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 20175

Article: 20175
Subject: Re: Xilinx vs Altera
From: raja <raja@elec.uq.edu.au>
Date: Sun, 30 Jan 2000 16:34:42 +1000
Links: << >>  << T >>  << A >>
This is a multi-part message in MIME format.
--------------9966CE9F98DDCDB17E66E923
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

i would suggest  xilinx foundation series( new version supports virtex
family too) rather than altera because  it has very good placement and
routing  ( floorplanner) than altera max plus

  all the best
kamal

Shawki Areibi wrote:

> We are introducing a new course on Digital Design at the School of
> Engineering
> at the University of Guelph. We would like to choose between the Altera
> MAXII-Plus and the Xilinx Foundation Series Software.
> Anyone with the experience of using the two systems? Which is better to
> use
> by the students (i.e simpler to use e.t.c)
> Thanks,
> Shawki Areibi
>
> --
> Shawki Areibi
> Assistant Professor
> School of Engineering
> University of Guelph
> Guelph, Ont, Canada N1G 2W1
> Tel: (519) 824-4120
> Fax: (519) 836-0227

--------------9966CE9F98DDCDB17E66E923
Content-Type: text/x-vcard; charset=us-ascii;
 name="raja.vcf"
Content-Transfer-Encoding: 7bit
Content-Description: Card for raja
Content-Disposition: attachment;
 filename="raja.vcf"

begin:vcard 
n:kamalanathan;Raja
tel;home:07  38761962
tel;work:07 33658849
x-mozilla-html:FALSE
adr:;;;;;;
version:2.1
email;internet:raja@elec.uq.edu.au
fn:Raja kamalanathan
end:vcard

--------------9966CE9F98DDCDB17E66E923--

Article: 20176
Subject: Re: Altera Quartus vs Xilinx Place and Route tools (help needed)
From: allan.herriman.hates.spam@fujitsu.com.au (Allan Herriman)
Date: Sun, 30 Jan 2000 06:46:30 GMT
Links: << >>  << T >>  << A >>
Ray,
	Peter Alfke kindly informed me via private email that my
assertion was totally wrong, and that the Virtex/Spartan2 DLLs have a
completely digital implementation.

Peter expressed some doubt that it was possible to implement a VCO
using a ring oscillator.  ISTR from my VLSI classes taken many moons
ago that such a VCO could be made by adjusting the supply voltage to
the delay elements in the ring oscillator.  This would still require
some analog elements (opamp + capacitor to make the loop filter), and
so would be more expensive than the DLL.

I'd be interested to know how Altera actually do it.

On a slightly different topic: *any* DLL or PLL will have some
sensitivity to noise on the supply votlage.  I haven't seen any
specifications for this.
(Recently I've built a few boards using Motorolla microprocessors with
onboard PLLs.  They recommended using separately filtered supplies for
the PLLs (which had separate supply pins for this purpose).)

Thanks,
Allan.

On Fri, 28 Jan 2000 17:41:52 GMT, Ray Andraka <randraka@ids.net>
wrote:

>I hadn't considered a VCO made as a ring oscillator.  I'll have to noodle on that.
>I'm pretty sure the DLL's loop filter function can be done digitally pretty easily
>(require more than one phase miss before moving to a new delay tap).
>
>Allan Herriman wrote:
>
>> On Thu, 27 Jan 2000 15:26:31 GMT, Ray Andraka <randraka@ids.net>
>> wrote:
>>
>> >Oh, one more thing,  The DLL is implemented basically as delay gates with
>> >feedback, which allows it to be done in the same process as the SRAM
>> >technology for the rest of the chip, with virtually nothing extra added to the
>> >process or yield testing.  It's a totally digital circuit.  That means the
>> >incremental cost of adding the DLL is virtually nothing after the NRE.
>> >
>> >Contrast that with a PLL, which by nature is an analog animal.  That means
>> >extra care is required in the fab, as there are parameters that have to be met
>> >which would not have to be met for a strictly digital circuit.
>>
>> Ray, I disagree with your statements about DLLs and PLLs.  Both are
>> implemented with delay lines made from gates.  Both use feedback to
>> adjust the gate delay.  Both have an analog loop filter (but the DLL
>> loop filter is simpler).
>>
>> About the only difference is that the PLL has the output of its delay
>> line fed back to the input to make a ring oscillator.  The DLL has its
>> input connected to the input clock signal.  This should give the DLL
>> better jitter performance, but the quoted specs (A vs X) are
>> identical.
>>
>> (BTW, I'm making some guesses about the implementations.  It is
>> possible that they have been implemented in other ways.)
>>
>> Regards,
>> Allan.
>>
>> >Magnus Homann wrote:
>> >
>> >> Steve Dewey <steve@s-dewey123.demon.co.uk> writes:
>> >>
>> >> > One point in this debate is that Xylinx seem to put the DLLs into ALL
>> >> > the Virtex and Spartan parts. Altera definitely only puts them into
>> >> > certain speed grades/packages/device sizes. So you have to pay extra and
>> >> > probably not use the speed grade, package, or device size that you
>> >> > thought.
>> >>
>> >> Well, wont you have to pay extra for the Xilinx DLLs all the time too?
>> >>
>> >> Homann
>> >> --
>> >> Magnus Homann, M.Sc. CS & E
>> >> d0asta@dtek.chalmers.se
>> >
>> >--
>> >-Ray Andraka, P.E.
>> >President, the Andraka Consulting Group, Inc.
>> >401/884-7930     Fax 401/884-7950
>> >email randraka@ids.net
>> >http://users.ids.net/~randraka
>> >
>> >
>
>--
>-Ray Andraka, P.E.
>President, the Andraka Consulting Group, Inc.
>401/884-7930     Fax 401/884-7950
>email randraka@ids.net
>http://users.ids.net/~randraka
>
>

Article: 20177
Subject: Re: Program Xilinx Through TI DSP Serial McBSP
From: "Joel Kolstad" <Joel.Kolstad@USA.Net>
Date: Sat, 29 Jan 2000 23:20:19 -0800
Links: << >>  << T >>  << A >>
Mike Horwath <mikeh@merge.com> wrote in message
news:3891ADB2.D28DAA9C@merge.com...
> Has anyone used TI serail McBSP to program Spartan XL.  If so how did
> you accomplish this?

We use the regular (unbuffered) mode of a 320VC549 to program an XC4005E.
The DSP is far faster than the maximum configuration rate of the 4005
anyway, so we didn't bother with the auto-buffering abilities of the DSP and
just sit in a tight loop waiting for each byte to be sent.

To do this, the only tricky signals are Init and CClk.  We used a couple of
CLBs off of an address decoder CPLD that we already had on the board to gate
CClk.  We use XIO off of the DSP for Init; the CPLD then stops providing
CClk to the FPGA until it sees the first frame signal from the DSP's serial
port, at which point it starts simply passing it through (again note that
timing is slow enough that you can get away with slightly crude methods such
as this).

To build the bitstream, Xilinx has a tool that takes the configuration
bitstream file and dumps it as ASCII hex.  It's configurable enough that you
can get it to generate a compilable file in just about any language you'd
care to, and also get it to produce output as bytes, words, long words, with
big endian or little endian bit ordering.  I think this program is called
"makesrc," but I could be completely mis-remembering that detail (it's all
magically done in a makefile in our environment).

Our current task is getting a PLX9054 to program a Virtex...

---Joel Kolstad



Article: 20178
Subject: Re: ARM core?
From: "Mark Harvey" <mark.harvey@iol.it>
Date: Sun, 30 Jan 2000 07:37:49 GMT
Links: << >>  << T >>  << A >>
it's a RISC microprocessor core - see the ARm website http://www.arm.com


¹Ú±¸¿ë <hereim@maru.comtec.re.kr> wrote in message
news:86rlav$ovu$1@news.nuri.net...
> hi, friends.
> what is the ARM core? Is there any site or Book for this question?
> thanks, good day!!
>
>


