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 69175

Article: 69175
Subject: EDK 3.2
From: vananth@gmail.com (Vinod)
Date: 29 Apr 2004 04:53:40 -0700
Links: << >>  << T >>  << A >>
Hi,
I am having problems starting up the Xilinx Platform Studio. I have
installed EDK 3.2 with service pack 2. When I run it, I get a message
saying

"The procedure entry point
?GetProjectToolBarID@Dco_PlToolBarManager@@SAIXZ could not be located
in the dynamic link library libDco_Plugin.dll"

Im running Windows XP professional with SP1. Also have Xilinx ISE
5.1.3i installed. My env variables seem to be pointing to the correct
location. I've tried reinstalling, but of no avail.

Any help will be appreciated.

Vinod

Article: 69176
Subject: Design development costs for FPGA on PCI board (sorry if slightly off-topic)
From: fpgadev@yahoo.com (AndyAtHome)
Date: 29 Apr 2004 05:13:04 -0700
Links: << >>  << T >>  << A >>
Dear All,

After investigating the price of FPGA PCI boards for a project, it
seems the cost of a board is typically 3 to 4 times the component cost
for mid-range FPGA's like Xilinx 2V3000.

Of course board manufacturers have overheads and need to make a profit
- my point isn't to complain about their Gross Profit margin.

Rather, at what point is it better to meet the initial capital
expenditure of having your own board designed, thereafter only having
to meet board manufacturing costs?

For example, just how much design time and cost does it take to create
a board with -

Quicklogic or PLXTech PCI Bridge
Single FPGA
A PROM 
Single bank of DDR memory.

Surely this type of requirement is common and would not require months
of development time or am I being naive? :)

Any comments appreciated...

Thanks,

Andy.

Article: 69177
Subject: basic question, virtex 2 pro
From: vananth@gmail.com (Vinod)
Date: 29 Apr 2004 06:31:39 -0700
Links: << >>  << T >>  << A >>
Hi,
I am using EDK 3.2 with ISE 5.1.3 on a WinXp machine, and targeting
the Virtex 2 pro board. Does Modelsim XE allow me to simulate code
running on the PowerPC along with other external blocks written in
VHDL as one whole unit?
I do not have a board yet, and was also wondering  whether such
simulations would be prohibitively long.

Links to informative websites would be appreciated

Thanks
Vinod

Article: 69178
Subject: Re: VHDL / Verilog circuits work in 1-V still correct?
From: johnjakson@yahoo.com (john jakson)
Date: 29 Apr 2004 07:26:11 -0700
Links: << >>  << T >>  << A >>
"Boki" <bokiteam@ms21.hinet.net> wrote in message news:<c6ov4e$pk1@netnews.hinet.net>...
> Dear All,
> 
> VHDL / Verilog circuits  work in 1-V still correct?
> 
> I want to combine 1-V analog circuits and VHDL circuits..
> 
> Thanks
> 
> Boki.

HDL will work with digital at any voltage its designed for.

For mixed signal with VHDL I think there is AVHDL for analog VHDL
mixed signal. Its been a while since I looked at that. Somebody in the
VHDL group might know.

regards

johnjakson_usa_com

Article: 69179
Subject: Re: I think I fried my I/O bank... (virtex-E question)
From: Austin Lesea <austin@xilinx.com>
Date: Thu, 29 Apr 2004 07:44:32 -0700
Links: << >>  << T >>  << A >>
Not,

Seems extremely unlikely.  It is true that without Vcco connected to 
anything, the ESD protection as well as the other protection circuits 
for large positive voltages is not well protected.

Is this attached to a long cable?  If so, you may blow out the IOBs even 
with the Vcco connected (inductive kickback can cause very large voltages).

For long cable applciations, clamp diodes are almost required.

