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 130000

Article: 130000
Subject: Re: infer block ram with mismatched port width
From: KJ <kkjennings@sbcglobal.net>
Date: Wed, 12 Mar 2008 12:38:30 -0700 (PDT)
Links: << >>  << T >>  << A >>
On Mar 12, 1:05=A0pm, "u_stad...@yahoo.de" <u_stad...@yahoo.de> wrote:
> hi
>
> i have a question on how to infer a block ram with mismatched ports.
<snip>
> could somebody help me out how to modify this so that the data width
> on port A is 32 bits and on port B 8 bits.
> sorry but i could not find an example for this or figure it out
> myself.

1. Infer 4 dual memories that are each 8 bits wide (i.e. the width of
the smaller port, in this case the 'B' side).
2. Use the lower two bits of the B side address gated with the 'B'
side write enable to generate 4 individual write enables for the above
mentioned 4 dual port memories.
3. The 'A' side write enable goes to each of the four dual port
memories.
4. During a 'B' side read, read from all 4 memories in parallel and
then again using the lower 2 'B' side address bits, mux the
appropriate byte to the 'B' side data out port.

There are other ways of doing this but the above is fairly forward to
follow.

The key ideas here are:
- If you want to use FPGA internal memory and it doesn't happen to
support mismatched data widths, then you have to instantiate multiple
smaller memories that have a data width that matches your smaller data
width.
- Wrap logic around those multiple memories to give the whole thing
the appearance of implementing dual port with mismatched data width
sizes.
Kevin Jennings

Article: 130001
Subject: Re: Could I develop a new gui using java based on the script language of ChipScope?
From: "Dwayne Dilbeck" <ddilbeck@yahoo.com>
Date: Wed, 12 Mar 2008 13:17:32 -0700
Links: << >>  << T >>  << A >>
I am going to address these out of order.
3) I can understand not having a budget.  Since my college was using a 
parallel port logic analyzer on the PC(cheap and limited input), I assumed 
you may have had access to some older technology like this. Just laying 
around in the labs.

2) The Spartan3e development kit is $150 each. Your students can certainly 
buy their own if they want.  That is only one engineering book in cost. Some 
of your students may want to do that.  You may want to list places where 
your students can order the boards.
Since, you are a Xilinx XUP, I doubt you will be the only professor looking 
to use the tools at hand. A good student may see this as a sound investment, 
especially if the boards you have available in the labs are limited in 
number and availability.

1) I agree the soft core will give you more flexibility for observation. I 
don't think that writing a new GUI for chipscope versus extra explanation 
written in the lab guides is worth the effort.

Have fun torturing....I mean educating your students. ;-)

"wicky" <wicky.zhang@gmail.com> wrote in message 
news:4ef30bf0-e18c-43f0-b082-05188f210ee9@i29g2000prf.googlegroups.com...
In the course about 8086/8088 processor, I just plan to introduce this
CPU and its bus transactions. if those younker are interested in more
about FPGA, i think "no problem", :-)

The reason why I choose FPGA for the course lab is:

1) Soft core 8086/8088 is more flexible, we could monitor the bus
transactions and even the cpu internal signal.   Furthermore, we could
introduce other processor system based on the same hardware platform.

2) We are a member of Xilinx XUP, we have several V2PRO and Spartan3E
boards.

3) We have no budget to prepare logic analyzer for every student

Btw: There is a opensource 8086/8088 for FPGA in this web:

http://www.ht-lab.com/freecores/cpu8086/cpu86.html

Many thanks!

Wicky



On Mar 12, 1:42 am, "Dwayne Dilbeck" <ddilb...@yahoo.com> wrote:
> Uhm....Do you really need chipscope for this? Is the purpose to introduce
> the students to FPGAs or to introduce them to the 8086/8088 processor and
> their bus transactions?
>
> Granted it has been a bit more than 10 years since I was in college. But I
> had multiple courses dealing with the 8086/8088 as well as courses with 
> the
> 6800 Motorola processor. I even had a DSP course with a TI processor.
> Seeing the BUS transaction during those classes was a simple lab 
> introducing
> us to a digital logic analyzer.
>
> Seems to me throwing the FPGA and Chipscope in when it may not need to be
> there is overly complex.
>
> If the purpose is to introduce them to FPGAs, then I think it is better to
> keep the students using the standard tools they can download from xilinx.
> Just have a good lab write up and there shouldn't be a problem. I remember
> many a lab where the tool used was complex, but the Lab kept us to a small
> subset of the functionality as an introduction to the tool. Adding a new
> yet simpler GUI, will only hinder the students wehn they start going to 
> job
> interviews and say they know how to use chipscope version "easy_mode".
>
> "wicky" <wicky.zh...@gmail.com> wrote in message
>
> news:c388a3b0-f07f-4572-8c61-a49743aaa57c@s13g2000prd.googlegroups.com...
>
>
>
> >I found Chipscope is too difficult to learn for college students, it
> > has too much options. I want to develop a simple gui software and
> > using it in a 8086/8088 FPGA embedded system. For example, students
> > can understand a bus transaction with just a simple mouse click in
> > this software, instead of setting so much options in the tranditional
> > Chipscope software. Can anyone give me some information about this
> > work? Thank you!
>
> > Best Regards,
>
> > Wicky- Hide quoted text -
>
> - Show quoted text -