Article: 20179
Subject: Re: Testbenches
From: "Mark Harvey" <mark.harvey@iol.it>
Date: Sun, 30 Jan 2000 07:42:32 GMT
Links: << >>  << T >>  << A >>
also try Stefan Doll's verification course at :
 http://www.i2.i-2000.com/~stefan/vcourse/html/



Madison <madisonfj@uswest.net> wrote in message
news:38927C20.36CA5CB9@uswest.net...
> I am seeking a source of comprehensive information on building
> testbenches for programmable logic simulations.  The reference material
> I presently have (both instructional texts and software documentation)
> give this matter very light treatment.  Are there books which cover
> testbenches exclusively or at least have thorough coverage of the topic?
>
> Thanks and Best Regards,
>
> Frank Madison
>


Article: 20180
Subject: Re: Xilinx vs Altera
From: danberquet@my-deja.com
Date: Sun, 30 Jan 2000 13:32:55 GMT
Links: << >>  << T >>  << A >>
I think that Xilinx foundation is better, because
they are less parameters to fix to obtain a
executable design with students, altough all
these parameters can be manipulated.
regards.
Daniel BERQUET
In article <3891C2B3.9FB4614E@uoguelph.ca>,
  Shawki Areibi <sareibi@uoguelph.ca> wrote:
> We are introducing a new course on Digital
Design at the School of
> Engineering
> at the University of Guelph. We would like to
choose between the Altera
> MAXII-Plus and the Xilinx Foundation Series
Software.
> Anyone with the experience of using the two
systems? Which is better to
> use
> by the students (i.e simpler to use e.t.c)
> Thanks,
> Shawki Areibi
>
> --
> Shawki Areibi
> Assistant Professor
> School of Engineering
> University of Guelph
> Guelph, Ont, Canada N1G 2W1
> Tel: (519) 824-4120
> Fax: (519) 836-0227
>
>



Sent via Deja.com http://www.deja.com/
Before you buy.
Article: 20181
Subject: Re: Lucent Orca designs
From: pmueller <pbcmuellerNOpbSPAM@hotmail.com.invalid>
Date: Sun, 30 Jan 2000 06:12:24 -0800
Links: << >>  << T >>  << A >>
I have tried out the new Foundry 9.4 with 2 different designs for
a or3t30 (98% full) and a or3t80 (84% full) (both running at 62.5
MHz).

Well, plain-spoken it was a DESASTER!!!

Compilation times increased, score was 10 times worse, timing
slack was 2 times worse..... (then with 9.35).
This version is for me useless..
Did anybody make the same experiences like me?

Patrick Muller


* Sent from RemarQ http://www.remarq.com The Internet's Discussion Network *
The fastest and easiest way to search and participate in Usenet - Free!

Article: 20182
Subject: Re: Which FPGA to learn with?
From: Dave Vanden Bout <devb@xess.com>
Date: Sun, 30 Jan 2000 10:37:45 -0500
Links: << >>  << T >>  << A >>
This is a multi-part message in MIME format.
--------------CE98D6D901F9B0D5FE8F47DA
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

As always, I will recommend our XSK-40 product which combines an XS40 FPGA Board with the Xilinx Student Edition for around $200.  The Foundation 1.5 software in the Student Edition is very similar to the WebPACK software (which is based on Foundation 2.1).  Go to http://www.xess.com/prod009.html for info on the XSK-40 or go to http://www.xess.com to examine the other boards we offer.

As always, I recommend a look at http://www.optimagic.com for a list of other sources for free/low-cost software and hardware.



Ralph Mason wrote:

> I have been using the Xilinx webpack to experiment with VHDL using  a couple
> of XL95108's and would now like move to a FPGA for larger designs.
>
> So far my designs have included an AVR so I am quite interested in the Atmel
> FPSLIC but I don't think they're readily available yet, and I'm not sure of
> what tools are available for them.
>
> So which I am looking for is a beginner friendly FPGA with a free or low
> cost VHDL compiler and programmer.  Preferably a low cost part that is easy
> to obtain one or two at a time.
>
> Thanks for any help.




