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 33050

Article: 33050
Subject: Re: Design entry
From: Ken McElvain <ken@synplicity.com>
Date: Mon, 16 Jul 2001 11:20:11 -0700
Links: << >>  << T >>  << A >>
There are several schematic entry systems that can write VHDL or
Verilog netlists.  Just include that generated HDL with the
HDL for the subblocks and I think you have what you asked for.

Look at Aldec for an example.  In Europe, there is a tool called
EASE from Translogic that generates VHDL.  I think these tools
also have state machine tools as well.

"Keith R. Williams" wrote:
> 
> In article <3B5053B0.8E3FF4D2@ecubics.com>, emu@ecubics.com says...
> > Noddy wrote:
> > >
> > > This brings me to my question? Should I rather be
> > > attempting to implement my designs in VHDL instead?
> >
> > Mix it !
> > A top level schematic, the units below in VHDL, Verilog, etc..
> > It's amazing, already magic how easy it is to explain what you're doing
> > to somebody else, when you can see all functional units, data flow on
> > one sheet.
> > And, sometimes you see your own design more clearly ;-)
> 
> That certainly would be an advantage.  However, I understand that
> synthesis in a mixed environment doesn't work well.  The redundancy
> isn't squished out beyond the black boxes.
> 
> > when you like to go to HDL only, just replace one sheet to a text page.
> 
> If I could do both schematics and (V)HDL with the same toolset with the
> flattening occurring before the "real" synthesis this would be nice.
> This soul also allow us old fuddy-duddy schematic wonks pick up on HDLs
> wile remaining productive. Listening Synplicity? ;-)
> 
> ----
>   Keith

-- 
Ken McElvain, CTO
Synplicity Inc.
(408)215-6060

Article: 33051
Subject: Re: Design entry
From: Keith R. Williams <krw@btv.ibm.com>
Date: Mon, 16 Jul 2001 15:17:00 -0400
Links: << >>  << T >>  << A >>
In article <3B53305B.68C195A5@synplicity.com>, ken@synplicity.com 
says...
> There are several schematic entry systems that can write VHDL or
> Verilog netlists.  Just include that generated HDL with the
> HDL for the subblocks and I think you have what you asked for.
> 
> Look at Aldec for an example.  In Europe, there is a tool called
> EASE from Translogic that generates VHDL.  I think these tools
> also have state machine tools as well.

I don't know if this is the same thing.  Synplicity, for instance, does 
a lot of work optimizing the logic for the target architecture.  It 
obviously can't see what it didn't generate so there may not be any 
optimizations across black boxes (even if two block boxes are adjacent 
in real life).  In fact I don't think there are any optimizations 
attempted across black boxes.

Anyway, I'm nervous adding more tools to the chain.  If the tools did 
schematic in the hierarchy, fine.  I agree that schematic is more 
"natural" for the high-level design and documentation.

----
  Keith

> "Keith R. Williams" wrote:
> > 
> > In article <3B5053B0.8E3FF4D2@ecubics.com>, emu@ecubics.com says...
> > > Noddy wrote:
> > > >
> > > > This brings me to my question? Should I rather be
> > > > attempting to implement my designs in VHDL instead?
> > >
> > > Mix it !
> > > A top level schematic, the units below in VHDL, Verilog, etc..
> > > It's amazing, already magic how easy it is to explain what you're doing
> > > to somebody else, when you can see all functional units, data flow on
> > > one sheet.
> > > And, sometimes you see your own design more clearly ;-)
> > 
> > That certainly would be an advantage.  However, I understand that
> > synthesis in a mixed environment doesn't work well.  The redundancy
> > isn't squished out beyond the black boxes.
> > 
> > > when you like to go to HDL only, just replace one sheet to a text page.
> > 
> > If I could do both schematics and (V)HDL with the same toolset with the
> > flattening occurring before the "real" synthesis this would be nice.
> > This soul also allow us old fuddy-duddy schematic wonks pick up on HDLs
> > wile remaining productive. Listening Synplicity? ;-)
> > 
> > ----
> >   Keith
> 
> 

Article: 33052
Subject: Re: Fixing routing in a Virtex FPGA
From: "Jeffrey Vallier" <jvallier_nospam_@gibson.com>
Date: Mon, 16 Jul 2001 12:52:43 -0700
Links: << >>  << T >>  << A >>

