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 4300

Article: 4300
Subject: Re: Reversible LFSR?
From: hatcher@asd470.dseg.ti.com (Rich Hatcher)
Date: 11 Oct 1996 18:54:27 GMT
Links: << >>  << T >>  << A >>
In article 1335@univision.com, "John L. Smith" <jsmith@univision.com> () writes:
> LFSRs make good cheap address generators for FIFOs, but how about
> an address generator for a LIFO? Has anyone ever looked at an LFSR
> that can run in both directions? One bit is lost from the end of the
> shift register each clock, but perhaps there is an interesting
> combination of xor's that can produce the neccessary bits for shifting
> reversibly in either direction? Just posing the question...
Good idea!

You can get a reverse sequence in at least one situation.  Given a
maximal-length LFSR, i.e. one with a sequence length of 2^n - 1 for n
flip-flops, you can form a new LFSR with the property that it counts in
the reverse sequence in the following way.

Assume the bit numbers of the LFSR start from n-1 down to 0 as shown
below.  The setup has the taps as shown.

Let the tap numbers be the set of values in range 0 to n where the
number of values is n+1 and one of the values always equals n and one
of them always equals 0.
                                           +-----+
                                           |     |<----------------+
                                       +---| XOR |                 |
                                       |   |     |<---+            |
                                       |   +-----+    |            |
    +-----+                       <----+              |            |
    |     |<--------    etc. ...                      |            |
 +--| XOR |                                           |            |
 |  |     |<--+                                       |            |
 |  +-----+   |                                       |            |
 |            |                                       |            |
 |            | n-1                                   | 1          | 0
 |     +-----------------------------------------------------+     |
 | n   |                                                     |     |
 +---->|            n-bit Right Shift Register               |-----+
       |                                                     |
       +-----------------------------------------------------+
            ^  
            |
Clock ------+  
               
Now form a new LFSR with all taps = n - (each tap value)

Then use the bits in reverse order.  The sequence of the new LFSR
will be the reverse of the original LFSR.

For example:  For a three bit register start with taps = (3, 1, 0)

The sequence is 001, 100, 010, 101, 110, 111, 011

Form the new LFSR with taps = (3-3, 3-1, 3-0) = (0, 2, 3)= (3, 2, 0)

The sequence is 110, 111, 011, 101, 010, 001, 100 but reverse the bit
order and you have 011, 111, 110, 101, 010, 100, 001 which is the
reverse sequence of the original register.

For a four bit LFSR, a maximal-length set of taps is (4, 1, 0) and the
reverse LFSR would have the taps (4, 3, 0)

For a five bit LFSR a maximal-length set of taps is (5, 2, 0) and the
reverse LFSR is (5, 3, 0)

These values come from a table of primitive polynomials for the Galois
field of 2.

As Philip Freidin stated, if you use a left/right shift register and
rearrange the XORs when you change from left to right then you can use
the above taps to get the reversible counter.

Up/Down 
enable     +--------------------------------------------+ 
---+------>|                                            |   
   |  +----|             XOR's with Up/Down enable      |-----+
   |  |    |                                            |     |
   |  |    +--------------------------------------------+     |
   |  |         ^                                 ^           |
   |  |         |                                 |           |
   |  |   +-----------------------------------------------+   |
   |  |   |                                               |   |
   |  +-->|        n-bit Right/Left Shift Register        |<--+
   +----->|                                               |
          +-----------------------------------------------+
            ^  
            |
Clock ------+   

   
For the three bit example it looks like:

Up/Down    +----------------------------------------+
enable     |                                        |
---+---------------------+                          |
   |       |             |                          |
   |       |             v           +-------+      |
   |       |          +-----+        |       |      |
   |       |          | sel |        |  XOR  |----->+
   |       |          |     |------->|       |      |
   |       |          | MUX |        +-------+      |
   |       |          +-----+           ^           |
   |       |          ^     ^           |           |
   |       |          | 2   | 1         | 0         |
   |       |  +---------------------------------+   |
   |       |  |                                 |   |
   |       +->| 3-bit Right/Left Shift Register |<--+
   +--------->|                                 |
              +---------------------------------+
               ^  
               |
Clock ---------+   

This assumes "up" means with bit 1 selected and shift to right.  When
shifting left, the bit numbers are reversed from the original
numbering in the taps.


Rich Hatcher

Article: 4301
Subject: Xilinx XACT Performance Appl. Note?
From: oakley@appliedmicro.ns.ca (Bruce Oakley)
Date: 11 Oct 1996 15:54:27 -0300
Links: << >>  << T >>  << A >>
To any Xilinx users:

We've recently derived great benefit from specifying detailed timing
constraints (ie. the XACT Performance utility) for an XC4013E design.
However, we found the documentation for this to be fairly limited ... the
syntax is reasonably clear, but we had to learn a lot ourselves about how
to use it effectively.

This seems like an ideal topic for an application note.  However, we've
been unable to find any on the web site.  I posted a question to the
email hotline, and received the usual deafening silence.  Has anybody
(perhaps one of the Xilinx employees who work here) ever seen such an
app. note?

Thanks,
Bruce.
oakley@appliedmicro.ns.ca

-- 
---------------------------------------------------
Bruce Oakley   [ oakley@appliedmicro.ns.ca ]
Hardware Design Engineer
Applied Microelectronics, Inc. (AMI)
Halifax, Nova Scotia, CANADA
---------------------------------------------------
Article: 4302
Subject: Re: VHDL for Xilinx designs?
From: David Emrich <emrich@exemplar.com>
Date: Fri, 11 Oct 1996 18:37:21 -0700
Links: << >>  << T >>  << A >>
Note that I'm with a tool vendor. But I think my answer will be helpful.

Nick Gent wrote:
> 
> I am taking (another) look at VHDL for FPGA design.  I am a complete novice
> at VHDL but have been doing FPGA design for years. I would be looking for a
> solution which runs under HP-UX on the HP 700-series.  The target will
> probably be Xilinx 5k and 4k series.
> 
> Our past attempts at synthesising Xilinx (3000) LCAs from VHDL apparently
> failed because it took too much manual intervention to get a decent end
> result. Have the tools improved significantly?

Yes.

> 
> Which vendors should I contact?

IMHO us.  www.exemplar.com. info@exemplar.com. From the UK: info@saros.co.uk.

> 
> With VHDL entry, is it still possible to make use of the static-timing
> driven routing facilities of the Xilinx PPR router?