Austin

NotTooSmart wrote:
> Quick question... I've been using a Virtex-E fpga on a development
> board that provides jumpers to connect or disconnect Vcco for each I/O
> bank.  Some of my inputs are LVDS, and one of my signal sources has
> LVDS output that swings up as high as ~2.4V at the highest.  On one of
> my last test runs I accidentally left the Vcco jumper uninstalled for
> the IO bank that this device was connected to, meaning that Vcco for
> that bank was left floating.  To make a long story short, much current
> was drawn, and now (probing the FPGA pins directly) Vcco for that I/O
> bank appears to be shorted to ground inside the chip package(~1.5 ohms
> from Vcco pin to Gnd pin).
> 
> To anyone out there who's familiar with Xilinx hardware, would the
> mistake I made above (no Vcco for a ~2.4V input LVDS signal) have
> caused the IO bank to become damaged?  I'd just like some confirmation
> that this is the cause, before I power up the FPGA board and fry
> another IO bank...

Article: 69180
Subject: Behaviour of Xilinx FPGA pins during Slave Serial Download.
From: abeaujean@gillam-fei.be (A Beaujean)
Date: 29 Apr 2004 08:16:53 -0700
Links: << >>  << T >>  << A >>
I am trying to understand some very specifics of the behaviour of the
I/O pins of the Xilinx SpartanII family, that I went through in a not
quite clean prototype setup.

The situation was as follows:

I/O pins defined as future outputs,

I/O pins connected to a DSP control bus and an old FLUKE (Philips)
state analyzer, probably presenting an impedance of a few tens of
Kohms to ground per input,

Configuration mode = all '1' (Serial Slave Mode, no pre-configuration
pullups), 6 FPGA's in a row.

By mistake, no pullups or pulldowns on the pins. Thus, a situation
where, before programming, voltage on the pins is floating and
probably dictated by leakage currents and the equivalent impedance of
the state analyzer,

Before programming : all I/O's are seen as "HIGH" (Vcc) by the State
Analyzer (confirmed by scope reading),

BUT : when programming in Slave Serial Mode, somewhere during
programming, ALL I/O pins are pulled to "LOW", then return back to
"HIGH" at the end of the programming.

A reasonable explanation is that the DSP input leakage current, high
enough to keep the pins at Vcc when not in programming mode, is not
large enough during the programming and is "eaten" by some current
pumped to ground by the FPGA.

Adding a 10K pullup to 3V3 is enough to always keep the voltage on the
pins at "HIGH" during the programming.

Has anyone got an explanation for this behaviour ?

A. Beaujean

Article: 69181
Subject: Re: Post-Place & Route Simulation with ISE
From: Brian Philofsky <brian.philofsky@no_xilinx_spam.com>
Date: Thu, 29 Apr 2004 09:29:21 -0600
Links: << >>  << T >>  << A >>


arkaitz wrote:
> Hi all,
> 
> I am trying to make a post-place & route simulation of a very simple
> design.
> 
> In the "top.vhd "file there is declared an entity that contains some
> generics and the architecture.
> 
> entity top is
>   generic(
>     gen1 : real := 3.0;
>     gen2 : std_logic;
>     gen3 : std_logic
>   );
>   port(
>     clk : in std_logic;
>     rst : in std_logic;
>     input : in std_logic;
>     output : out std_logic
>   );
> end top;
> 
> When I ask ISE to simulate a PPR simulation with modelsim it generates
> a file called "top_timesim.vhd". Then generates a "tb_top.tdo" where
> there are some modelsim commands that compile the "top.vhd" and
> "tb_top.vhd" files.
> 
> Modelsim generates a "not default binding for component top" error. I
> have seen the top_timesim.vhd file and the generics of my top are
> deleted!
> 
> Why can be this? Is there any way to avoid this?


