinbraceusenet@ihatespam99hotmail.com>
Date: Tue, 16 Apr 2002 15:32:32 -0500
Links: << >>  << T >>  << A >>
Virtex and Spartan-II do support PCI33_5, but Virtex-E and Spartan-IIE
don't.



Kevin Brace (In general, don't respond to me directly, and respond
within the newsgroup.)



Sean wrote:
> 
> Different documentation seems to say different things about what IO
> standards are supported in Spartan-II.  One table says it doesn't
> support PCI33_5, but another document about the Virtex (the Xilinx
> site takes you to that doc since Spartan-II and Virtex have the same
> IO) says it does support PCI33_5.  Does anyone know if it actually
> does or not?

Article: 42140
Subject: A problem with PAR in ISE WebPACK 4.2
From: Kevin Brace <ihatespam99kevinbraceusenet@ihatespam99hotmail.com>
Date: Tue, 16 Apr 2002 15:39:59 -0500
Links: << >>  << T >>  << A >>
Hi, I am currently having a problem with Xilinx ISE WebPACK 4.2WP0.0's
PAR.
In ISE WebPACK 4.1, when I specified the number of routing iterations to
whatever number, PAR routed it whatever many times even if it thinks the
design has no chance of meeting timing requirements.
However, somehow things have changed in ISE WebPACK 4.2, and now even if
I specify the number of routing iterations, let's say to ten iterations,
PAR always gives up after two routing iterations.
Has anyone noticed this behavior, and is there a way to override the
current behavior?
Some may suggest that I should manually floorplan my design, but I
already did that (I am using UCF flow.), and what I am trying to do is
to run multiple routing iterations after manually floorplanning my
design.
Others may suggest that I should go back to ISE WebPACK 4.1, but the
newer version of ModelSim XE-Starter that comes with ISE WebPACK 4.2
'seems' to run faster than the version that comes with ISE WebPACK 4.1,
so I don't want to go back to ISE WebPACK 4.1.



Thanks,



Kevin Brace (In general, don't respond to me directly, and respond
within the newsgroup.)

Article: 42141
Subject: Re: creating my own RPMs(?) or similar
From: Ray Andraka <ray@andraka.com>
Date: Tue, 16 Apr 2002 21:18:59 GMT
Links: << >>  << T >>  << A >>
You can build placed macros in your source: to do so you need to
structurally instantiate anything you want to place and then put RLOCs on
it.  That works fine for 3.3sp8, however you will find out that the
floorplanner in v4.x wrecks your virtex/virtexE/spartanII/spartanIIe RPMs
because of a nasty bug that moves placed stuff around willy-nilly.  Word
right now is that it won't be fixed until the next major release, so if you
go this route you are confined to 3.3 software.  Ask your FAE for an update
to the timing files if you do stick with 3.3

Theron Hicks wrote:

> Hi,
>     My design is somewhat structural at the highest level.  I would like
> to floorplan the lower level modules and then have that floorplan (and
> perhaps even routing) be retained at the higher level.  Is that
> possible?  Is there some hidden trap that I need to be aware of (other
> than the potential problem with the blocks not being able to be placed)?
>
> If this is a reasonable and wise idea, how do I do so?  My code is VHDL
> and my target device is the Spartan2E.  I am currently using ISE4.1 but
> I have ISE4.2 available.  (I just haven't installed it yet, as I would
> rather avoid changing software in mid project.)
>
> Thanks,
> Theron Hicks

--
--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, 1759



Article: 42142
Subject: Telecom Bus info
From: lucky_hero@yahoo.com (VP)
Date: 16 Apr 2002 14:25:21 -0700
Links: << >>  << T >>  << A >>
Hi,

I am looking for spec/details on Telecom Bus (TelecomBus). I searched
on the web and all I could get is a number of links to PMC Sierra
which  offers Telecom bus bridges, converters, serializers etc but
does not offer much details about Telecom bus protocol. Neither could
I find anything on the IEEE standards website.

I am designing an ATM Cell Processor in FPGA which has Telecom bus
interface. I would greatly appreciate if someone can point me to
relevant info on Telecom Bus.

