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 23950

Article: 23950
Subject: Re: Altera fitter woes
From: Ken McElvain <ken@synplicity.com>
Date: Mon, 17 Jul 2000 16:50:32 -0700
Links: << >>  << T >>  << A >>
Two possibilities:
    1) Make sure that you have requested "fast I/Os" in the global logic
synthesis options
    1) Preset flops can't be merged into the I/Os in Flex devices.  Use
either no reset at all or a clear.


martin@gubbins.demon.co.uk wrote:

> I've been banging my head against the Max+Plus2 fitter all day.  I
> can't get some of my I/O's to meet a 7.5ns tco - which I have
> specified in the ACF file.  It would appear that the routing delay
> from my mux to my output is too long.  No problem I'll stick another
> DFF on the output of the mux to break the route into two.
>
> What does the fitter do? Slap the extra FF right next to the mux
> output FF in the same LAB.  Neatly using the extremely fast local-LAB
> interconnect, but still having the same long route to the pin that I
> was trying to avoid!
>
> Aaargh!  Anyone got any ideas on how to make the fitter do something
> sensible?
>
> The device is a 10KE100EBC356-1, MP2 v 9.6, using the Quartus fitter
> if it makes any odds...
>
> Any help greatly appreciated!
>
> Cheers,
> Martin
> Martin Thompson
> martin@gubbins.demon.co.uk
> http://www.gubbins.demon.co.uk/

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


Article: 23951
Subject: Re: Xilinx Foundation 2.1 Run Times
From: krw@attglobal.net (Keith R. Williams)
Date: 18 Jul 2000 00:45:44 GMT
Links: << >>  << T >>  << A >>
On Mon, 17 Jul 2000 12:54:36, Keith Wootten 
<Keith@wootten.demon.co.uk> wrote:

> Hi
> 
> My present XCS40XL design takes about 25 minutes from editing to .bit
> file generation.  I'm using a 200MHz Pentium Pro with 64MB RAM and
> Windoze 95.

I wanna know how you're getting it so low!  My design will take 
upwards of an hour on a PIII-650 and then another hour or with 
the "re-entrant router".  I had to strip the design down to the 
bare-bones to get it to route at all.  My DC setup stuff, much of
the design, I had to do in shift-registers because it wouldn't 
route otherwise.  BTW, 3.1i improved things considerbly, though 
it still take hours (and it is very unpredictable). 

> What sort of ballpark speed improvement could I get by using a faster PC
> and/or more RAM?  Has anyone run the same design on different machines?  

With 2.1i I found that it was CPU limited, at least with 256K or 
greater memory.  The processing times scaled pretty lineraly 
between a 333MHz PII and a 650MHz PIII.  I was rather surprised 
actually.

> I need to balance run time improvements against the hassle of upgrading
> my PC.

Understood.  I run my other design tools on my antique PII-333 
(or on my laptop - depending) and have the PIII-650 sitting there
to do PnR (it's the host for the target of the design, so it's 
not being used at the time).  A super-fast system doesn't do a 
lot for Synplicity, in fact I've used my P5-120 laptop without 
too much agrivation, but a fast machine is critical for the 
Xilinx PnR tools.  It's worth it to transfer files between two 
machines, even over a sneraker-net.

----
  Keith
Article: 23952
Subject: Re: FPGA Intro
From: "Tony Burch" <tony@BurchED.com.au>
Date: Tue, 18 Jul 2000 14:53:43 +1000
Links: << >>  << T >>  << A >>
Nick Young wrote in message ...
>I am new to FPGA's and want to learn how to use them.  This includes
>designing, simulation and implementation.  Can anyone tell me of a good
>introduction/tutorial on the topic, (on-line if possible).
>I am new to FPGA's and want to learn how to use them.  This includes
>designing, simulation and implementation.  Can anyone tell me of a good
>introduction/tutorial on the topic, (on-line if possible).

Hi Nick,  Burch Electronic Designs have some low cost kits that
may be of interest for the implementation part of your learning.

The kits are intended for real-world prototyping and for quickly
trying out ideas in the lab with a basic and easy to use starting
platform.  However, they are also great starting platforms
for student projects.  We have Xilinx, Altera, Atmel, Actel and Lucent
kits.  Prices range from US$66 to US$89.

Check out the kits at www.BurchED.com.au

The Altera kit is popular for students, basically because the
Altera Max+PlusII Baseline software, which works great with
the kit, is freely downloadable from the Altera website.
See a picture and specs of the BED-ALTERA-BASE+ kit at
http://www.burched.com.au/bedalterabase.html
We have also written a free "jumpstart guide" for using
the Altera tools at http://www.burched.com.au/downloads.html

I must also mention that if you are looking for a learning
platform that already has things like an external micro,
memory, peripheral connectors, 7 segment / LED displays etc.
already on the board, then the BurchED kits are less
appropriate that some other kits on the market.
However, with the BurchED kits, you can always add
the LEDs, switches, and peripheral connectors that you need.
That's the idea - only add what you need for your
application / prototype.