--------------CE98D6D901F9B0D5FE8F47DA
Content-Type: text/x-vcard; charset=us-ascii;
 name="devb.vcf"
Content-Transfer-Encoding: 7bit
Content-Description: Card for Dave Vanden Bout
Content-Disposition: attachment;
 filename="devb.vcf"

begin:vcard 
n:Vanden Bout;David
tel;fax:(919) 387-1302
tel;work:(919) 387-0076
x-mozilla-html:FALSE
url:http://www.xess.com
org:XESS Corp.
adr:;;2608 Sweetgum Drive;Apex;NC;27502;USA
version:2.1
email;internet:devb@xess.com
title:FPGA Product Manager
x-mozilla-cpt:;28560
fn:Dave Vanden Bout
end:vcard

--------------CE98D6D901F9B0D5FE8F47DA--

Article: 20183
Subject: Re: Spartan II availability and pricing
From: Brian Dipert <bdipert@NOSPAM.pacbell.net>
Date: Sun, 30 Jan 2000 09:08:28 -0800
Links: << >>  << T >>  << A >>
>EDN mag, Jan 6th issue, p. 24 had some unattributed Spartan-2 price projections:
>XC2S15 3.95, XC2S30 4.95, XC2S50 7.95, XC2S100 9.95, XC2S150 12.95
>The fine print states that this pricing is for 250,000 units, lowest
>speed and cheapest package, end of 2000. They'll be throwing in a free
>pizza and beverage with every million-dollar order, too :)
>
>regards,
>Tom Burgess

You forgot about the free Ginsu knife set ;-) Sorry, Tom, I can only
print the prices they're willing to give me. My mind-reading abilities
haven't been sufficiently fine-tuned yet to allow me extract
additional price/quantity combinations from the brains of Xilinx's
marketing folks, especially when I get briefed on the product over the
telephone instead of face-to-face.....(long-distance, don'tchaknow)

Brian Dipert
Technical Editor: Memory, Multimedia and Programmable Logic
EDN Magazine: The Design Magazine Of The Electronics Industry
http://www.ednmag.com
Contributing Editor; CommVerge Magazine
http://www.commvergemag.com
1864 52nd Street
Sacramento, CA   95819
(916) 454-5242 (voice), (916) 454-5101 (fax)
***REMOVE 'NOSPAM.' FROM EMAIL ADDRESS TO REPLY***
mailto:bdipert@NOSPAM.pacbell.net
Visit me at http://members.aol.com/bdipert
Article: 20184
Subject: Which FPGA to learn with?
From: "Ralph Mason" <ralphmason@geocities.com>
Date: Sun, 30 Jan 2000 22:18:00 -0000
Links: << >>  << T >>  << A >>
I have been using the Xilinx webpack to experiment with VHDL using  a couple
of XL95108's and would now like move to a FPGA for larger designs.

So far my designs have included an AVR so I am quite interested in the Atmel
FPSLIC but I don't think they're readily available yet, and I'm not sure of
what tools are available for them.

So which I am looking for is a beginner friendly FPGA with a free or low
cost VHDL compiler and programmer.  Preferably a low cost part that is easy
to obtain one or two at a time.

Thanks for any help.




Article: 20185
Subject: Re: ADC to DSP... FIFO?
From: russell shaw <russell@webaxs.net>
Date: Mon, 31 Jan 2000 10:54:43 +1100
Links: << >>  << T >>  << A >>
Hi,

Thanks RC, i'll add lucent to my green book (red book: try again, black book:
don't bother).