Unfortunately, there is no way for the software to retain the original 
generics you defined in the code (at least not yet) and thus when netgen 
(the Xilinx ISE netlister) creates the entity for your top module, it 
can not recreate those generics in the declaration.  Fortunately though, 
there is an easy way around this without having to edit the file or the 
test bench.  If you run netgen with the -a switch (In the ISE GUI it is 
the advanced simulation model property, Generate Architecture Only) the 
netlist produced will not contain an entity declaration and only contain 
the architecture.  What this allows you to do is retain the original 
entity declaration with the defined generics from your RTL code and bind 
the structural timing architecture to that.  If you have already run a 
functional simulation using that RTL code for that project, then all you 
would need to do is flip this switch and that should fix the problem for 
you.  As long as all of your ports are defined as 
std_logic/std_logic_vector, you should not have any problems with this. 
  This is also an easy way to get around any top-level ports declared as 
a buffer rather than an out although I still highly suggest not to 
declare any port as a buffer in VHDL.

Good luck,

--  Brian


> 
> Thanks in advance,
> 
> Arkaitz.
> 
> Note: I have rewriten the generics in "top_timesim.vhd" file and now
> it works well.


Article: 69182
Subject: Re: good starter kit
From: "cpex" <cpex@NOSPAMmyrealbox.com>
Date: Thu, 29 Apr 2004 15:49:43 GMT
Links: << >>  << T >>  << A >>

It seems digilentinc actually makes the kit I refrenced earlier.

Let me rephrase my question are there any starter kits to be had in the
$20-$30 range or is 49.99 the min i am going to find

Josiah

"Yves Deweerdt" <yves@news.be> wrote in message
news:c6qps4$r7t$1@reader13.wxs.nl...
> cpex wrote:
> > Hello,
> >
> > I am a computer engineering student and I am looking to do a project
which
> > will require a FPGA or CPLD. I will need something with > 70 general IO
> > pins. I am looking for a development board that will give me an
expansion
> > port to plug it into my project. I want something less than $100
perferably
> > less than $50. I am currently considering the CPLD design kit from
XILINX
> > which has  XC2C256-7TQ144 CoolRunner-II CPLD and XC9572XL-10VQ44 CPLD on
it
> > as it seems to be a good starter package. Are there other quick dirty
> > solutions that will allow me to easily program a device and integrate it
> > with my project? I do not need something that has tons of LEDS and
> > pushbuttons and tons of extra proto space.
> >
> > Thank You
> > Josiah Vivona
> >
> >
>
> Have a look at http://www.digilentinc.com
> They have some nice FPGA and CPLD boards. Most of the boards contain
> only the necessary with nearly every IO of the FPGA or CPLD going to a
> connector. If needed you can add some of their IO-boards which contain
> the leds, push buttons, lcds, ...
>
> Kind regards,
>
> Yves
>



Article: 69183
Subject: Re: Post-Place & Route Simulation with ISE
From: "paris" <273986malaka@email.it>
Date: Thu, 29 Apr 2004 18:30:04 +0200
Links: << >>  << T >>  << A >>

"arkaitz" <arkagaz@yahoo.com> escribió en el mensaje
news:c1408b8c.0404290218.5106d484@posting.google.com...
> Hi all,
>
> I am trying to make a post-place & route simulation of a very simple
> design.
>
> In the "top.vhd "file there is declared an entity that contains some
> generics and the architecture.
>
> entity top is
>   generic(
>     gen1 : real := 3.0;
>     gen2 : std_logic;
>     gen3 : std_logic
>   );
>   port(
>     clk : in std_logic;
>     rst : in std_logic;
>     input : in std_logic;
>     output : out std_logic
>   );
> end top;
>
> When I ask ISE to simulate a PPR simulation with modelsim it generates
> a file called "top_timesim.vhd". Then generates a "tb_top.tdo" where
> there are some modelsim commands that compile the "top.vhd" and
> "tb_top.vhd" files.
>
> Modelsim generates a "not default binding for component top" error. I
> have seen the top_timesim.vhd file and the generics of my top are
> deleted!
>
> Why can be this? Is there any way to avoid this?
>
> Thanks in advance,
>
> Arkaitz.
>
> Note: I have rewriten the generics in "top_timesim.vhd" file and now
> it works well.