Yes.  Exemplar's Leonardo writes Timespecs from the user provided timing constraints.
Exemplar's Galileo has some Timespec support as well. And Xilinx provides a couple of
very useful perl scripts maketnm & addtnm  that help you add TNMs to the XNF if you choose
to do the Timespecs from the Xilinx constraint file. They are probably on www.xilinx.com.

> 
> Has anyone tried mixing ABEL-HDL with VHDL? What would be the issues here?

While you can mix any form of design entry you like, it is best to stick with
VHDL and/or Verilog.

This allows you to simulate each block as you design it, then to simulate the entire design,
then after place & route, to use the same testbench to simulate the routed design.

Furthermore your test is then written in the same HDL as your design. The simulation phase
can take far more time than the design phase, so in simulation, like design (maybe more so),
it is important to use a good methodology, to maximize your productivity.

There is a post further back from someone looking for macros for the primitive commands from
his non-HDL gate level simulator.  While it's a fine gate level simulator, it has a terrible
language for developing tests.  For example it has no functions or procedures that accept
parameters.  It's much easier to simulate in an HDL.

While it is simplest (and generally best) to choose either VHDL or Verilog, Exemplar's Leonardo
can synthesize a design hierarchy of mixed VHDL and Verilog, and Model Tech's VSystem Plus can
simulate it.

If you mix blocks from different tools you need to make sure that busses are expanded into
bits in XNF using the same bit delimeters so that XNFMERGE can merge the design hierarchy.
And the tools must write the same XNF version (check the LCANET record at the start of the
XNF file).  Solutions to the above two problems are usual very simple (tool settings,
or PERL, SED or AWK scripts).

You can keep legacy ABEL designs units if you like.  But it can make it harder to adopt an
all HDL methodology.  Converting a few blocks from ABEL to VHDL should be pretty easy.

Regards,
David Emrich
emrich@exemplar.com

> 
> Any information will be gratefully received!
> 
> Nick
> ============================================================================
> Nick Gent
> 
> Queensferry Telecom Operation          Email:     nickg@hpsqf.sqf.hp.com
> Hewlett-Packard                        Telephone: +44 131 331 7644
> South Queensferry EH30 9TG             Fax:       +44 131 331 7488
> Scotland
> ============================================================================
Article: 4303
Subject: Re: Xilinx Startup symbol instantiation in VHDL ... ?
From: David Emrich <emrich@exemplar.com>
Date: Fri, 11 Oct 1996 19:11:57 -0700
Links: << >>  << T >>  << A >>
This is a multi-part message in MIME format.

--------------2781E494446B9B3D794BDF32
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

> Rynier van der Watt wrote:
> > 
> > Hi
> > 
> > Does anybody know how to instantiate the Xilinx Startup symbol in VHDL
> > for Viewlogic's Viewsynthesis?
> > 
> > I am currently using the Workview Office for Windows95 tools and a
> > XC4025E device.
> > 
> > Thank you very much in advance.
> > 
> > Rynier van der Watt
> > CSIR South Africa
> > 
> > Email: arvdwatt@csir.co.za
> 
> -- 
> 
> I give here an example how I implemented it for Examplar Galileo.
> It synthesizes Global Set Reset correctly, but there is still a
> problem with Global Three State. Maybe Viewsynthesis does it better.

There is no problem when I run it here - did you use the
-xnf_use_startup option?  I ran your design (attached). The XNF is
also attached. The Galileo command line is at the top.  The Startup
GSR & GTS are both fine.

If you only want GSR, and not GTS, we recommend against instantiation,
since we can handle global reset for you.

If one signal will cause all storage elements to be reset, -infer_gsr
will implement the Startup in your Xilinx design (or the GSR in your
ORCA design), and remove that GSR signal from the reset logic to 
the storage elements, so it isn't routed.

In cases where the signal you want for GSR does not reset all storage
elements in your HDL design, we cannot extract the reset signal, but you
can specify it with -global_sr=<active high reset signal name>.

This will drive GSR with the specified signal and take it out of the
reset logic.

Regards,
David Emrich
Exemplar Logic
emrich@exemplar.com

> 
> You have to instantiate a STARTUP component in your top level entity,
> with inputs gsr and gts. Then connect the reset and output enable
> input signals to STARTUP by PORT MAP.
> 
> 
> -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
> Hans Spaargaren			(for: Philips Medical Systems Nederland B.V.)
> Origin
> Engineering Services, CAD/CAM-E
> Building QG-037                       Internet:
> hspaarga@best.ms.philips.com
> P.O. Box 10.000                       Voice:		+31 40 2763018
> 5680 DA  Best    		      Fax:		+31 40 2765641
> the Netherlands				
> -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
>

--------------2781E494446B9B3D794BDF32
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Content-Disposition: inline; filename="c.vhd"

 -- This VHDL file is generated by EASE/VHDL from TRANSLOGIC BV,
 -- the 'Entity Architecture Schematics Editor for VHDL' tool.
 --
 -- EASE/VHDL Version 2.4 (Revision -8).
 -- Time stamp  : Thu Oct 10 07:47:40 1996.
 --
 -- Designed by : .
 -- Company     : .
 -- Design info : .
 ---------------------------------------------------------------------------
 
 ---------------------------------------------------------------------------
 -- Entity declaration of 'core'.
 ---------------------------------------------------------------------------
 
 LIBRARY ieee, synergy ;
 USE ieee.std_logic_1164.ALL ;
 USE ieee.std_logic_1164.ALL ;
-- USE synergy.signed_arith.ALL ;
 USE work.ALL ;
 
 ENTITY core IS
   PORT(
     clk    : IN     std_ulogic ;
     en_out : IN     std_ulogic ;
     reset  : IN     std_ulogic ;
     a      : IN     std_ulogic ;
     q1     : OUT    std_ulogic ) ;
 END core ;
 
 ---------------------------------------------------------------------------
 -- Architecture 'a0' of 'core'.
 ---------------------------------------------------------------------------
 
 
 ARCHITECTURE a0 OF core IS
 
 signal q1_a:  std_ulogic;
 
 BEGIN
 
 process(reset,clk,a)
 begin
    if reset = '1' then
       q1_a <= '0';
    elsif clk = '1' and clk'event then
       q1_a <= a;
    end if;
 end process;
 
 process(q1_a,en_out)
 begin
    if en_out = '1' then
       q1 <= q1_a;
    else
       q1 <= 'Z';
    end if;
 end process;
 
