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 134250

Article: 134250
Subject: Re: Xilinx FFT core's IFFT function not working? Dun Xilinx TEST
From: Tom <tom.derham@gmail.com>
Date: Fri, 1 Aug 2008 07:50:30 -0700 (PDT)
Links: << >>  << T >>  << A >>
Hi

> =A0 =A0I have checked. the first input sample is fed in when index for xn
> is 3.

No it isn't... the first input sample should be time aligned to when
xn_index is 0.
This occurs a few cycles (3 or so? i forget) after you assert Start
high.

>  Did you do this the last time?

No I haven't done what you said because I implemented in actual
hardware, with real ADC (although double-checked against a digitially
synthesized waveform), real data buffers on input and output, feeding
output of the FFT out of the FPGA into a PC where it was post-
processed in Matlab. I found no such problems that you report. I used
the streaming core with various values of N (points), 24-bit for
input, output and twiddles.

I think the problems you seem to have found are unlikely to be related
to quantization error... more likely a timing/simulation problem. It
is extremely unlikely this bug exists in the hardware core itself,
since it is used very widely in many applications. It is just about
possible that it is a bug existing in the simulation files used by
Modelsim. However, to be honest, it is much more likely it is due to
something in your own implementation of the simulation testbench, or
misunderstanding of the timing requirements for the core.

Not sure I can assist you much more, if you are really stuck then open
a Webcase with Xilinx and send them the simulation files you are
using. One other thing - do check whether your scaling schedule is
appropriate, and also whether you are expecting output in natural or
bit/byte reversed order. Finally check you have not accidently enabled
"cyclic prefix" (intended for use with OFDM transmitters), if you are
using v5 of the core.

-T

Article: 134251
Subject: Re: Fixed point number hardware implementation
From: "MikeWhy" <boat042-nospam@yahoo.com>
Date: Fri, 1 Aug 2008 10:07:09 -0500
Links: << >>  << T >>  << A >>
"kami" <kamran.wadood@yahoo.co.uk> wrote in message 
news:wJOdnV4QFvPZ7w_V4p2dnAA@giganews.com...
> >"kami" <kamran.wadood@yahoo.co.uk> wrote in message
>>news:JJudnS58eI-AdwzV4p2dnAA@giganews.com...
>>> Now, I was wondering what will be the best option if I wanted to
>>> demonstrate the filter operation more clearly. I mean these are just
>>> numbers that I am showing on the 7-seg display. But if I want to give
> a
>>> sinusoidal input and get a filtered waveform at the output.
>>
>>It wouldn't take much to drive a VGA display.
>>
> Do you mean using a logic analyzer? But my friend is struggling to develop
> a logic analyzer in FPGA. So, how do you so it won't take much for VGA
> display?
> or do you mean using a cable (i've heard of a cable that can be connected
> to FPGA and then the other end to the PC) is that what you meant?

A VGA display needs only a few signal lines to drive it, fairly easy for the 
FPGA. I was maybe wrong to suggest it. Setting up a memory controller for 
the frame buffer and writing the line draw routines is too much of a project 
and distraction early on. Forget I mentioned it.

Directly displaying the filtered waveform won't show the filtering as well a 
spectrogram. Your ears are also pretty sensitive to spectral changes. Audio 
will be the simplest demonstration. A DAC can drive earphones directly.



Article: 134252
Subject: Re: question about fifo
From: John_H <newsgroup@johnhandwork.com>
Date: Fri, 1 Aug 2008 08:36:17 -0700 (PDT)
Links: << >>  << T >>  << A >>
fmostafa wrote:
> hi all;
>
> I am using Edk  to create  IP  peripheral which contains RD and WR
> Fifos. My question is,  if the required data to be written to the read
> fifo is undefined  it will not be stored in the fifo or what.
>
> fatma

Please define undefined data, electrically.