i dont think you should use "generics" in a "top" entity that you want to
synthesize (correct me if im wrong). I've heard that some tools accept them
and use the default value if specified.
what i do is to "wrap" the generic component, and instantiate it into a
"wrapper" that doesnt contain any generics and where all generics have been
fixed to constants defined in some package. It also eases simulation before
and after synthesis, as you only have to switch wrappers (or architectures)
in the testbench.



Article: 69184
Subject: Re: VHDL / Verilog circuits work in 1-V still correct?
From: bokiteam@ms21.hinet.net (boki)
Date: 29 Apr 2004 10:03:44 -0700
Links: << >>  << T >>  << A >>
johnjakson@yahoo.com (john jakson) wrote in message news:<adb3971c.0404290626.3d334519@posting.google.com>...
> "Boki" <bokiteam@ms21.hinet.net> wrote in message news:<c6ov4e$pk1@netnews.hinet.net>...
> > Dear All,
> > 
> > VHDL / Verilog circuits  work in 1-V still correct?
> > 
> > I want to combine 1-V analog circuits and VHDL circuits..
> > 
> > Thanks
> > 
> > Boki.
> 
> HDL will work with digital at any voltage its designed for.
> 
> For mixed signal with VHDL I think there is AVHDL for analog VHDL
> mixed signal. Its been a while since I looked at that. Somebody in the
> VHDL group might know.
> 
> regards
> 
> johnjakson_usa_com


Thank you for information.

Best regards,

Boki.

Article: 69185
Subject: Re: basic question, virtex 2 pro
From: "Symon" <symon_brewer@hotmail.com>
Date: Thu, 29 Apr 2004 10:14:23 -0700
Links: << >>  << T >>  << A >>
IIRC, ModelSIM XE doesn't support the SWIFT models needed to simulate the
PowerPCs. Perhaps IBM will give you the VHDL? Perhaps not! ;-)
Cheers, Syms.
"Vinod" <vananth@gmail.com> wrote in message
news:6cd79f5.0404290531.ffccd75@posting.google.com...
> Hi,
> I am using EDK 3.2 with ISE 5.1.3 on a WinXp machine, and targeting
> the Virtex 2 pro board. Does Modelsim XE allow me to simulate code
> running on the PowerPC along with other external blocks written in
> VHDL as one whole unit?
> I do not have a board yet, and was also wondering  whether such
> simulations would be prohibitively long.
>
> Links to informative websites would be appreciated
>
> Thanks
> Vinod



Article: 69186
Subject: package choice, temperature and obsolesence issues with a xilinx fpga
From: "jean-francois hasson" <jfhasson@club-internet.fr>
Date: Thu, 29 Apr 2004 19:48:43 +0200
Links: << >>  << T >>  << A >>
Hi,

I am involved in a project dealing with fpgas and the pcb may only be used
on one side for component placement (actives an passives). Therefore I am
considering using a QFP package instead of a BGA package for decoupling
purposes mainly. My application runs at 40 MHz and the IO need is small
(about 100 IOs). I am considering either a 208 QFP or a 256 FBGA for a
Xilinx part like spartan IIE. The other issue is about obsolescence. I
noticed the QFP package is still available for Spartan III. Is it a good
indicator for the life of this package for Spartan IIE or is this idea
irrelevant ? Finally, the temperature of the environment will border 85°C so
I believe the regular industrial grade parts we use may be out of our specs.
What is the rule for Xilinx parts like spartan IIE or else ? Since the
junction temperature is 100°C what happens if the part is in an ambient
temperature above 85°C ? Will it keep working or will performance be
degraded ? Does it depend on the process of the FPGA for a same grade ? Is
there a derating one can apply if the FPGA is out of specs but not about to
be destoyed because junction temperature was not reached ?
Any idea is welcome !!!