Best regards
Tony Burch
www.BurchED.com.au





Article: 23953
Subject: download to a xilinx fpga
From: "Andreas Wüstefeld" <andreas.wuestefeld@ertec.com>
Date: Tue, 18 Jul 2000 09:11:40 +0200
Links: << >>  << T >>  << A >>
Hi,

How do I have to connect the peripherial components to the FPGA in order to
correctly download the .bit file with the parrallel cabel III.
My error message is:

    Checking boundary-scan chain integrity...ERROR:JTag - Boundary-scan
chain test failed at bit position     '2' on instance
'<project_name>(Device1)'.
    Check that the cable, system and device JTAG TAP connections are
correct,....

Where can I get information about configuring the clocks in the options
menue of the Foundation 2.i?
I'm using a XCS40.

thanks Andreas


Article: 23954
Subject: Re: Altera fitter woes
From: martinthompson@my-deja.com
Date: Tue, 18 Jul 2000 07:39:24 GMT
Links: << >>  << T >>  << A >>
Hi Ken, I'm posting from Deja at work as I haven't got 'proper' Usenet access
from here!

I have ticked the fast I/O option, and the pins are cleared at reset.... but
the pins are bidirectional (I knew I'd missed something important out!) and
the fitter is using the IOC for the input reg (presumably to meet the tsu
time).

Thanks for the suggestions though!

Cheers,
Martin


In article <39739BC8.3D541A06@synplicity.com>,
  Ken McElvain <ken@synplicity.com> wrote:
> Two possibilities:
>     1) Make sure that you have requested "fast I/Os" in the global logic
> synthesis options
>     1) Preset flops can't be merged into the I/Os in Flex devices.  Use
> either no reset at all or a clear.
>



Sent via Deja.com http://www.deja.com/
Before you buy.
Article: 23955
Subject: Re: Altera fitter woes
From: martinthompson@my-deja.com
Date: Tue, 18 Jul 2000 07:39:25 GMT
Links: << >>  << T >>  << A >>
Hi Ken,
I'm posting from Deja at work as I haven't
got 'proper' Usenet access from here!

I have ticked the fast I/O option, and the pins
are cleared at reset.... but
the pins are bidirectional (I knew I'd missed
something important out!) and the fitter is using
the IOC for the input reg (presumably to meet the
tsu time).

Thanks for the suggestions though!

Cheers,
Martin
In article <39739BC8.3D541A06@synplicity.com>,
  Ken McElvain <ken@synplicity.com> wrote:
> Two possibilities:
>     1) Make sure that you have requested "fast
I/Os" in the global logic
> synthesis options
>     1) Preset flops can't be merged into the
I/Os in Flex devices.  Use
> either no reset at all or a clear.
>


Sent via Deja.com http://www.deja.com/
Before you buy.
Article: 23956
Subject: CHES 2000 Program
From: Christof Paar <christof@ece.wpi.edu>
Date: Tue, 18 Jul 2000 07:30:15 -0400
Links: << >>  << T >>  << A >>

                PRELIMINARY SCHEDULE 
                     CHES 2000

-------------------------------------------------------
Workshop on Cryptographic Hardware and Embedded Systems
                     
            Worcester Polytechnic Institute
     Worcester, Massachusetts, August 17-18, 2000
        http://ece.wpi.edu/Research/crypt/ches

for registration information, check CHES web site above
-------------------------------------------------------


===============  WEDNESDAY, AUGUST 16th  ==============  

5:00 - 10:00 pm  REGISTRATION AND RECEPTION

Atwater Kent Building, WPI
(Shuttle Service will be provided between WPI and the Crowne Plaza Hotel)
            


===============  THURSDAY, AUGUST 17th  ===============

7:00 - 8:45 am  REGISTRATION AND CONTINENTAL BREAKFAST

Atwater Kent Building, WPI
(Shuttle Service will be provided between WPI and the Crowne Plaza Hotel)

            
8:45 - 9:15 am WELCOME

Welcome by Jack Carney (Provost, WPI) 
Introductory remarks by Cetin Koc and Christof Paar


9:15 - 9:55 am  INVITED TALK

Alfred Menezes, University of Waterloo, Canada
Elliptic curve cryptography in constrained environments.


9:55 - 10:55 am  IMPLEMENTATION OF ELLIPTIC CURVE CRYPTOSYSTEMS

S. Okada, N. Torii, K. Ito, and M. Takenaka.
Implementation of elliptic curve cryptographic coprocessor over GF(2^m) on FPGA.

G. Orlando and C. Paar.
A high-performance reconfigurable elliptic curve processor for GF(2^m).

J. W. Chung, S. S. Gyoo, and L. P. Joong.
Fast implementation of elliptic curve defined over GF(p^m) on CalmRISC with 
MAC2424 coprocessor.


