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 12300

Article: 12300
Subject: Re: Synthesis: Exemplar or Synopsys
From: ems@nospam.riverside-machines.com
Date: Thu, 08 Oct 1998 11:39:08 GMT
Links: << >>  << T >>  << A >>
On Thu, 08 Oct 1998 04:50:37 GMT,
allan.herriman.hates.spam@fujitsu.com.au (Allan Herriman) wrote:

>Will it be 1076.6 compliant?

forget it - 1076.6 is a total waste of paper (and will hopefully never
be ratified). it is, to say the least, ambivalent about whether it's
based on '87 or '93. for all practical purposes, it's a limited subset
of '87 (and, in fact, FPGA express is already pretty close to being
compliant anyway).

>I got a rude suprise once when trying to port some code from Galileo
>to FPGA Express.  Couldn't convince the Synopsys tools to believe that
>a two dimensional array of std_logic was a valid type for a generic
>parameter.

1076.6 doesn't allow multi-dimensional arrays either.

evan

Article: 12301
Subject: Re: USAGE of XILINX "FROM:TO" for VHDL and IMPLEMENTATION
From: ems@nospam.riverside-machines.com
Date: Thu, 08 Oct 1998 11:39:46 GMT
Links: << >>  << T >>  << A >>
On Wed, 7 Oct 1998 14:55:00 -0700, "Andy Peters"
<apeters@noao.edu.NOSPAM> wrote:

>The "gotcha" with PERIOD is that you have to place it on the clock net AFTER
>the BUFGx, so you're outta luck if you're using VHDL (unless you want to
>look at the .xnf file and find out what the buffered clock net is called).
>It's noted in the timing constraints presentation that that little "feature"
>is a known problem that will be fixed eventually.

or just instantiate your BUFGs, in which case it's easy. i personally
prefer this - it's portable among all tools, and i don't like the
synthesiser putting in BUFGs for me.

evan

Article: 12302
Subject: Re: Degradation of results from Xilinx F1.3 -> F1.4 -> F1.5
From: ems@nospam.riverside-machines.com
Date: Thu, 08 Oct 1998 11:41:37 GMT
Links: << >>  << T >>  << A >>
here's a different spin on all this.

(1) you've got two conflicting constraints: a period constraint at
25ns, and a FROM:TO (which you don't show) at 20ns. this isn't a good
idea, and leads to your prioritisation problem and conflicting
frequency results. in general, always use PERIOD, and just use FROM:TO
to specify specific over-rides, ie. signals originating in a different
clock domain, multi-cycles, or whatever.

(2) there wasn't a specific problem with PERIOD in 1.4; just a generic
problem with any timespecs being traced through a BUFG (i haven't got
1.5 yet, and i don't know if the generic problem has been fixed).

(3) you create a timegroup which includes both the input to the clock
buffer, and the output. this is again asking for problems - why not
just:

> NET clk* TNM = fsa_grp;
> TIMESPEC TS01 = PERIOD fsa_grp 25ns;

or just on the input clock assuming the 1.5 fix?
and remove the FROM:TO spec.

(4) i would have guessed that a design which fails timespecs will
still run ok at the lower reported frequency, but it's just a guess
(but maybe the stubs are the result of an inadequate cleanup after
failing?)

(5) i'm pretty sure that you can ignore the clock skew warning, as
long as you're sure that there's no problem on that net. i've had the
same warning on a design in which the offending net wasn't used as a
clock anywhere.

evan

Article: 12303
Subject: Altera MAXPLUS2 V9 slow.
From: "Eric Pearson" <ecp@focus-systems.nospam.on.ca>
Date: 8 Oct 1998 13:07:26 GMT
Links: << >>  << T >>  << A >>
Has anyone noticed problems in re-compiling designs that worked in v8.0 of
maxplus2 under v9.01 and found comilation times went throught the roof even
to the point excess ( I aborted after 100 hours v9 vs 1 hour under v8 ).

Eric Pearson - Focus Systems

Article: 12304
Subject: Re: Synthesis: Exemplar or Synopsys
From: janovetz@tempest.ece.uiuc.edu (Jake Janovetz)
Date: 8 Oct 1998 08:13:06 -0500
Links: << >>  << T >>  << A >>
Derek Palmer <derekp@synopsys.com> writes:

>Let's make rumor fact!
>FPGA Express 3.0 will have VHDL 93 support.
>Available on the Synopsys web site at the end of Oct.