 END a0 ;
 
 ---------------------------------------------------------------------------
 -- Entity declaration of 'xxx'.
 ---------------------------------------------------------------------------
 
 
 LIBRARY ieee, synergy ;
 USE ieee.std_logic_1164.ALL ;
 USE ieee.std_logic_1164.ALL ;
-- USE synergy.signed_arith.ALL ;
 USE work.ALL ;
 
 ENTITY xxx IS
   PORT(
     a      : IN     std_ulogic ;
     reset  : IN     std_ulogic ;
     en_out : IN     std_ulogic ;
     q1     : OUT    std_ulogic ;
     clk    : IN     std_ulogic ) ;
 END xxx ;
 
 ---------------------------------------------------------------------------
 -- Architecture 'a0' of 'xxx'.
 ---------------------------------------------------------------------------
 
 ARCHITECTURE a0 OF xxx IS
 
  COMPONENT core
   PORT(
     clk    : IN     std_ulogic ;
     en_out : IN     std_ulogic ;
     reset  : IN     std_ulogic ;
     a      : IN     std_ulogic ;
     q1     : OUT    std_ulogic ) ;
  END COMPONENT ;
  COMPONENT startup                --  <--------------------------
   PORT(
     gsr : IN     std_ulogic ;
     gts : IN     std_ulogic ) ;
  END COMPONENT ;
  SIGNAL clock : std_ulogic ;
  SIGNAL rst   : std_ulogic ;
  SIGNAL ain   : std_ulogic ;
  SIGNAL ena   : std_ulogic ;
  
 BEGIN
 
  clock <= clk ;
  rst <= reset ;
  ain <= a ;
  ena <= en_out ;
 
  u0: core
     PORT MAP(
       clk => clock,
       en_out => ena,
       reset => rst,
       a => ain,
       q1 => q1 ) ;
 
  u4: startup                      -- <----------------------
     PORT MAP(
       gsr => rst,
       gts => ena ) ;
 
 END a0 ; -- of xxx
 
 

--------------2781E494446B9B3D794BDF32
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Content-Disposition: inline; filename="c_xi4.xnf"

LCANET, 5
USER, FPGA, gc c.vhd c_xi4.xnf -so=xi4 -ta=xi4 -xnf_use_startup 
PROG, gc, V3.2.5, "Fri Oct 11 18:45:55 1996"
USER,AREA,0
USER,PACKED_ESTIMATE,0
USER,DELAY,16.000000
USER,LEVELS,1
EXT, A, I,
EXT, RESET, I,
EXT, EN_OUT, I,
EXT, CLK, I,
EXT, Q1, T,
SYM, U4, STARTUP,LIBVER=2.0.0
PIN, GSR, I, RESET_int, , 
PIN, GTS, I, EN_OUT_int, , 
END
SYM, i90, OUTFFT,SCHNM=OFDT,LIBVER=2.0.0,INIT=R
PIN, C, I, CLK_int, , 
PIN, D, I, A_int, , 
PIN, T, I, n545, , 
PIN, O, O, Q1, 7.000000, 
END
SYM, IBUF_CLK_int, IBUF,SCHNM=IBUF,LIBVER=2.0.0
PIN, I, I, CLK, , 
PIN, O, O, CLK_int, 3.000000, 
END
SYM, IBUF_EN_OUT_int, IBUF,SCHNM=IBUF,LIBVER=2.0.0
PIN, I, I, EN_OUT, , 
PIN, O, O, EN_OUT_int, 3.000000, 
END
SYM, IBUF_RESET_int, IBUF,SCHNM=IBUF,LIBVER=2.0.0
PIN, I, I, RESET, , 
PIN, O, O, RESET_int, 3.000000, 
END
SYM, IBUF_A_int, IBUF,SCHNM=IBUF,LIBVER=2.0.0
PIN, I, I, A, , 
PIN, O, O, A_int, 3.000000, 
END
SYM, i545, INV,SCHNM=INV,LIBVER=2.0.0
PIN, I, I, EN_OUT_int, , 
PIN, O, O, n545, , 
END
EOF

--------------2781E494446B9B3D794BDF32--

Article: 4304
Subject: Async with FPGA?
From: veakblad@Glue.umd.edu (David T. Wang)
Date: 12 Oct 1996 05:02:51 GMT
Links: << >>  << T >>  << A >>
Hi, I'm looking for a way to shrink some Async circuit, has anyone done
any work with implementing some Async stuff with an FPGA?  I have an 
application which needs to respond to an input within a small time period, 
~7 to 10 ns.  unfortunately I can't just use a fast clock, so I was
thinking of trying to design some Asyn stuff in an FPGA.  any comments/
suggestions?     Thanks.

						David 
Article: 4305
Subject: Re: Async with FPGA?
From: zz80@digiserve.com (extra z to stop junk mail)
Date: Sat, 12 Oct 1996 09:28:02 GMT
Links: << >>  << T >>  << A >>
This is dangerous, since the routing delays are likely to be different
every time you run place/route.
Peter.

>Hi, I'm looking for a way to shrink some Async circuit, has anyone done
>any work with implementing some Async stuff with an FPGA?  I have an 
>application which needs to respond to an input within a small time period, 
>~7 to 10 ns.  unfortunately I can't just use a fast clock, so I was
>thinking of trying to design some Asyn stuff in an FPGA.  any comments/
>suggestions?     Thanks.
>
>						David 

Article: 4306
Subject: Re: Async with FPGA?
From: Scott Kroeger <Scott.Kroeger@mei.com>
Date: Sat, 12 Oct 1996 08:14:32 -0500
Links: << >>  << T >>  << A >>
David T. Wang wrote:
> 
> Hi, I'm looking for a way to shrink some Async circuit, has anyone done
> any work with implementing some Async stuff with an FPGA?  I have an
> application which needs to respond to an input within a small time period,
> ~7 to 10 ns.  unfortunately I can't just use a fast clock, so I was
> thinking of trying to design some Asyn stuff in an FPGA.  any comments/
> suggestions?     Thanks.

Your question is too vague to enable a concise answer.  That said:

We've proabably all done some async logic (accidently or purposely) and
often come to regret it.  If you can't an async solution, at least lock
the placement of your async logic blocks to minimize changes in routing
delay over the course of your design effort.  If you are careful, and
keep the async portion of your design as simple as possible you
shouldn't get into trouble.