Article: 130002
Subject: Re: SiliconBlue enters the FPGA fray
From: Antti <Antti.Lukats@googlemail.com>
Date: Wed, 12 Mar 2008 14:29:06 -0700 (PDT)
Links: << >>  << T >>  << A >>
On 10 Mrz., 18:35, austin <aus...@xilinx.com> wrote:
> John_H,
>
> Aye, that is the rub:  success.
>
> Austin

Austin,

if their silicon works at all, and is low cost as advertized, it will
defenetly find some happy customers just because the PACKAGE options
alone:

3500 cells, 10 Kbyte RAM   ==> 3x4 mm !!!
15000 cells, 32 Kbyte RAM ==> 5x6 mm

if I compare that with Spartan-3A family ??

Actel is supposed to have now IGLOO15 in 4x4 package but that as tiny
as hell, so doesnt really count, so as do not Altera MAX2's in 6x6

Doesnt even matter how bad the tools, for this size:package density
options the silicon will find instant customers very quickly in my
opinion.

So would Xilinx sell S4 to new customers if only the package options
would be better than S3 of course.

Antti

Article: 130003
Subject: Re: Matlab, RS-232, Ethernet
From: lm317t <lm317t@gmail.com>
Date: Wed, 12 Mar 2008 14:31:58 -0700 (PDT)
Links: << >>  << T >>  << A >>
On Mar 12, 4:27 am, StYm <satyam.dwiv...@gmail.com> wrote:
> On Mar 12, 2:34 am, sky46...@trline4.org wrote:
>
>
>
> > satyam <satyam.dwiv...@gmail.com> wrote:
> > >I want to interface matlab with the Xilinx Virtex-II pro board. Intent
> > >is to give input from matlab to the FPGA and to read the ouput of FPGA
> > >in matlab.
> > >Problem is in interfacing speed. I need high speed interface, of the
> > >order of 2 mega bits per second (Mbps). Seems RS-232 will be
> > >inadequate for my purpose. From Documents interface through ethernet
> > >seems to be a viable option but I am not sure. To summarize I want
> > >answers and suggestions on following:
> > >1). FPGA to PC communication by ethernet ?
>
> > Yes works.
>
> > >2). What can be the maximum speed ?
>
> > 1000Mbps depending on your ethernet chip(s).
>
> > >3). How to transfer data on ethernet by matlab ?
>
> > C socket programming
>
> > >4). Is it possible to write inputs (60 Mega bits) to some memory on
> > >FPGA board and then read it from there to do the computation ?
>
> > If your ethernet PHY manages 100 Mbps in full duplex, then yes.
Ethernet is a complex and expensive protocol (in terms of time) to
implement.  How about the opencores  USB 1.1 PHY/IP and using libusb
on the PC side?  It seems to me this would be much less time consuming
than ethernet.
>
> > >Please let me know if you have any suggestion for me.
>
> > Do you need realtime or synchronous operation?
>
> > Btw, there's lots of good stuff to be found via the google force luke ;)
>
> Dear Dave and sky46, thank you for responding.
>
> Seems Matlab has something in intrument control toolbox. I need to
> explore this.
> How about ethernet core for FPGA. The core provided by xilinx is too
> expensive to buy. Have  found an ethernet from opencore website.
> Trying to make it work.


Article: 130004
Subject: Re: SiliconBlue enters the FPGA fray
From: Peter Alfke <peter@xilinx.com>
Date: Wed, 12 Mar 2008 15:32:31 -0700 (PDT)
Links: << >>  << T >>  << A >>
On Mar 12, 2:29=A0pm, Antti <Antti.Luk...@googlemail.com> wrote:
> On 10 Mrz., 18:35, austin <aus...@xilinx.com> wrote:
>
> > John_H,
>
> > Aye, that is the rub: =A0success.
>
> > Austin
>
> Austin,
>
> if their silicon works at all, and is low cost as advertized, it will
> defenetly find some happy customers just because the PACKAGE options
> alone:
>
> 3500 cells, 10 Kbyte RAM =A0 =3D=3D> 3x4 mm !!!
> 15000 cells, 32 Kbyte RAM =3D=3D> 5x6 mm
>
> if I compare that with Spartan-3A family ??
>
> Actel is supposed to have now IGLOO15 in 4x4 package but that as tiny
> as hell, so doesnt really count, so as do not Altera MAX2's in 6x6
>
> Doesnt even matter how bad the tools, for this size:package density
> options the silicon will find instant customers very quickly in my
> opinion.
>
> So would Xilinx sell S4 to new customers if only the package options
> would be better than S3 of course.
>
> Antti

