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 43575

Article: 43575
Subject: Re: Small FIFOs in Spartan
From: "Falk Brunner" <Falk.Brunner@gmx.de>
Date: Fri, 24 May 2002 17:11:59 +0200
Links: << >>  << T >>  << A >>
"Børge Strand" <borge.strand.remove.if.not.spamming@sintef.no> schrieb im
Newsbeitrag news:1022230585.37631@halvan.trd.sintef.no...
> I will need some really small FIFOs for my next project. The word length
> will be 24 bits, but the depth of the FIFO is not requirred to be more
than
> two to sixteen.
>
> Could dedicated parts of the Spartan be used for such small FIFOs? I plan
to
> program the thing in VHDL.

If this is for educational purpose, you can use the dual port capability of
the Spartan LUTs.
If it is "just" to solve a problem, use coregenerator to create the FIFO.
But for Spartan (do you really mean Spartan, not Spartan-II?) this is only
possible with Foundation 3.3 or older, since the synchronous FIFO for
Spartan is missing in newer version of coregen (Xilinx is pushing its
customers to leave the old parts by stoping to support them as usual)

--
MfG
Falk





Article: 43576
Subject: Targeting Virtex with WebPack
From: kjhales <kjhales@catalpatechnology.com>
Date: Fri, 24 May 2002 11:17:31 -0400
Links: << >>  << T >>  << A >>
I have a customer who has off-loaded a small portion of a project to my
company.  Originally the project was to use a Virtex E part, but it got
changed to Virtex along the way (not under my control). I've always used
Altera products in the past; this is my first Xilinx project

In the near future, I will have more capable tools to work with, but I
was hoping to get a head start using WebPack. Unfortunately, Virtex is
not one of the supported device families. I understand Spartan 2
bitstreams are compatible (yes, I've read Xilinx's positions on this).

Here's the question:
The Virtex part is in a 240 pin package, and the Spartan 2 line does not
support this package. Can I target a Spartan 2 in 144 pin package, and
load that bitstream into a 240 pin Virtex?
Spartan 2 docs seem to show how pins for same chip in different packages
correlate, but Virtex docs don't show this (or I've not found it). If
the bitstream works, how do I match up pins between the 144 and 240
packages?

Thank you.

--
Kevin Hales
Catalpa Technology, Inc.
302 E. Davis St. Ste 211
Culpeper, VA 22701
540-727-8005




Article: 43577
Subject: Re: Small FIFOs in Spartan
From: John_H <johnhandwork@mail.com>
Date: Fri, 24 May 2002 15:23:42 GMT
Links: << >>  << T >>  << A >>
For new designs, the Spartan-II or Spartan-IIE may be more cost effective.

The CLB SelectRAM FIFO elements don't need to be in a single row so the smallest
device in the families are the XC2S15 (8x12 CLB array) or the XC2S50E (16x24 CLB
array) for the newer family.

One 24 bit word, 2 to 16 words deep would require 24 *slices* or 12 CLBs using a
DPRAM, the native SRL16 elements in the Spartan-II(E) series would only require
half that.  In either case you'll need a little extra logic (or pins) for address
manipulation.



Utku Ozcan wrote:

> "Børge Strand" wrote:
>
> > I will need some really small FIFOs for my next project. The word length
> > will be 24 bits, but the depth of the FIFO is not requirred to be more than
> > two to sixteen.
> >
> > Could dedicated parts of the Spartan be used for such small FIFOs? I plan to
> > program the thing in VHDL.
> >
> > Regards,
> > Børge
>
> Let's calculate:
>
> your FIFO will use a DP RAM of 24 x 16 max. a DP RAM is composed of
> 16x1 RAM16X1D.v primitives. Thus:
>
> NUM_OF_DPRAMS_OF_BORGE = (24 x 16) / (16 x 1) = 24 CLBs.
> This gives you 24 CLBs laid horizontally on the chip.
>
> This DPRAM can be generated with Logiblox but your chip must be greater
> than or equal to Xilinx Spartan/XL XCS30XL (24x24 CLB matrix).
>
> Otherwise, you have to divide your words into smaller words.
> This is not bad, but uses twice routing resources the one above has.
> More routing resources cause longer PAR times.
>
> I'm using a 32x8 FIFO on a Xilinx Spartan-XL XCS40XL-4-PQ208C.
>
> Utku


