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 135300

Article: 135300
Subject: Re: Weird DCM problem with external deskew
From: Brian Drummond <brian_drummond@btconnect.com>
Date: Thu, 25 Sep 2008 00:35:43 +0100
Links: << >>  << T >>  << A >>
On Wed, 24 Sep 2008 08:10:30 -0700 (PDT), Benjamin Couillard
<benjamin.couillard@gmail.com> wrote:

>Hi, I'm currently developping a FPGA application where I implemented a
>SDRAM controller with a DCM to remove external skew.
>
>So basically, I checked the XAPP462 application note, and  I designed
>the deskew circuit with 1X feedback instead of 2X feedback. When I
>program the FPGA using the JTAG cable, the DCM locks and everything
>works fine (i.e. I can read and write the SDRAM correctly). However,
>when I program the FPGA with the onboard MPU, the DCM doesn't lock at
>all. It seems to me that it's probably some kind of timing issue with
>the DCM.

What is the clock source for the deskew DCM?

If it has already been through more than one DCM, the accumulation of
jitter may prevent the last one locking. One DCM ahead of the de-skew
DCM should be OK, provided the original clock is a high quality source.

- Brian


Article: 135301
Subject: Re: Weird DCM problem with external deskew
From: Benjamin Couillard <benjamin.couillard@gmail.com>
Date: Wed, 24 Sep 2008 18:02:42 -0700 (PDT)
Links: << >>  << T >>  << A >>
the clock source is the output of another DCM, so there's one problem,
my reset circuit didn't take that in account. I changed the reset
connection so that the reset input of the deskew DCM (the 2nd one in
the chain) is "FIRST_DCM_LOCKED" not. However, even with that change,
I still get the same behavior. The 2nd DCM locks when I program the
FPGA with the platform cable and it doesn't when I program it with the
MPU.

But I still gotta check some other things like the MPU code.

Thanks


Article: 135302
Subject: Re: OFDM band switch ...
From: "Kappasm" <_NOSPAM_78kappa78_NO_SPAM@virgilio.it>
Date: Thu, 25 Sep 2008 09:06:45 +0200
Links: << >>  << T >>  << A >>
Hi Jerzy Gbur.

Thanks for your replay.

> You can design new OFDM symbol, without part of carriers. I did not 
> understand it anyway as achieve it. you might be more accurate ?

I sincerely not thought I had at this solution.

I have an IFFT of 2048 point. Mapping pilot at input of IFFT for +/-DC 
offset, [0 - 852 Data pilot] [853 - 1194 Null pilot] [1195 - 2047 Data 
pilot]. Process IFFT and i obtain a time domain data, with sample time of 
7/64e6 I obtain 8 MHz band.

Keeping the same IFFT point to 2048 as carriers can remove ?

Regards,

Kappasm.



Article: 135303
Subject: wishbone interface
From: mkr <mahenreddy@gmail.com>
Date: Thu, 25 Sep 2008 05:39:47 -0700 (PDT)
Links: << >>  << T >>  << A >>
I ma trying to make my first design (UART) wishbone compliant and
looked at wishbone specification and several examples with source code
on the net. Though I understand there are some restrictions on
multiplexing of address and data bus, tri-state bus etc.. in many
cases I hardly see any additional logic in the wishhbone compliant
designs other than naming the external signals as per the wishbone
standard and some additional documentation like wishbone datasheet
etc.. Is that all about Wishbone compliance?

TIA
mkr

Article: 135304
Subject: Re: Peter says Good Bye
From: Andreas Ehliar <ehliar-nospam@isy.liu.se>
Date: Thu, 25 Sep 2008 13:58:22 +0000 (UTC)
Links: << >>  << T >>  << A >>
On 2008-09-25, Steve Knapp <steveD.O.TknappA.Tprevailing-technologyD.O.Tcom> wrote:
> Peter, thank you again for being "progressive" and championing the "old" 
> customer-centric Xilinx ways.
>
>      Informed customer + responsive, communicative vendor = Long-time 
> customer and successful vendor.