Rickman wrote:
> 
> Russell,
> 
> I am not trying to push the Lucent stuff, but they do give away their
> toolset which includes the front end (Viewlogic) and the back end tools
> as well as VHDL support (or so I hear, I have not tried to use it). You
> just have to ask your FAE to provide you with a copy. My FAE (Rick
> Golabowski of Impact) has been very supportive (especially on snow days
> ;). Try your distributor based FAEs.
> 
> I like the Xilinx tools which work well if they are a bit hard to learn.
> But I have found the Lucent tools to be very useful as well and I am
> getting decent support from my local FAE as well as some advice from Bob
> Wagner via postings here.
> 
> One common point to all of the FPGA tools I have looked at; they all
> seem to be going in the direction of HDL support. Every time I ask
> questions the responder seems to assume that I am using an HDL ignoring
> the possibility of my entry method being schematic. I am using schematic
> entry because I often have pieces of my design that need to be very
> highly optimized which I just don't have enough control via VHDL to make
> work. So rather than do a mixed design, I just use schematic capture. I
> also have tight schedules which I have found fit better with predictable
> schematic capture.
> 
> My concern is that I may end up like the dodo bird when it comes to
> support. A lot of the support engineers already don't use schematic
> tools and so can't answer questions about them. So if you do any serious
> design in FPGAs, you may want to go for the HDL tools and keep schematic
> as an adjunct for the tough pieces.
> 
> 
> Rick Collins
> 
> rick.collins@XYarius.com
> 
> remove the XY to email me.
> 
> 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

-- 
*******************************************
*   Russell Shaw, B.Eng, M.Eng(Research)  *
*   Electronics Consultant                *
*      email: russell@webaxs.net          *
*      Australia                          *
*******************************************
Article: 20186
Subject: Announcement: Xilinx on Linux HowTo
From: "B. Joshua Rosen" <bjrosen@polybus.com>
Date: Sun, 30 Jan 2000 20:06:15 -0500
Links: << >>  << T >>  << A >>
I've just put up a HowTo page for running Xilinx place and route tools under Linux.

http://www.polybus.com/xilinx_on_linux.html
Article: 20187
Subject: Re: Why Sinplicity make combinatorial loops from latches ?
From: "Steven K. Knapp" <sknapp@triscend.com>
Date: Sun, 30 Jan 2000 19:08:30 -0800
Links: << >>  << T >>  << A >>
I had an old article in Personal Engineering magazine (January 1998) that
covered this topic.  See http://www.optimagic.com/acrobat/pein_0198.pdf.
Some of the figures are a bit difficult to make out.

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



"Bonio Lopez" <bonio.lopezNOboSPAM@gmx.ch.invalid> wrote in message
news:2a42f762.e1d74b29@usw-ex0101-003.remarq.com...
> Hi friends,
> i am trying to synthesize one external mC core with Sinplicity and
> Leonardo.
> I have got very strange results by synplisity.
> The Leonardo make really good job,
> but simpisity synthesize some memory cells (quite many of them) as
> combinatorial
> loops.
> Have somebody any thinks about this phenomena.
> Sorry, that i can't send you source code.
>
>
>
> * Sent from RemarQ http://www.remarq.com The Internet's Discussion Network
*
> The fastest and easiest way to search and participate in Usenet - Free!
>


Article: 20188
Subject: Re: Altera Quartus vs Xilinx Place and Route tools (help needed)
From: allan.herriman.hates.spam@fujitsu.com.au (Allan Herriman)
Date: Mon, 31 Jan 2000 03:15:20 GMT
Links: << >>  << T >>  << A >>
On Fri, 28 Jan 2000 17:41:52 GMT, Ray Andraka <randraka@ids.net>
wrote:

>I hadn't considered a VCO made as a ring oscillator.  I'll have to noodle on that.

Here's a description of a ring oscillator used as a VCO in a PLL.

http://www.cypress.com/pub/appnotes/robo.pdf
(see figure 29, on page 14)

Regards,
Allan.
Article: 20189
Subject: Re: Lattice isp & FPGA
From: "Steven K. Knapp" <sknapp@optimagic.com>
Date: Sun, 30 Jan 2000 19:23:46 -0800
Links: << >>  << T >>  << A >>
The Programmable Logic Jump Station at http://www.optimagic.com might be a
good place to start.  It has links for practically all topics.  You might
check out the FAQ section as it has an overview on the various programmable
logic architectures.  It hasn't been updated for awhile but the basic
concepts still apply.