10:55 - 11:15 am BREAK


11:15 - 12:35 pm POWER AND TIMING ANALYSIS ATTACKS

A. Shamir.
Protecting smart cards from passive power analysis with detached power supplies.

R. Mayer-Sommer.
Smartly analyzing the simplicity and the power of simple power analysis on 
Smartcards.

M. A. Hasan.
Power analysis attacks and algorithmic approaches to their countermeasures for 
Koblitz curve cryptosystems.

W. Schindler.
A timing attack against RSA with the Chinese Remainder Theorem.


12:35 - 2:00 pm LUNCH


2:00 - 3:20 pm  HARDWARE IMPLEMENTATION OF BLOCK CIPHERS

A. Dandalis, V. K. Prasanna, and J. D. P. Rolim.
A comparative study of performance of AES final candidates using FPGAs.

C. Patterson.
A dynamic FPGA implementation of the Serpent Block Cipher.

S. Trimberger, R. Pang, and A. Singh.
A 12 Gbps DES Encryptor/Decryptor core in an FPGA.

H. Leitold, W. Mayerwieser, U. Payer, K. C. Posch, R. Posch, and J. 
Wolkerstorfer.
A 155 Mbps triple-DES network encryptor.


3:20 - 3:40 pm  BREAK


3:40 - 5:00 pm  HARDWARE ARCHITECTURES

J. Goodman and A. Chandrakasan.
An energy efficient reconfigurable public-key cryptography processor 
architecture.

J. Groszschaedl.
High speed RSA hardware based on dynamic true single phase clocked logic.

C. Walter.
Data integrity in hardware for modular arithmetic.

T. Kato, S. Ito, J. Anzai, and N. Matsuzaki.
A design for modular exponentiation coprocessor in mobile telecommunication 
terminals.


6:00 - 9:00 pm  BANQUET



===============  FRIDAY, AUGUST 18TH  ==============

8:00 - 9:00 am  REGISTRATION AND CONTINENTAL BREAKFAST

Atwater Kent Building, WPI
(Shuttle Service will be provided between WPI and the Crowne Plaza Hotel)


9:00 - 9:40 am  INVITED TALK

David Naccache, Gemplus, France.
How to explain side channel leakage to your kids.


9:40 - 10:40 am POWER ANALYSIS ATTACKS

J.-S. Coron and L. Goubin.
On Boolean and arithmetic masking against differential power analysis.

T. S. Messerges.
Using second-order power analysis to attack DPA resistant software.

C. Clavier, J.-S. Coron, and N. Dabbous.
Differential power analysis in the presence of hardware countermeasures.



10:40 - 11:00 am BREAK


11:00 - 12:00 pm ARITHMETIC ARCHITECTURES

H. Wu.
Montgomery multiplier and squarer in GF(2^m).

E. Savas, A. F. Tenca, and C. K. Koc.
A scalable and unified multiplier architecture for finite fields GF(p) and 
GF(2^m).

G. Hachez and J.-J. Quisquater.
Montgomery exponentiation with no final subtraction: Improved results.


12:00 - 1:30 pm  LUNCH


1:30 - 2:10 pm  PHYSICAL SECURITY AND CRYPTANLYSIS

S. H. Weingart.
Physical security devices for computer subsystems: A survey of attacks and 
defenses.

T. Pornin and J. Stern.
Software-Hardware Trade-offs: application to A5/1 Cryptanalysis.


2:10 - 2:50 pm  NEW SCHEMES AND ALGORITHMS

J. Hoffstein and J. Silverman.
MiniPASS: Authentication and digital signatures in a constrained environment.

M. Joye, P. Paillier, and S. Vaudenay.
Efficient generation of prime numbers.



Article: 23957
Subject: Re: Silicon Valley Housing Nightmare?
From: Scott Bilik <scott@vautomationNOSPAM.com>
Date: 18 Jul 2000 08:49:27 -0400
Links: << >>  << T >>  << A >>
>>>> Don Husby wrote:

    Don> This is way off topic.
    Don> I'm thinking it's time to leave my cushy science job in Chicago and
    Don> move to the West Coast. The scariest aspect of this is the nightmare
    Don> stories I've heard about the difficulties of finding a place to live
    Don> in Silicon Valley and the Bay area. Is it really that bad? Are there
    Don> areas that are better/worse than others? Will I be spending hours
    Don> per day in my car trying to get to work?

>>>>> "B" == B Joshua Rosen <bjrosen@polybus.com> writes:

    B> You may want to consider the East coast instead. Massachusetts is
    B> booming, there are wall to wall startups and the big companies like
    B> Sun, Cisco, Lucent are all building huge facilities. The housing is
    B> expensive compared to anywhere except Silicon Valley, by Valley
    B> standards its cheap, you should be able to buy a decent house for
    B> $350, there are million dollar houses but at least they are
    B> 5000sq/feet and have a couple of acres of land, in Silicon Valley a
    B> million dollars gets you a phone booth. The commutes are reasonable,
    B> the Chinese food isn't as good as the Valley but the Italian food is
    B> better.