Article: 43578
Subject: Re: SDRAM controler for Virtex-II
From: Laurent Gauch <laurent.gauch@amontec.com>
Date: Fri, 24 May 2002 17:37:15 +0200
Links: << >>  << T >>  << A >>
Sorry, my last link was for verilog sources

try for vhdl one:
http://www.latticesemi.com/account/_download.cfm?AMID=2886

Or use the search on http://www.latticesemi.com/

Laurent Gauch
Amontec

Laurent Gauch wrote:

> Philippe,
> 
> try this free HDL source of sdram controller:
> http://www.latticesemi.com/account/_download.cfm?AMID=3866
> 
> Since I want to includ it on a spartanII in next months, could you give 
> me a feedback, if you use it.
> 
> Laurent Gauch
> 
> Philippe Robert wrote:
> 
>> Hi,
>>
>> I am after an SDRAM controller in VHDL for Virtex-II.
>> Does anyone know where I could get one from ?
>>
>> Thanks.
>> Philippe.
>>
>>
>>
> 


Article: 43579
Subject: Re: FPGA, VHDL : RAM initialization
From: m0 <>
Date: Fri, 24 May 2002 07:38:18 -0800
Links: << >>  << T >>  << A >>
? = normal
?? = typo
??? = drunk
???? = stupid

Article: 43580
Subject: Xilinx Pull Ups/Dpwns
From: "D Brown" <dbrown123@shaw.ca>
Date: Fri, 24 May 2002 09:43:48 -0600
Links: << >>  << T >>  << A >>
I notice that in the ISE4.2 Constraints Editor you can specifiy whether an
input or output has a pull up or pull down on it. What is the default if you
don't specify anything? I'm using an XCS10XL.
Thanks,
Dave




Article: 43581
Subject: Re: JTAG ICE or programmer
From: Laurent Gauch <laurent.gauch@amontec.com>
Date: Fri, 24 May 2002 17:50:12 +0200
Links: << >>  << T >>  << A >>
That will be not very easy to write the same software to programm Xilinx 
and Altera FPGAs and Atmel AVRs. That's the job of www.jtag.com . Price 
about $30K for the jtag software, but I know why!
Same hardware is possible using the universal POD called Chameleon POD
http:\\www.amontec.com\chameleon.shtml

Laurent

Frank Scherler wrote:

> Hello 
> 
> I am looking for a JTAG ICE. I don't wanne spend much money on it.  
> I want to be able to have a look on the schematics and to be able to
> compile the software on my own. I am looking for hardware to programm
> Xilinx and Altera FPGAs and Atmel AVRs with one Software and one
> hardware. If it would be possible to debug software over JTAG with
> this hardware it would be great. Dose anyone know any Project with
> this topic?
> 
> regards
> Frank
> 


Article: 43582
Subject: Re: Xilinx STARTUP and PADS problems
From: Steven Elzinga <steven.elzinga@xilinx.com>
Date: Fri, 24 May 2002 09:55:43 -0600
Links: << >>  << T >>  << A >>
Some comments:

Przemyslaw Wegrzyn wrote:

> I use Xilinx ISE 4.1 (FPGA Express 3.6.1) + Active-HDL.
> Target chip: XC4003E
>
> 1. I/OPADS problem.
>
> Should I place PADS or PORTS next to I/OBUFs ? PADS are a little
> uncomfortable to me, as I can't use such entity for simulation,
> thus I use ports everywhere. I've tried to use FPGA Express option to
> inert I/O pads but it doesn't seem to work.
> Finally I've found similar option in the implementation tools ("Create
> I/O Pads from Ports"), and it works, but I'm not
> sure if it's a right solution.