Thanks

Article: 42143
Subject: Beta Testers Need for HOTMan
From: "Steve Casselman" <sc.nospam@vcc.com>
Date: Tue, 16 Apr 2002 21:50:34 GMT
Links: << >>  << T >>  << A >>
HOTMan is a program/API for management and delivery of Xilinx Virtex/XC4000
bitstreams. It can run locally or remotely and includes bitstream
compression and bitstream version control. It uses C++ for low level access
to the hardware and Java for the GUI and networking. It has source code
examples of programming through the Xilinx download Parallel Cable III (I'm
told Cable IV is backwards compatible) in both Serial Slave and Jtag 1149.1.
It currently runs under any of the NT flavors and is being ported to Solaris
and Linux. With this system it is easy to program applications that help
test and debug your hardware. Not only does it deliver bitstreams but  it
also enables you to add your own commands to the GUI interface and gives you
a quick and reliable method of remote test and upgrade.

We need some beta testers to use this software and report back any bugs or
other anomalies. If you are interested please mailto:jas@vcc.com .

Thanks

Steve Casselman, CEO
Virtual Computer Corporation

--
Remove the .nospam for the correct reply address



Article: 42144
Subject: Re: Synario v2.3
From: Jim Granville <jim.granville@designtools.co.nz>
Date: Wed, 17 Apr 2002 10:34:41 +1200
Links: << >>  << T >>  << A >>
Apeak wrote:
> 
> Is there a copy of Synario v2.3 available either for free or purchased?
> I start learning about programming the GAL22V10 and someone recommended
> Synario v2.3.

 Synario is no longer a standalone tool.

 You could go to the Atmel web, and get AWINCUPL.EXE, which
is CUPL, and this supports 22V10, and std SPLDs ( 16V8 + ).

 CUPL can run via the IDE, or with a programmers editor, 
and includes conditional compile controls & optimise choices.

 CUPL can also create JED test vectors for post-pgm functional 
test in most programmers.

 - jg

Article: 42145
Subject: FPGA Timing Problem
From: John Larkin <jjlarkin@highlandSNIPTHIStechnology.com>
Date: Tue, 16 Apr 2002 19:32:04 -0700
Links: << >>  << T >>  << A >>

Why is it that all the production people like to get to work real
early and then leave early, while the engineers arrive late and work
late, so that when you really need a big chip replaced there's nobody
around to do it for you?

John



Article: 42146
Subject: Re: Using SRL16E Xilinx primitive.
From: andrew.bridger@paragon.co.nz (Andrew Bridger)
Date: 16 Apr 2002 21:02:48 -0700
Links: << >>  << T >>  << A >>
Hi Austin and Newsgroup,
Sorry, the part is an XC2S50PQ208.  XST, 4.2i sp1.  No daisy chain,
just a single device, and there are no other devices on the JTAG port.
I have tried with the M0, M1, M2, programming mode pins set to 010,
slave parallel mode and with them set to 100, boundary scan mode. 
Same results both times.  The FPGA is not configured
properly(exhibited by simple single SRL16E not working correctly and
other portions of design faulty) when it is configured for a second
time over JTAG.

I had a look through xapp188 - configuration and readback of sp2 fpgas
using boundary scan.  On page 9, fig 5, it implies to me that the
shutdown sequence is run when the FPGA is configured for a second time
over JTAG.  So the configuration memory doesn't get cleared the second
time its programmed.  Well, thats one difference I noted, dunno if
thats significant.

Ok, I'll submit a case.