"Rotem Gazit" <rotemg@mysticom.com> wrote in message
news:86b060d0.0107160944.73e343@posting.google.com...
> Hi,
> Is there a simple way to fix manual or automatic routing so that on
> the next PR cycle it would not change ?
> The only way I know to do this is by instantiating predefined macro
> and I am looking for an easier way.
> Thanks,
>
> Rotem.
>
> Rotem Gazit
> MystiCom LTD
> mailto:rotemg@mysticom.com
> http://www.mysticom.com/
>

If you are done with your code and only playing with P&R, you can freeze the
design with the floorplanner and have the tools use that placement for
subsequent P&R iterations.  there's various options dialog boxes, depending
on which flavour of tools you are using.

If you are have not finalized your HDL and need to resynthesize, you're in
trouble. As a FAE explained to me (and correct me if I'm wrong) the P&R
tools use a random seed to start mapping the design into the chip following
translation. even if you did freeze the P&R w/floorplanner, resynthesis
renames nets and there's the source of the problem with that approach.

If your HDL is solid, some judicious use of constraints in the UCF can be
very helpful.

regards,

jeff

***********************************************
Jeffrey Vallier            Sr. FW Engineer
Gibson Guitar Corp.  GMICS Division
1283 F Old Mtn View/Alviso Rd.
Sunnyvale, CA 94089 408 734 4394
***********************************************



Article: 33053
Subject: Re: Fixing routing in a Virtex FPGA
From: Ray Andraka <ray@andraka.com>
Date: Mon, 16 Jul 2001 20:21:23 GMT
Links: << >>  << T >>  << A >>
You can freeze placement, but not routing.  To specify routing, you are
unfortunately stuck with either using hrad (fpga editor) macros or using
Jbits.  Sorry, no easy solutions here.

Rotem Gazit wrote:

> Hi,
> Is there a simple way to fix manual or automatic routing so that on
> the next PR cycle it would not change ?
> The only way I know to do this is by instantiating predefined macro
> and I am looking for an easier way.
> Thanks,
>
> Rotem.
>
> Rotem Gazit
> MystiCom LTD
> mailto:rotemg@mysticom.com
> http://www.mysticom.com/

--
-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



Article: 33054
Subject: Re: Fixing routing in a Virtex FPGA
From: Eric Crabill <eric.crabill@xilinx.com>
Date: Mon, 16 Jul 2001 13:39:39 -0700
Links: << >>  << T >>  << A >>

Hi,

There is a way to do what you want, but it will require some
time investment (perhaps more than is warranted by your design).
So it does not qualify as "easier".  We use this technique to
lock routes in the Xilinx PCI products to guarantee performance.
If you are using an HDL synthesis tool, forget it...

If you are looking to simply preserve as much of the placement
and routing between design iterations, I suggest you try using
guide mode in PAR.

par -gm exact -gf previously_routed_design.ncd
or
par -gm leverage -gf previously_routed_design.ncd

You can read the Xilinx software manuals online to understand
the difference between exact and leverage modes.

Long Live Viewdraw,
Eric

Rotem Gazit wrote:

> Hi,
> Is there a simple way to fix manual or automatic routing
> so that on the next PR cycle it would not change?   The only
> way I know to do this is by instantiating predefined macro
> and I am looking for an easier way.
> Thanks, 
> Rotem

Article: 33055
Subject: Re: Which Chip Family?
From: Ray Andraka <ray@andraka.com>
Date: Mon, 16 Jul 2001 20:45:29 GMT
Links: << >>  << T >>  << A >>


Falk Brunner wrote:

> Ray Andraka schrieb:
> >
> > but including Spartan II doesn't make sense.  For new designs, if you are
>
> I think it makes sense, since the Spartan-II devices are cheaper.

My point is that for designs that fit the XC2S, virtex and XC2S are
interchangeable (same bitstream), so you have a low cost target for the virtex
architecture (the virtex E architecture is essentially the same with a few
enhancements).  You are correct, I would not recommend Virtex for a new design
unless it was a QPRO application.  In any event it is a nitpick.