Selecting the implementation option "Create I/O Pads from Ports" will
insert IPAD/OPAD primitves on your ports in the EDIF netlist.  The FPGA
Express option "Insert I/O Pads" will insert both the IPAD/OPAD as well as
the IBUF/OBUF primitives except in certain cases when FPGA Express infers
a BUFG for the clock line.  You will know when you have reached this
'special' condition when your design goes through the map portion of the
implementation tools as map will trim away everything associated with that
clock line.  The reason is that FPGA Express is not inserting an IPAD/BUFG
combination (required for 4K).  When the design then goes through
implementation map sees this portion of logic as being not connected.

FYI for Virtex devices you do not need IPADs/OPADs although they are valid
primitives for the part.  Furthermore, the "Insert I/O Pads" option in
FPGA Express will only insert IBUF/OBUF type components.

>
>
> 2. How to use GSR ?
> I've placed STARTUP component on my toplevel schematic, connected IBUF
> and port to it, it seems to be ok -
> I can see these components connected together in FPGA Editor. But GSR
> option is disabled in STARTUP block :(
> I don't know what to do next. There's a net that sets/resets all the
> flipflops (I've checked it in post-synth EDIF netlist),
> but I don't know what to do with it - place IBUF/port on it ?

Excuse me if I am missing the question on this one as I do not know what
could be the problem.  Here are some possible scenarios for using the
STARTUP block:

If the design is all HDL then having an asynchronous reset will let FPGA
Express automatically infer the STARTUP block.

If the design is all HDL with a black box core then an option needs to be
set to tell FPGA Express to ignore black boxes when connecting the STARTUP
block.

Connect your reset line to the output of the instantiated IBUF.


>
>
> 3. Clocks
>
> I use 2 clocks in my design, one external and OSC4 output. I pass the
> clocks through BUFGS.
> During implementation I get the following warning:
>
> Timing - Clock nets using non-dedicated resources were found in this
> design....
>
> Why ? Both clocks uses dedicated global nets !

A possbility would be a gated clock or a latch that got inferred.  A quick
way to check the validity of the warning message is to examine your design
in FPGA Editor and follow the clock line.

>
>
> -=Czaj-nick=-

Steve



Article: 43583
Subject: Re: Time for a new computer. Suggestions?
From: rickman <spamgoeshere4@yahoo.com>
Date: Fri, 24 May 2002 13:14:24 -0400
Links: << >>  << T >>  << A >>
Hal Murray wrote:
> 
> >Agreed,  The raid is in the current system for reliability.  I've suffered 3
> >or 4 disk failures over the years and the time spent recovering from backups
> >is well worth the extra cost of having a raid system.
> 
> Don't forget to back things up (somehow) anyway.  RAID won't protect
> you from software/mushware/operator errors.

Just to add my two cents worth.  I use a manual disk mirror approach for
protection against HW, SW and "mushware" issues.  The second HD is an
exact duplicate of the first and is updated on a regular basis.  We do
this once a week, but the weak point in the system is the operator
reliability.  If I forget, I have lost some of my protection.  

By not having automatic duplication, we have some protection against
accidentally deleted file and other issues from software installation
and such.  If we decide the backup is eaiser to work with than the
messed up current drive, we swap the drives and copy the backup to the
old original.  

This has saved us in some situations where a software crash wiped out a
significant part of the OS, once when we were infected with a virus and
many times when we did something to a file that we regretted.  

So RAID may not be the best option if you have the discipline to do your
backups.  

-- 

Rick "rickman" 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      URL http://www.arius.com
4 King Ave                               301-682-7772 Voice
Frederick, MD 21701-3110                 301-682-7666 FAX

Article: 43584
Subject: Re: Xilinx Pull Ups/Dpwns
From: "gosensgo@goober_lumictech.com" <gosensgo@GOOBERlumictech.com>
Date: Fri, 24 May 2002 17:27:07 GMT
Links: << >>  << T >>  << A >>
The default would be neither.