>Derek Palmer
>FPGA Express Team
>Synopsys Corporation.


And it's only 1998.  Talk about support!

   Cheers,
   Jake


>Andy Peters wrote:

>>     Hans Lindkvist wrote in message <3614BD09.E32A0E14@ldecs.ericsson.se>...
>>     Rickman wrote:
>>         Jake Janovetz wrote:
>>         One possibly big difference, depending on whether you have existing
>>         code, is that Synopsys does not support VHDL-93. They seem to be
>> pretty
>>         stuck in VHDL-87.
>>         Otherwise, I can only stay that Synopsys is what comes with Xilinx
>>         Foundation, so a lot of Xilinx users have it!   ;)
>>
>> I've heard from a Synopsys representative that vhdl-93 support is close.
>> I just got the Xilinx Foundation 1.5 tools w/Synopsys FPGA Express, and
>> there's support for *some* VHDL'93 things, like rising_edge() and
>> falling_edge() and myprocess : process (yadda) IS.
>>
>> A start.
>>
>> -andy



-- 
   janovetz@uiuc.edu    | Once you have flown, you will walk the earth with
 University of Illinois | your eyes turned skyward, for there you have been,
                        | there you long to return.     -- da Vinci
        PP-ASEL         | http://www.ews.uiuc.edu/~janovetz/index.html
Article: 12305
Subject: Re: Help Desperately Needed with Altera Microprocessor Design.
From: "Alun Morris" <bob@planetmars.demon.co.uk>
Date: Thu, 08 Oct 1998 14:12:30 GMT
Links: << >>  << T >>  << A >>
Have a look at the Atlas section on Altera's site (www.altera.com) and if
this doesn't help email your problem to sos@altera.com.

Alun

Article: 12306
Subject: Re: Xilinx Foundation Base
From: msimon@tefbbs.com
Date: Thu, 08 Oct 1998 14:56:09 GMT
Links: << >>  << T >>  << A >>
Schools can get them for free.

Simon
=================================================
"F. Arnold" <fa218@ural2.hszk.bme.hu> wrote:

>Hi!
>
>Where or How can I free get Xilinx Foundation Base?
>
>Thanks,
>    Arnold
>
>

Design Your Own MicroProcessor(tm) http://www/tefbbs.com/spacetime/index.htm
Article: 12307
Subject: Re: Design security again - the Actel solution
From: "Steven K. Knapp" <sknapp@optimagic.com>
Date: Thu, 8 Oct 1998 08:25:02 -0700
Links: << >>  << T >>  << A >>
At these "costs", the weak link is probably the design engineer.  Hiring him
or her away (or the whole team) at a fraction of the cost would be the easy
solution. :)

Also, if the devices are _really_ that secure, Actel should offer a big cash
prize to "crack" a secured device.  A $250,000 prize should be risk-free for
Actel according to their arguments.  Imagine the marketing messages they
could make then.

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


>Daniel K Elftmann wrote:
>
>> Some actual number of antifuses for the Actel devices are as follows:
>> Arch     Fuses    Cost@$500
>> A1010 -> 112,000  $56,000,000
>> A1020 -> 186,000  $93,000,000
>> A1225 -> 250,000 $125,000,000
>> A1240 -> 400,000 $200,000,000
>> A1280 -> 750,000 $375,000,000
>>
>> Still not prices that the typical hacker can afford in my opinion.
>>
>
>> >



Article: 12308
Subject: Xilinx Foundation Base
From: "F. Arnold" <fa218@ural2.hszk.bme.hu>
Date: Thu, 8 Oct 1998 17:42:41 +0200
Links: << >>  << T >>  << A >>
Hi!

Where or How can I free get Xilinx Foundation Base?

Thanks,
    Arnold


Article: 12309
Subject: Verilog Vs VHDL
From: craig_jacobs@asl-tk.com
Date: Thu, 08 Oct 1998 16:15:54 GMT
Links: << >>  << T >>  << A >>
Hello, I have been programming Altera FPGA's for several years using the
altera HDL and graphic designs.  We are considering transferring new and some
existing designs to either VHDL or Verilog.  I don't know anything about
either language, and I would like a list of pro's and con's about each.