>
>
> > not in the Spartan II, I would go with the Virtex E right now.  The Virtex
>
> As I said.
>
> > design into space, you're stuck with Virtex.  Personally, I'd forgo the
> > 4K/Spartan (original) architecture for new designs, as the cost per gate and
> > speeds are better for the later families, and the later families have a
>
> Yes, but there are lots of applications that dont need the Multi-million
> gate FPGAs. If you have some small on-on boards which needs just some
> glue logic, small fifos, some uC registers etc., Spartan(XL) is just
> fine and cheap. Not everyone is working everyday on cutting edge,
> breakthrough designs. ;-)
>

Price and capability makes SpartanII much more attractive than Spartan.  I would
not recommend the original spartan for a new design in most cases.  The exception
would be when you already have a substantial part of the design that is tailored
or targeted to a 4K architecture.

single unit pricing From www.avnetmarshall.com/dynamic/search:

XC2S15-5VQ100  $8.96
XCS10XL-4VQ100 $12.76

These are the cheapest I saw for each part.  The XC2S15 gives you 384 LUTs plus 4
block RAMs at 3/4 the price of the XCS10, which has 392 LUTs.  You give up all of
8 LUTs, but gain 4 block RAMs, SRL16 capability (which is big in my book), faster
clocks, later technology and a longer time horizon.  I'll give up the 8 LUTs for
all that, especially considering the price difference.  Besides, I can use the 4
dual port block rams as 8 big LUTs if I wanted to.


--
-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



Article: 33056
(removed)


Article: 33057
(removed)


Article: 33058
Subject: New Book: Readings in Hardware/Software Co-design
From: webmaster@mkp.com (Morgan Kaufmann Publishers)
Date: 16 Jul 2001 14:59:06 -0700
Links: << >>  << T >>  << A >>
Morgan Kaufmann Publishers - New Title Announcement

Readings in Hardware/Software Co-design
by Giovanni De Micheli, Rolf Ernst, Wayne Wolf
http://www.mkp.com/books_catalog/catalog.asp?ISBN=1-55860-702-1

Part of the Morgan Kaufmann Series in Systems on Silicon
http://www.mkp.com/sos/

Embedded system designers are constantly looking for new tools and
techniques to help satisfy the exploding demand for consumer
information appliances and specialized industrial products. One
critical barrier to the timely release of embedded system products is
integrating the design of the hardware and software systems.
Hardware/software co-design is a set of methodologies and techniques
specifically created to support the concurrent design of both systems,
effectively reducing multiple iterations and major redesigns. In
addition to its critical role in the development of embedded systems,
many experts believe that co-design will be a key design methodology
for Systems-on-a-Chip.

Readings in Hardware/Software Co-Design presents the papers that have
shaped the hardware/software co-design field since its inception in
the early 90s. Field experts -- Giovanni De Micheli, Rolf Ernst, and
Wayne Wolf -- introduce sections of the book, and provide context for
the paper that follow. This collection provides professionals,
researchers and graduate students with a single reference source for
this critical aspect of computing design.

_______________________________________________________________
Morgan Kaufmann Publishers
San Francisco, California
http://www.mkp.com
orders@mkp.com

Article: 33059
Subject: Re: Fixing routing in a Virtex FPGA
From: John_H <johnhandwork@mail.com>
Date: Mon, 16 Jul 2001 22:17:12 GMT
Links: << >>  << T >>  << A >>
I actually had good success a while back with nets defined in my guide file
for a Spartan-II design (Virtex architecture).  I got rid of the nets I
didn't care about with the FPGA editor and manually routed the net I
wanted.  The results were pretty consistent for this critical net.  Newer
service packs have eliminated the need for this particular net's hand
routing, producing better results all around.

The only big troubles I could see is if the synthesis tool used renames
elements in the guide file.  This is the biggest problem in getting a
"minimum change" design.  If all I want is to add two test points I end up
with 20% of my edif elements renamed.  I hate it when that happens!

- John


Ray Andraka wrote:

> You can freeze placement, but not routing.  To specify routing, you are
> unfortunately stuck with either using hrad (fpga editor) macros or using
> Jbits.  Sorry, no easy solutions here.
>
> Rotem Gazit wrote:
>
> > Hi,
> > Is there a simple way to fix manual or automatic routing so that on
> > the next PR cycle it would not change ?
> > The only way I know to do this is by instantiating predefined macro
> > and I am looking for an easier way.
> > Thanks,
> >
> > Rotem.
> >
> > Rotem Gazit
> > MystiCom LTD
> > mailto:rotemg@mysticom.com
> > http://www.mysticom.com/
>
> --
> -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


Article: 33060
Subject: Re: Design entry
From: husby_d@yahoo.com (Don Husby)
Date: 16 Jul 2001 15:20:28 -0700
Links: << >>  << T >>  << A >>
John_H <johnhandwork@mail.com> 
> HDL, not schematics, yes!  yes!
> Verilog and VHDL are both appropriate to the task.
> It's a little like the age old question (probably still valid in
> your corner of this world:  "Coke or Pepsi?"