I would second this. I was raised in the Chicago suburbs for the first 22
years of my life. I moved to 'Silicon Valley East' a little over 5 years
ago. Actually I wanted to make a small plug for living in southern New
Hampshire. There are _many_ startups both here and in the north of Boston
area that are _struggling_ to find EE's and software guys. I know we at
VAutomation certainly are! 

Nashua,NH was twice picked as Money magazine's best city in the US. It's
funny to listen to local folks talk about how expensive property is getting
because it's all relative. Nice 3-4 bedroom homes in the Nashua area run
between $150K and $175K. What does that buy in the Bay area, a toilet?!
Cross the MA border, though, and it gets more difficult to find something
_that_ low. Even so, they're still less than the Bay area. 

If anyone would like to crunch HDL code on FPGA's with us, give me a
call.... please! In 25 words or less: VHDL, ModelSim, Leonardo & Design
Compiler, Xilinx Virtex, processor and interface core development, a great
group of engineers, and now stock shares in ARC. Fun stuff!

-- 
Scott Bilik                              VAutomation Inc.
http://www.vautomation.com               402 Amherst St #100
(603) 882-2282 x24                       Nashua NH 03063
Article: 23958
Subject: Re: Altera fitter woes
From: "Eric Pearson" <ecp@focus-systems.nospam.on.ca>
Date: Tue, 18 Jul 2000 10:14:05 -0400
Links: << >>  << T >>  << A >>
Hi Martin...

To meet timing in that situation, I explicitly place both input and output
DFFs.
Place the output DFF in the IOB and the input DFF in the nearest cell. The
setup time
is actually shorter in this case (due to the bonus input delay on the IOBs).

I generally lock down all I/O flipflops, because our system is
re-configurable, yet must always meet the same external timing. This way I
can just worry about register-register timing.

Eric


martinthompson@my-deja.com wrote in message <8l11je$tsc$1@nnrp2.deja.com>...
>Hi Ken,
>I'm posting from Deja at work as I haven't
>got 'proper' Usenet access from here!
>
>I have ticked the fast I/O option, and the pins
>are cleared at reset.... but
>the pins are bidirectional (I knew I'd missed
>something important out!) and the fitter is using
>the IOC for the input reg (presumably to meet the
>tsu time).
>
>Thanks for the suggestions though!
>
>Cheers,
>Martin
>In article <39739BC8.3D541A06@synplicity.com>,
>  Ken McElvain <ken@synplicity.com> wrote:
>> Two possibilities:
>>     1) Make sure that you have requested "fast
>I/Os" in the global logic
>> synthesis options
>>     1) Preset flops can't be merged into the
>I/Os in Flex devices.  Use
>> either no reset at all or a clear.
>>
>
>
>Sent via Deja.com http://www.deja.com/
>Before you buy.


Article: 23959
Subject: xilinx prom 17s05lvc Need programming specification
From: "Lino de Martin" <lino.de.martin@claranet.fr>
Date: Tue, 18 Jul 2000 16:15:05 +0200
Links: << >>  << T >>  << A >>
Bonjour,

I can't read the "manufacturer/ID code" of a xilinx 17s05lvc prom.

I power up the prom with Vcc=Vpp= 5.0v then
I put  ce and oe high and apply 12.25V at Vpp then
2 pulses on clk then
5.0V at Vpp then clk=1 to enter in programming mode.
I put oe low and pulse 4600 clk then oe high then ce low and I read :
0011 0110 0111 1111 in place of
1100 1001 1000 0111 = C987
C9 for xilinx manufacturer ID,
8 for S05XL prom
7 for the algo code.

What's wrong ?

Merci de vos reponses.
lino.de.martin@claranet.fr



Article: 23960
Subject: Re: Altera fitter woes
From: "Eric Pearson" <ecp@focus-systems.nospam.on.ca>
Date: Tue, 18 Jul 2000 10:15:25 -0400
Links: << >>  << T >>  << A >>
Late breaking news...

turn off the quartus fitter if you want to make DFF location assignments.

<sigh>

Eric Pearson


Eric Pearson wrote in message ...
>Hi Martin...
>
>To meet timing in that situation, I explicitly place both input and output
>DFFs.
>Place the output DFF in the IOB and the input DFF in the nearest cell. The
>setup time
>is actually shorter in this case (due to the bonus input delay on the
IOBs).
>
>I generally lock down all I/O flipflops, because our system is
>re-configurable, yet must always meet the same external timing. This way I
>can just worry about register-register timing.
>
>Eric
>
>
>martinthompson@my-deja.com wrote in message
<8l11je$tsc$1@nnrp2.deja.com>...
>>Hi Ken,
>>I'm posting from Deja at work as I haven't
>>got 'proper' Usenet access from here!
>>
>>I have ticked the fast I/O option, and the pins
>>are cleared at reset.... but
>>the pins are bidirectional (I knew I'd missed
>>something important out!) and the fitter is using
>>the IOC for the input reg (presumably to meet the
>>tsu time).
>>
>>Thanks for the suggestions though!
>>
>>Cheers,
>>Martin
>>In article <39739BC8.3D541A06@synplicity.com>,
>>  Ken McElvain <ken@synplicity.com> wrote:
>>> Two possibilities:
>>>     1) Make sure that you have requested "fast
>>I/Os" in the global logic
>>> synthesis options
>>>     1) Preset flops can't be merged into the
>>I/Os in Flex devices.  Use
>>> either no reset at all or a clear.
>>>
>>
>>
>>Sent via Deja.com http://www.deja.com/
>>Before you buy.
>
>