JF



Article: 69187
Subject: Re: good starter kit
From: Jim Lewis <Jim@SynthWorks.com>
Date: Thu, 29 Apr 2004 11:01:53 -0700
Links: << >>  << T >>  << A >>
First $50 is a quite good price on a board.
It probably represents an educational discount
on the parts.

Before you go buy the $50 board, I would do an
accessment to see if the part meets your needs or
do you need to spend a little more ($75/$85), and
get a part with more capabilities.

Since you are a student, perhaps it might help to put
it in perspective of other things you spend money on.
How many waters do you have to drink instead of beer
or soda to be able to afford the more expensive board
or how many bags of chips does the cost of the board
represent.  It is not as an absurd analysis as you think -
chips and soda are much more expensive than you think.


Cheers,
Jim

P.S.
When I was in college, I had an active hobby in kayaking.
I always thought of expenditures in terms of how many
kayak paddles it would buy.   Did not eat many bags of
chips.  I can still count the number of sodas I drink in
a year on one hand and the ones I drink are usually free.
I can alternately recommend that you pop your own popcorn
instead of chip.

cpex wrote:
> It seems digilentinc actually makes the kit I refrenced earlier.
> 
> Let me rephrase my question are there any starter kits to be had in the
> $20-$30 range or is 49.99 the min i am going to find
> 
> Josiah
> 
> "Yves Deweerdt" <yves@news.be> wrote in message
> news:c6qps4$r7t$1@reader13.wxs.nl...
> 
>>cpex wrote:
>>
>>>Hello,
>>>
>>>I am a computer engineering student and I am looking to do a project
> 
> which
> 
>>>will require a FPGA or CPLD. I will need something with > 70 general IO
>>>pins. I am looking for a development board that will give me an
> 
> expansion
> 
>>>port to plug it into my project. I want something less than $100
> 
> perferably
> 
>>>less than $50. I am currently considering the CPLD design kit from
> 
> XILINX
> 
>>>which has  XC2C256-7TQ144 CoolRunner-II CPLD and XC9572XL-10VQ44 CPLD on
> 
> it
> 
>>>as it seems to be a good starter package. Are there other quick dirty
>>>solutions that will allow me to easily program a device and integrate it
>>>with my project? I do not need something that has tons of LEDS and
>>>pushbuttons and tons of extra proto space.
>>>
>>>Thank You
>>>Josiah Vivona
>>>
>>>
>>
>>Have a look at http://www.digilentinc.com
>>They have some nice FPGA and CPLD boards. Most of the boards contain
>>only the necessary with nearly every IO of the FPGA or CPLD going to a
>>connector. If needed you can add some of their IO-boards which contain
>>the leds, push buttons, lcds, ...
>>
>>Kind regards,
>>
>>Yves
>>
> 
> 
> 


-- 
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Jim Lewis
Director of Training             mailto:Jim@SynthWorks.com
SynthWorks Design Inc.           http://www.SynthWorks.com
1-503-590-4787

Expert VHDL Training for Hardware Design and Verification
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

Article: 69188
Subject: Re: package choice, temperature and obsolesence issues with a xilinx
From: Austin Lesea <austin@xilinx.com>
Date: Thu, 29 Apr 2004 11:59:24 -0700
Links: << >>  << T >>  << A >>
JF,

Wow.  Not enough questions?

See below, I will try to do my best.

Austin

jean-francois hasson wrote:
> Hi,
> 
> I am involved in a project dealing with fpgas and the pcb may only be used
> on one side for component placement (actives an passives). Therefore I am
> considering using a QFP package instead of a BGA package for decoupling
> purposes mainly.