At the risk of starting a long religious debate, my experience
has been that VHDL is better suited for high speed FPGA design.
To push an FPGA to the edge, you pretty much need attributes
to control mapping and placement, and Verilog doesn't (yet)
have a standard way to implement attributes.  Also, VHDL's "generate"
construct is almost necessary for attaching attributes when using
wide busses.

I'm not a big fan of VHDL either.  I keep hoping that some day
someone will invent a *good* HDL.

Article: 33061
Subject: Re: Book Recommendation (bit different)
From: safahmy@hotmail.com (SAF)
Date: 16 Jul 2001 15:43:27 -0700
Links: << >>  << T >>  << A >>
OK,

After spending 3 hours in the biggest bookshops in town (the
world-famous Foyles, Waterstones and some new joint), I managed to
find the following two titles, which I bought:

Yalamanchili, Sudhakar (2001) "Introductory VHDL: From Simulation to
Synthesis" includes Xilinx Student Edition 2.1i - Only 25! (about
$37) It seems a pretty good book too.

Zwolinski, Mark (1997) "Digital System Design with VHDL" - 30 (about
$45) Good because it stradles the digital systems and VHDL borders,
explaining both.

I was surprised how little there was available. There were quite a few
research-oriented Titles by Kluwer Press, but with prices like 90
(about $135), and being quite old, forget it.

The Yalamanchili book popped out of nowhere in a smaller bookshop with
only one shelf on electronics. While the 7 or 8 shelves at Foyles were
full of stuff on CMOS, Digital IC design and lots of old titles.

Anyway, I'll let you all know how I get on with these titles. I'll
have a look at the Xilinx stuff too.

Thanks all.

Article: 33062
(removed)


Article: 33063
Subject: I NEED XILINX FOUNDATION PROFESSIONAL
From: yoram@puc.cl (Yoram Rovner)
Date: 16 Jul 2001 17:14:00 -0700
Links: << >>  << T >>  << A >>
Hello:

I'm a student from Chile starting an FPGA project. I need to get the Xilinx
Foundation Pro and I cant effort it(it is very expensive).
If somebody can tell me where i can get that software or a similar, please write me.

Sincerily

Yoram Rovner
yoram@puc.cl
Santiago de Chile

Article: 33064
Subject: Re: Book Recommendation (bit different)
From: Gonzalo Arana <gonzaloa@sinectis.com.ar>
Date: Mon, 16 Jul 2001 21:27:13 -0300
Links: << >>  << T >>  << A >>
Hi,

SAF wrote:
> 
> OK,
> 
> After spending 3 hours in the biggest bookshops in town (the
> world-famous Foyles, Waterstones and some new joint), I managed to
> find the following two titles, which I bought:
> 
> Yalamanchili, Sudhakar (2001) "Introductory VHDL: From Simulation to
> Synthesis" includes Xilinx Student Edition 2.1i - Only 25! (about
> $37) It seems a pretty good book too.

Is it this one (ISBN: 0130809829)?
http://www.amazon.com/exec/obidos/tg/stores/detail/-/books/0130809829/contents/ref=pm_dp_ln_b_2/104-5678861-7531964

> 
> Zwolinski, Mark (1997) "Digital System Design with VHDL" - 30 (about
> $45) Good because it stradles the digital systems and VHDL borders,
> explaining both.

Is it this book (ISBN: 0201360632)?
http://www.amazon.com/exec/obidos/tg/stores/detail/-/books/0201360632/contents/ref=pm_dp_ln_b_2/104-5678861-7531964

> 

You got them really cheap!
Good for you!

Article: 33065
Subject: Re: Byteblasting an ACEX in running system?
From: Russell Shaw <rjshaw@iprimus.com.au>
Date: Tue, 17 Jul 2001 10:34:42 +1000
Links: << >>  << T >>  << A >>
I tried out a simple counter design and it programmed and worked ok.
Only general IO pins had active signals from the dsp upon them.