Agreed. Another point that is not immediately obvious is the fact that a posting
from comp.arch.fpga is often the first hit you get when you search for some kind
of FPGA related issue. So many people beside Usenet readers will benefit from
vendor participation on comp.arch.fpga.

While I understand that Peter have other things to do with his time, I hope
that at least some people from Xilinx will continue their presence here.

The response from Xilinx here is probably the biggest issue why I landed in
Xilinx-land and haven't yet bothered to look deeply at Altera, Actel or Lattice.
Although I must admit that I'm probably not the biggest customer that Xilinx
have :)

/Andreas

Article: 135305
Subject: Re: wishbone interface
From: argee <nope@nope.com>
Date: Thu, 25 Sep 2008 16:10:11 +0200
Links: << >>  << T >>  << A >>
mkr wrote:
> I ma trying to make my first design (UART) wishbone compliant and
> looked at wishbone specification and several examples with source code
> on the net. Though I understand there are some restrictions on
> multiplexing of address and data bus, tri-state bus etc.. in many
> cases I hardly see any additional logic in the wishhbone compliant
> designs other than naming the external signals as per the wishbone
> standard and some additional documentation like wishbone datasheet
> etc.. Is that all about Wishbone compliance?
> 

Hi,

I suggest you take a closer look at the "Simple 8-bit SLAVE Output Port" 
example in the Tutorial appendix of the WISHBONE specification. For 
simple slave modules, that's pretty much all there is to it.

HTH

RG

Article: 135306
Subject: Re: Xilinx Mode Select Pins
From: Barry <barry374@gmail.com>
Date: Thu, 25 Sep 2008 08:05:32 -0700 (PDT)
Links: << >>  << T >>  << A >>
On Sep 24, 10:22=A0am, emeb <ebromba...@gmail.com> wrote:
> On Sep 24, 5:45=A0am, Andy <jonesa...@comcast.net> wrote:
>
> > On Sep 23, 5:54=A0pm, "Steve Knapp" <steveD.O.TknappA.Tprevailing-
>
> > technologyD.O.Tcom> wrote:
> > > IIRC, JTAG is always functional, even if using Master Serial configur=
ation
> > > mode. =A0
>
> > Normal JTAG functions are always available. JTAG configuration
> > functions (on some devices, including at least V2) are not available
> > unless the mode pins are set to JTAG, or a valid configuration has
> > already been completed. Don't ask me how I know... ;^)
>
> Yeow! Does anyone know if this is true for V5? I've got a board going
> to layout with the mode pins grounded for master serial, but I want to
> be able to do JTAG as well just in case. This is on a BGA part, so if
> I need to tweak these pins, I'll have to route them out or be screwed.
>
> Eric

JTAG certainly works for ChipScope with mode pins grounded.  I am
doing that right now.

Article: 135307
Subject: Re: Xilinx Mode Select Pins
From: emeb <ebrombaugh@gmail.com>
Date: Thu, 25 Sep 2008 08:29:21 -0700 (PDT)
Links: << >>  << T >>  << A >>
On Sep 25, 8:05=A0am, Barry <barry...@gmail.com> wrote:
> On Sep 24, 10:22=A0am, emeb <ebromba...@gmail.com> wrote:
>
>
>
> > On Sep 24, 5:45=A0am, Andy <jonesa...@comcast.net> wrote:
>
> > > On Sep 23, 5:54=A0pm, "Steve Knapp" <steveD.O.TknappA.Tprevailing-
>
> > > technologyD.O.Tcom> wrote:
> > > > IIRC, JTAG is always functional, even if using Master Serial config=
uration
> > > > mode. =A0
>
> > > Normal JTAG functions are always available. JTAG configuration
> > > functions (on some devices, including at least V2) are not available
> > > unless the mode pins are set to JTAG, or a valid configuration has
> > > already been completed. Don't ask me how I know... ;^)
>
> > Yeow! Does anyone know if this is true for V5? I've got a board going
> > to layout with the mode pins grounded for master serial, but I want to
> > be able to do JTAG as well just in case. This is on a BGA part, so if
> > I need to tweak these pins, I'll have to route them out or be screwed.
>
> > Eric
>
> JTAG certainly works for ChipScope with mode pins grounded. =A0I am
> doing that right now.