Antti, when a much smaller company (like Actel) wants to compete and
survive in this field, it must pick a protected niche. (Ultra- small
size, lowest power, non-volatility, different logic-to-pin ratio, you
name it!).
It is up to Xilinx and Altera to evaluate these "niches" and assess
whether they are attractive for us. Sometimes they are, sometimes they
are not. What is right for Actel is not necessarily right for Xilinx
and Altera. The user, however, benefits from the variety made possible
by the wide differences between companies, and their genetic
diversification.
"Survival of the fittest" combined with a diverse gene-pool. It works
for plants and animals...
Peter Alfke

Article: 130005
Subject: Re: Could I develop a new gui using java based on the script language
From: SKatsyuba@gmail.com
Date: Wed, 12 Mar 2008 16:23:31 -0700 (PDT)
Links: << >>  << T >>  << A >>
Wicky,

You can try to use Scanseer boundary-scan software instead of
ChipScope. Scanseer can display waveforms for FPGA pins and it's quite
easy-to-use. But, unfortunately, it can't show internal FPGA's
signals.

Scanseer home is http://www.scanseer.com

Kind regards,
SK

On Mar 11, 4:58 pm, wicky <wicky.zh...@gmail.com> wrote:
> I found Chipscope is too difficult to learn for college students, it
> has too much options. I want to develop a simple gui software and
> using it in a 8086/8088 FPGA embedded system. For example, students
> can understand a bus transaction with just a simple mouse click in
> this software, instead of setting so much options in the tranditional
> Chipscope software. Can anyone give me some information about this
> work? Thank you!
>
> Best Regards,
>
> Wicky

Article: 130006
Subject: Re: SiliconBlue enters the FPGA fray
From: Antti <Antti.Lukats@googlemail.com>
Date: Wed, 12 Mar 2008 16:24:59 -0700 (PDT)
Links: << >>  << T >>  << A >>
On 12 Mrz., 23:32, Peter Alfke <pe...@xilinx.com> wrote:
> On Mar 12, 2:29 pm, Antti <Antti.Luk...@googlemail.com> wrote:
>
>
>
> > On 10 Mrz., 18:35, austin <aus...@xilinx.com> wrote:
>
> > > John_H,
>
> > > Aye, that is the rub:  success.
>
> > > Austin
>
> > Austin,
>
> > if their silicon works at all, and is low cost as advertized, it will
> > defenetly find some happy customers just because the PACKAGE options
> > alone:
>
> > 3500 cells, 10 Kbyte RAM   ==> 3x4 mm !!!
> > 15000 cells, 32 Kbyte RAM ==> 5x6 mm
>
> > if I compare that with Spartan-3A family ??
>
> > Actel is supposed to have now IGLOO15 in 4x4 package but that as tiny
> > as hell, so doesnt really count, so as do not Altera MAX2's in 6x6
>
> > Doesnt even matter how bad the tools, for this size:package density
> > options the silicon will find instant customers very quickly in my
> > opinion.
>
> > So would Xilinx sell S4 to new customers if only the package options
> > would be better than S3 of course.
>
> > Antti
>
> Antti, when a much smaller company (like Actel) wants to compete and
> survive in this field, it must pick a protected niche. (Ultra- small
> size, lowest power, non-volatility, different logic-to-pin ratio, you
> name it!).
> It is up to Xilinx and Altera to evaluate these "niches" and assess
> whether they are attractive for us. Sometimes they are, sometimes they
> are not. What is right for Actel is not necessarily right for Xilinx
> and Altera. The user, however, benefits from the variety made possible
> by the wide differences between companies, and their genetic
> diversification.
> "Survival of the fittest" combined with a diverse gene-pool. It works
> for plants and animals...
> Peter Alfke

Peter,

would you describe "CONSUMER" as a niche market?

sorry, I could not resist.

Xilinx, Altera, Actel, Lattice have nothing to compete with the
SiliconBlue, the only one possible having competing solutions is
QuickLogic (ArticLink). Sure SiliconBlue is not targetting the general
purpose FPGA market, so you can call it niche. But consumer market is
a pretty damn big niche!

Antti



Article: 130007
Subject: Re: Cyclone III FPGA Starter Kit: As USB device? Using JTAG terminal
From: sdf <drop669@gmail.com>
Date: Wed, 12 Mar 2008 16:42:41 -0700 (PDT)
Links: << >>  << T >>  << A >>
On Mar 12, 8:11=A0pm, ghel...@lycos.com wrote:
> On Mar 12, 3:42 am, sdf <drop...@gmail.com> wrote:
>
> > Hi.
> > Got this board.
> > Two questions.
> > 1. Is it possible to use this board as prototype of some USB device
> > without any solder rewiring?
> > 2. Is it possible to have, let's say, nios2-terminal.exe on computer
> > side, computer connected to board by USB and have *something* but not
> > Nios2 to communicate with nios2-terminal.exe? In any words, it's
> > possible to use JTAG wire as RS232 port without using Nios2 CPU?
>
> 1: No. =A0The USB port is only for programming.
>
> 2: No. =A0Well, not really no, but you would need to replicate the way
> Altera/NIOS tunnels the serial port through JTAG, and I think it's
> highly unlikely that you will find documentation on that.

I found that: USB <-> FT245BL <-> EPM3128ATC100 <- JTAG ->
EP3C25F324NES
I probably will need to do something with EPM3128 and use JTAG link..
if someone have the same intentions as me, it will be interesting to
hear any suggestions.