Craig Jacobs
Systems Engineer
Advanced Product Development
Automotive Systems Lab

-----------== Posted via Deja News, The Discussion Network ==----------
http://www.dejanews.com/       Search, Read, Discuss, or Start Your Own    
Article: 12310
Subject: Re: Degradation of results from Xilinx F1.3 -> F1.4 -> F1.5
From: "Andy Peters" <apeters@noao.edu.NOSPAM>
Date: Thu, 8 Oct 1998 09:58:26 -0700
Links: << >>  << T >>  << A >>
ems@nospam.riverside-machines.com wrote in message
<361ca485.10782062@news.dial.pipex.com>...
>here's a different spin on all this.
>
>(1) you've got two conflicting constraints: a period constraint at
>25ns, and a FROM:TO (which you don't show) at 20ns. this isn't a good
>idea, and leads to your prioritisation problem and conflicting
>frequency results. in general, always use PERIOD, and just use FROM:TO
>to specify specific over-rides, ie. signals originating in a different
>clock domain, multi-cycles, or whatever.


Good idea.  Now, I wish that the Synopsys constraint-entry spreadsheet knew
from PERIOD (and TIG, and THRU, and TPSYNC...). It generates FROM:TOs for
all possible combinations of clocking and inputs: input pads to rising edge
clocks, input pads to falling edge clocks, rising-edge clocks to
rising-edge, falling to falling, rising to falling, falling to rising,
falling to output, rising to output, input to output (whew!).  So you end up
with about twenty FROM:TO constraints when all you'd need is one PERIOD
constraint and your two-phase clock stuff is all taken care of.

>(2) there wasn't a specific problem with PERIOD in 1.4; just a generic
>problem with any timespecs being traced through a BUFG (i haven't got
>1.5 yet, and i don't know if the generic problem has been fixed).


I think the "fix" is the added TNM_NET which lets you trace from input pads
through the BUFG.


>(4) i would have guessed that a design which fails timespecs will
>still run ok at the lower reported frequency, but it's just a guess
>(but maybe the stubs are the result of an inadequate cleanup after
>failing?)


I always thought that it should run at the slower speed if it finishes P+R,
but since we set timing constraints for a reason, I've never slowed my clock
down to try it!


--
Andy Peters
Sr. Electrical Engineer
National Optical Astronomy Observatories
950 N Cherry Ave
Tucson, AZ 85719
520-318-8191
apeters@noao.edu
>


Article: 12311
Subject: Re: FIR Filter Design
From: "Andy Peters" <apeters@noao.edu.NOSPAM>
Date: Thu, 8 Oct 1998 10:03:00 -0700
Links: << >>  << T >>  << A >>
Kartheepan, Madasamy wrote in message <361C5B37.7ED02EF@t-three.com>...

>Thank you for the last two informative emails. I definitely understand the
need
>to have a different design philosophy when tackling hardware
implementations of
>signal processing functions.A little off topic, I want to bounce this off
you.
>I have been thinking of  putting  Viterbi algorithm for finding MLS onto an
>FPGA. Have you come across similar implementations and what is your take on
the
>feasibility of putting the algorithm handling 32 states (5 memory elements)
on
>a 100K FPGA?. I would appreciate any input from you,

Just a comment: are you going to put the memory to store your search data
on-chip or in off-chip RAM?

I know that there are custom Viterbi decoder chips out there; not sure what
they do for memory.

--
Andy Peters
Sr. Electrical Engineer
National Optical Astronomy Observatories
950 N Cherry Ave
Tucson, AZ 85719
520-318-8191
apeters@noao.edu



Article: 12312
Subject: Xilinx Foundation forgets the pin assignment. Bug?
From: Ingo Froehlich <Ingo.Froehlich@exp2.physik.uni-giessen.de>
Date: Thu, 08 Oct 1998 19:04:17 +0200
Links: << >>  << T >>  << A >>
Hello,

I am using F1.4 Xilinx Foundation Software to programm a XC95216 CPLD.
The abel code has a pin assignment like 

name pin #;

but the fitter report tells me a different assignment then I have
written it down in the code. In the design manager, the option "Use
Design Location Constrains" is ON (At the end, it is same assignment if
its OFF). The edif-file look correct. 