Ben, Barry,

Yes - I have several systems with mode pins grounded that work both in
Master Serial and JTAG modes. I believe that Andy's point was that if
for some reason the Master Serial configuration fails, then JTAG
doesn't work as a backup on systems up to & including V2. My question
is this: for V5 systems, if Master Serial fails, will JTAG work as a
backup? Have either of you tried JTAG when there isn't a valid
configuration loaded into the config ROM?

Eric

Article: 135308
Subject: Re: FPGA Lab Liquidation Sale
From: John_H <newsgroup@johnhandwork.com>
Date: Thu, 25 Sep 2008 08:52:05 -0700 (PDT)
Links: << >>  << T >>  << A >>
No PayPal ordering method?  What a shame.

Some interesting stuff - thanks for making it available.

Article: 135309
Subject: Re: Weird DCM problem with external deskew
From: Ed McGettigan <ed.mcgettigan@xilinx.com>
Date: Thu, 25 Sep 2008 09:32:11 -0700
Links: << >>  << T >>  << A >>
Benjamin Couillard wrote:
> the clock source is the output of another DCM, so there's one problem,
> my reset circuit didn't take that in account. I changed the reset
> connection so that the reset input of the deskew DCM (the 2nd one in
> the chain) is "FIRST_DCM_LOCKED" not. However, even with that change,
> I still get the same behavior. The 2nd DCM locks when I program the
> FPGA with the platform cable and it doesn't when I program it with the
> MPU.
> 
> But I still gotta check some other things like the MPU code.

Are you using any BUFGMUX cells?  Which family is the design in?

Ed McGettigan
--
Xilinx Inc.

Article: 135310
Subject: Re: Peter says Good Bye
From: "Steve Knapp" <steveD.O.TknappA.Tprevailing-technologyD.O.Tcom>
Date: Thu, 25 Sep 2008 09:34:37 -0700
Links: << >>  << T >>  << A >>

"przemek klosowski" <przemek.klosowski@gmail.nospam> wrote in message 
news:gYiCk.957$Jw.827@nwrddc02.gnilink.net...

<snip>

> Xilinx should understand that accumulated influence of people like you
> and Austin may be more important than a legion of marketeers. They should
> make you an editor, and give you a bunch of them to ghost-write the docs
> as you see fit.

Agreed.  This isn't just a Xilinx issue.  I've been amazed by the number of 
companies with similar thinking, with a thought process that goes something 
like this.

* <insert name> is too valuable (or expensive) to waste <his/her> time 
communicating with customers. <He/She> has more important things to do.

* <insert name> doesn't always use the "politically correct" marketing terms 
or is sometimes too open and honest about product issues.

* We already know what is best for our customers.  Participating in an open 
forum doesn't provide much return on investment.  A glossy four-page 
brochure is far more important.

Peter, thank you again for being "progressive" and championing the "old" 
customer-centric Xilinx ways.

     Informed customer + responsive, communicative vendor = Long-time 
customer and successful vendor.