Martin Schoeberl wrote:
> 
> Sould be no problem if the dsp has no problems with tri state pins.
> But how are the config pins driven on the board original?
> You could get a contention with the byte blaster.
> 
> Martin
> --
> Whant to see the evolution of a Java processor?
> 
>          http://www.jopdesign.com
> 
> "Russell Shaw" <rjshaw@iprimus.com.au> schrieb im Newsbeitrag
> news:3B4D8E87.31046A07@iprimus.com.au...
> > Hi all,
> >
> > I've got a dsp driving an altera ep1k30, and was wondering if its
> > ok to reprogram that chip even tho the dsp is still driving it.
> > I've got a byteblaster connected to it. I read somewhere that
> > all(?) the pins go to tri-state when programming (passive serial).
> >

Article: 33066
Subject: Re: Design entry
From: Ken McElvain <ken@synplicity.com>
Date: Mon, 16 Jul 2001 18:24:36 -0700
Links: << >>  << T >>  << A >>

"Keith R. Williams" wrote:
> 
> In article <3B53305B.68C195A5@synplicity.com>, ken@synplicity.com
> says...
> > There are several schematic entry systems that can write VHDL or
> > Verilog netlists.  Just include that generated HDL with the
> > HDL for the subblocks and I think you have what you asked for.
> >
> > Look at Aldec for an example.  In Europe, there is a tool called
> > EASE from Translogic that generates VHDL.  I think these tools
> > also have state machine tools as well.
> 
> I don't know if this is the same thing.  Synplicity, for instance, does
> a lot of work optimizing the logic for the target architecture.  It
> obviously can't see what it didn't generate so there may not be any
> optimizations across black boxes (even if two block boxes are adjacent
> in real life).  In fact I don't think there are any optimizations
> attempted across black boxes.
> 
> Anyway, I'm nervous adding more tools to the chain.  If the tools did
> schematic in the hierarchy, fine.  I agree that schematic is more
> "natural" for the high-level design and documentation.

Synplify would see the generated(from the schematic) top level VHDL/verilog 
because you would include it in the Synplify project.  The lower
level VHDL/Verilog files would then be optimized together.
You don't need to make the submodules black boxes.

> 
> ----
>   Keith
> 
> > "Keith R. Williams" wrote:
> > >
> > > In article <3B5053B0.8E3FF4D2@ecubics.com>, emu@ecubics.com says...
> > > > Noddy wrote:
> > > > >
> > > > > This brings me to my question? Should I rather be
> > > > > attempting to implement my designs in VHDL instead?
> > > >
> > > > Mix it !
> > > > A top level schematic, the units below in VHDL, Verilog, etc..
> > > > It's amazing, already magic how easy it is to explain what you're doing
> > > > to somebody else, when you can see all functional units, data flow on
> > > > one sheet.
> > > > And, sometimes you see your own design more clearly ;-)
> > >
> > > That certainly would be an advantage.  However, I understand that
> > > synthesis in a mixed environment doesn't work well.  The redundancy
> > > isn't squished out beyond the black boxes.
> > >
> > > > when you like to go to HDL only, just replace one sheet to a text page.
> > >
> > > If I could do both schematics and (V)HDL with the same toolset with the
> > > flattening occurring before the "real" synthesis this would be nice.
> > > This soul also allow us old fuddy-duddy schematic wonks pick up on HDLs
> > > wile remaining productive. Listening Synplicity? ;-)
> > >
> > > ----
> > >   Keith
> >
> >

Article: 33067
Subject: Re: I NEED XILINX FOUNDATION PROFESSIONAL
From: Eric Smith <eric-no-spam-for-me@brouhaha.com>
Date: 16 Jul 2001 18:51:07 -0700
Links: << >>  << T >>  << A >>
yoram@puc.cl (Yoram Rovner) writes:
> I'm a student from Chile starting an FPGA project. I need to get the
> Xilinx Foundation Pro and I cant effort it(it is very expensive).  If
> somebody can tell me where i can get that software or a similar,

Can you get your University to buy it?  Perhaps Xilinx has special
deals for educational institutions.