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

"#XUE ZHONG#" <P144850500@ntu.edu.sg> wrote in message
news:0CF260C495FED111A6610000F866308D09C8C10E@mail3.ntu.edu.sg...
> Hi folks,
>
> In face, I am familiar with Lattice's ispPLD chip 1000, 2000, 3000
> series:
> Such as 1016, 2032, etc. including there structures, and programming.
>
> How about the difference between FPGA and the same. Where can I find
> (website)
> the information about FPGA, and learn it quickly, I need to develop
> circurts with FPGA
> in the near future.
>
> Thanks.
>


Article: 20190
Subject: Re: Which FPGA to learn with?
From: "Joel Kolstad" <Joel.Kolstad@USA.Net>
Date: Sun, 30 Jan 2000 23:34:33 -0800
Links: << >>  << T >>  << A >>
Dave Vanden Bout <devb@xess.com> wrote in message
news:38945AC9.41B6AA23@xess.com...
> As always, I will recommend our XSK-40 product which combines an XS40 FPGA
Board with the Xilinx Student Edition for around $200.

Does this come with programming software (for the XS40 board) that works
under Windows NT or Linux yet?  When I looked at it a couple of years ago,
the provided software appeared a little on the amateurish side; hopefully a
lot has changed over time.

---Joel Kolstad



Article: 20191
Subject: Re: Announcement: Xilinx on Linux HowTo
From: Erik de Castro Lopo <erikd@zip.com.au>
Date: Mon, 31 Jan 2000 08:51:18 +0000
Links: << >>  << T >>  << A >>
B. Joshua Rosen wrote:
> 
> I've just put up a HowTo page for running Xilinx place and route tools under Linux.
> 
> http://www.polybus.com/xilinx_on_linux.html

Alleluyah!!!!!!!

I was thinking about trying this out to see if its possible
but you've beat me to it. This is fantastic news. I may finally
be able to dump WinNT at work like I already have at home.

Erik
-- 
+-------------------------------------------------+
     Erik de Castro Lopo     erikd@zip.com.au
+-------------------------------------------------+
"There are two ways of constructing a software design. One way is
 to make it so simple that there are obviously no deficiencies
 and the other is to make it so complicated that there are no
 obvious deficiencies."  --  C A R Hoare
Article: 20192
Subject: Re: Xilinx programming from a Linux PC
From: Phil Endecott <phil_endecott@spamcop.net>
Date: Mon, 31 Jan 2000 10:05:03 +0000
Links: << >>  << T >>  << A >>
Hi Everyone,

Many thanks to all of you who replied to my original posting about
downloading a Xilinx design from a Linux box: you've given me several
strategies to try out.  I'm going to try Larry Doolittle's Linux port of
the XESS utility first.  I note that this has to run as root in order to
access the IO port: has anyone looked into adding calls to setuid() so
that it uses the real user id to read the input files?  Has anyone else
used this with the VCC board or another Virtex device?  Anything else I
should know?

Cheers,

--Phil.
Article: 20193
Subject: Re: ARM core?
From: Sprow <rps102@york.ac.uk>
Date: Mon, 31 Jan 2000 11:08:11 +0000
Links: << >>  << T >>  << A >>


On Fri, 28 Jan 2000, Tim Tyler wrote:

> : what is the ARM core? Is there any site or Book for this question?
> 
> http://www.arm.com/
> 
> ARM Architecture Reference Manual - D. Jaggar;
> ARM System Architecture - S.Furber;

I can recommend the Steve Furber book - he was one of the Acorn team
involved at the time of the development of the Acorn Risc Machine (later
renamed to Advanced Risc Machine - ARM)

> The ARM RISC Chip, A Programmer's Guide A. van Someron & C. Atack. > -- 
> 

Sprow.