===============================================
Steven K. Knapp
Prevailing Technology, Inc.
Web:  www.prevailing-technology.com


Article: 135311
Subject: Re: Weird DCM problem with external deskew
From: Benjamin Couillard <benjamin.couillard@gmail.com>
Date: Thu, 25 Sep 2008 09:38:12 -0700 (PDT)
Links: << >>  << T >>  << A >>
On 25 sep, 12:32, Ed McGettigan <ed.mcgetti...@xilinx.com> wrote:
> Benjamin Couillard wrote:
> > the clock source is the output of another DCM, so there's one problem,
> > my reset circuit didn't take that in account. I changed the reset
> > connection so that the reset input of the deskew DCM (the 2nd one in
> > the chain) is "FIRST_DCM_LOCKED" not. However, even with that change,
> > I still get the same behavior. The 2nd DCM locks when I program the
> > FPGA with the platform cable and it doesn't when I program it with the
> > MPU.
>
> > But I still gotta check some other things like the MPU code.
>
> Are you using any BUFGMUX cells? =A0Which family is the design in?
>
> Ed McGettigan
> --
> Xilinx Inc.

I solved the problem, the first DCM in the chain wasn't properly
configured. The expected input frequency was set to 29 MHz instead of
49 MHz.

 Now everything works fine.

Article: 135312
Subject: Re: decimal to ieee 754 single precision floating point
From: Kevin Neilson <kevin_neilson@removethiscomcast.net>
Date: Thu, 25 Sep 2008 15:32:53 -0600
Links: << >>  << T >>  << A >>
jack.harvard@googlemail.com wrote:
> what's the best way to transform decimal numbers to 32bits wide ieee
> 754 single precision floating point numbers? possibly using perl or
> verilog. i did some searching, there are converters like this
> http://babbage.cs.qc.edu/IEEE-754/Decimal.html, but not for large
> amount of data. thanks, -j

I've never had cause to use it, but there are Verilog functions 
$bitstoreal and $realtobits which convert reals to the IEEE-754 format 
and vice versa.  (Note: these use the 64-bit double format, not 
single-precision.)  One main use of these is to allow a user to pass 
reals as vectors through 64-bit ports (since a port can't have a real 
format).  If you have Matlab, you can also use "format hex".
-Kevin

Article: 135313
Subject: Please recommend good textbook or technical report about FPGA
From: KJ <lkjrsy@gmail.com>
Date: Thu, 25 Sep 2008 14:49:43 -0700 (PDT)
Links: << >>  << T >>  << A >>
Hello everyone.

The first thing I need to do is programming the data transfer between
CPU and PCI based FPGA.
OS is Linux. For now, I don't know how to approach this.

Is there anyone who recommend the useful textbook and technical report
for this kind of project?
Useful links also would be very appreciated.


Article: 135314
Subject: Re: LED lights flashing while LCD shows chars, Spartan-3A
From: m m <msmeerkat@gmail.com>
Date: Thu, 25 Sep 2008 16:30:04 -0700 (PDT)
Links: << >>  << T >>  << A >>
Thanks for the hint.

I tried with a different design/programming, this time declaring a
FSM, and now it is working properly, without the LEDs blinking.

I never knew what was not working properly of last design.

Oh and I found the LCD's datasheet (Sitronix ST7066U)
http://www.sitronix.com.tw/sitronix/product.nsf/Doc/ST7066U?OpenDocument
I honestly have not checked it completely. I am now working with other
phase.


Thank you,
m  m   _s
__________________________

On Sep 19, 5:51=A0pm, "Steve Knapp" <steveD.O.TknappA.Tprevailing-
technologyD.O.Tcom> wrote:

> I looked at the Spartan-3A Starter Kit board schematic, but it does not
> appear that the character LCD pins are shared with the LEDs.http://www.xi=
linx.com/support/documentation/boards_and_kits/ug334.pdf
>
> Do you have other outputs in the design? =A0Are those output pins defined=
 a
> LOCation in the UCF file? =A0It could be that output pins with an unassig=
ned
> pin location are landing on the LED pins.
>
> =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D
> Steven K. Knapp
> Prevailing Technology, Inc.
> Web: =A0www.prevailing-technology.com


Article: 135315
Subject: Re: Please recommend good textbook or technical report about FPGA coprocessor
From: Andreas Ehliar <ehliar-nospam@isy.liu.se>
Date: Fri, 26 Sep 2008 01:03:49 +0000 (UTC)
Links: << >>  << T >>  << A >>
On 2008-09-26, Mark McDougall <markm@vl.com.au> wrote:
> You probably should read O'Reilly's "Linux Device Drivers". IIRC there is
> some way to map PCI memory into user space without writing a driver, but
> it's been a long time...