"D Brown" <dbrown123@shaw.ca> wrote in message
news:acln3d$917$1@pallas.novatel.ca...
> I notice that in the ISE4.2 Constraints Editor you can specifiy whether an
> input or output has a pull up or pull down on it. What is the default if
you
> don't specify anything? I'm using an XCS10XL.
> Thanks,
> Dave
>
>
>



Article: 43585
Subject: Re: Time for a new computer. Suggestions?
From: nweaver@CSUA.Berkeley.EDU (Nicholas Weaver)
Date: Fri, 24 May 2002 17:40:53 +0000 (UTC)
Links: << >>  << T >>  << A >>
In article <3CEE74F0.1CC15C51@yahoo.com>,
rickman  <spamgoeshere4@yahoo.com> wrote:
>Hal Murray wrote:
>> 
>> >Agreed,  The raid is in the current system for reliability.  I've suffered 3
>> >or 4 disk failures over the years and the time spent recovering from backups
>> >is well worth the extra cost of having a raid system.
>> 
>> Don't forget to back things up (somehow) anyway.  RAID won't protect
>> you from software/mushware/operator errors.
>
>Just to add my two cents worth.  I use a manual disk mirror approach for
>protection against HW, SW and "mushware" issues.  The second HD is an
>exact duplicate of the first and is updated on a regular basis.  We do
>this once a week, but the weak point in the system is the operator
>reliability.  If I forget, I have lost some of my protection.  
>
>By not having automatic duplication, we have some protection against
>accidentally deleted file and other issues from software installation
>and such.  If we decide the backup is eaiser to work with than the
>messed up current drive, we swap the drives and copy the backup to the
>old original.  
>
>This has saved us in some situations where a software crash wiped out a
>significant part of the OS, once when we were infected with a virus and
>many times when we did something to a file that we regretted.  
>
>So RAID may not be the best option if you have the discipline to do your
>backups.  

My 2 cents worth:

Retrospect.  It is a wonderful mac backup tool, doing both full and
incremental backups, both on the machine and to other client machines
over the network.  My dad has had it save his bacon on his production
mac system many times: he uses both a nightly incremental backup of
everything and a nightly copying of critical files onto other drives.

They finally have a Windows version as well, which if it works half as
good as the Mac version, makes it worth its weight in gold.
-- 
Nicholas C. Weaver                                 nweaver@cs.berkeley.edu

Article: 43586
Subject: Xilinx "Real PCI-X" core capabilities
From: no-george-spam@cmf.nrl.navy.mil (George Un-Spam Schmitt)
Date: Fri, 24 May 2002 19:28:20 +0000 (UTC)
Links: << >>  << T >>  << A >>
I am building a card which will initially be used as a PCI 64/66 card
and eventually as a PCI-X card (also 64/66).  The card will use a
Xilinx as the PCI/PCI-X interface.

In the datasheet for the Xilinx "Real PCI-X" core there is a listing
of the various modes in which the core can operate.  PCI 64/33 is
listed but PCI 64/66 is not.  Am I correct to assume that if I place a
device utilizing this core into a PCI bus running 64-bits wide at
66-MHz that it will not be able to run at 66 MHz?  My interest in
using the PCI-X core for my PCI 64/66 design is based on a desire to
use the same core (and card) for both modes.

According to the datasheets for the PCI-X and PCI 64/66 cores, both
can operate on the XCV300E.  Can both cores operate with the same
pinout on this device?  Until I buy either of them, it seems difficult
to get information like this from the Xilinx website.

Thanks.