Article: 130008
Subject: Re: SiliconBlue enters the FPGA fray
From: Jim Granville <no.spam@designtools.maps.co.nz>
Date: Thu, 13 Mar 2008 13:41:07 +1300
Links: << >>  << T >>  << A >>
Peter Alfke wrote:
> Antti, when a much smaller company (like Actel) wants to compete and
> survive in this field, it must pick a protected niche. (Ultra- small
> size, lowest power, non-volatility, different logic-to-pin ratio, you
> name it!).
> It is up to Xilinx and Altera to evaluate these "niches" and assess
> whether they are attractive for us. Sometimes they are, sometimes they
> are not. What is right for Actel is not necessarily right for Xilinx
> and Altera. The user, however, benefits from the variety made possible
> by the wide differences between companies, and their genetic
> diversification.
> "Survival of the fittest" combined with a diverse gene-pool. It works
> for plants and animals...

and, of course, Xilinx could always Buy them out. They did that
to get Coolrunner, and Coolrunner is now looking 'long in the tooth'
[in FPGA terms] with no recent advances.
MAX IIZ and MachXO are in a space Xilinx is missing.

-jg


Article: 130009
Subject: Re: SiliconBlue enters the FPGA fray
From: Jim Granville <no.spam@designtools.maps.co.nz>
Date: Thu, 13 Mar 2008 13:42:25 +1300
Links: << >>  << T >>  << A >>
Antti wrote:

> Austin,
> 
> if their silicon works at all, and is low cost as advertized, it will
> defenetly find some happy customers just because the PACKAGE options
> alone:
> 
> 3500 cells, 10 Kbyte RAM   ==> 3x4 mm !!!
> 15000 cells, 32 Kbyte RAM ==> 5x6 mm

Hi Antti,

I have not seen price promises, or package info yet.
Got any links for this ? (or send me a PDF ?)

-jg


Article: 130010
Subject: microblaze to blockram - Byte-Writes
From: kislo <kislo02@student.sdu.dk>
Date: Wed, 12 Mar 2008 18:17:29 -0700 (PDT)
Links: << >>  << T >>  << A >>
I am trying to interface a 32 bit blockram to microblaze (spartan 3E),
using the User-Address ip support in the "Create and import
peripheral". I have instantiated a black box blockram 32x256 and it
works fine with read and write operations of 32bit datawidth. But how
can i perform 8bit write and read operations? i know there are the BE
(byte_enable) bus, but i cant see how this can work with a single
block ram .. i would need 4 block ram=B4s inorder to perform byte write
and reads or what? i can see that in the never FPGA`s (according to
Block Memory Generator 2.6 .pdf) that block ram supports Byte-Writes,
which is EXACTLY what i need .. but im using the spartan 3E which
dosent support this feature .. are there any work around so i dont
need 4 block ram=B4s of 8x256 instead of a single 32x256 ? ... is there
a way to create more than one block ram instance that only uses one
block-ram in hardware?

Regards
Kim

Article: 130011
Subject: Using xilinx XAUI core in Ethernet design. What is the exact frame
From: "mynewlifever@yahoo.com.cn" <mynewlifever@yahoo.com.cn>
Date: Wed, 12 Mar 2008 19:23:14 -0700 (PDT)
Links: << >>  << T >>  << A >>
We use xilinx xaui core in our Ethernet design, we choose XGMII as our
internal interface. And now, I am confused. I found that the data
between START(0xfb) andTERMINATE(0xfd) will all be transmit through
XAUI. So, if we use xaui core in Ethernet design, do we need to
transmit preamble, sfd and FCS? Now, I think may be we need preamble
and sfd to meet XGMII standard, but what about FCS. In 802.3ae
chapter46, it said CRC is done by RS.Please help me to make sure frame
format, Thank you!

Article: 130012
Subject: Re: Could I develop a new gui using java based on the script language
From: wicky <wicky.zhang@gmail.com>
Date: Wed, 12 Mar 2008 20:38:16 -0700 (PDT)
Links: << >>  << T >>  << A >>
Sorry, I don't think that I describe my idea clearly because of my
poor english.

First, the XUP lab now serves for other course such as "introduction
to VHDL" and so on. Now I hope we can utilize this lab more
effectively. From the 8086/8088 lab on, we may port a serials of
course to this lab and form a more close relation among those courses.

As for 8086/8088 lab which is not so much relative with fpga
technology, a more simple debugging software will facilitate to most
students. This course will focus on 8086/8088 system only, but with a
soft core 8086/8088 lab, I think it will be helpful for the successive
FPGA relative courses in which students will use those special
software such as chipscope.

Many thanks!