Article: 134253
Subject: Re: ISE new file wizard
From: langwadt@fonz.dk
Date: Fri, 1 Aug 2008 09:13:29 -0700 (PDT)
Links: << >>  << T >>  << A >>
On 31 Jul., 15:26, nei...@pipstechnology.co.uk wrote:
> On 31 Jul, 11:13, Brian Drummond <brian_drumm...@btconnect.com> wrote:
>
> > On Wed, 30 Jul 2008 06:41:15 -0700 (PDT), Dave <dhsch...@gmail.com>
> > wrote:
>
> > >Does anyone know of a way to modify which VHDL libraries ISE
> > >automatically uses in VHDL files created by the new file wizard? It
> > >seems to always use std_logic_arith and std_logic_unsigned, and I
> > >can't find a setting anywhere to modify this. I remember from my days
> > >of using Quartus that there was a text are in the settings menus where
> > >this could be changed. There's got to be something similar in ISE,
> > >right?
>
> > If enouch people raise WebCases on this, Xilinx might eventually do
> > something about it.
>
> > - Brian
>
> I raised a webcase on this back in 2006, and was told a change request
> (CR # 425340) has already been filed and is scheduled to be fixed in a
> later software release.  I was running 8.2 at the time, so we've
> already gone through a couple of software releases and still no fix.
>
> Maybe one of the Xilinx people here could chase up the CR and give us
> an idea of when it will be fixed.
>
> Neill.

oddly enough if you poke around in the xilinx directory you will find
templates for some things, see for example \xilinx\data\testfix.tft


-Lasse

Article: 134254
Subject: Re: ISE new file wizard
From: Kevin Neilson <kevin_neilson@removethiscomcast.net>
Date: Fri, 01 Aug 2008 12:42:18 -0600
Links: << >>  << T >>  << A >>
langwadt@fonz.dk wrote:
> On 31 Jul., 15:26, nei...@pipstechnology.co.uk wrote:
>> On 31 Jul, 11:13, Brian Drummond <brian_drumm...@btconnect.com> wrote:
>>
>>> On Wed, 30 Jul 2008 06:41:15 -0700 (PDT), Dave <dhsch...@gmail.com>
>>> wrote:
>>>> Does anyone know of a way to modify which VHDL libraries ISE
>>>> automatically uses in VHDL files created by the new file wizard? It
>>>> seems to always use std_logic_arith and std_logic_unsigned, and I
>>>> can't find a setting anywhere to modify this. I remember from my days
>>>> of using Quartus that there was a text are in the settings menus where
>>>> this could be changed. There's got to be something similar in ISE,
>>>> right?
>>> If enouch people raise WebCases on this, Xilinx might eventually do
>>> something about it.
>>> - Brian
>> I raised a webcase on this back in 2006, and was told a change request
>> (CR # 425340) has already been filed and is scheduled to be fixed in a
>> later software release.  I was running 8.2 at the time, so we've
>> already gone through a couple of software releases and still no fix.
>>
>> Maybe one of the Xilinx people here could chase up the CR and give us
>> an idea of when it will be fixed.
>>
>> Neill.
> 
> oddly enough if you poke around in the xilinx directory you will find
> templates for some things, see for example \xilinx\data\testfix.tft
> 
> 
> -Lasse
The CR is still open.  I'm not sure how this feature is any better than 
creating your own source file and adding it to the project.
-Kevin

Article: 134255
Subject: Re: ISE new file wizard
From: Dave <dhschetz@gmail.com>
Date: Fri, 1 Aug 2008 12:07:14 -0700 (PDT)
Links: << >>  << T >>  << A >>
On Aug 1, 2:42 pm, Kevin Neilson <kevin_neil...@removethiscomcast.net>
wrote:
> langw...@fonz.dk wrote:
> > On 31 Jul., 15:26, nei...@pipstechnology.co.uk wrote:
> >> On 31 Jul, 11:13, Brian Drummond <brian_drumm...@btconnect.com> wrote:
>
> >>> On Wed, 30 Jul 2008 06:41:15 -0700 (PDT), Dave <dhsch...@gmail.com>
> >>> wrote:
> >>>> Does anyone know of a way to modify which VHDL libraries ISE
> >>>> automatically uses in VHDL files created by the new file wizard? It
> >>>> seems to always use std_logic_arith and std_logic_unsigned, and I
> >>>> can't find a setting anywhere to modify this. I remember from my days
> >>>> of using Quartus that there was a text are in the settings menus where
> >>>> this could be changed. There's got to be something similar in ISE,
> >>>> right?
> >>> If enouch people raise WebCases on this, Xilinx might eventually do
> >>> something about it.
> >>> - Brian
> >> I raised a webcase on this back in 2006, and was told a change request
> >> (CR # 425340) has already been filed and is scheduled to be fixed in a
> >> later software release.  I was running 8.2 at the time, so we've
> >> already gone through a couple of software releases and still no fix.
>
> >> Maybe one of the Xilinx people here could chase up the CR and give us
> >> an idea of when it will be fixed.
>
> >> Neill.
>
> > oddly enough if you poke around in the xilinx directory you will find
> > templates for some things, see for example \xilinx\data\testfix.tft
>
> > -Lasse
>
> The CR is still open.  I'm not sure how this feature is any better than
> creating your own source file and adding it to the project.
> -Kevin