-George


Article: 43587
Subject: Re: Time for a new computer. Suggestions?
From: Ray Andraka <ray@andraka.com>
Date: Fri, 24 May 2002 21:10:27 GMT
Links: << >>  << T >>  << A >>
I currently do my backups onto a drive on another machine.  The hard part is
remembering to do the copy.   I don't trust one machine to hold the data
uncorrupted.  I had a case a year or so ago where my raid card bellied up and took
out the data on the raid array.  I am still concerned about a virus getting into the
network and wiping both, so I keep not-frequent-enough-backups of the accounting and
project files onto CDs.  I've dabbled with tapes a couple of times, but frankly have
never had much success with them.  I had attempted to use an IDE drive on my machine
which I did disk images too on a frequent basis, and intended to disable in the BIOS
except when doing the back-up.  Good intention, but too inconvenient to keep it off
line.

Nicholas Weaver wrote:

> In article <3CEE74F0.1CC15C51@yahoo.com>,
> rickman  <spamgoeshere4@yahoo.com> wrote:
> >Hal Murray wrote:
> >>
> >> >Agreed,  The raid is in the current system for reliability.  I've suffered 3
> >> >or 4 disk failures over the years and the time spent recovering from backups
> >> >is well worth the extra cost of having a raid system.
> >>
> >> Don't forget to back things up (somehow) anyway.  RAID won't protect
> >> you from software/mushware/operator errors.
> >
> >Just to add my two cents worth.  I use a manual disk mirror approach for
> >protection against HW, SW and "mushware" issues.  The second HD is an
> >exact duplicate of the first and is updated on a regular basis.  We do
> >this once a week, but the weak point in the system is the operator
> >reliability.  If I forget, I have lost some of my protection.
> >
> >By not having automatic duplication, we have some protection against
> >accidentally deleted file and other issues from software installation
> >and such.  If we decide the backup is eaiser to work with than the
> >messed up current drive, we swap the drives and copy the backup to the
> >old original.
> >
> >This has saved us in some situations where a software crash wiped out a
> >significant part of the OS, once when we were infected with a virus and
> >many times when we did something to a file that we regretted.
> >
> >So RAID may not be the best option if you have the discipline to do your
> >backups.
>
> My 2 cents worth:
>
> Retrospect.  It is a wonderful mac backup tool, doing both full and
> incremental backups, both on the machine and to other client machines
> over the network.  My dad has had it save his bacon on his production
> mac system many times: he uses both a nightly incremental backup of
> everything and a nightly copying of critical files onto other drives.
>
> They finally have a Windows version as well, which if it works half as
> good as the Mac version, makes it worth its weight in gold.
> --
> Nicholas C. Weaver                                 nweaver@cs.berkeley.edu

--
--Ray Andraka, P.E.
President, the Andraka Consulting Group, Inc.
401/884-7930     Fax 401/884-7950
email ray@andraka.com
http://www.andraka.com

 "They that give up essential liberty to obtain a little
  temporary safety deserve neither liberty nor safety."
                                          -Benjamin Franklin, 1759



Article: 43588
Subject: IO simulations
From: rickman <spamgoeshere4@yahoo.com>
Date: Fri, 24 May 2002 18:09:08 -0400
Links: << >>  << T >>  << A >>
Checking the Xilinx web site it looks like you can download IBIS models
for various devices including the Spartan IIe devices.  However, these
files do you no good if you don't have IBIS simulator which seem to cost
even more than a logic or HDL simulator.  

I also found that there are spice models available for the VirtexII
parts.  But not for any other devices.  Since Spice is much more
available, any idea why the Spice models are not available for parts
other than the VII?  


-- 

Rick "rickman" 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      URL http://www.arius.com
4 King Ave                               301-682-7772 Voice
Frederick, MD 21701-3110                 301-682-7666 FAX

Article: 43589
Subject: Re: Building a relaxation oscillator with a Xilinx 9536XL
From: Frank Wirtz <frankw@xilinx.com>
Date: Fri, 24 May 2002 16:11:43 -0600
Links: << >>  << T >>  << A >>
What sort of frequency are you after????