A QFP package has significantly more inductance, and is harder to get a 
good decoupling design done right.  A fgXXX package (fg256) is a much 
lower inductance value package, and with all components on top, is not 
impossible at all to do a power distribution decoupling scheme well. 
You may need two more layers in the pcb stackup to do it right (in 
either case).  If you are having problems with your layout, you may send 
it into the Hotline, and someone will look at it for you, and advise 
you.  My Lab gets invovled in many pcb cases.  I would prefer you learn 
how to do it right and succeed the first time!

  My application runs at 40 MHz and the IO need is small
> (about 100 IOs). I am considering either a 208 QFP or a 256 FBGA for a
> Xilinx part like spartan IIE. The other issue is about obsolescence. I
> noticed the QFP package is still available for Spartan III. Is it a good
> indicator for the life of this package for Spartan IIE or is this idea
> irrelevant ?

The QFP will be around for quite awhile.  It is too popular, and too 
high a seller.  For extreme SI applications, it is not used, however.

  Finally, the temperature of the environment will border 85°C so
> I believe the regular industrial grade parts we use may be out of our specs.

85C to 100C is 15C junction to ambient.  In a QFP package, that might 
limit you to about 1 watt of power dissipation in the part!  This will 
be tough.

> What is the rule for Xilinx parts like spartan IIE or else ?

The 'rule' is that above 85C for commercial, and above 100C for 
Industrial, the device is not expected to meet specifications (but it 
will still function up to 125C).

  Since the
> junction temperature is 100°C what happens if the part is in an ambient
> temperature above 85°C ?

Specifications may not be met.

  Will it keep working or will performance be
> degraded ?

Functionality will continue to ~125C junction, and beyond that we make 
no statements or assurances.

  Does it depend on the process of the FPGA for a same grade ?

No.

  Is
> there a derating one can apply if the FPGA is out of specs but not about to
> be destoyed because junction temperature was not reached ?

In the old days, when everything was simple, simple derating formulas 
could be applied to the CLB, routing, etc.  With DCMs, BRAMs, etc. this 
no longer works at all.  Perhaps you need the automotive version of the 
parts which is characterized (has a speeds file) for 125C abs max? 
Contact your disti or Xilinx FAE.

> Any idea is welcome !!!

Contact your disti or Xilinx FAE.  They can be more helpful than I.

> 
> JF
> 
> 

Article: 69189
Subject: Re: good starter kit
From: "Symon" <symon_brewer@hotmail.com>
Date: Thu, 29 Apr 2004 12:02:08 -0700
Links: << >>  << T >>  << A >>

"Jim Lewis" <Jim@SynthWorks.com> wrote in message
news:1092goikgfsf93@corp.supernews.com...
<snip>
> Since you are a student, perhaps it might help to put
> it in perspective of other things you spend money on.
> How many waters do you have to drink instead of beer
> or soda to be able to afford the more expensive board
<snip>
> P.S.
> When I was in college, I had an active hobby in kayaking.
> I always thought of expenditures in terms of how many
> kayak paddles it would buy.   Did not eat many bags of
> chips.  I can still count the number of sodas I drink in
> a year on one hand and the ones I drink are usually free.


Jim,
I notice you didn't economise on beer! Good choice!
cheers, Syms.







Article: 69190
Subject: best machine setup for ISE ??
From: Matthew E Rosenthal <mer2@andrew.cmu.edu>
Date: Thu, 29 Apr 2004 15:07:03 -0400 (EDT)
Links: << >>  << T >>  << A >>
Hi all,
My boss has authorized me to purchase a new machine(dell) to be used for
synthesis, translate, map and P&R.  I'm looking for some suggestions on
which processor   P3 vs P4 vs AMD
how much memory   1gig or more
bus speed
which brand
server vs end user machine

Thanks for your help.  I need to order the machine ASAP.