It seems like there's a few fairly simple changes like this that
Xilinx has been "meaning to do" for quite a while, but never
implementing. With FPGAs growing insanely large and complex, I wonder
if silicon features will be less of a determining factor for market
share than tool quality. I'm not at all sure X is beating A in that
category...

Article: 134256
Subject: What's the deal with PSoC programmers?
From: andersod2 <thechrisanderson@gmail.com>
Date: Fri, 1 Aug 2008 13:22:37 -0700 (PDT)
Links: << >>  << T >>  << A >>

Hi, just wondering about these...what are they, what hardware is
involved, what are the advantages etc...

thanks

Article: 134257
Subject: Re: ISE new file wizard
From: Andy Peters <google@latke.net>
Date: Fri, 1 Aug 2008 17:55:27 -0700 (PDT)
Links: << >>  << T >>  << A >>
On Jul 31, 6:26=A0am, nei...@pipstechnology.co.uk wrote:
> On 31 Jul, 11:13, Brian Drummond <brian_drumm...@btconnect.com> wrote:
>
> > On Wed, 30 Jul 2008 06:41:15 -0700 (PDT), Dave <dhsch...@gmail.com>
> > wrote:
>
> > >Does anyone know of a way to modify which VHDL libraries ISE
> > >automatically uses in VHDL files created by the new file wizard? It
> > >seems to always use std_logic_arith and std_logic_unsigned, and I
> > >can't find a setting anywhere to modify this. I remember from my days
> > >of using Quartus that there was a text are in the settings menus where
> > >this could be changed. There's got to be something similar in ISE,
> > >right?
>
> > If enouch people raise WebCases on this, Xilinx might eventually do
> > something about it.
>
> > - Brian
>
> I raised a webcase on this back in 2006, and was told a change request
> (CR # 425340) has already been filed and is scheduled to be fixed in a
> later software release. =A0I was running 8.2 at the time, so we've
> already gone through a couple of software releases and still no fix.
>
> Maybe one of the Xilinx people here could chase up the CR and give us
> an idea of when it will be fixed.

FWIW, I opened up a WebCase and was told that Change Request CR
#474759 was assigned to this same problem.

Of course there is no way to determine what happens to a change
request after it has been filed. I expect it was routed to /dev/null.

-a

Article: 134258
Subject: Re: question about fifo
From: Peter Alfke <peter@xilinx.com>
Date: Fri, 1 Aug 2008 18:02:36 -0700 (PDT)
Links: << >>  << T >>  << A >>
On Aug 1, 7:05=A0am, fmostafa <fatma.abouele...@ugent.be> wrote:
> hi all;
>
> I am using Edk =A0to create =A0IP =A0peripheral which contains RD and WR
> Fifos. My question is, =A0if the required data to be written to the read
> fifo is undefined =A0it will not be stored in the fifo or what.
>
> fatma

Fatma, a FIFO is (conceptually) very simple: You store your data word
by enabling the write clock, and you pull the data word out in the
same sequence that you wrote it, by enabling the read clock. Use free-
running clocks, and manipulate the Clock Enable. Remember: Data is
never undefined, it is what it is, 0 or 1.
Peter Alfke

Article: 134259
Subject: Re: Is there a totally command-line driven way to use Xilinx Webpack?
From: andersod2 <thechrisanderson@gmail.com>
Date: Fri, 1 Aug 2008 18:34:53 -0700 (PDT)
Links: << >>  << T >>  << A >>

It seems odd that the oscillator for that board is only 16Mhz...isn't
the internal oscillator for the 3A even faster than that?