Wicky


On Mar 13, 4:17=A0am, "Dwayne Dilbeck" <ddilb...@yahoo.com> wrote:
> I am going to address these out of order.
> 3) I can understand not having a budget. =A0Since my college was using a
> parallel port logic analyzer on the PC(cheap and limited input), I assumed=

> you may have had access to some older technology like this. Just laying
> around in the labs.
>
> 2) The Spartan3e development kit is $150 each. Your students can certainly=

> buy their own if they want. =A0That is only one engineering book in cost. =
Some
> of your students may want to do that. =A0You may want to list places where=

> your students can order the boards.
> Since, you are a Xilinx XUP, I doubt you will be the only professor lookin=
g
> to use the tools at hand. A good student may see this as a sound investmen=
t,
> especially if the boards you have available in the labs are limited in
> number and availability.

>
> 1) I agree the soft core will give you more flexibility for observation. I=

> don't think that writing a new GUI for chipscope versus extra explanation
> written in the lab guides is worth the effort.
>
> Have fun torturing....I mean educating your students. ;-)
>
> "wicky" <wicky.zh...@gmail.com> wrote in message
>
> news:4ef30bf0-e18c-43f0-b082-05188f210ee9@i29g2000prf.googlegroups.com...
> In the course about 8086/8088 processor, I just plan to introduce this
> CPU and its bus transactions. if those younker are interested in more
> about FPGA, i think "no problem", :-)
>
> The reason why I choose FPGA for the course lab is:
>
> 1) Soft core 8086/8088 is more flexible, we could monitor the bus
> transactions and even the cpu internal signal. =A0 Furthermore, we could
> introduce other processor system based on the same hardware platform.
>
> 2) We are a member of Xilinx XUP, we have several V2PRO and Spartan3E
> boards.
>
> 3) We have no budget to prepare logic analyzer for every student
>
> Btw: There is a opensource 8086/8088 for FPGA in this web:
>
> http://www.ht-lab.com/freecores/cpu8086/cpu86.html
>
> Many thanks!
>
> Wicky
>
> On Mar 12, 1:42 am, "Dwayne Dilbeck" <ddilb...@yahoo.com> wrote:
>
>
>
> > Uhm....Do you really need chipscope for this? Is the purpose to introduc=
e
> > the students to FPGAs or to introduce them to the 8086/8088 processor an=
d
> > their bus transactions?
>
> > Granted it has been a bit more than 10 years since I was in college. But=
 I
> > had multiple courses dealing with the 8086/8088 as well as courses with
> > the
> > 6800 Motorola processor. I even had a DSP course with a TI processor.
> > Seeing the BUS transaction during those classes was a simple lab
> > introducing
> > us to a digital logic analyzer.
>
> > Seems to me throwing the FPGA and Chipscope in when it may not need to b=
e
> > there is overly complex.
>
> > If the purpose is to introduce them to FPGAs, then I think it is better =
to
> > keep the students using the standard tools they can download from xilinx=
.
> > Just have a good lab write up and there shouldn't be a problem. I rememb=
er
> > many a lab where the tool used was complex, but the Lab kept us to a sma=
ll
> > subset of the functionality as an introduction to the tool. Adding a new=

> > yet simpler GUI, will only hinder the students wehn they start going to
> > job
> > interviews and say they know how to use chipscope version "easy_mode".
>
> > "wicky" <wicky.zh...@gmail.com> wrote in message
>
> >news:c388a3b0-f07f-4572-8c61-a49743aaa57c@s13g2000prd.googlegroups.com...=

>
> > >I found Chipscope is too difficult to learn for college students, it
> > > has too much options. I want to develop a simple gui software and
> > > using it in a 8086/8088 FPGA embedded system. For example, students
> > > can understand a bus transaction with just a simple mouse click in
> > > this software, instead of setting so much options in the tranditional
> > > Chipscope software. Can anyone give me some information about this
> > > work? Thank you!
>
> > > Best Regards,
>
> > > Wicky- Hide quoted text -
>
> > - Show quoted text -- Hide quoted text -
>
> - Show quoted text -


Article: 130013
Subject: Re: Virtex-5 FX when ? (III)
From: "Xilinx User" <anonymous@net.com>
Date: Wed, 12 Mar 2008 20:52:17 -0700
Links: << >>  << T >>  << A >>
No Virtex-5/FX has been, or ever will be, released.

...

If you mean the Virtex-5/FXT, then that's a whole 'nother story :)
Last November, I heard "soon", as in, synchronized with the next major
release of ISE (10.1?)

I hope Xilinx introduces Systemverilog synthesis with ISE 10.1 -- competitor 
Altera has
been taking the lead for so long now in that area...
(whether anyone actually uses Systemverilog synthesis in a large 
FPGA-project,
 is a debate for another day...)

"Udo" <WeikEngOff@aol.com> wrote in message 
news:3faa5be2-29b9-4e51-b4fc-a0b7f94e9bd6@u69g2000hse.googlegroups.com...
> Hello Antti, Peter and ...,
>
> yep, meanwhile many, many months later - the same question...
> V5-FX?
>
> Thanks and greetings
> Udo 



Article: 130014
Subject: Re: Cyclone III FPGA Starter Kit: As USB device? Using JTAG terminal without Nios2?
From: "Xilinx User" <anonymous@net.com>
Date: Wed, 12 Mar 2008 20:53:53 -0700
Links: << >>  << T >>  << A >>
I thought Altera uses one of FTDI 's USB/RS-232 bridge-chips in their
programming solution.