Article: 23961
Subject: Re: SerialProm programmer
From: "Lino de Martin" <lino.de.martin@claranet.fr>
Date: Tue, 18 Jul 2000 16:23:55 +0200
Links: << >>  << T >>  << A >>
Bonjour,

I see a tiny programmer at http://www.mvd-fpga.com/

lino.de.martin@claranet.com



Article: 23962
Subject: Re: Altera fitter woes
From: Martin <martin.j.thompson@trw.com>
Date: Tue, 18 Jul 2000 14:50:42 GMT
Links: << >>  << T >>  << A >>
Hmm, I was hoping to avoid turning off the Quartus fitter, as it takes about
three times as long to fit.  Ho hum!  Thanks for the suggestion, I'll give it
a try.

Cheers,
Martin



In article <DAZc5.22241$227.416217@nnrp1.uunet.ca>,
  "Eric Pearson" <ecp@focus-systems.nospam.on.ca> wrote:
> Late breaking news...
>
> turn off the quartus fitter if you want to make DFF location assignments.
>
> <sigh>
>
> Eric Pearson
>
> Eric Pearson wrote in message ...
> >Hi Martin...
> >
> >To meet timing in that situation, I explicitly place both input and output
> >DFFs.
> >Place the output DFF in the IOB and the input DFF in the nearest cell. The
> >setup time
> >is actually shorter in this case (due to the bonus input delay on the
> IOBs).
> >
> >I generally lock down all I/O flipflops, because our system is
> >re-configurable, yet must always meet the same external timing. This way I
> >can just worry about register-register timing.
> >
> >Eric
> >
> >
> >martinthompson@my-deja.com wrote in message
> <8l11je$tsc$1@nnrp2.deja.com>...
> >>Hi Ken,
> >>I'm posting from Deja at work as I haven't
> >>got 'proper' Usenet access from here!
> >>
> >>I have ticked the fast I/O option, and the pins
> >>are cleared at reset.... but
> >>the pins are bidirectional (I knew I'd missed
> >>something important out!) and the fitter is using
> >>the IOC for the input reg (presumably to meet the
> >>tsu time).
> >>
> >>Thanks for the suggestions though!
> >>
> >>Cheers,
> >>Martin
> >>In article <39739BC8.3D541A06@synplicity.com>,
> >>  Ken McElvain <ken@synplicity.com> wrote:
> >>> Two possibilities:
> >>>     1) Make sure that you have requested "fast
> >>I/Os" in the global logic
> >>> synthesis options
> >>>     1) Preset flops can't be merged into the
> >>I/Os in Flex devices.  Use
> >>> either no reset at all or a clear.
> >>>
> >>
> >>
> >>Sent via Deja.com http://www.deja.com/
> >>Before you buy.
> >
> >
>
>


Sent via Deja.com http://www.deja.com/
Before you buy.
Article: 23963
Subject: Re: Quartus
From: "Jaap H. Mol" <jh_mol@wxs.nl>
Date: Tue, 18 Jul 2000 22:20:15 +0200
Links: << >>  << T >>  << A >>

--------------D737631738975EBB9988E6E0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit



Alan wrote:

> In article <3969D673.573FE3B5@voila.fr>, christelle <van-den-
> broucke@voila.fr> writes
> >Hi all,
> >
> >Is there anyone using Quartus?
> >I've heard that there are lots of problems with this tool...
> >
>
> In my experience, don't bother with a UNIX ported version (Ultra 60 -
> 700Mbytes RAM) - it's the only program that I've known to core dump on a
> regular basis :(
>
> Quartus appears to run well on a 900MHz PC with 1 Gbyte of RAM. Anything
> less and again I have found it to be painfully slow and, more serious,
> prone to various unexplained error messages/crashes.
>
> I'm currently attempting to fit a 400k gate equivalent design with 380
> I/O pins into the APEX 20KE600 device. It appears have gone in and met
> timing but it will be a two/three weeks before I can establish if the
> post fitting simulations work as expected.