Article: 134260
Subject: Re: Is there a totally command-line driven way to use Xilinx Webpack?
From: "beky4kr@gmail.com" <beky4kr@gmail.com>
Date: Fri, 1 Aug 2008 22:26:47 -0700 (PDT)
Links: << >>  << T >>  << A >>
On 31 =D7=99=D7=95=D7=9C=D7=99, 06:31, andersod2 <thechrisander...@gmail.co=
m> wrote:
> Am using the free ISE Webpack, and wanted to know if it's possible
> just to use the command line and code/config files...that counts for
> things like package assignments and simulator stuff...can it be done
> at all, or easily?
>
> Thanks for your insights...

http://bknpk.no-ip.biz/leon_i2c_syn/leon_i2c_syn.html

Article: 134261
Subject: cpu,fpga, clock, dac, initialize sequence
From: nubont@gmail.com
Date: Sat, 2 Aug 2008 07:34:09 -0700 (PDT)
Links: << >>  << T >>  << A >>
i'm dealing a board which has cpu,fpga, clock, adc etc. onboard.
the clock is for board timing for both the fpga and adc.

the init sequence was to load the fpga logic first, then the clock,
then the adc.
but fpga guys say it's wrong. should i init clock and adc first, then
the fpga?
what's the point here?

Article: 134262
Subject: Re: cpu,fpga, clock, dac, initialize sequence
From: Gabor <gabor@alacron.com>
Date: Sat, 2 Aug 2008 09:39:28 -0700 (PDT)
Links: << >>  << T >>  << A >>
On Aug 2, 10:34 am, nub...@gmail.com wrote:
> i'm dealing a board which has cpu,fpga, clock, adc etc. onboard.
> the clock is for board timing for both the fpga and adc.
>
> the init sequence was to load the fpga logic first, then the clock,
> then the adc.
> but fpga guys say it's wrong. should i init clock and adc first, then
> the fpga?
> what's the point here?

It's hard to see what you you should do in which order without
more information on the system.  What "clock", what "adc"?
Is the FPGA in any way involved in the initialization of these
circuits?  If not, what is?  A microcomputer?  What sort of
fpga?

I'm guessing the "fpga guy" knows more about how the system
should come up...

Regards,
Gabor

Article: 134263
Subject: Re: What's the deal with PSoC programmers?
From: emeb <ebrombaugh@gmail.com>
Date: Sat, 2 Aug 2008 15:37:24 -0700 (PDT)
Links: << >>  << T >>  << A >>
On Aug 1, 1:22 pm, andersod2 <thechrisander...@gmail.com> wrote:
> Hi, just wondering about these...what are they, what hardware is
> involved, what are the advantages etc...
>
> thanks

PSoC is neither fish nor fowl. It's a strange combination of an 8-bit
MCU (think 6502 levels of performance) with partially reconfigurable
analog and digital arrays. It's great for low-cost applications
requiring processing of slowly varying analog signals.

The reconfigurable arrays are loaded at reset time with configuration
info by the MCU from tables in flash memory that are defined with
Cypress's design tools. Much of the run-time management of the various
functions implemented in the arrays is dependent on interrupt service
routines that are (mostly) hidden from the designer by the tools. The
designer then ties these blocks together with hand-written assembly,
or high-level point-n-click design flows.

Analog functions include gain blocks, filters, ADCs and DACs. All
functions are built using op-amps and switched capacitor blocks.

Digital functions include counters, random number generators, UARTs,
I2C slaves, etc.

Neat parts if you've got a fairly low-end application. Not really
FPGAs as such.

Eric

Article: 134264
Subject: Re: Is there a totally command-line driven way to use Xilinx
From: phil hays <philhays@dont.spam>
Date: Sat, 02 Aug 2008 23:01:57 GMT
Links: << >>  << T >>  << A >>
andersod2 wrote:

> Am using the free ISE Webpack, and wanted to know if it's possible just
> to use the command line and code/config files...that counts for things
> like package assignments and simulator stuff...can it be done at all, or
> easily?

Yes.

The simplest is to take the .cmd_log file and use it as a script file. 
Under windows, rename this file to bat, and type that name at a command 
prompt to build the design. This is simple to set up, but fairly hard to 
modify. There are several other files that have been set up by ISE to 
control synthesis. These other files have the XST settings (not simple to 
modify) and the list of files used (the order the files are listed might 
be critical). The script might not work on Linux if produced on windows, 
and vice versa.