If you are doing something simple like address decode (one of the more
popular uses of async logic) you will benefit from partial decode, where
only the minimum necessary set of address lines are decoded.  This
results in aliases of the selected device but minimizes the width (and
indirectly depth) of the decode logic if your device does not have wide
decoder logic.

Regards,
Scott
Article: 4307
Subject: Re: help: max+ edif input problems
From: jedelman@epi043.gesd (John Edelman)
Date: 12 Oct 1996 13:34:00 GMT
Links: << >>  << T >>  << A >>
	I have had a simular problem with MAXPLUS 6.2 and MENTOR 8.25 QUICKSIMII.
	Found that the edif format changed from
		version 5.4 (old style edif)
		version 6.0 (new style edit)
	I had to use a old version of MAXPLUS 5.4, but in your case you have to
	use a newer version 6.0 and above (latest is 7.0)

						Good luck
						John Edelman 
---
Design -> Develop -> Fab -> Assembly -> Checkout -> Test -> Sell-off -> RETIRE

Article: 4308
Subject: Re: Viewlogic v4.1 Plotter.exe cmd line usage?
From: zz80@digiserve.com (extra z to stop junk mail)
Date: Sat, 12 Oct 1996 20:03:30 GMT
Links: << >>  << T >>  << A >>

>The trick is to ask workview to keep a log (record) of what you are doing
>first, like open a single schematic, using the plot function, generate a
>postscript, close the schematic.

How do you get Viewlogic to keep a logfile?

The only command I could find is one which "saves" a log, and the
content of that is just menu commands activated by the mouse, plus
keyboard input.

Please remember I am using a Xilinx-only version of Viewlogic, not the
$25,000 full version.

Peter.
Article: 4309
Subject: Re: Viewlogic v4.1 Plotter.exe cmd line usage?
From: pak@cse.ucsc.edu (Pak K. Chan)
Date: 12 Oct 1996 21:19:55 GMT
Links: << >>  << T >>  << A >>
>How do you get Viewlogic to keep a logfile?
>The only command I could find is one which "saves" a log, and the

You should use "macro" at the viewdraw command line,
viewdraw will create a macro file.


Article: 4310
Subject: Re: Async with FPGA?
From: Ray Andraka <randraka@ids.net>
Date: Sun, 13 Oct 1996 10:09:32 -0700
Links: << >>  << T >>  << A >>
> >Hi, I'm looking for a way to shrink some Async circuit, has anyone done
> >any work with implementing some Async stuff with an FPGA?  I have an
> >application which needs to respond to an input within a small time period,
> >~7 to 10 ns.  unfortunately I can't just use a fast clock, so I was
> >thinking of trying to design some Asyn stuff in an FPGA.  any comments/
> >suggestions?     Thanks.
> >

David,

It sounds like what you are really asking is whether it is feasible to 
reliably implement an async state machine in an FPGA (ie a state machine 
that changes state on input transitions rather than on a system clock).
To answer your question, YES async state machines can be done in FPGAs.  
You need to be extra careful about delay paths vs placement.  Lots
of care is needed in the design to make sure all possible input 
conditions (with all possible timing combinations)result in predictable  
and consistent state changes. Rigorous attention to async state machine 
design practices should keep you out of trouble.  That said, I do not
recommend async design in FPGAs for the novice.  There are plenty of 
pitfalls both for the inexperienced FPGA designer and the inexperienced 
async state machine designer, and it is a genuine bear to troubleshoot or 
even verify.

-Ray Andraka, P.E.
Chairman, the Andraka Consulting Group
401/884-7930   FAX 401/884-7950
mailto:randraka@ids.net
http://www.ids.net/~randraka/
 
The Andraka Consulting Group is a digital hardware design firm 
specializing in high performance FPGA designs.  Services include 
complete design, development, simulation, and integration of these 
devices and the surrounding circuits.  We also evaluate,troubleshoot, 
and improve existing designs. Please call or write for a free 
brochure or visit our web site.
Article: 4311
Subject: Re: Info/opinions wanted for PCI interface in an FPGA
From: waynet@goodnet.com (Wayne Turner)
Date: Sun, 13 Oct 96 18:05:00 GMT
Links: << >>  << T >>  << A >>
Make sure that the part you choose is capable of meeting the PCI spec WRT 
setup times and clock-to-out.  I know that most Altera devices are 
PCI-compliant; don't know about Xilinx...



In article <53lunp$fvi@mordred.gatech.edu>, dan.bartram@gtri.gatech.edu (Dan 
Bartram) wrote:
>I'm planning on a PCI interface design and would like some first-hand
>experiences and opinions on implementing this interface in an FPGA,
>especially a Xilinx part.  If you have done this, any comments would be
>appreciated.
>
>thank you,
>
>Dan Bartram
>
>****************************************************************************
>Dan Bartram, Jr.
>Internet:  dan.bartram@.gtri.gatech.edu
>****************************************************************************
Article: 4312
Subject: 50 Mhz 24 CHANNEL LOGIC ANALYZER only $199
From: alex00009@aol.com (Alex 00009)
Date: 13 Oct 1996 16:09:16 -0400
Links: << >>  << T >>  << A >>
            50 Mhz 24 CHANNEL LOGIC ANALYZER only $199
	
	Simultaneous 50MHz sampling rate on all 24 channels
	PC based instrument with full color graphical user interface	
	57,600 baud RS232 link	
	Documented RS232 protocol
	Point and click timing analysis	
	Point and click state analysis
	FREE state machine option	
	FREE external clock option
	FREE terminal interface	
	FREE emulator source code
	Expandable to 192 channels	
	Programmable trigger
	Up to 32K samples per channel	
	Fully assembled and tested
	30 day money back guarantee	
	ONE YEAR WARRANTY

Get this powerful new tool at this incredibly low price!
Call ProBoard Circuits at (409)762-5436 for more information.
Article: 4313
Subject: Seeking 16V8: Vcc=3.0-5.0V: Zero standby power.
From: John Vickers <john@rhizik.demon.co.uk>
Date: Mon, 14 Oct 1996 10:28:32 +0100
Links: << >>  << T >>  << A >>
Hi.
I'm looking for a small PLD (preferably with 16V8 fusemap)
which will operate over this supply range.  Speed could
be anyhing.  35ns would be fine.

Does anyone make such a device ?

The power comes from a battery-backed (3.6VNiMh) supply,
which is 5V when the box is powered up, and 3.6-4.3V
when running off the battery.