Article: 20194
Subject: Re: Which FPGA to learn with?
From: Dave Vanden Bout <devb@xess.com>
Date: Mon, 31 Jan 2000 08:11:21 -0500
Links: << >>  << T >>  << A >>
This is a multi-part message in MIME format.
--------------D05E331F891685464BA50A8E
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

Joel Kolstad wrote:

> Dave Vanden Bout <devb@xess.com> wrote in message
> news:38945AC9.41B6AA23@xess.com...
> > As always, I will recommend our XSK-40 product which combines an XS40 FPGA
> Board with the Xilinx Student Edition for around $200.
>
> Does this come with programming software (for the XS40 board) that works
> under Windows NT or Linux yet?  When I looked at it a couple of years ago,
> the provided software appeared a little on the amateurish side; hopefully a
> lot has changed over time.
>
> ---Joel Kolstad

Our software for testing, programming, and debugging the XS40 Board runs under Win95/98/NT in both a DOS window and using the GUI.  You can read more about that at http://www.xess.com/gxstools-v3_0.pdf.

We have made the underlying code for these tools publicly available so others can port them as they see fit.  Larry Doolittle and others have made ports to Linux.  We will have no official Linux port of our tools until Xilinx has an official Linux port of Foundation.



--------------D05E331F891685464BA50A8E
Content-Type: text/x-vcard; charset=us-ascii;
 name="devb.vcf"
Content-Transfer-Encoding: 7bit
Content-Description: Card for Dave Vanden Bout
Content-Disposition: attachment;
 filename="devb.vcf"

begin:vcard 
n:Vanden Bout;David
tel;fax:(919) 387-1302
tel;work:(919) 387-0076
x-mozilla-html:FALSE
url:http://www.xess.com
org:XESS Corp.
adr:;;2608 Sweetgum Drive;Apex;NC;27502;USA
version:2.1
email;internet:devb@xess.com
title:FPGA Product Manager
x-mozilla-cpt:;28560
fn:Dave Vanden Bout
end:vcard

--------------D05E331F891685464BA50A8E--

Article: 20195
Subject: Re: Altera Quartus vs Xilinx Place and Route tools (help needed)
From: husby@fnal.gov (Don Husby)
Date: Mon, 31 Jan 2000 14:39:14 GMT
Links: << >>  << T >>  << A >>
Ray Andraka <randraka@ids.net> wrote:
> I hadn't considered a VCO made as a ring oscillator.

Take a look at the Lucent 3T parts.  Their clock synchronizer
can be used in either DLL or PLL mode.  It's implemented as a
multi-tap voltage-controlled delay line.  


--
Don Husby <husby@fnal.gov>             http://www-ese.fnal.gov/people/husby
Fermi National Accelerator Lab                          Phone: 630-840-3668
Batavia, IL 60510                                         Fax: 630-840-5406
Article: 20196
Subject: Re: Lucent Orca designs
From: husby@fnal.gov (Don Husby)
Date: Mon, 31 Jan 2000 15:24:05 GMT
Links: << >>  << T >>  << A >>
Bummer.  Thanks for the warning.
I guess it's not too surprising since they made some major
changes to the constraints file processing.

pmueller <pbcmuellerNOpbSPAM@hotmail.com.invalid> wrote:
> I have tried out the new Foundry 9.4 with 2 different designs for
> a or3t30 (98% full) and a or3t80 (84% full) (both running at 62.5
> MHz).
> 
> Well, plain-spoken it was a DESASTER!!!
> 
> Compilation times increased, score was 10 times worse,

Does the part meet your timing constraints?
Did you use the same timing analyzer for both measurements?
(IE: did you use the 9.35 analyzer to check the 9.4 compilation
or the 9.4 analyzer to check the 9.35 compliation?)
Lucent may have just changed the way they compute the score.

Also note that some new paths-types were added to the analysis:
For example, 9.35 does not check paths through an open latch
(a serious flaw that I have been whining about for years.)
9.4 does (according to the release notes).  It also is supposed
to handle timing between multiple clock domains differently (better).

> timing slack was 2 times worse..... (then with 9.35).

Unfortunately, the software doesn't try to optimize timing slack.
To get an optimal design, I usually rachet-up the clock speed until
it no-longer routes.  For example, I try to place a 62.5 MHz design
using an 80 MHz constraint.  Even if a few paths fail to meet that
constraint, the whole design runs easily at 62.5.