You could also use a makefile. This is more complex to setup, but has the 
advantage of running only what is needed, and can be made OS independent. 
Starting with the script file and XST files generated above, convert the 
script to a makefile. Install the Cygwin package including gnu make. Then 
to run your make file at a command prompt type "make".

Or you could use Tcl. Tcl is a built in scripting language. To get 
started, ISE under the project menu has "generate Tcl script". Edit the 
script as needed, or call it from another Tcl script. Tcl is a language 
used for scripting by both ISE many other CAD tools, such as Synplicity, 
ModelSim and PlanAhead. I've talked about this before:


http://groups.google.com/group/comp.arch.fpga/msg/a6b1e400dd24d509?hl=en


-- 
Phil Hays

Article: 134265
Subject: Re: question about fifo
From: "jtw" <wrightjt @hotmail.invalid>
Date: Sat, 2 Aug 2008 19:53:22 -0700
Links: << >>  << T >>  << A >>


"John_H" <newsgroup@johnhandwork.com> wrote in message 
news:88fb24da-8f2b-4f3f-833a-059b79fc0eae@k30g2000hse.googlegroups.com...
> fmostafa wrote:
>> hi all;
>>
>> I am using Edk  to create  IP  peripheral which contains RD and WR
>> Fifos. My question is,  if the required data to be written to the read
>> fifo is undefined  it will not be stored in the fifo or what.
>>
>> fatma
>
> Please define undefined data, electrically.

??

Undefined data is not defined, by definition.

It is a simulation state, not an electrical state.

JTW 



Article: 134266
Subject: Re: Is there a totally command-line driven way to use Xilinx Webpack?
From: andersod2 <thechrisanderson@gmail.com>
Date: Sat, 2 Aug 2008 20:44:21 -0700 (PDT)
Links: << >>  << T >>  << A >>

Thanks for your very informative answer Phil.  I was wondering about
using cygwin...good to know.  Also, I just found out that many of the
command line tools have pdf's that explain their operation...I forget
the directory, but just search for "ngdbuild" in the xilinx directory
and you'll see a bunch of pdfs.


Article: 134267
Subject: Re: What's the deal with PSoC programmers?
From: rickman <gnuarm@gmail.com>
Date: Sun, 3 Aug 2008 08:16:21 -0700 (PDT)
Links: << >>  << T >>  << A >>
On Aug 2, 6:37 pm, emeb <ebromba...@gmail.com> wrote:
> On Aug 1, 1:22 pm, andersod2 <thechrisander...@gmail.com> wrote:
>
> > Hi, just wondering about these...what are they, what hardware is
> > involved, what are the advantages etc...
>
> > thanks
>
> PSoC is neither fish nor fowl. It's a strange combination of an 8-bit
> MCU (think 6502 levels of performance) with partially reconfigurable
> analog and digital arrays. It's great for low-cost applications
> requiring processing of slowly varying analog signals.

Speaking of PSOC (will saying the name three times bring the CEO of
Cypress forth?) what ever happened to their NEW, IMPROVED PSOC based
on the ARM (CM3, IIRC)?  My understanding is that the CEO said, thou
shalt make a new PSOC with "NO EXCUSES" in the analog!  I think it was
originally supposed to be out at end of year 2006 or 2007 (can't
remember which) and then put off for a bit.  This is eight months
later and I still haven't heard a peep about it.  I expect it has gone
the way of the T9000 transputer?  I guess that is what happens when
you are told to make it happen and "NO EXCUSES"!

Rick

Article: 134268
Subject: Re: What's the deal with PSoC programmers?
From: emeb <ebrombaugh@gmail.com>
Date: Sun, 3 Aug 2008 09:31:57 -0700 (PDT)
Links: << >>  << T >>  << A >>
On Aug 3, 8:16 am, rickman <gnu...@gmail.com> wrote:

> Speaking of PSOC (will saying the name three times bring the CEO of
> Cypress forth?) what ever happened to their NEW, IMPROVED PSOC based
> on the ARM (CM3, IIRC)?  My understanding is that the CEO said, thou
> shalt make a new PSOC with "NO EXCUSES" in the analog!  I think it was
> originally supposed to be out at end of year 2006 or 2007 (can't
> remember which) and then put off for a bit.  This is eight months
> later and I still haven't heard a peep about it.  I expect it has gone
> the way of the T9000 transputer?  I guess that is what happens when
> you are told to make it happen and "NO EXCUSES"!