<ghelbig@lycos.com> wrote in message 
news:d5104cd3-98c1-449d-85ff-ebc074207564@e23g2000prf.googlegroups.com...
> On Mar 12, 3:42 am, sdf <drop...@gmail.com> wrote:
>> Hi.
>> Got this board.
>> Two questions.
>> 1. Is it possible to use this board as prototype of some USB device
>> without any solder rewiring?
>> 2. Is it possible to have, let's say, nios2-terminal.exe on computer
>> side, computer connected to board by USB and have *something* but not
>> Nios2 to communicate with nios2-terminal.exe? In any words, it's
>> possible to use JTAG wire as RS232 port without using Nios2 CPU?
>
> 1: No.  The USB port is only for programming.
>
> 2: No.  Well, not really no, but you would need to replicate the way
> Altera/NIOS tunnels the serial port through JTAG, and I think it's
> highly unlikely that you will find documentation on that.
>
> Sorry. 



Article: 130015
Subject: Xilinx ISE Evaluation DVD 10.1 request...
From: "Xilinx User" <anonymous@net.com>
Date: Wed, 12 Mar 2008 21:03:22 -0700
Links: << >>  << T >>  << A >>
Please, please, please fix the DVD installation-wizard, so that
you can install the 32-bit (Webpack) tools on a 64-bit platform.
(This goes for both MS Windows and Linux.)

I really hate having to manually traverse the bin/lin32 subdir
(or bin/nt) on the DVD, because the toplevel installer determines
I have a 64-bit OS/platform, then cryptically tells me "registration
ID not valid!"  (I'd be nice if the installer said the Evaluation-ID
is only good for the 32-bit version of the tools, rather than
producing a totally misleading error-message.)

An little checkbox in the installer-GUI that lets the user override
the installer's platform-autodetect would be really nice :
"install 32-bit tools?" 



Article: 130016
Subject: Re: Could I develop a new gui using java based on the script language of ChipScope?
From: Andreas Ehliar <ehliar-nospam@isy.liu.se>
Date: Thu, 13 Mar 2008 04:21:43 +0000 (UTC)
Links: << >>  << T >>  << A >>
On 2008-03-12, wicky <wicky.zhang@gmail.com> wrote:
> In the course about 8086/8088 processor, I just plan to introduce this
> CPU and its bus transactions. if those younker are interested in more
> about FPGA, i think "no problem", :-)
>
> The reason why I choose FPGA for the course lab is:
>
> 1) Soft core 8086/8088 is more flexible, we could monitor the bus
> transactions and even the cpu internal signal.   Furthermore, we could
> introduce other processor system based on the same hardware platform.
>
> 2) We are a member of Xilinx XUP, we have several V2PRO and Spartan3E
> boards.
>
> 3) We have no budget to prepare logic analyzer for every student
>
> Btw: There is a opensource 8086/8088 for FPGA in this web:
>
> http://www.ht-lab.com/freecores/cpu8086/cpu86.html


That is an interesting idea, I have been thinking about exactly the
same thing actually as we currently use a rather old platform for
our bus transaction lab here which make it hard to demonstrate more
interesting features such as bus mastering. (Although we would probably
use the tg68 core instead, as our current course material is heavily
based on the MC68000.

Anyway, if you know that you are only interested in bus transactions you
could use a synthesized logic analyzer that is instantiated in your design.

Look at for example http://www.ise.pw.edu.pl/~wzab/fpgadbg/ for something
which seems fairly reasonable although I haven't tried it myself.


If you need to select signals dynamically I agree with other posters that
chipscope shouldn't be that hard to setup/use.


However, if you find that you have very special needs that cannot be taken
care of by chipscope it is actually possible to make your own logicanalyzer
that you can dynamically insert into a design. I have an example of how
to do that at the following URL:

http://www.da.isy.liu.se/~ehliar/pyxdl/

And a screenshot of the tool:
http://www.da.isy.liu.se/~ehliar/stuff/logicanalyzer.png

The current version presents an ASCII interface to the logic analyzer
but it should be quite simple to write a waveform viewer although I
haven't had time to worry about that. (The Python code is quite ugly but
I thought that it was better to release the code than to just let it
collect dust in my home directory.)



/Andreas

Article: 130017
Subject: Re: Design complexity in Logic cells - Virtex-5 FPGA
From: bkurtz@engineer.com
Date: Wed, 12 Mar 2008 21:29:08 -0700 (PDT)
Links: << >>  << T >>  << A >>
Using registers is probably the best way to go.  You avoid the
marketing mathematics of any given vendor, and to be honest with fmax
being primarily limited by routing now, it seems that designs have
very few levels of logic meaning the logic:register ratio is close to
1:1 for high performance IP.