Regards
Andrew


Austin Lesea <austin.lesea@xilinx.com> wrote in message news:<3CBC34F8.15E46326@xilinx.com>...
> Andrew,
> 
> I would need to know which part to comment.
> 
> Sometimes the startup sequence at the end of configuration may not behave correctly if there are other older
> devices in a daisy chain (ie Virtex II with Spartan II) as early bitgen option defaults were not correct for
> Virtex II and daisy chaining (since been corrected in subsequent service packs).
> 
> As well, Virtex II ES parts required a bitgen switch set for generating the correct bitfile for the ES
> material.
> 
> Often when we encounter one of these 'cycle the power' mysteries, we do not follow it to the end and find out
> what was really going on:  I would advise against that.  File a case on the hotline, and follow through.  I
> despise mysteries!
>

Article: 42147
Subject: Re: FPGA Timing Problem
From: "Leo Havm°ller" <leh@-nospam-rtx.dk>
Date: Wed, 17 Apr 2002 06:38:14 +0200
Links: << >>  << T >>  << A >>
"John Larkin" <jjlarkin@highlandSNIPTHIStechnology.com> wrote in message
news:Xt68PBLDZOsrEop6+udYs5ye81GL@4ax.com...
>
> Why is it that all the production people like to get to work real
> early and then leave early, while the engineers arrive late and work
> late, so that when you really need a big chip replaced there's nobody
> around to do it for you?

Im an engineer - i arrive early and work late ;-)

Leo Havm°ller.



Article: 42148
Subject: Re: Source code for a NIOS instruction set simulator?
From: "Johnsonw10" <johnsonw10_NOSPAM@hotmail.com>
Date: Wed, 17 Apr 2002 04:43:16 GMT
Links: << >>  << T >>  << A >>
Quartus Megawizard generates Verilog or VHDL files for the NIOS core for
simulation. It may be slower than an instruction set functional model.

Jim

"Mats Brorsson" <Mats.Brorsson@imit.kth.se> wrote in message
news:a9gnr5$3f7$1@news.kth.se...
> Hi,
>
> I'm considering using the NIOS processor from Altera for lab exercises in
a
> computer organization course and would then like to have an instruction
set
> simulator for this processor. With the Altera Excalibur NIOS development
kit
> there is a simulator, but I can't find the source code for it. Anyone
knows
> of another simulator for this architecture?
>
> Please respond by email since I do not regularly read the news groups.
>
> Regards
>
> Mats Brorsson
> Mats.Brorsson@imit.kth.se
>
>
>



Article: 42149
Subject: Re: bad experience with Xilinx ISE 4.1i and Xilinx hotline suppot
From: andreas <andreas4fun@subdimension.com>
Date: Wed, 17 Apr 2002 06:59:30 +0200
Links: << >>  << T >>  << A >>
On Tue, 09 Apr 2002 03:50:25 GMT, Ray Andraka <ray@andraka.com> wrote:
Hi,
I recently discovered a problem that not all applicable paths covered
by a timespec were checked by the software. 

I had a multichannel design and the same timespec for all channels.
But in the report I saw that the parts covered were only half for a
certain channel. There were paths not meeting the timespec which
didn't get reported !!!

And concerning the post P&R simulation, there are a lot of cases were
timing errors will be not found by this simulation. The simulator uses
the max. time in all paths, also the clock paths. So a design might
work in the simulator but since there is a skew between paths in
practise the real implementation might have problems!

Regards
Andreas

>You mention all timing constraints were met.  What about constraint coverage?
>It is entierly possible to have all constraints met with only a small portion
>of the design actually constrained.  If your design is passing a functional
>simulation and timing, but does not work, the next step is to carefully examine
>your timing constraints to make sure you haven't got something important
>unconstrained, followed by doing a post-route simulation to verify that the
>routed design does indeed match the input design.  There have been a few cases
>where the xilinx tool is the culprit, but much more often it is the result of a
>change in the synthesizer.
>
>al wrote:
>
>> We were using 3.1i and works fairly stable with no major problems. Last
>> week, we loaded up the 4.1i and decided to give it a try. We used the same
>> set of VHDL design files, UCF constraint and batch file. At first, tool
>> seems working fine -- we are able to save compile time by about 40% and all
>> timing constraints are meet. Well, not so happy yet -- the output binary
>> doesn't not work in our hardware, seems like there is some sort of timing
>> problem somewhere. We called up xilix hotline, this support guy by the name
>> of Justin have no idea of what he is talking about -- keep telling me to do
>> a post route timing simulation on my 1.6 million gate design. Anyway, bad
>> tool + bad support = unhappy customer.





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
2017JanFebMarApr2017

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