A 4.8V NiCad/NiMh battery would be rather inconvenient:
e.g. need to supply 6V for charging.
-- 
John Vickers                                            Tel:    +44 1223 560129
Hardware / Software Person                              Fax:    +44 1223 563698
The Magic Board Company                                 john@rhizik.demon.co.uk
Article: 4314
Subject: Re: ORCA 2C10A - RAM placement advice
From: husby@fnal.gov (Don Husby)
Date: 14 Oct 1996 16:22:13 GMT
Links: << >>  << T >>  << A >>
<richard@milliped.demon.co.uk> wrote:
> Any ideas on how I can improve the placement/routing of the four address
> lines?  Any advice on hand placement?  Of course, I've already made the nets
> critical.  Is there the equivalent of the Xilinx longlines?  So far the

The simple questions:
Yes there is a long-line attribute.  ORCA typically uses the exact same
attributes as Xilinx.  (The only one that it still doesn't have is HBLKNM
which is a major problem for ORCA.)

It also helps to set a minimum skew of about 3ns on a net.  This will
force it to try to make all delays the same which usually means using long 
lines.

The hard question:
It's near impossible to route all of your address signals on
long lines.  For a dual-port memory, you really have 8 address lines in 
addition to a write-enable.  You also are probably using a common 2-bit
tri-state output bus (per PFU) and a common input bus.  Due to the
weird twisting thing that happens on the long lines, and the fact that
using a horizontal TBUF will block a vertical long-line input, it's near
impossible to use all of the vertical long lines well.  The half-lines work 
better.   Here are some rules of thumb:

Make sure the clock is routed in the same direction as the tristate data
bus (Usually horizontally).  Make sure your OE signals go in the opposite
direction (vertical) and on a long-line.  This allows part of the clock 
distribution resources to be used for routing OE.  This also requires you to 
do some hand routing.  Routing a few clock pins will set the pattern and PAR
will do the rest.

It's necessary to route the WE signal on a half-line.  WE can only
go in on A4/B4 and will get screwed up on the twisted Long Lines.

If possible, put RAM only in even (or only odd) rows and/or columns to avoid 
some twisted-long-line conflicts.  Doing both will eliminate almost all 
conflicts, but will make the array twice as big.

Take a shot at pre-placing TBUFS.  Typically, PAR wants to move the TBUF
away from the PFU that drives it (Probably to avoid the twisting problem).
I've had good luck pre-placing all of the TBUFs at the same site as their 
driving PFUs providing that you only use 2 horizontal full-length lines:
using more will block your vertical inputs.  The two lines should be adjacent
so that at least one line will have an alternate path on the local nodes. (You 
pretty much have to look in EPIC to see what I'm talking about here.  At any 
PFU, two of the TBUFs can be driven by two local nodes.  The other two
have only a single input).
  If you need 4 TBUFs (If you're using 4-bit single-port RAM) then 2 outputs
should go on long-lines and 2 should go on alternate half-length lines.  Again 
the reason for this is only explainable by pointing at an EPIC screen.


I have a simple floor planning tool, XPLACE, for ORCA that allows you
to place arrays of TBUFs and PFUs easily.  I put this at our FTP site
with a bunch of other tools:
  ftp://eseserver1.fnal.gov/pub/docx/xilinx/awks.zip  (1 Megabyte)

if you just want the documentation for these tools:
  ftp://eseserver1.fnal.gov/pub/docx/xilinx/mans.zip  (10K byte)

You might also want to look at some of the other tools which are
useful for xilinx and ORCA designs.

Article: 4315
Subject: Re: Async with FPGA?
From: "John L. Smith" <jsmith@univision.com>
Date: Mon, 14 Oct 1996 09:59:05 -0700
Links: << >>  << T >>  << A >>
David T. Wang wrote:
> 
> Hi, I'm looking for a way to shrink some Async circuit, has anyone done
> any work with implementing some Async stuff with an FPGA?  I have an
> application which needs to respond to an input within a small time period,
> ~7 to 10 ns.  unfortunately I can't just use a fast clock, so I was
> thinking of trying to design some Asyn stuff in an FPGA.  any comments/
> suggestions?     Thanks.
> 
>                                                 David

Xilinx application note XAPP028.001 shows a phase/frequency detector
for phase locked loops. This is an example of a simple async state
machine, although the note does not show the derivation. So, like
everyone else has said, it is possible to implement ASMs in FPGAs,
but can be tricky.
-- 
John L. Smith, Pr. Engr.     | Sometimes we are inclined to class
Univision Technologies, Inc. | those who are once-and-a-half witted
6 Fortune Dr.                | with the half-witted, because we
Billerica, MA 01821-3917     | appreciate only a third part of their wit.
jsmith@univision.com         | - Henry David Thoreau
Article: 4316
Subject: COURSES: High Level Design Using VHDL, Beaverton, Oregon
From: lindab@qualis.qualis.com (Linda Boyd)
Date: 14 Oct 1996 17:56:02 GMT
Links: << >>  << T >>  << A >>

                              A N N O U N C E

Qualis Design Corporation is offering additional sessions of our popular 
courses "High Level Design Using VHDL", and "Advanced Techniques Using VHDL"
at our Beaverton, Oregon, Training Center.

Our course "High Level Design Using VHDL" presents a comprehensive
introduction to VHDL and teaches the student how to approach complex
design tasks using High Level Design methods.  The "Elite Class" course
"Advanced Techniques Using VHDL" leverages off the student's foundation
in VHDL to teach advanced design and verification methods.  Both courses
provide high leverage knowledge for board, ASIC and FPGA designers.
For additional information about the material covered in these leading-edge
courses, see the course descriptions below.

These courses can also be held at your facilities in a private, one-on-one
setting -- contact us for more information.

Schedule
--------
The Portland course schedule through December follows:

       Course Title                        Course Dates      Status
       -------------------------------------------------------------------
       High Level Design Using VHDL:       Nov 04 - Nov 08   open


       Advanced Techniques Using VHDL:     Nov 13 - Nov 15   open


For information about the material covered in these leading-edge courses,
see the course descriptions below.

The Qualis Difference
---------------------
We know what it's like to work under the pressure of aggressive schedules and
immense technical challenges.  We believe that High Level Design methods
and technology, such as HDL-based verification and synthesis, are the key
to tackling those challenges and conquering today's design problems.
Our courses can really make a difference in your day-to-day work life
by showing you the high leverage points of VHDL and High Level Design.
Here's how we do it:

 -- The Qualis "Best In Class" Instructor Team draws upon the absolute best
    VHDL and Verilog consulting and training talent available.  Our
    instructors, top-notch Engineers with cutting-edge design experience,
    know how to relate the course material to your real-world design problems.