Article: 130018
Subject: Re: Matlab, RS-232, Ethernet
From: StYm <satyam.dwivedi@gmail.com>
Date: Thu, 13 Mar 2008 00:24:52 -0700 (PDT)
Links: << >>  << T >>  << A >>
On Mar 13, 2:31 am, lm317t <lm3...@gmail.com> wrote:
> On Mar 12, 4:27 am, StYm <satyam.dwiv...@gmail.com> wrote:
>
> > On Mar 12, 2:34 am, sky46...@trline4.org wrote:
>
> > > satyam <satyam.dwiv...@gmail.com> wrote:
> > > >I want to interface matlab with the Xilinx Virtex-II pro board. Intent
> > > >is to give input from matlab to the FPGA and to read the ouput of FPGA
> > > >in matlab.
> > > >Problem is in interfacing speed. I need high speed interface, of the
> > > >order of 2 mega bits per second (Mbps). Seems RS-232 will be
> > > >inadequate for my purpose. From Documents interface through ethernet
> > > >seems to be a viable option but I am not sure. To summarize I want
> > > >answers and suggestions on following:
> > > >1). FPGA to PC communication by ethernet ?
>
> > > Yes works.
>
> > > >2). What can be the maximum speed ?
>
> > > 1000Mbps depending on your ethernet chip(s).
>
> > > >3). How to transfer data on ethernet by matlab ?
>
> > > C socket programming
>
> > > >4). Is it possible to write inputs (60 Mega bits) to some memory on
> > > >FPGA board and then read it from there to do the computation ?
>
> > > If your ethernet PHY manages 100 Mbps in full duplex, then yes.
>
> Ethernet is a complex and expensive protocol (in terms of time) to
> implement.  How about the opencores  USB 1.1 PHY/IP and using libusb
> on the PC side?  It seems to me this would be much less time consuming
> than ethernet.
>
>
Dear lm317t,

can you give me instructions as to how to setup USB 1.1 PHY/IP ? What
files to download, from where ?? and how to make it work ?

Thank you in advance.
StYm


> > > >Please let me know if you have any suggestion for me.
>
> > > Do you need realtime or synchronous operation?
>
> > > Btw, there's lots of good stuff to be found via the google force luke ;)
>
> > Dear Dave and sky46, thank you for responding.
>
> > Seems Matlab has something in intrument control toolbox. I need to
> > explore this.
> > How about ethernet core for FPGA. The core provided by xilinx is too
> > expensive to buy. Have  found an ethernet from opencore website.
> > Trying to make it work.


Article: 130019
Subject: Almost offtopic about HDL optimizing.
From: sdf <drop669@gmail.com>
Date: Thu, 13 Mar 2008 01:45:52 -0700 (PDT)
Links: << >>  << T >>  << A >>
Hi.
I see how Quartus synthesizer removes unused parts of algorithm
implemented in HDL, optimizes it and so on.
Other synthesizers probably do the same.
My question is almost offtopic: is there any method to do the same in
software code?
Is the starting point is in compiler theory or there're also another
topics?

Article: 130020
Subject: Re: Almost offtopic about HDL optimizing.
From: Kolja Sulimma <ksulimma@googlemail.com>
Date: Thu, 13 Mar 2008 02:05:57 -0700 (PDT)
Links: << >>  << T >>  << A >>
On 13 Mrz., 09:45, sdf <drop...@gmail.com> wrote:
> Hi.
> I see how Quartus synthesizer removes unused parts of algorithm
> implemented in HDL, optimizes it and so on.
> Other synthesizers probably do the same.
> My question is almost offtopic: is there any method to do the same in
> software code?
> Is the starting point is in compiler theory or there're also another
> topics?

Compiler optimizations are done differently. Each method it self is a
lot
more primitive than what is done for hardware optimization, but a
decent compiler
uses a larger set of these optimization strategies.

I did many years of research in circuit optimization and allways
wondered whether it
would be fruitful to apply more computational intensive algorithms to
critical code in software.

A starting point could be the work done in compiling software
descriptions to hardware. UC Berkeley
for example simulates the code and postulates theories on the data
range of variables. (e.g. int x is
allways smaller than 100). They then use solvers to proof these
assumptions. If proven, the information
can be used for optimization. This approach is computationally very
expensive.

Also: It turns out that many variable are only written once during
program execution. This means that there
is a big optimization potential to optimize a program after startup.
Java and .NET do this to some extend
but in principle you could also do with C and other languages.

Kolja Sulimma









Article: 130021
Subject: Re: Design complexity in Logic cells - Virtex-5 FPGA
From: Kolja Sulimma <ksulimma@googlemail.com>
Date: Thu, 13 Mar 2008 02:08:15 -0700 (PDT)
Links: << >>  << T >>  << A >>
On 13 Mrz., 05:29, bku...@engineer.com wrote:
> meaning the logic:register ratio is close to
> 1:1 for high performance IP.

As a counter example: The ration is almost 2:1 in the example given
by the OP.

Kolja Sulimma

Article: 130022
Subject: Re: Almost offtopic about HDL optimizing.
From: sdf <drop669@gmail.com>
Date: Thu, 13 Mar 2008 02:33:09 -0700 (PDT)
Links: << >>  << T >>  << A >>
On Mar 13, 11:05=A0am, Kolja Sulimma <ksuli...@googlemail.com> wrote:
> Also: It turns out that many variable are only written once during
> program execution. This means that there
> is a big optimization potential to optimize a program after startup.
> Java and .NET do this to some extend
> but in principle you could also do with C and other languages.