Hi,

You're probably referring to the fact that post-layout (timing) simulation
takes for ages, because
of the backannotated (SDF) delays, and all associated timing-checks.... ;-)
Trying using (static) timing analysis instead (Quartus Timing Analyzer), to
verify the TIMING requirements of your design. No testbench and/or vectors
needed for this, and 100% coverage of all timing paths (not just what
you believe to be critical....)
If you don't trust the functionality anymore after fitting/optimization by
Quartus, export a
VHDL/Verilog netlist, and simulate without backannotating timing.

Best regards,

Jaap



>
> --
> Alan
> mailto:news4me@amacleod.clara.co.uk

--------------D737631738975EBB9988E6E0
Content-Type: text/html; charset=us-ascii
Content-Transfer-Encoding: 7bit

<!doctype html public "-//w3c//dtd html 4.0 transitional//en">
<html>
&nbsp;
<p>Alan wrote:
<blockquote TYPE=CITE>In article &lt;3969D673.573FE3B5@voila.fr>, christelle
&lt;van-den-
<br>broucke@voila.fr> writes
<br>>Hi all,
<br>>
<br>>Is there anyone using Quartus?
<br>>I've heard that there are lots of problems with this tool...
<br>>
<p>In my experience, don't bother with a UNIX ported version (Ultra 60
-
<br>700Mbytes RAM) - it's the only program that I've known to core dump
on a
<br>regular basis :(
<p>Quartus appears to run well on a 900MHz PC with 1 Gbyte of RAM. Anything
<br>less and again I have found it to be painfully slow and, more serious,
<br>prone to various unexplained error messages/crashes.
<p>I'm currently attempting to fit a 400k gate equivalent design with 380
<br>I/O pins into the APEX 20KE600 device. It appears have gone in and
met
<br>timing but it will be a two/three weeks before I can establish if the
<br>post fitting simulations work as expected.</blockquote>
Hi,
<p>You're probably referring to the fact that post-layout (timing) <u>simulation</u>
takes for ages, because
<br>of the backannotated (SDF) delays, and all associated timing-checks....
;-)
<br>Trying using (static) timing <u>analysis</u> instead (Quartus Timing
Analyzer), to verify the TIMING requirements of your design. No testbench
and/or vectors needed for this, and 100% coverage of all timing paths (not
just what
<br>you believe to be critical....)
<br>If you don't trust the <u>functionality</u> anymore after fitting/optimization
by Quartus, export a
<br>VHDL/Verilog netlist, and <u>simulate without</u> backannotating timing.
<p>Best regards,
<p>Jaap
<br>&nbsp;
<br>&nbsp;
<blockquote TYPE=CITE>&nbsp;
<br>--
<br>Alan
<br><a href="mailto:news4me@amacleod.clara.co.uk">mailto:news4me@amacleod.clara.co.uk</a></blockquote>
</html>

--------------D737631738975EBB9988E6E0--

Article: 23964
Subject: 104 Page Collective DAC'00 Trip Report Up
From: jcooley@world.std.com (John Cooley)
Date: Tue, 18 Jul 2000 20:43:16 GMT
Links: << >>  << T >>  << A >>
I just wanted to give ya'll a heads up that the massive 104 page DAC'00 Trip
Report is up on http://www.DeepChip.com for viewing.  It's a collective 
report based on 113 survey responses and/or individual DAC trip reports (it 
consists of quotes from those 113 engineers.)  Enjoy.

    - John Cooley

-----------------------------------------------------------------------------
  __))    "Glass ceilings?  Name ANY ex-goat farmer who's made management!"
 /_ oo  
  (_ \   Holliston Poor Farm                                   - John Cooley
%//  \"  Holliston, MA 01746-6222           part time Sheep & ex-Goat Farmer
%%%  $   jcooley@world.std.com       full time contract ASIC & FPGA Designer
Article: 23965
Subject: Re: Dual Port RAM
From: fliptron@netcom.com (Philip Freidin)
Date: 18 Jul 2000 21:53:10 GMT
Links: << >>  << T >>  << A >>
In article <bq6ums8i08fove3ebmjsuh8e2engqvgqlb@4ax.com>,
Muzaffer Kal  <muzaffer@kal.st> wrote:
>you can always use two blocks of memory and while one side writes to
>one of the blocks, the other side reads the other block. When both
>sides are done, you switch the blocks.

Usually call a ping-pong memory.

>Of course this assumes consumer
>is faster than producer. As long as you have enough bandwidth, true
>multi-port access is almost never needed. (In fact I can't think of
>any case where true multi-port access would be needed, any ideas ?)


Sure.

Register files in CPUs where you need to fetch multiple operands
every cycle. Even more so with architectures that can have multiple 
function units complete in the same cycle, and there are multiple results 
that need to be written back to the register file.

Buffer memories, where the incomming and/or outgoing data is bursty, and
the average data rate of incoming is less than or equal to the outgoing 
data rate. This situation can preclude a ping-pong topology, unless 
the buffers are very large. Depends on the statistics.

FIFOs are another structure that fairly much mandate a dual-port
memory internally. Especially a FIFO that has separate input
and output clocks.