Frank.

Len Chisholm wrote:

> Hi all,
>
> One of the Xilinx FPGA demo boards has a relaxation oscillator circuit
> attached to a 3000 series FPGA, where there are external RC networks
> attached to a pair of cross-coupled I/O blocks.
>
> Is there any reason why this would be a bad idea for a 9536XL ? The
> datasheet says that there is 50mV hysteresis on the inputs in the XL
> family but not the standard 9500s. I've cobbled a circuit together
> which seems to run OK, but I'm looking for any gotchas which I've
> overlooked.
>
> I need a not-particularly-accurate clock signal to run a state
> machine, and if I can get an oscillator for 'free' then that's a good
> thing.
>
> Thanks,
>
> Len Chisholm.


Article: 43590
Subject: Re: Frequency synthesiser
From: mohaase@nilenet.com (Marty)
Date: 24 May 2002 22:28:50 -0700
Links: << >>  << T >>  << A >>
My coment is based on info from the AD 9854 data sheet.  There is a
short discussion of jitter from DDSs and PLLs.  The PLL contributes
some jitter to the final output, but is less than the jitter of a PLL
by itself.

I seem to recall an article from about 10 or more years ago that
stated that no output from the phase accumulator would provide a good
square wave. That is probably why AD connects the comparator to the
filtered sinewave output of the DDS.

Marty

Ray Andraka <ray@andraka.com> wrote in message news:<3CEE572B.D07F18FC@andraka.com>...
> Depends on the number of bits you use from the DDS.  If you use the DDS to create
> a sinewave, convert it to analog, filter it then use a comparator it can be better
> than a PLL.  If you just take the MSB to create a square wave, then your jitter is
> up to a cycle of your master clock.
> 
> Marty wrote:
> 
> > Jitter from a DDS is lower than a PLL!
> >
> > Marty
> >
> 
> --Ray Andraka, P.E.
> President, the Andraka Consulting Group, Inc.
> 401/884-7930     Fax 401/884-7950
> email ray@andraka.com
> http://www.andraka.com
> 
>  "They that give up essential liberty to obtain a little
>   temporary safety deserve neither liberty nor safety."
>                                           -Benjamin Franklin, 1759

Article: 43591
Subject: evaluation boards for virtex
From: "Geiger Hope" <isuzu@softhome.net>
Date: Sat, 25 May 2002 08:05:32 +0000 (UTC)
Links: << >>  << T >>  << A >>
I am working with Altera products since two years. 
I have my first project with Xilinx, now.
We are going to use Virtex(-E) device. In order to get 
start to Xilinx, I need help with evaluation boards 
with Virtex-e.

Actually, I am looking for the Xilinx counterparts of the
following Altera boards,

1-) APEX PCI Development Kit
(http://www.altera.com/products/devkits/altera/kit-apex_dev_kit.html)

2-) Altera SOPC board. 

Thanks in advance.


-- 
Posted via Mailgate.ORG Server - http://www.Mailgate.ORG

Article: 43592
Subject: Re: Xilinx chip scope: Comments
From: "H.L" <alphaboran@yahoo.no-spam.com>
Date: Sat, 25 May 2002 11:32:03 +0300
Links: << >>  << T >>  << A >>
Hi Rick,

your comments on ChipScope helped me a lot. I will go for it!

Thanks for your help!