I wrote a post about this on comp.arch.fpga some time ago:

http://groups.google.com/group/comp.arch.fpga/browse_thread/thread/5171491963302b63/daf1deb1296c46eb?lnk=st&q=group%3Acomp.arch.fpga+linux+ehliar#daf1deb1296c46eb

Essentially what you do is that you open /dev/mem and mmap it at the
appropriate location.

/Andreas

Article: 135316
Subject: Re: Please recommend good textbook or technical report about FPGA
From: Mark McDougall <markm@vl.com.au>
Date: Fri, 26 Sep 2008 12:58:08 +1000
Links: << >>  << T >>  << A >>
KJ wrote:

> The first thing I need to do is programming the data transfer between
> CPU and PCI based FPGA.
> OS is Linux. For now, I don't know how to approach this.

PCI-based FPGA? Do you need to implemented PCI in the FPGA, or is there
already a PCI slave on the FPGA board?

> Is there anyone who recommend the useful textbook and technical report
> for this kind of project?
> Useful links also would be very appreciated.

Depends on your answer to the above question.

You probably should read O'Reilly's "Linux Device Drivers". IIRC there is
some way to map PCI memory into user space without writing a driver, but
it's been a long time...

Regards,

-- 
Mark McDougall, Engineer
Virtual Logic Pty Ltd, <http://www.vl.com.au>
21-25 King St, Rockdale, 2216
Ph: +612-9599-3255 Fax: +612-9599-3266

Article: 135317
Subject: Having problems with using flash in EDK
From: Goli <togoli@gmail.com>
Date: Thu, 25 Sep 2008 20:04:50 -0700 (PDT)
Links: << >>  << T >>  << A >>
Hi,

I am using EDK10.1, and I have Atmel 32MB flash on my board,
(AT49BV322D). But whenever I try to program the flash through the
Device configuration, Program Flash tab it says, Unable to query part
layout using CFI.

I have checked all the connections and even probed the signals, they
all seem to be fine.

Is there anything special that I have to do?

--
Goli

Article: 135318
Subject: Re: duty cycle significance
From: glen herrmannsfeldt <gah@ugcs.caltech.edu>
Date: Thu, 25 Sep 2008 20:54:46 -0800
Links: << >>  << T >>  << A >>
Andreas Ehliar wrote:
(I wrote)

>>The Intel 8086 and 8088 use a 33% duty cycle clock for maximum
>>clock speed.  Something strange in that design.

> My guess would be that this is caused by a latch based design.

That would do it, but it might also be related to dynamic
logic.  Unlike many others, Intel processors such as the
8080 and 8086 used dynamic logic, somewhat like DRAM.
(It might just be that registers are DRAM, or it might
be much more.)

The 8080 has a two phase clock with different times
for each phase.  (That is, two clock input pins.)

The 8086 has a minimum clock frequency of 2MHz due to
the use of dynamic logic.

-- glen


Article: 135319
Subject: Re: wishbone interface
From: mkr <mahenreddy@gmail.com>
Date: Thu, 25 Sep 2008 22:57:32 -0700 (PDT)
Links: << >>  << T >>  << A >>
RG, thanks.

I looked at the 8-bit slave o/p port and 16X8 bit slave memory given
in the appendix of the spec. The 16X8 bit slave memory  has only one
AND gate besides the required two 16X4 bit SRAM modules. Even if you
don't need it to make wb compliant you pretty much need to have
everything (data in , data out , address in, Clock in , ack , write
enable  and associated logic) and design it the same way whther you
want it wb compatible or not. The only obvious difference is naming of
interface signals as per wb spec. I know I am missing something but
not able to figure out.