Philip Freidin

Article: 23966
Subject: FPGA Conferences
From: "EKC" <NOSPAMalpha3.1@ix.netcom.com>
Date: Tue, 18 Jul 2000 22:09:41 GMT
Links: << >>  << T >>  << A >>
    Does anyone know whether or not there are any PLD-related conferences
scheduled to take place in the US before mid-September? If so, which ones?


Thanks,

EKC


Article: 23967
Subject: Re: download to a xilinx fpga
From: "Tony Burch" <tony@BurchED.com.au>
Date: Wed, 19 Jul 2000 10:28:59 +1000
Links: << >>  << T >>  << A >>
Andreas Wüstefeld wrote in message <8l1045$8ob$16$1@news.t-online.com>...
>Hi,
>
>How do I have to connect the peripherial components to the FPGA in order to
>correctly download the .bit file with the parrallel cabel III.
>My error message is:
>
>    Checking boundary-scan chain integrity...ERROR:JTag - Boundary-scan
...

Hi Andreas,
 I believe that your problem is that you are using the JTAG Programmer,
instead of the Hardware Debugger.
For an expanded answer see
http://www.burched.com.au/jtag.html

Best regards
Tony Burch
www.BurchED.com.au



Article: 23968
Subject: Re: 104 Page Collective DAC'00 Trip Report Up
From: "Robert Carney" <bobcarney@worldnet.att.net>
Date: Wed, 19 Jul 2000 03:12:07 GMT
Links: << >>  << T >>  << A >>
Just read your DAC report, John, and it was great.   The
amount of detail and the "unabridged" nature of the opinions
expressed made it the best DAC report I've ever read.
Now I'm concerned that you might get sucked into a chief
editor's job at one of the industry mags.  Keep up the good
work.

John Cooley <jcooley@world.std.com> wrote in message
news:Fxwuw5.AEB@world.std.com...
> I just wanted to give ya'll a heads up that the massive 104 page DAC'00 Trip
> Report is up on http://www.DeepChip.com for viewing.  It's a collective
> report based on 113 survey responses and/or individual DAC trip reports (it
> consists of quotes from those 113 engineers.)  Enjoy.
>
>     - John Cooley
>
> ----------------------------------------------------------------------------
-
>   __))    "Glass ceilings?  Name ANY ex-goat farmer who's made management!"
>  /_ oo
>   (_ \   Holliston Poor Farm                                   - John Cooley
> %//  \"  Holliston, MA 01746-6222           part time Sheep & ex-Goat Farmer
> %%%  $   jcooley@world.std.com       full time contract ASIC & FPGA Designer


Article: 23969
Subject: Re: FPGA Conferences
From: Simmler Harald <simmler@ti.uni-mannheim.de>
Date: Wed, 19 Jul 2000 07:56:54 +0200
Links: << >>  << T >>  << A >>

--------------0C398DDE2D3E98DD5F99CB9A
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

Have a look at the optimagic webside.
 CPLD, FPGA: The Programmable Logic Jump Station


EKC wrote:

>     Does anyone know whether or not there are any PLD-related conferences
> scheduled to take place in the US before mid-September? If so, which ones?
>
> Thanks,
>
> EKC

--

---------------------------------------------------------------------------
Harald Simmler                               Lehrstuhl fuer Informatik V
                                             Universitaet Mannheim
Tel:    +49-621-181-2632   ! NEW !           B6, 26
Fax:    +49-621-181-2634   ! NEW !           D-68131 Mannheim
eMail:   simmler@ti.uni-mannheim.de          Germany


--------------0C398DDE2D3E98DD5F99CB9A
Content-Type: text/html; charset=us-ascii
Content-Transfer-Encoding: 7bit

<!doctype html public "-//w3c//dtd html 4.0 transitional//en">
<html>
Have a look at the optimagic webside.
<br>&nbsp;<a href="http://www.optimagic.com/">CPLD, FPGA: The Programmable
Logic Jump Station</a>
<br>&nbsp;
<p>EKC wrote:
<blockquote TYPE=CITE>&nbsp;&nbsp;&nbsp; Does anyone know whether or not
there are any PLD-related conferences
<br>scheduled to take place in the US before mid-September? If so, which
ones?
<p>Thanks,
<p>EKC</blockquote>

<p>--
<p>---------------------------------------------------------------------------
<br>Harald Simmler&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
Lehrstuhl fuer Informatik V
<br>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
Universitaet Mannheim
<br>Tel:&nbsp;&nbsp;&nbsp; +49-621-181-2632&nbsp;&nbsp; ! NEW !&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
B6, 26
<br>Fax:&nbsp;&nbsp;&nbsp; +49-621-181-2634&nbsp;&nbsp; ! NEW !&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
D-68131 Mannheim
<br>eMail:&nbsp;&nbsp; simmler@ti.uni-mannheim.de&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
Germany
<br>&nbsp;</html>