--
Don Husby <husby@fnal.gov>             http://www-ese.fnal.gov/people/husby
Fermi National Accelerator Lab                          Phone: 630-840-3668
Batavia, IL 60510                                         Fax: 630-840-5406
Article: 20197
Subject: Re: EEPROM based FPGAs
From: "Steven K. Knapp" <sknapp@optimagic.com>
Date: Mon, 31 Jan 2000 07:41:22 -0800
Links: << >>  << T >>  << A >>
Because EEPROM technology requires additional mask steps during
manufacturing, EEPROM-based devices are more expensive to manufacture, not
to mention more expensive to test.

The incremental cost is probably insignificant on a small device but become
intolerable as the die size increases.

Also, generally SRAM is a leading-edge process.  It's one of the first to be
migrated to a new, more aggressive process technology.  A manufacturing fab
needs SRAM to be up and running before implementing more complex processes
like EEPROM.  The more aggressive fab process is one of the reasons that
SRAM-based devices are generally faster and denser.

The most direct evidence of this is to see what process is used by the
largest and fastest devices like Xilinx Virtex and Altera Apex.  Both Xilinx
and Altera are racing down the process curve from 0.25u, to 0.18u, to 0.15u,
to ....


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



"George" <g_roberts75@hotmail.com> wrote in message
news:86nge5$sq2$1@news.qub.ac.uk...
> Hi Folks,
>
> EEPROM based FPGAs are in-circuit re-programmable FPGAs, which are, unlike
> SRAM based FPGAs, not volatile. Their disadvantage, I presume, is that
they
> are slower. Could you think about any other disadvantage? or advantage?
> compared to SRAM based ones.
>
> Cheers.
>
>


Article: 20198
Subject: Actel proAsic availability, experiences?
From: "Adrian Dunn" <adunn@domosys.com>
Date: Mon, 31 Jan 2000 16:39:33 GMT
Links: << >>  << T >>  << A >>
Has anyone in this group worked with these devices, and if so, could they
share their experiences?

We used Gatefield FPGAs (GF200 and GF250) to prototype our last ASIC design,
and were quite happy with the design flow and performance that these FPGAs
offered.  We have been planning to use the new Actel proAsic devices to
prototype our next ASIC design, however it has been very difficult to obtain
information from them about the availability of their higher density
devices, let alone pricing.  We understant the the 130 series is available,
however we need the additional logic tiles found in the 180 series.

Any opinions as to the viability of these FPGAs in the market, long term?
I'm getting the impression that many of their higher density devices are
vapour-ware until a market materializes for them, and perhaps the market
just isn't there.  The quantity we will use to prototype our ASIC is
obviously very small, and they can't survive on small fry such as us.

Adrian


Article: 20199
Subject: Re: Altera Quartus vs Xilinx Place and Route tools (help needed)
From: Ray Andraka <randraka@ids.net>
Date: Mon, 31 Jan 2000 16:52:41 GMT
Links: << >>  << T >>  << A >>
I thought the point to using a DLL was to get away from the analog circuit.  I
know the xilinx DLL is entirely digital, and I have played a little with making
DLLs in the XC4000 series parts using the carry chain as a tapped delay.
Theoretically speaking, I suppose one could make a ring oscillator with the
same type of structure to make an entirely digital PLL (without the many-x
clock normally associated with DPLLs).  Practically, though, I don't see how it
would work because the delays in the selection logic are in the ring, and would
tend to swamp the tapped delay.  To make the selection logic delay
inconsequential, the adjustable part of the delay chain would have to be too
long to get the frequencies out that we need in modern designs.


Don Husby wrote:

> Ray Andraka <randraka@ids.net> wrote:
> > I hadn't considered a VCO made as a ring oscillator.
>
> Take a look at the Lucent 3T parts.  Their clock synchronizer
> can be used in either DLL or PLL mode.  It's implemented as a
> multi-tap voltage-controlled delay line.
>
> --
> Don Husby <husby@fnal.gov>             http://www-ese.fnal.gov/people/husby
> Fermi National Accelerator Lab                          Phone: 630-840-3668
> Batavia, IL 60510                                         Fax: 630-840-5406

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




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