I designed a simple UART as my first project and trying it to make it
wb compliant.  I read the wb spec but I don't see the need for any
changes, to make it wb compliant other than changing the interface
signal names. On the system side UART has clk,_in rst_in, Data in/out,
reg address in and write enable signals and the logic is obvious from
signal names. As far as the interface is concerned I don't find any
difference between my current non wb  UART and the example 16x8 memory
in the spec except the signal names.

I am stuck here and not able to move forward to the next design.

- mkr

On Sep 25, 7:10=A0pm, argee <n...@nope.com> wrote:
> mkr wrote:
> > I ma trying to make my first design (UART) wishbone compliant and
> > looked at wishbone specification and several examples with source code
> > on the net. Though I understand there are some restrictions on
> > multiplexing of address and data bus, tri-state bus etc.. in many
> > cases I hardly see any additional logic in the wishhbone compliant
> > designs other than naming the external signals as per the wishbone
> > standard and some additional documentation like wishbone datasheet
> > etc.. Is that all about Wishbone compliance?
>
> Hi,
>
> I suggest you take a closer look at the "Simple 8-bit SLAVE Output Port"
> example in the Tutorial appendix of the WISHBONE specification. For
> simple slave modules, that's pretty much all there is to it.
>
> HTH
>
> RG


Article: 135320
Subject: Re: wishbone interface
From: Martin Thompson <martin.j.thompson@trw.com>
Date: Fri, 26 Sep 2008 10:24:33 +0100
Links: << >>  << T >>  << A >>
mkr <mahenreddy@gmail.com> writes:

> RG, thanks.
>
> I looked at the 8-bit slave o/p port and 16X8 bit slave memory given
> in the appendix of the spec. The 16X8 bit slave memory  has only one
> AND gate besides the required two 16X4 bit SRAM modules. Even if you
> don't need it to make wb compliant you pretty much need to have
> everything (data in , data out , address in, Clock in , ack , write
> enable  and associated logic) and design it the same way whther you
> want it wb compatible or not. The only obvious difference is naming of
> interface signals as per wb spec. I know I am missing something but
> not able to figure out.
>

I'm not sure you are missing anything.  For simple slaves, it really
is that simple.

Cheers,
Martin

-- 
martin.j.thompson@trw.com 
TRW Conekt - Consultancy in Engineering, Knowledge and Technology
http://www.conekt.net/electronics.html

Article: 135321
Subject: MicroBlaze SMP system DEMO
From: Pablo H <pablo.huerta@gmail.com>
Date: Fri, 26 Sep 2008 03:46:24 -0700 (PDT)
Links: << >>  << T >>  << A >>
I have recently developed a MicroBlaze SMP system for FPGAs and want
to share it with the community.

The system can be used as an example of how to build SMP systems on
FPGA and how to write applications for it.

In the software side, a modified version of xilkernel I have developed
is used to write multi-threaded applications that run on the SMP
system.

If you want to take a look at the system or to the modified xilkernel
version, you can download it from http://www.escet.urjc.es/~phuerta/SMP_project.htm

The system has 2 Microblazes and is for use on the XUP V2P board, but
there will be demos for other boards and with more processors soon.

Any comments, suggestions and questions are welcomed !