--------------0C398DDE2D3E98DD5F99CB9A--

Article: 23970
Subject: Re: FPGA Conferences
From: Simmler Harald <simmler@ti.uni-mannheim.de>
Date: Wed, 19 Jul 2000 07:57:04 +0200
Links: << >>  << T >>  << A >>

--------------CFEFC5E74F2E7D73965A5A0A
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

Have a look at the optimagic webside.
 CPLD, FPGA: The Programmable Logic Jump Station


EKC wrote:

>     Does anyone know whether or not there are any PLD-related conferences
> scheduled to take place in the US before mid-September? If so, which ones?
>

--------------CFEFC5E74F2E7D73965A5A0A
Content-Type: text/html; charset=us-ascii
Content-Transfer-Encoding: 7bit

<!doctype html public "-//w3c//dtd html 4.0 transitional//en">
<html>
Have a look at the optimagic webside.
<br>&nbsp;<a href="http://www.optimagic.com/">CPLD, FPGA: The Programmable
Logic Jump Station</a>
<br>&nbsp;
<p>EKC wrote:
<blockquote TYPE=CITE>&nbsp;&nbsp;&nbsp; Does anyone know whether or not
there are any PLD-related conferences
<br>scheduled to take place in the US before mid-September? If so, which
ones?
<br>&nbsp;</blockquote>
</html>

--------------CFEFC5E74F2E7D73965A5A0A--

Article: 23971
Subject: IP CORE, 8250 core with 16byte fifo which only uses 100CLB's
From: "David Dart" <davidd@nulec.com.au>
Date: Wed, 19 Jul 2000 16:35:36 +1000
Links: << >>  << T >>  << A >>
Hi All,
          does any one know of a core which has the 8250 with a 16 byte fifo
(Or equivalent) and only uses approx 100 CLB's. I need to place 3 of these
in a XCS30XL with some room to spare.

thanks in advance.



--
David Dart,
 email davidd@nulec.com.au



Article: 23972
Subject: Re: Dual Port RAM
From: rickman <spamgoeshere4@yahoo.com>
Date: Wed, 19 Jul 2000 02:35:59 -0400
Links: << >>  << T >>  << A >>
Philip Freidin wrote:
> 
> In article <bq6ums8i08fove3ebmjsuh8e2engqvgqlb@4ax.com>,
> Muzaffer Kal  <muzaffer@kal.st> wrote:
> >you can always use two blocks of memory and while one side writes to
> >one of the blocks, the other side reads the other block. When both
> >sides are done, you switch the blocks.
> 
> Usually call a ping-pong memory.
> 
> >Of course this assumes consumer
> >is faster than producer. As long as you have enough bandwidth, true
> >multi-port access is almost never needed. (In fact I can't think of
> >any case where true multi-port access would be needed, any ideas ?)
> 
> Sure.
> 
> Register files in CPUs where you need to fetch multiple operands
> every cycle. Even more so with architectures that can have multiple
> function units complete in the same cycle, and there are multiple results
> that need to be written back to the register file.
> 
> Buffer memories, where the incomming and/or outgoing data is bursty, and
> the average data rate of incoming is less than or equal to the outgoing
> data rate. This situation can preclude a ping-pong topology, unless
> the buffers are very large. Depends on the statistics.

Actually, you can use a ping-pong buffer arrangement as long as you add
a "pang" buffer (a third block). This allows you to collect as much as
twice the normal data rate while empting the first buffer collected. If
you can't meet that requirement, then you will have a problem sizing
your buffers for any arrangement. 

 
> FIFOs are another structure that fairly much mandate a dual-port
> memory internally. Especially a FIFO that has separate input
> and output clocks.
> 
> Philip Freidin

-- 

Rick Collins

rick.collins@XYarius.com

Ignore the reply address. To email me use the above address with the XY
removed.



Arius - A Signal Processing Solutions Company
Specializing in DSP and FPGA design

Arius
4 King Ave
Frederick, MD 21701-3110
301-682-7772 Voice
301-682-7666 FAX

Internet URL http://www.arius.com
Article: 23973
Subject: Experts-Exchange
From: "Randy Given" <GivenRandy@aol.com>
Date: Wed, 19 Jul 2000 07:59:24 -0400
Links: << >>  << T >>  << A >>
If you think Usenet Newsgroups are helpful, wait until you try
Experts-Exchange.  It's a great place to ask technical computer
questions (software or hardware).  It's also a great place to give
advice if you know the answers.

It's free!  If you would, mention that GivenRandy referred you -- I
won't get any money, just some extra question points so that I can ask
more questions myself.  Hope to see you there!

http://www.experts-exchange.com/



Article: 23974
Subject: Re: Dual Port RAM
From: "Ian J. Smith" <ianjs@lucent.com>
Date: Wed, 19 Jul 2000 13:50:05 +0100
Links: << >>  << T >>  << A >>
We've had great success with Xilinx DP RAMs here. The tools are rather good too.
Altera still have some catching up to do.


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