Thanks

Matt

Article: 69191
Subject: Re: best machine setup for ISE ??
From: Matthew E Rosenthal <mer2@andrew.cmu.edu>
Date: Thu, 29 Apr 2004 15:32:29 -0400 (EDT)
Links: << >>  << T >>  << A >>
I will be using a V2pro  XC2VP50

On Thu, 29 Apr 2004, Matthew E Rosenthal wrote:

> Hi all,
> My boss has authorized me to purchase a new machine(dell) to be used for
> synthesis, translate, map and P&R.  I'm looking for some suggestions on
> which processor   P3 vs P4 vs AMD
> how much memory   1gig or more
> bus speed
> which brand
> server vs end user machine
>
> Thanks for your help.  I need to order the machine ASAP.
>
> Thanks
>
> Matt
>

Article: 69192
Subject: Re: Xilinx edk/modelsim/ VHDL question
From: wpiman@aol.com (MS)
Date: 29 Apr 2004 12:34:25 -0700
Links: << >>  << T >>  << A >>
Goran,
Yeah- I tried that too.  But I did find my problem.

I was performing the sims on the testbench file expecting it to read
the configuration.  I ran the one that worked and I realized I was
running "vsim system_conf"- and it occured to me- you have to execute
the configuration file- not the testbench.

Oh well- there is a day I will never get back.  I made a
system_stub_conf that pointed to system_conf and a testbench_conf that
pointed to that.  Modelsim did seem to like too much heirarchy in its
statements- although I have seen example where the heirarchy goes way
down.

Thanks for your help,
Mike

Goran Bilski <goran@xilinx.com> wrote in message news:<c6qfho$24m2@cliff.xsj.xilinx.com>...
> Hi,
> 
> If you compile the system_init file into "work", you can use the 
> configuration specified in the file directly.
> A configuration can use another configuration.
> 
> ex.
> I have created a testbench "system_tb" for the system generated from XPS:
> 
> In my configuration statement I have this:
> 
> configuration tb_cfg of system_tb is
>   for VHDL_TEST
>     for all : system use configuration work.system_conf;
>     end for;   
>   end for;
> end tb_cfg;
> 
> You should be able to do this for your system:
> 
> configuration conf_name of testbenchname is
>    for testbench_arch
>       for all : system_stub use configuration work.system_conf
>       end for;
>    end for;
> end conf_name;
> 
> 
> MS wrote:
> 
> >I am using the Xilinx EDK to perform simulations of the embedded
> >PowerPC on a V2Pro.  I have had success using simply the EDK with
> >Modelsim but when I change the flow to the ISE as an embedded project-
> >I am having trouble getting the boot ROM to be read in as a
> >configuration in Modelsim.
> >
> >The trouble has to do with assigning the blockRAM configurations to
> >the blockmemories once I wrapped the system_stub in a testbench.  I
> >was able to get this all to work with ISE when I did not embed the
> >processor, but once it is embedded- the blockRAM moves a level down in
> >the heirarchy and then the configuration statement does not compile. 
> >My understanding in VHDL is it is like this.... (I am improvising from
> >memory here)
> >
> >configuration name of testbenchname is
> >   for behavior -- arch name
> >      for uut: system_stub -- unit under test
> >         for STRUCUTRE -- arch of system_stub
> >            for system_i: system -- declared system
> >               for structure
> >                  for all: blockram_types: use work.blockram_conf;
> >bunch of end fors and end of configuration
> >
> >I stole this from the system_init file and added my new levels of
> >heirarchy.
> >
> >It complains to me that system is not a valid component- but in the
> >work directory I see it there clear as day.  Is there a limit to how
> >far down in the heirarchy you can assign configurations?  I do not
> >believe so.  I tried making multiple level configurations and have had
> >no luck either.
> >
> >If I blow this away- the PowerPC fetches from FFFFFFC and gets zero
> >back.  When done properly, it fetches the first opcode and I am off an
> >running.
> >
> >Thanks,
> >MS
> >  
> >