 -- Our courses are intense, hands-on events using the latest EDA tools
    and hardware.  Everything you need to learn quickly and efficiently is
    provided -- you supply the brain, we'll supply everything else.

 -- Our courses are like no other in the EDA industry.  Engineers and
    Managers who attend our courses will learn what's important and why, and
    where to focus their time and resources for maximum leverage from HDLs
    and design tools.  And, unlike other vendor courses, our courses are
    *dynamic* -- we constantly update our material with the latest in
    High Level Design techniques and information, so you're assured of
    learning the latest in the field.

--  Our courses are respected in the industry.  We have taught our
    High Level Design courses to dozens of companies and hundreds of
    Engineers and Managers.  Our student references attest to the
    outstanding quality and real-world usefulness of our classes.

About Qualis Design Corporation
-------------------------------
Founded in 1992, Qualis Design Corporation has quickly become the
leading independent provider of High Level Design consulting and
training services.  The company provides services to leading-edge
high technology firms worldwide, including Intel, Hewlett-Packard,
Tektronix, Xerox, TRW, and Northern Telecom.  Qualis' corporate
headquarters are located in Beaverton, Oregon.

Don't miss this opportunity to learn the latest in High Level Design from
the best in the industry.  For course syllabi and registration information,
contact us at:

                         Linda Boyd, Training Registrar
                           Qualis Design Corporation
                        8705 SW Nimbus Avenue, Suite 118
                          Beaverton, Oregon 97008 USA

                            Phone: +1-503-644-9700
                             FAX: +1-503-644-1583
                         E-mail: training@qualis.com
                         World Wide Web: coming soon!


Brief Course Descriptions
--------------------------

                    High Level Design Using VHDL

                          Course Overview

           Copyright (c) 1995, 1996 Qualis Design Corporation

"High Level Design Using VHDL" is a fast paced, 5-day hands-on,
multimedia course designed not only to teach High Level Design
techniques and the VHDL language, but to make class participants
immediately productive in a system design environment using
state-of-the-art simulation and synthesis tools.

After an introduction to VHDL, the course deviates from the
traditional bottom-up, gates-to-behavioral modeling presentation of
other VHDL courses and reverses the flow, teaching top-down
design practices, with early special emphasis on coding for synthesis,
efficient testbench generation and advanced design verification
techniques. These skills are reinforced throughout the week while
teaching VHDL from a High Level Design perspective.

The course labs are designed to accommodate the learning aptitudes of
a wide range of students with diverse design experiences. Each lab is
structured into three parts:

     1.        Fundamental Concepts Review and Experience
     2.        Recognition of Common Mistakes and Correcting Problems
     3.        Additional Material for Advanced Students

All students complete parts one and two of each lab.  Part three is
for students who finish early and want to learn additional material.
This lab structure caters to all student skill levels and provides
excellent opportunities to expand one's knowledge of VHDL simulation
and synthesis techniques.

Each day of class is divided into multiple interactive lecture and
lab sessions.  Students have access to individual Sun Sparcstations,
the VHDL simulation environment of their choice, and the Synopsys
DC Expert synthesis environment for use during the lab sessions.
The course material is presented using a projection system that allows
30% more material to be presented in a given amount of time with vivid,
interest-grabbing color slides.

Full Course Syllabus Available
------------------------------
A full course syllabus listing all topics covered in this course is available.
Contact us for more information.


------------------------------------------------------------------------------

                   Advanced Techniques Using VHDL

                          Course Overview

             Copyright (c) 1996 Qualis Design Corporation

"Advanced Techniques Using VHDL" is a fast paced, 3-day hands-on,
multimedia course designed to bring Engineers with experience in
using VHDL to an unparalleled level of efficiency.

After a review of the more advanced behavioral constructs and
features of VHDL, the course immediately illustrates how these
constructs can turn an ordinary-looking environment into a design
simulation and verification powerhouse through the use of
bus-functional models, test harnesses, abstract regressionable
testbenches and behavioral models.

Each day of class includes interactive lecture sessions with
8 challenging labs.  Students will have access to individual Sun
Sparcstations and the VHDL simulation environment of their
choice.  The course material is presented using a projection
system that allows 30% more material to be presented in a given amount
of time with vivid, interest-grabbing color slides.

Student Questions
-----------------
Prior to the first day of the course, registered students are
encouraged to submit questions related to their current usage
of VHDL to the instructor.  Questions or topics that illustrate
modeling techniques or features of the language will be included
in the course presentation.

Course Prerequisites
--------------------
Due to the fast-paced nature of this class and the caliber of the
students attending, all registrants must have prior experience with
the VHDL language, or previously attended the Qualis course
"High Level Design Using VHDL".  Knowledge of a structured
programming language, such as C or Pascal, is recommended.

Full Course Syllabus Available
------------------------------
A full course syllabus listing all topics covered in this course is available.
Contact us for more information.

------------------------------------------------------------------------------

                           Qualis Design Corporation
                        8705 SW Nimbus Avenue, Suite 118
                          Beaverton, Oregon 97008 USA

                            Phone: +1-503-644-9700
                             FAX: +1-503-643-1583
                         E-mail: training@qualis.com
                         World Wide Web: coming soon!

"DC Expert" is a trademark of Synopsys, Inc.
"Verilog" is a registered trademark of Cadence Design Systems, Inc.

Copyright (c) 1995, 1996, Qualis Design Corporation.  All Rights Reserved.

Article: 4317
Subject: COURSES: High Level Design Using Verilog, Beaverton, Oregon
From: lindab@qualis.qualis.com (Linda Boyd)
Date: 14 Oct 1996 17:57:52 GMT
Links: << >>  << T >>  << A >>

                            A N N O U N C E

Qualis Design Corporation is offering additional sessions of our
popular courses, "High Level Design Using Verilog", and "Advanced Techniques
Using Verilog" at our Beaverton, Oregon, Training Center.

Our course "High Level Design Using Verilog" presents a comprehensive
introduction to Verilog and teaches the student how to approach complex
design tasks using High Level Design methods.  The "Elite Class" course
"Advanced Techniques Using Verilog" leverages off the student's foundation
in Verilog to teach advanced design and verification methods.  Both
courses provide high leverage knowledge for board, ASIC and FPGA designers.
For additional information about the material covered in these leading-edge
courses, see the course descriptions below.