It is like this option has no effect.

Do anyone know this problem and how to solve it?

Regards, Ingo Froehlich
Article: 12313
Subject: LCELL delay of Altera 10K's
From: "John Huang" <hungi@tpts4.seed.net.tw>
Date: Fri, 9 Oct 1998 01:08:15 +0800
Links: << >>  << T >>  << A >>
Hi all:

    Does anyone know the delay time of one LCELL,
if I implement an VHDL statement

  U1: LCELL(CLK1, CLK2);

  the CLK2 will be delay how many ns?

    John Huang


Article: 12314
Subject: Re: Degradation of results from Xilinx F1.3 -> F1.4 -> F1.5
From: Nestor Caouras <nestor@ece.concordia.ca>
Date: Thu, 08 Oct 1998 14:57:27 -0400
Links: << >>  << T >>  << A >>
> 
> here's a different spin on all this.
> 
> (1) you've got two conflicting constraints: a period constraint at
> 25ns, and a FROM:TO (which you don't show) at 20ns. this isn't a good
> idea, and leads to your prioritisation problem and conflicting
> frequency results. in general, always use PERIOD, and just use FROM:TO
> to specify specific over-rides, ie. signals originating in a different
> clock domain, multi-cycles, or whatever.
> 

What you suggested is exactly what I tried this morning before I checked
the newsgroup.  I simply removed the FROM:TO constraint and only kept
one PERIOD constraint on the clock net that is attached at the output of
the BUFG (which has been instantiated manually).  I also had to increase
the period to 30 ns in order to achieve an operating frequency of about
33MHz with no timing errors.

> 
> (4) i would have guessed that a design which fails timespecs will
> still run ok at the lower reported frequency, but it's just a guess
> (but maybe the stubs are the result of an inadequate cleanup after
> failing?)

The new implementation does not report any stubs, fortunately!  You are
probably right that the cleanup might not have been done adequately
after ppr.


> 
> (5) i'm pretty sure that you can ignore the clock skew warning, as
> long as you're sure that there's no problem on that net. i've had the
> same warning on a design in which the offending net wasn't used as a
> clock anywhere.

I finally pinpointed the warning to a part of my design driven by a
clock which is 1/8th of the fastest input clock. I guess this warning
can't be avoided in this case since the 1/8 clock has to be implemented
somehow in the chip (using F/G/H maps).  But this is fine since the
required frequency is much lower.  It's odd though that F1.4 and
previous don't produce this warning.

If I have problems with skew, I'll just set a MAXSKEW attribute on that
clock net and hope that it will fix them


Thanks for your input.

Nestor
Article: 12315
Subject: Re: Degradation of results from Xilinx F1.3 -> F1.4 -> F1.5
From: Nestor Caouras <nestor@ece.concordia.ca>
Date: Thu, 08 Oct 1998 15:03:23 -0400
Links: << >>  << T >>  << A >>
A few more notes...

> >
> >What I did is use the xc4000 library from Synplify (I don't know what
> >tool you are using) to instantiate the BUFG and connect it as:
> >
> >fsa -> BUFG -> clk_net
> 
> Is Synplify smart enough to automatically insert clock buffers? FPGA Express
> (which I use) does; so does Metamor.

Synplify can insert them automatically, but given all the different
types of small problems I have seen and read about when implementation
in FPGAs, I always instantiate special components (BUFG's, or other)
manual in the VHDL code or in the schematic. Also, the STARTUP block is
also inserted automatically in Synplify, provided that option is
selected.  I still manually instantiate the STARTUP block because I like
to see it in the schematic that Synplify generates. Components that
Synplify instantiates automatically are not displayed on the RTL view it
produces, but includes them in the edif or xnf it generates.


> 
> >and use the following constraints to cover all the nets renamed to
> >clk_net_c or something similar (luckily Synplify only appends characters
> >to the VHDL net names):
> >
> >NET fsa* TNM = fsa_grp;
> >NET clk* TNM = fsa_grp;
> >TIMESPEC TS01 = PERIOD fsa_grp 25ns;
> >
> >I think I may need to use the new timing name net constraint as:
> >
> >NET fsa* TNM_NET = fsa_grp;
> >NET clk* TNM_NET = fsa_grp;
> >TIMESPEC TS01 = PERIOD fsa_grp 25ns;
> 
> Try it and let us know what happens.