Best Regards,
Harris

"Rick Filipkiewicz" <rick@algor.co.uk> wrote in message
news:3CED84C8.E94491D3@algor.co.uk...
>
>
> "H.L" wrote:
>
> > Hello all,
> > I have read few weeks ago in xilinx site about the Chip Scope, seems
like a
> > good solution for testing a FPGA. I am thinking to purchase it but first
I
> > would like some comments about this tool from someone who have used it.
Can
> > someone help me on this?
> >
> > Greetings,
> > Harris
>
> Very good tool and a steal @ $450 (well it saved my bacon a couple of
months
> ago so I may be biased).
>
> Some restrictions and caveats:
>
> o The GUI is pretty clanky but just about useable. For example there's no
way
> to reset the triggers(s) to all `X' after you've been using one of the
trigger
> types that doesn't support `X' values.
>
> o The s/w resident on the PC sometimes fails to connect to the ILA core
through
> the JTAG if there's some heavy compute process running in the background.
The
> symptom is the s/w complaining of an illegal version - I've seen v0.0 and
> v15.0.
>
> o I've heard that MultiLinx USB is very flakey/fragile, use the
Parallel-III/IV
> JTAG cable instead. In fact knowing this is what held me back from using
> Chipsope a while ago when V1.x didn't support Paralle-III.
>
> Things missing that you'd find in a real LA:
>
> o It requires a continuous sample clock of some description. i.e. you
cannot
> async sample some number of events, hit stop before the buffer's full, and
> expect to see at least the events you captured. I got around this by
creating a
> sort of background sample clock that ticked if no real one had arrived for
a
> while.
>
> o There's no filter on pre-trigger events. In other words you can't say
``trig
> on event E while storing <filtered subset of states>''. Again its possible
to
> get around this by designing some extra h/e to wrap around the ILA core
but
> changing the pre-filter condition => rebuilding the FPGA.
>
> Other than that its an verygood-to-excellent tool and works as advertised.
If
> Xilinx would fix up some of the points above it might even head into magic
> class.
>



Article: 43593
Subject: Re: Small FIFOs in Spartan
From: "Ulf Samuelsson" <ulf@atmel.REMOVE.com>
Date: Sat, 25 May 2002 10:43:32 +0200
Links: << >>  << T >>  << A >>
"Børge Strand" <borge.strand.remove.if.not.spamming@sintef.no> skrev i
meddelandet news:1022230585.37631@halvan.trd.sintef.no...
> I will need some really small FIFOs for my next project. The word length
> will be 24 bits, but the depth of the FIFO is not requirred to be more
than
> two to sixteen.
>
> Could dedicated parts of the Spartan be used for such small FIFOs? I plan
to
> program the thing in VHDL.
>
> Regards,
> Børge
>
>

This is an excellent application for the Atmel AT40K.
For each 16 macroblocks, you have a small Dual Port RAM 32 x 4.
The H/W Macrogenerator can be used to create a 32 x 24 FIFO.
You will have 16 of the 32x4 DP SRAMs in the smallest 5 k gate FPGA,
(Only 12 can be used as DPRAM though)
If you need more than two FIFOs then there are larger FPGAs available.

Maybe you could also use the FPSLIC which has an internal AVR processor.

--
Best Regards
Ulf at atmel dot com
These comments are intended to be my own opinion and they
may, or may not be shared by my employer, Atmel Sweden.




Article: 43594
Subject: Re: Time for a new computer. Suggestions?
From: "Ulf Samuelsson" <ulf@atmel.REMOVE.com>
Date: Sat, 25 May 2002 10:55:48 +0200
Links: << >>  << T >>  << A >>
"Ray Andraka" <ray@andraka.com> skrev i meddelandet
news:3CED5561.82DA8B37@andraka.com...
> I agree, but then there is a limit to the amount of RAM that can go in a
system,
> and that is often greater than what the OS can use.  I've got 2GB in my
current
> system: that frequently gets exceeded during  synplicity's compile phase
and Xilinx
> PAR.  As soon as paging starts, it slows to a crawl.
>
> hamish@cloud.net.au wrote:
>
> >

Tyan S2468 will allow for 4 GB of SDRAM or wait for Hammer's w 64 bit
addressing.

--
Best Regards
Ulf at atmel dot com
These comments are intended to be my own opinion and they
may, or may not be shared by my employer, Atmel Sweden.