These courses can also be held at your facilities in a private, one-on-one
setting -- contact us for more information.

Schedule
--------
The Portland course schedule through December follows:

       Course Title                        Course Dates      Status
       -------------------------------------------------------------------

       High Level Design Using Verilog:    Oct 21 - Oct 25   filling fast!
                                           Nov 18 - Nov 22   open


       Advanced Techniques Using Verilog:  Oct 28 - Oct 30   open
                                           Dec 09 - Dec 11   open

For information about the material covered in these leading-edge courses,
see the course descriptions below.

The Qualis Difference
---------------------
We know what it's like to work under the pressure of aggressive schedules and
immense technical challenges.  We believe that High Level Design methods
and technology, such as HDL-based verification and synthesis, are the key
to tackling those challenges and conquering today's design problems.
Our courses can really make a difference in your day-to-day work life
by showing you the high leverage points of Verilog and High Level Design.
Here's how we do it:

 -- The Qualis "Best In Class" Instructor Team draws upon the absolute best
    Verilog and VHDL consulting and training talent available.  Our
    instructors, top-notch Engineers with cutting-edge design experience,
    know how to relate the course material to your real-world design problems.

 -- Our courses are intense, hands-on events using the latest EDA tools
    and hardware.  Everything you need to learn quickly and efficiently is
    provided -- you supply the brain, we'll supply everything else.

 -- Our courses are like no other in the EDA industry.  Engineers and
    Managers who attend our courses will learn what's important and why, and
    where to focus their time and resources for maximum leverage from HDLs
    and design tools.  And, unlike other vendor courses, our courses are
    *dynamic* -- we constantly update our material with the latest in
    High Level Design techniques and information, so you're assured of
    learning the latest in the field.

--  Our courses are respected in the industry.  We have taught our
    High Level Design courses to dozens of companies and hundreds of
    Engineers and Managers.  Our student references attest to the
    outstanding quality and real-world usefulness of our classes.

About Qualis Design Corporation
-------------------------------
Founded in 1992, Qualis Design Corporation has quickly become the
leading independent provider of High Level Design consulting and
training services.  The company provides services to leading-edge
high technology firms worldwide, including Intel, Hewlett-Packard,
Tektronix, Xerox, TRW, and Northern Telecom.  Qualis' corporate
headquarters are located in Beaverton, Oregon.

Don't miss this opportunity to learn the latest in High Level Design from
the best in the industry.  For course syllabi and registration information,
contact us at:

                        Linda Boyd, Training Registrar
                           Qualis Design Corporation
                       8705 SW Nimbus Avenue, Suite 118
                          Beaverton, Oregon 97008 USA

                            Phone: +1-503-644-9700
                             FAX: +1-503-643-1583
                         E-mail: training@qualis.com
                         World Wide Web: coming soon!

Brief Course Descriptions
--------------------------

                   High Level Design Using Verilog

                          Course Overview

           Copyright (c) 1995, 1996 Qualis Design Corporation

"High Level Design Using Verilog" is a fast paced, 5-day hands-on,
multimedia course designed not only to teach High Level Design
techniques and the Verilog language, but to make class participants
immediately productive in a system design environment using
state-of-the-art simulation and synthesis tools.

After an introduction to Verilog, the course deviates from the
traditional bottom-up, gates-to-behavioral modeling presentation of
other Verilog courses and reverses the flow, teaching top-down
design practices, with early special emphasis on coding for synthesis,
efficient testbench generation and advanced design verification
techniques. These skills are reinforced throughout the week while
teaching Verilog from a High Level Design perspective.

The course labs are designed to accommodate the learning aptitudes of
a wide range of students with diverse design experiences. Each lab is
structured into three parts:

     1.        Fundamental Concepts Review and Experience
     2.        Recognition of Common Mistakes and Correcting Problems
     3.        Additional Material for Advanced Students

All students complete parts one and two of each lab.  Part three is
for students who finish early and want to learn additional material.
This lab structure caters to all student skill levels and provides
excellent opportunities to expand one's knowledge of Verilog simulation
and synthesis techniques.

Each day of class is divided into multiple interactive lecture and
lab sessions.  Students have access to individual Sun Sparcstations,
the Verilog simulation environment of their choice, and the Synopsys
DC Expert synthesis environment for use during the lab sessions.
The course material is presented using a projection system that allows
30% more material to be presented in a given amount of time with vivid,
interest-grabbing color slides.

Full Course Syllabus Available
------------------------------
A full course syllabus listing all topics covered in this course is available.
Contact us for more information.


------------------------------------------------------------------------------

                  Advanced Techniques Using Verilog

                          Course Overview

             Copyright (c) 1996 Qualis Design Corporation

"Advanced Techniques Using Verilog" is a fast paced, 3-day hands-on,
multimedia course designed to bring Engineers with experience in
using Verilog to an unparalleled level of efficiency.

After a review of the more advanced behavioral constructs and
features of Verilog, the course immediately illustrates how these
constructs can turn an ordinary-looking environment into a design
simulation and verification powerhouse through the use of
bus-functional models, test harnesses, abstract regressionable
testbenches and behavioral models.

Each day of class includes interactive lecture sessions with
8 challenging labs.  Students will have access to individual Sun
Sparcstations and the Verilog simulation environment of their
choice.  The course material is presented using a projection
system that allows 30% more material to be presented in a given amount
of time with vivid, interest-grabbing color slides.

Student Questions
-----------------
Prior to the first day of the course, registered students are
encouraged to submit questions related to their current usage
of Verilog to the instructor.  Questions or topics that illustrate
modeling techniques or features of the language will be included
in the course presentation.

Course Prerequisites
--------------------
Due to the fast paced nature of this class and the caliber of the
students attending, all registrants must have prior experience with
the Verilog language, or previously attended the Qualis course
"High Level Design Using Verilog".  Knowledge of a structured
programming language, such as C or Pascal, is recommended.

Full Course Syllabus Available
------------------------------
A full course syllabus listing all topics covered in this course is available.
Contact us for more information.

------------------------------------------------------------------------------

                           Qualis Design Corporation
                       8705 SW Nimbus Avenue, Suite 118
                          Beaverton, Oregon 97008 USA

                            Phone: +1-503-644-9700
                             FAX: +1-503-643-1583
                         E-mail: training@qualis.com
                         World Wide Web: coming soon!