Not much difference.  The operating frequency still ended up being
around 33MHz.


Thanks again.

Nestor
Article: 12316
Subject: Xilinx F1.5/FPGA Express wackiness
From: "Andy Peters" <apeters@noao.edu.NOSPAM>
Date: Thu, 8 Oct 1998 12:11:10 -0700
Links: << >>  << T >>  << A >>
Hey, all:

I've been futzing with F1.5 and FPGA Express for about a week now...here's
some observations:

1) Supposedly there's a way to keep revs of source code with implementation
revs but I can't figure out how to do it.

2) In Synthesis Options for FPGA Express, when I checked the box labelled
"Export Timing Constraint," I got SIXTY-TWO timespecs, ranging from the
sublime (pads to pads) to the ridiculous (output tristate flipflops to
output tristate flipflops????).

3) After synthesizing, click on the implementation button and then choose
"Options" for physical implementation settings.  It won't let you choose a
Simulation Program Template other than Foundation EDIF.  Even if you select
Generic VHDL or ModelSim VHDL, once you start the implementation it throws
your choice away and chooses Foundation EDIF.

4) The could-be-nifty Constraints Editor GUI (not accessible from the
Project Manager; rather it's buried in the program manager/start menu under
Programs->Xilinx->Accessories->Constraints Editor) requires as input both a
.UCF AND a .NGD file.  Unfortunately, the .NGD file isn't available until
after you do an implementation.

5) Something got into my .UCF file and changed my timing ignore for
bidirectional data bus pins
    TIMESPEC "TS_FALSE" = FROM "FFS" THRU "ramthru" TO "FFS"  TIG;

to

    TIMESPEC "TS_FALSE" = FROM "FFS"  TO "FFS"  TIG;

which had the amusing-yet-useless effect of cause all timing constraints to
be ignored by PPR.

-andy

--------------------------------------------------------------------------
Andy Peters
Sr. Electrical Engineer
National Optical Astronomy Observatories
950 N Cherry Ave
Tucson, AZ 85719
520-318-8191
apeters@noao.edu
"Stupidity is like hydrogen: it's universal.  It's everywhere and in
everything."
    -- Frank Zappa


Article: 12317
Subject: Re: LCELL delay of Altera 10K's
From: yilmaza@removethis.geocities.com (emmanuel jolly)
Date: Thu, 08 Oct 1998 19:58:39 GMT
Links: << >>  << T >>  << A >>
LCell time depends on Component family and speed grade (10K, 10KA,
10KE, -4, -3, -2, -1).
It is not a good design rule to use LCELL to implement delay. Use it
only if you have no other solution

Regards

On Fri, 9 Oct 1998 01:08:15 +0800, "John Huang"
<hungi@tpts4.seed.net.tw> wrote:

>Hi all:
>
>    Does anyone know the delay time of one LCELL,
>if I implement an VHDL statement
>
>  U1: LCELL(CLK1, CLK2);
>
>  the CLK2 will be delay how many ns?
>
>    John Huang
>
>

---
Pour me contacter enlever "removethis." de mon adresse Email.
To contact me, remove "removethis." from my Email address.
Article: 12318
Subject: Re: Altera MAXPLUS2 V9 slow.
From: yilmaza@removethis.geocities.com (emmanuel jolly)
Date: Thu, 08 Oct 1998 20:03:37 GMT
Links: << >>  << T >>  << A >>
I have compared compilation times between version 8.14, 8.2, 8.3 and
9.01 and have seen no significative difference with one of my designs
written in AHDL. (95% full 10K50A compiles in approx. 7' on a PII333
with 192MBytes of RAM and Win NT4). However I must point out that I
have no timing constraints on my design.
I think the compilation time can increase if you have timing
constraints.
Regards.
E. JOLLY

On 8 Oct 1998 13:07:26 GMT, "Eric Pearson"
<ecp@focus-systems.nospam.on.ca> wrote:

>Has anyone noticed problems in re-compiling designs that worked in v8.0 of
>maxplus2 under v9.01 and found comilation times went throught the roof even
>to the point excess ( I aborted after 100 hours v9 vs 1 hour under v8 ).
>
>Eric Pearson - Focus Systems
>

---
Pour me contacter enlever "removethis." de mon adresse Email.
To contact me, remove "removethis." from my Email address.
Article: 12319
Subject: FPGA core design
From: jamil.khatib@pemail.net (J. Khatib)
Date: Thu, 08 Oct 1998 20:38:03 GMT
Links: << >>  << T >>  << A >>
Please help 

I want to design IP cores for FPGAs. Please can you point to where to
start from, what should I read, where are the basics of IP cores
design methods, what are the standards,where could I get the tools for
this reason and anything related to this topic

thanks in advance
Article: 12320
Subject: Re: FPGA core design
From: msimon@tefbbs.com
Date: Thu, 08 Oct 1998 21:04:10 GMT
Links: << >>  << T >>  << A >>


I have a nifty - I think - tool for designing cores.

Check out:


Design Your Own MicroProcessor(tm) http://www/tefbbs.com/spacetime/index.htm
Article: 12321
Subject: Software tool
From: "Ido Kleinman" <kleinn@mail.biu.ac.il>
Date: Fri, 9 Oct 1998 00:42:49 +0200
Links: << >>  << T >>  << A >>
Dear all,

I am looking for a good VHDL development (compiling, simulation, gate-level
sim/FSM support, Testbench generator) software tool. I need it to have a
comfortable interface to maxplus2 for exporting compiled designs for
synthesis in Altera's devices.
I've been looking around lately and I've evaluated Aldec's "Active-VHDL" and
quite happy with it, but I have a slow internet connection, therefore,
before I start downloading tens of megabytes again, I would like to know if
there are any other compact software tools worth looking at...?

Anyone got any experience with Aldec's tool?
Are the big ones (Synopsys, Examplar..) worth the investment?


--


 Ido Kleinman.
 kleinn@mail.biu.ac.il




Article: 12322
Subject: Software tool
From: "Ido Kleinman" <kleinn@mail.biu.ac.il>
Date: Fri, 9 Oct 1998 00:42:49 +0200
Links: << >>  << T >>  << A >>
Dear all,

I am looking for a good VHDL development (compiling, simulation, gate-level
sim/FSM support, Testbench generator) software tool. I need it to have a
comfortable interface to maxplus2 for exporting compiled designs for
synthesis in Altera's devices.
I've been looking around lately and I've evaluated Aldec's "Active-VHDL" and
quite happy with it, but I have a slow internet connection, therefore,
before I start downloading tens of megabytes again, I would like to know if
there are any other compact software tools worth looking at...?

Anyone got any experience with Aldec's tool?
Are the big ones (Synopsys, Examplar..) worth the investment?


--


 Ido Kleinman.
 kleinn@mail.biu.ac.il




Article: 12323
Subject: Re: FPGA core design
From: "Ken Coffman" <kcoffman@intermec.com>
Date: 8 Oct 1998 23:18:09 GMT
Links: << >>  << T >>  << A >>
Start with the Reuse Methodology Manual, you can get it from
www.computerliteracy.com

J. Khatib <jamil.khatib@pemail.net> wrote in article
<361e229f.261872@news.planet.edu>...
> Please help 
> 
> I want to design IP cores for FPGAs. Please can you point to where to
> start from, what should I read, where are the basics of IP cores
> design methods, what are the standards,where could I get the tools for
> this reason and anything related to this topic
> 
> thanks in advance
> 
Article: 12324
Subject: VHDL'93 in MaxPlus
From: "Ido Kleinman" <kleinn@mail.biu.ac.il>
Date: Fri, 9 Oct 1998 01:38:30 +0200
Links: << >>  << T >>  << A >>
Hi,
Anyone knows how do I activate VHDL93 compiling ability in MaxPlus2?

I wrote some code with direct component instanciation that would compile
nice in Aldec, but won't in Maxplus...

MaxPlus specifies many VHDL93 constructs in its help so I guess the VHDL93
compiling ability exists. (I have the full software, no "unavailable
features"...)

--


 Ido Kleinman.
 kleinn@mail.biu.ac.il






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