Article: 43595
Subject: Re: Synchronous Single Clock Designs
From: 9+3@supereva.it (Fabio G.)
Date: Sat, 25 May 2002 12:57:53 GMT
Links: << >>  << T >>  << A >>
Rick Filipkiewicz <rick@algor.co.uk> ha scritto:

>that makes ASIC vendors dislike tristates

So in ASIC data bus can not be realized?

--
Per rispondermi via email sostituisci il risultato
dell'operazione (in lettere) dall'indirizzo.

Article: 43596
Subject: Re: SDRAM controler for Virtex-II
From: spam_hater_7@email.com (Spam Hater)
Date: Sat, 25 May 2002 14:31:04 GMT
Links: << >>  << T >>  << A >>

FWIW, both of those links are 'protected', and require
an NDA with Lattice...



On Fri, 24 May 2002 17:37:15 +0200, Laurent Gauch
<laurent.gauch@amontec.com> wrote:

>Sorry, my last link was for verilog sources
>
>try for vhdl one:
>http://www.latticesemi.com/account/_download.cfm?AMID=2886
>
>Or use the search on http://www.latticesemi.com/
>
>Laurent Gauch
>Amontec
>
>Laurent Gauch wrote:
>
>> Philippe,
>> 
>> try this free HDL source of sdram controller:
>> http://www.latticesemi.com/account/_download.cfm?AMID=3866
>> 
>> Since I want to includ it on a spartanII in next months, could you give 
>> me a feedback, if you use it.
>> 
>> Laurent Gauch
>> 
>> Philippe Robert wrote:
>> 
>>> Hi,
>>>
>>> I am after an SDRAM controller in VHDL for Virtex-II.
>>> Does anyone know where I could get one from ?
>>>
>>> Thanks.
>>> Philippe.
>>>
>>>
>>>
>> 
>


Article: 43597
Subject: Re: SDRAM controler for Virtex-II
From: "Antonio Pasini" <pasini.a@libero.it>
Date: Sat, 25 May 2002 17:12:08 GMT
Links: << >>  << T >>  << A >>
The Xilinx app. note for sdr sdram controller on Virtex could be a good
start; I also have to design a simple controller for Spartan II.

http://www.xilinx.com/apps/xappsumm.htm#xapp134

"Philippe Robert" <PhilippeR@sundance.com> ha scritto nel messaggio
news:3cee31f1$1@shiva.ukisp.net...
> Hi,
>
> I am after an SDRAM controller in VHDL for Virtex-II.
> Does anyone know where I could get one from ?
>
> Thanks.
> Philippe.
>
>



Article: 43598
Subject: Documentation on XDL reports?
From: Neil Steiner <neil.steiner@vt.edu>
Date: Sat, 25 May 2002 19:13:07 -0400
Links: << >>  << T >>  << A >>

From archived postings to comp.arch.fpga I've figured out that some
documentation is available at $XILINX/help/data/xdl/xdl.html. 
Unfortunately the BNF included there only covers converted designs and
not XDL reports (c.f. "xdl -report virtex2 xc2v40").  Can anybody
suggest any pointers?

Article: 43599
Subject: Re: Time for a new computer. Suggestions?
From: Duane Clark <junkmail@junkmail.com>
Date: Sat, 25 May 2002 17:14:57 -0700
Links: << >>  << T >>  << A >>
Jeff Mock wrote:
> 
> For me the linux environment is great.  I don't use any of
> the Xilinx GUI tools, just the command line tools.

By the way, in case you (or someone) are interested in running 
fpga_editor under wine, here is a patch that cleans up most of the 
remaining wine bugs that affect it. It is against the recent CVS version 
of wine, and I will attempt to keep it up to date.

http://www.leewardfpga.com/fpga.diff

Stick that in the wine top level directory, and execute:
patch -p0 < fpga.diff

I run quite a few programs under wine, and it does not appear to harm 
any of them (some run better). But as always, my warranty only extends 
to the price you paid for the patch :-)

-- 
My real email is akamail.com@dclark (or something like that).




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