I'm not sure about "Foundation Pro", but if you think you need such a
thing presumably you've already determined that you can't get by with
WebPack ISE (free download from www.xilinx.com) or the student edition
of Foundation (inexpensive, can be purchased in bookstores including
amazon.com).

I'm somewhat surprised because WebPack ISE can handle fairly large
designs (up to XC2S200, XCV300E, and XC2V250), but if your design really
won't fit into one of those, and you can't partition it for multiple
designs, you're probably out of luck.

Maybe if you told us more about your project we could offer more
advice?

Best regards,
Eric

Article: 33068
(removed)


Article: 33069
Subject: Re: How to fan out signals to bus lines in Xilinx Foundation Schematic Editor?
From: Hong <lvhong@aol.com>
Date: Mon, 16 Jul 2001 21:53:52 -0700
Links: << >>  << T >>  << A >>
Alex,
You are right on the 2.1i, it might has problem as you said! Just a reference, the 3.3i works fine what you try to do "BUS". 

However, to workaround it (2.1i), you could try the Foundation logiBLOX/Macro program, it might works? The difference is that the compiler might not run DRC in the FPGA primitive converting process (EDIF writer). 
For example, try to open BUFT in symbol-editor, you might see I/O's types are funny like I=in ,T=input and O=output. The O=output type is wrong, it should be 3 state or tristate type. This is explained where the problem comes from, of course, SW wouldn't be happy in compilation when must we tie all outputs together in BUS circuitry , which is supposed to be 3 state, it is failed DRC! BUGGY!!!

To answer about Other tool solution is that you can use the EDIF/XNF (primitive) files from Orcad/viewlogic schematic (if you own one).
Hope this helps,
Hong

Article: 33070
Subject: PROBLEM!!!
From: "Tomek" <tbednar@poczta.onet.pl>
Date: Tue, 17 Jul 2001 08:33:54 +0200
Links: << >>  << T >>  << A >>
Hi,
I have a problem. I want to work Xilinx Foundation under Windows 2000, but
program lmutil.exe return bad hostid.
I have license on my net card but lmutil return no correct net card number.
I worked under WinNT and everything was OK.
What can I do to run Xilinx Foundation?

Tom



Article: 33071
Subject: Utopia Interface
From: "Craig Abramson" <craig.abramson@xilinx.com>
Date: Tue, 17 Jul 2001 00:46:21 -0600
Links: << >>  << T >>  << A >>
Hello:

This is my first time posting to this newsgroup.  If this is not an
appropriate post, my apologies.  Anyway, here goes:

Does anyone know of a public domain Extended Utopia core?

There, I did it.

Thanks!

Craig A.



Article: 33072
(removed)


Article: 33073
Subject: Re: processor core
From: Ben Franchuk <bfranchuk@jetnet.ab.ca>
Date: Tue, 17 Jul 2001 01:00:42 -0600
Links: << >>  << T >>  << A >>
Brian Philofsky wrote:
> 
> John,
> 
>    If you are planning to target a Xilinx FPGA, there is a list of
> free and pay-for uP cores on the Xilinx website in the Processor
> Central sections:

And here is  a core for Altera FPGA's. Still untested but gives idea of the
size of FPGA's needed for small cpu designs.

"Pre-historic Cpu's" http://www.jetnet.ab.ca/users/bfranchuk
Now with schematics.
Ben.

Article: 33074
Subject: clock versus just a pad
From: Vladimir Dergachev <volodya@mindspring.com>
Date: Tue, 17 Jul 2001 05:20:03 -0400
Links: << >>  << T >>  << A >>

Hi all, I am just starting with FPGAs am very puzzled by behivour of my 
development board. I am using ATSK40 with Atmel's AT4K chip on it. 
The board has several buttons, manual clock and a 32khz clock input. 

My test design is a series of flip-flops which form a counter.  When I hook 
up the clock input of the counter to the buttons (which are not debounced)
I get expected behaviour - counter is incremented by 1 when the button is 
pressed. (though I am wondering how come this is happening as buttons are 
not debounced..) However when I hook up to clock io pad I get random values.

I have a suspicion that using clock pad somehow changes the circuit (maybe 
flip flops get their clock inputs reassigned ?), but I dont' know where in 
options to change that (Figaro), nor how to see what gets assigned to clock 
inputs.

                     thanks

                            Vladimir Dergachev





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