"DC Expert" is a trademark of Synopsys, Inc.
"Verilog" is a registered trademark of Cadence Design Systems, Inc.

Copyright (c) 1995, 1996, Qualis Design Corporation.  All Rights Reserved.

Article: 4318
Subject: LPM standard support?
From: ast-AE@actcom.co.il (Moshe Zalcberg)
Date: Mon, 14 Oct 1996 20:27:44 GMT
Links: << >>  << T >>  << A >>
I've heard lots on LPM from Altera stuff and lately I've found on the 
EDIF web that LPM is indeed somekind of standard. 
What tool/fpga/asic vendors actually support the LPM standard? 

MZ 

Article: 4319
Subject: Re: Xilinx XACT Performance Appl. Note?
From: Marc Baker <marc.baker@xilinx.com>
Date: Mon, 14 Oct 1996 13:37:25 -0700
Links: << >>  << T >>  << A >>
An older application note, XAPP043, "Improving XC4000 Design
Performance," has some additional information on using
XACT-Performance.  However, the app note is limited to
the older path-type specs and doesn't include the newer
group-type specs.  Xilinx plans to update the app note.

The app note can be found in Acrobat format at

http://www.xilinx.com/xapp/xapp043.pdf

The documentation for XACT-Performance can be found at

ftp://ftp.xilinx.com/pub/documentation/dsref1.pdf
Article: 4320
Subject: Re: Anyone using Altera MaxPlus VHDL ???
From: "Petrus Pelser" <petrus@fpgatec.co.za>
Date: Mon, 14 Oct 1996 22:25:02 GMT
Links: << >>  << T >>  << A >>


Mike Treseler <tres@tc.fluke.COM> wrote in article
<Dz0wvA.H6p@tc.fluke.COM>...
> sja@world.std.com (Stuart J Adams) writes:
> 
> >I am considering getting Altera's MaxPlus VHDL synthesis
> >add-on. Was wondering if anyone is using this and how it
> >compares to other synthesis tools such as Synplicity,
> >Synopsys, or Exemplar ???
> 
> 
> >Thanks,
> >  Stuart Adams
> >  sja@world.std.com
> 
> >-- 
> 
> >Stuart Adams
> >Chief Engineer
> >Bright Star Engineering
> 
> We use Exemplar, but I have spent a day on MaxPlus VHDL.
> Maxplus is very easy to use and has well integrated tools.
> The downside is Altera-only synthesis and no support of VHDL
> testbenches.
> 
> ----------------------------------------------------------------------
>      tres@tc.fluke.com                       Mike Treseler
>                                              Fluke Networks Division
> voice:  (206) 356-5409                       Fluke Corporation
> fax:    (206) 356-5043                       P.O. Box 9090
>                                              Everett, WA  98206-9090
> ----------------------------------------------------------------------
> 
> 

I use the Max+Plus 2 VHDL compiler for all my Altera designs. It might not
be the best, but it works fine. There is a few unsupported features or
deviations from the IEEE spec, but for most of them there are easy work
arounds. These are more VHDL compiler issues, when you get to the VHDL
synthesis issues MaxPlus2 does the job.

Feel free to contact me for more info

Petrus Pelser
petrus@fpgatec.co.za

Article: 4321
Subject: Update on Atmel AT17C128 Problem
From: Andrew Siska <asiska@interaccess.com>
Date: Mon, 14 Oct 1996 20:54:05 -0700
Links: << >>  << T >>  << A >>
Hello and thanks to all who gave advice related to my AT17C128 startup
problem.

The problem was related to my programmer's buffer containing data from a
previous data file which occupied the reset polarity bits for the Atmel
device.  Once I cleared the buffer (hex 0s), loaded the file, and
programmed the part - it fired up properly.  No more OTP serial proms
will be trashed during my development phase ($$)!


BTW, if anyone plans (I would recommend they do) on using the Atmel
parts for prototyping they should get Atmels 17Cxxx FAQ from their local
FAE or download it from their web site at www.atmel.com.

Thanks for your input,

Andy

Embedded Systems Electronics, Inc.
(630) 960-0052
Article: 4322
Subject: Re: LPM standard support?
From: Veli-Matti Karppinen <ventti@fincitec.fi>
Date: Tue, 15 Oct 1996 08:32:58 +0200
Links: << >>  << T >>  << A >>
Moshe Zalcberg wrote:
> 
> I've heard lots on LPM from Altera stuff and lately I've found on the
> EDIF web that LPM is indeed somekind of standard.
> What tool/fpga/asic vendors actually support the LPM standard?
> 
> MZ

Moshe,

You will find the LPM support atleast from VeriBest, for more info
look at http://www.veribest.com

Regards,

Veli-Matti Karppinen
-- 
*************************************************************************
**			    	 				       **
** Veli-Matti Karppinen	      		Internet: ventti@fincitec.fi   **
** Fincitec Oy				         		       **
** P.B. 11	               		tel. +358-698-21490	       **
** FIN-94601 KEMI,FINLAND    		fax. +358-698-21561	       **
**								       **
*************************************************************************
Article: 4323
Subject: Xilinx xchecker.exe and Windows NT
From: kobler@mppmu.mpg.de (Thomas Kobler)
Date: 15 Oct 1996 16:43:23 GMT
Links: << >>  << T >>  << A >>
Hello!

I would like to ask if anybody had success in using the Xilinx
xchecker cable and xchecker.exe V5.1 od V5.2 under Windows
NT3.51 to download a LCA? 

To me it keeps crashing or it ignoring the download cable almost
every time.

Any hints?

			

Article: 4324
Subject: Re: Update on Atmel AT17C128 Problem
From: dan.bartram@gtri.gatech.edu (Dan Bartram)
Date: Tue, 15 Oct 96 22:26:22 GMT
Links: << >>  << T >>  << A >>
In article <32630ADD.74E8@interaccess.com>, asiska@interaccess.com wrote:
>Hello and thanks to all who gave advice related to my AT17C128 startup
>problem.

>No more OTP serial proms
>will be trashed during my development phase ($$)!

I thought the AT17C128 serial proms were EEPROMS.  At least the ones
I used were.

Perhaps I missed the original posts, but I would definitely use the EE 
versions during development.



****************************************************************************
Dan Bartram, Jr.
Internet:  dan.bartram@.gtri.gatech.edu
****************************************************************************


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