(TJ Rogers appears in a puff of smoke)

I hadn't heard of that. It's an interesting exercise to consider what
an 'ideal' PSoC would consist of. An ARM-based MCU would be nice, as
well as more flexible digital and analog arrays. There certainly are
some attractive new ARM architectures out there (Cortex for instance)
which are apparently more easily integrated. Cypress' recent additions
to the PSoC line have tended to focus more on their capacitive touch
senosors though - apparently they made a lot of money on those back at
the start of the iPod days. Until they see a big pot-o-cash at the end
of the design process though, I'd count an advanced PSoC as a low
probability.

Eric

Article: 134269
Subject: Re: What's the deal with PSoC programmers?
From: rickman <gnuarm@gmail.com>
Date: Sun, 3 Aug 2008 10:21:03 -0700 (PDT)
Links: << >>  << T >>  << A >>
On Aug 3, 12:31 pm, emeb <ebromba...@gmail.com> wrote:
> On Aug 3, 8:16 am, rickman <gnu...@gmail.com> wrote:
>
> > Speaking of PSOC (will saying the name three times bring the CEO of
> > Cypress forth?) what ever happened to their NEW, IMPROVED PSOC based
> > on the ARM (CM3, IIRC)?  My understanding is that the CEO said, thou
> > shalt make a new PSOC with "NO EXCUSES" in the analog!  I think it was
> > originally supposed to be out at end of year 2006 or 2007 (can't
> > remember which) and then put off for a bit.  This is eight months
> > later and I still haven't heard a peep about it.  I expect it has gone
> > the way of the T9000 transputer?  I guess that is what happens when
> > you are told to make it happen and "NO EXCUSES"!
>
> (TJ Rogers appears in a puff of smoke)
>
> I hadn't heard of that. It's an interesting exercise to consider what
> an 'ideal' PSoC would consist of. An ARM-based MCU would be nice, as
> well as more flexible digital and analog arrays. There certainly are
> some attractive new ARM architectures out there (Cortex for instance)
> which are apparently more easily integrated. Cypress' recent additions
> to the PSoC line have tended to focus more on their capacitive touch
> senosors though - apparently they made a lot of money on those back at
> the start of the iPod days. Until they see a big pot-o-cash at the end
> of the design process though, I'd count an advanced PSoC as a low
> probability.


I have to say I am getting tired of awaiting this "new" product.  I
think it is getting "old" in the planning.

I did find this one public mention in a stock analyst's report.

http://www.thestreet.com/files/k/tc/pdf/200807231712.pdf

He gives some generally bullish info on Cypress with the PSOC part
holding their own and the PSOC3 anticipated.

Time will tell... if you can hold on long enough!

Rick

Article: 134270
Subject: Re: Fixed point number hardware implementation
From: "kami" <kamran.wadood@yahoo.co.uk>
Date: Sun, 03 Aug 2008 15:14:53 -0500
Links: << >>  << T >>  << A >>
>"kami" <kamran.wadood@yahoo.co.uk> wrote in message 
>news:wJOdnV4QFvPZ7w_V4p2dnAA@giganews.com...
>> >"kami" <kamran.wadood@yahoo.co.uk> wrote in message
>>>news:JJudnS58eI-AdwzV4p2dnAA@giganews.com...
>>>> Now, I was wondering what will be the best option if I wanted to
>>>> demonstrate the filter operation more clearly. I mean these are just
>>>> numbers that I am showing on the 7-seg display. But if I want to
give
>> a
>>>> sinusoidal input and get a filtered waveform at the output.
>>>
>>>It wouldn't take much to drive a VGA display.
>>>
>> Do you mean using a logic analyzer? But my friend is struggling to
develop
>> a logic analyzer in FPGA. So, how do you so it won't take much for VGA
>> display?
>> or do you mean using a cable (i've heard of a cable that can be
connected
>> to FPGA and then the other end to the PC) is that what you meant?
>
>A VGA display needs only a few signal lines to drive it, fairly easy for
the 
>FPGA. I was maybe wrong to suggest it. Setting up a memory controller for