It looks like a paradox, but it looks like functional programming.
Paradox is in likeness of such low-level computation device and such
high-level methods...
I just noticed that if to transform algorithm written in C to
something which looks like FP program, it is then can be more easily
rewritten into HDL, manually, of course.

Article: 130023
Subject: MAXDELAY="1.0"
From: "water9580@yahoo.com" <water9580@yahoo.com>
Date: Thu, 13 Mar 2008 02:38:44 -0700 (PDT)
Links: << >>  << T >>  << A >>
I generate PCI-Express PHY with GTP wizard.the datawidth is selected
8bit or 16bit. but ,both the phy wrapper MAXDELAY constrain  are
"1.0" ! as you know,it is hard to meet the timing. I expect the 16 bit
datawidth can reduce Clk Freq.

     (* MAXDELAY="1.0" *) wire            tile0_txcharisk0_r;
     (* MAXDELAY="1.0" *) wire            tile0_txdetectrx0_r;
     (* MAXDELAY="1.0" *) wire            tile0_txelecidle0_r;
     (* MAXDELAY="1.0" *) wire            tile0_txchardispmode0_r;
     (* MAXDELAY="1.0" *) wire    [7:0]   tile0_txdata0_r;
     (* MAXDELAY="1.0" *) wire            tile0_txchardispmode1_r;
     (* MAXDELAY="1.0" *) wire            tile0_txcharisk1_r;
     (* MAXDELAY="1.0" *) wire            tile0_txdetectrx1_r;
     (* MAXDELAY="1.0" *) wire    [7:0]   tile0_txdata1_r;
     (* MAXDELAY="1.0" *) wire            tile0_txelecidle1_r;

can it modified 2.0 or commented out?

Article: 130024
Subject: Re: Matlab, RS-232, Ethernet
From: lm317t <lm317t@gmail.com>
Date: Thu, 13 Mar 2008 05:14:34 -0700 (PDT)
Links: << >>  << T >>  << A >>
USB 1.1 PHY core in Verilog:
http://www.opencores.org/projects.cgi/web/usb_phy/overview
You can do your own IP from usb specs or use the USB 1.1 IP on
opencores.org.

How to use an AVR to implement USB gives lots of helpful info on USB
protocol
http://www.obdev.at/products/avrusb/index.html

Libusb for the PC:
http://sf.net/projects/libusb

Libusb doesn't run on windows, just mac, linux, bsd

Also the super simple route is to use a USB-UART chip like the ft232rl
from digikey.  They go up to 3 Mbps.  No driver writing for you if you
go this route.

On Mar 13, 3:24 am, StYm <satyam.dwiv...@gmail.com> wrote:
> On Mar 13, 2:31 am, lm317t <lm3...@gmail.com> wrote:
>
> > On Mar 12, 4:27 am, StYm <satyam.dwiv...@gmail.com> wrote:
>
> > > On Mar 12, 2:34 am, sky46...@trline4.org wrote:
>
> > > > satyam <satyam.dwiv...@gmail.com> wrote:
> > > > >I want to interface matlab with the Xilinx Virtex-II pro board. Intent
> > > > >is to give input from matlab to the FPGA and to read the ouput of FPGA
> > > > >in matlab.
> > > > >Problem is in interfacing speed. I need high speed interface, of the
> > > > >order of 2 mega bits per second (Mbps). Seems RS-232 will be
> > > > >inadequate for my purpose. From Documents interface through ethernet
> > > > >seems to be a viable option but I am not sure. To summarize I want
> > > > >answers and suggestions on following:
> > > > >1). FPGA to PC communication by ethernet ?
>
> > > > Yes works.
>
> > > > >2). What can be the maximum speed ?
>
> > > > 1000Mbps depending on your ethernet chip(s).
>
> > > > >3). How to transfer data on ethernet by matlab ?
>
> > > > C socket programming
>
> > > > >4). Is it possible to write inputs (60 Mega bits) to some memory on
> > > > >FPGA board and then read it from there to do the computation ?
>
> > > > If your ethernet PHY manages 100 Mbps in full duplex, then yes.
>
> > Ethernet is a complex and expensive protocol (in terms of time) to
> > implement.  How about the opencores  USB 1.1 PHY/IP and using libusb
> > on the PC side?  It seems to me this would be much less time consuming
> > than ethernet.
>
> Dear lm317t,
>
> can you give me instructions as to how to setup USB 1.1 PHY/IP ? What
> files to download, from where ?? and how to make it work ?
>
> Thank you in advance.
> StYm
>
> > > > >Please let me know if you have any suggestion for me.
>
> > > > Do you need realtime or synchronous operation?
>
> > > > Btw, there's lots of good stuff to be found via the google force luke ;)
>
> > > Dear Dave and sky46, thank you for responding.
>
> > > Seems Matlab has something in intrument control toolbox. I need to
> > > explore this.
> > > How about ethernet core for FPGA. The core provided by xilinx is too
> > > expensive to buy. Have  found an ethernet from opencore website.
> > > Trying to make it work.




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