Regards,

Pablo H

Article: 135322
Subject: Re: OFDM band switch ...
From: "jerzy.gbur@gmail.com" <jerzy.gbur@gmail.com>
Date: Fri, 26 Sep 2008 04:32:17 -0700 (PDT)
Links: << >>  << T >>  << A >>
On 25 Wrz, 09:06, "Kappasm" <_NOSPAM_78kappa78_NO_S...@virgilio.it>
wrote:
> Hi Jerzy Gbur.
>
> Thanks for your replay.
>
> > You can design new OFDM symbol, without part of carriers. I did not
> > understand it anyway as achieve it. you might be more accurate ?
>
> I sincerely not thought I had at this solution.
>
> I have an IFFT of 2048 point. Mapping pilot at input of IFFT for +/-DC
> offset, [0 - 852 Data pilot] [853 - 1194 Null pilot] [1195 - 2047 Data
> pilot]. Process IFFT and i obtain a time domain data, with sample time of
> 7/64e6 I obtain 8 MHz band.
>
> Keeping the same IFFT point to 2048 as carriers can remove ?

I'm not sure I understood. But what I propose is, for narrow band set,
for example [0 - 511 Data] [512 - 1535 Nul] [1536 - 2047 Data], It
depend what band you exactly want.
Lenght of IFFT transform should stay the same.

IMHO it's most simple way to filfull your requirements.

Best Regards,

Jerzy Gbur

Article: 135323
Subject: Re: Use of divided clocks inside modules
From: "Symon" <symon_brewer@hotmail.com>
Date: Fri, 26 Sep 2008 12:34:43 +0100
Links: << >>  << T >>  << A >>
Andy wrote:
> On Sep 23, 11:42 am, Gael Paul <gael.p...@gmail.com> wrote:
>> 5. Once you have converted your generated clock, you will probably
>> need to add multi-cycle paths constraints to avoid false critical
>> paths.
>
> Avoid using multi-cycle path constraints unless you absolutely have to
> use them, or are getting unacceptably long run-times trying to meet
> single cycle timing on those paths. The reason is that multi-cycle and
> false path constraints can be tricky to specify correctly, and make
> sure that only those paths that are truly multi-cycle or false are
> constrained (actually relaxed) as such.
>
> Andy

Hi Andy,

I've noticed that a lot of people on this newsgroup have a morbid fear of 
mutli-cycle clock constraints.

Personally, I use them all the time and have few problems; certainly fewer 
problems than I would have if I tried to do the same thing with multiple 
clocks. In a situation where a block of logic can run at (say) half the 
clock rate, I give it an enable and set the constraints accordingly using 
the
NET "clock_enable_signal*" TNM=TS1;
thing in the Xilinx tools. (BTW, The asterisk is useful if the enable gets 
replicated by the synthesis tools) This lets the P&R tools concentrate 
effort where it's needed.

Of course, avoiding multi-cycle constraints by using a divided clock would 
be bad.

Can you give an example where you've run into problems?

Thanks, Symon. 



Article: 135324
Subject: Re: decimal to ieee 754 single precision floating point
From: Gabor <gabor@alacron.com>
Date: Fri, 26 Sep 2008 05:33:03 -0700 (PDT)
Links: << >>  << T >>  << A >>
On Sep 25, 5:32=A0pm, Kevin Neilson
<kevin_neil...@removethiscomcast.net> wrote:
> jack.harv...@googlemail.com wrote:
> > what's the best way to transform decimal numbers to 32bits wide ieee
> > 754 single precision floating point numbers? possibly using perl or
> > verilog. i did some searching, there are converters like this
> >http://babbage.cs.qc.edu/IEEE-754/Decimal.html, but not for large
> > amount of data. thanks, -j
>
> I've never had cause to use it, but there are Verilog functions
> $bitstoreal and $realtobits which convert reals to the IEEE-754 format
> and vice versa. =A0(Note: these use the 64-bit double format, not
> single-precision.) =A0One main use of these is to allow a user to pass
> reals as vectors through 64-bit ports (since a port can't have a real
> format). =A0If you have Matlab, you can also use "format hex".
> -Kevin

If you have a C compiler you can write a simple program that
includes your data file as a comma-delimited list like:

float foo[] =3D {
#include numberlist.csv
};

and then write some simple code to coerce the values to unsigned int
and print them out.

Just a thought....



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