>the frame buffer and writing the line draw routines is too much of a
project 
>and distraction early on. Forget I mentioned it.
>
>Directly displaying the filtered waveform won't show the filtering as
well a 
>spectrogram. Your ears are also pretty sensitive to spectral changes.
Audio 
>will be the simplest demonstration. A DAC can drive earphones directly.
>
>
Can you explain a little bit more about this application? where and how am
I going to give input and how do I get the output? Is it just another (one
or more) VHDL file I need to write?
Thanks


Article: 134271
Subject: Why PCI9054 fails to assert pci interrupt when local interrupt input is pulled down? Any advice? Thanks
From: "Leon" <dechenxu@yahoo.com.cn>
Date: Mon, 04 Aug 2008 02:38:41 -0500
Links: << >>  << T >>  << A >>
Hi all,

I am doing a project involving PCI9054. 
 
I meet a problem of interrupt. I wanna  assert a PCI interrupt using the
LINT#  pin which is set as input with the INTCSR register being setting
as:
0X0f000900. But when I pull down the level of LINT# that is drived by
FPGA, the INTCSR becomes 0X0f000100, that is, INTCSR[11](local interrupt
input enable) is cleared and INTCSR[15](local input interrupt active) fails
to be set 1. And also the interrupt handler is not called. I have no idea
what the reason. 
 
However, when I set INTCSR as 0X0f000800 (PCI interrupt is not enabled)
and then pull down LINT#, the INTCSR can come to the ideal value:
0X0f008800. I dont know its the reason of WinDriver diagnostic program or
the PCI9054 itself. 
 
Any advice will be helpful, Thank u very much!
 
Leon,

Article: 134272
Subject: Cordic Core for Virtex 5
From: Venkat <venkat.japan@gmail.com>
Date: Mon, 4 Aug 2008 01:04:38 -0700 (PDT)
Links: << >>  << T >>  << A >>
Hello all,

Is it possible to use Cordic IP Core of Xilinx for Virtex 5 devices?
The only solution I could come across for this problem is from the
following URL:

http://www.xilinx.com/support/answers/29055.htm

Anyone who has tested this way or alternate solution please let me
know.

Thanks in advance,

Venkat.

Article: 134273
Subject: fixed FFT point implementation woes
From: chrisdekoh@gmail.com
Date: Mon, 4 Aug 2008 01:07:27 -0700 (PDT)
Links: << >>  << T >>  << A >>
Hi,
   I am trying to create a mex file that does fixed point FFT (512
point?). I am generally ok with creating a C code in single precision
to compute the the result and have ensured that it works fine as i get
the correct result. (i read it back and plotted it in matlab.)
    the problem comes in when i try to convert it to fixed point. My
input data is 24 bits integer. In addition, I take the sine and cosine
twiddles to also be 24 bits... I do the cos(2*pi*n/N) and sin(2*pi/N)
for n from 0 to 255 the multply the result by ((2^23)-1) and store it
in a look up table.

    What i really dun quite understand is how the scaling factors
work. In FFT books, they generally say that the output from each scale
should maybe be scaled by 1 bit to pre-empt bit growth problems. Also,
for unscaled bit-growth, it generally grows by log2(N) for an N point
FFT. But how does that work?
    What i see here is the following:

1) 24 bit input data. If i dun convert it to anything, it becomes Q.24
format right?
2) 24 bit input data x 24 bit twiddle factor (Q.24 also). this becomes
48 bits. to prevent overflow, i scale by 24 say.

But this is too much and the output frequency spectrum becomes a mess.
I would forsee plenty of precision loss. If i scale by 1 from the
output of each stage, then the FFT gets clipped.

any ideas?what did i go wrong here?  :(

     could someone kindly help me with this? thanks for your help.

Article: 134274
Subject: Schematic Capture tutorials/books?
From: laserbeak43 <laserbeak43@gmail.com>
Date: Mon, 4 Aug 2008 03:31:37 -0700 (PDT)
Links: << >>  << T >>  << A >>
Hi,
I'm new to FPGAs I just bought a Spartan 3E Starter Kit and I'm
looking for documentation on Schematic capture with ISE or whatever
would work for a good tutorial. It seems VERY hard to find ANY
documentation though. I did find a starter video from digilent, but
that was very short and seemed to be more of an intro to dataflow
programming.
Can anyone point me in the right direction?

Thanks



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