Article: 69193
Subject: Re: best machine setup for ISE ??
From: nweaver@ribbit.CS.Berkeley.EDU (Nicholas C. Weaver)
Date: Thu, 29 Apr 2004 19:54:49 +0000 (UTC)
Links: << >>  << T >>  << A >>
In article <Pine.LNX.4.58-035.0404291458250.11736@unix48.andrew.cmu.edu>,
Matthew E Rosenthal  <mer2@andrew.cmu.edu> wrote:
>Hi all,
>My boss has authorized me to purchase a new machine(dell) to be used for
>synthesis, translate, map and P&R.  I'm looking for some suggestions on
>which processor   P3 vs P4 vs AMD

P4, hyperthreaded turned on.  Allows you to do some low-intensity work
while the tools are grinding away, without paying the extra cost for a
multiprocessor system.

>how much memory   1gig or more.

As Much As Possible (tm).  ECC definatly preferred (Soft Errors Happen
(tm)).

>bus speed

HT P4s are all the 400 MHz DDR bus, which is what you want.

>which brand
>server vs end user machine

High end workstation.

Disk is less important, you want reasonably fast but reliability
counts for more.
-- 
Nicholas C. Weaver                                 nweaver@cs.berkeley.edu

Article: 69194
Subject: Re: Altera ByteBlaster II schematic
From: "Pierre de Vos" <pierre.devos@webmail.co.za>
Date: Thu, 29 Apr 2004 22:00:58 +0200
Links: << >>  << T >>  << A >>
Click on the disk icon in the second column


"ted" <edaudio2000@yahoo.co.uk> wrote in message
news:c54bf83f.0404280333.3f03ac24@posting.google.com...
> Leon
>
> I have tried the website, I even registered, but somehow, I cannot
> click on the "byteblaster.pdf" link to download. How do you get to
> download the file??
>
> Thanks
> Ted
>
>
> > I just came across this schematic for the ByteBlaster II:
> >
> > http://www.mcu.cz/atm/index.php?direction=0&order=&directory=EDA
> >
> > Someone has probably worked out the circuit by tracing the connections
on a
> > unit as Altera hasn't published it, AFAIK.
> >
> > Leon



Article: 69195
Subject: Re: Design development costs for FPGA on PCI board (sorry if slightly off-topic)
From: Bassman59a@yahoo.com (Andy Peters)
Date: 29 Apr 2004 13:15:50 -0700
Links: << >>  << T >>  << A >>
fpgadev@yahoo.com (AndyAtHome) wrote in message news:<29ab33c4.0404290413.2b5a2be5@posting.google.com>...
> Dear All,
> 
> After investigating the price of FPGA PCI boards for a project, it
> seems the cost of a board is typically 3 to 4 times the component cost
> for mid-range FPGA's like Xilinx 2V3000.
> 
> Of course board manufacturers have overheads and need to make a profit
> - my point isn't to complain about their Gross Profit margin.
> 
> Rather, at what point is it better to meet the initial capital
> expenditure of having your own board designed, thereafter only having
> to meet board manufacturing costs?
> 
> For example, just how much design time and cost does it take to create
> a board with -
> 
> Quicklogic or PLXTech PCI Bridge
> Single FPGA
> A PROM 
> Single bank of DDR memory.
> 
> Surely this type of requirement is common and would not require months
> of development time or am I being naive? :)
> 
> Any comments appreciated...

Well, it really depends on how many boards you need, and whether the
hardware design on the off-the-shelf board meets your requirements.

Regarding the former, it's simply the classic make/buy decision.  If
you need a handful boards, buying them makes sense because you can
concentrate your development efforts into the FPGA and you know (or at
least hope!) that the board design itself is stable.  If you need a
lot of boards, then it may very well make sense to spin your own.

Rolling your own board requires a different skill set from designing
an FPGA, and if you don't have those skills (and the software tools),
then you have to hire them and that cost may bust your budget.  You
also have to learn about (or have in place) the entire manufacturing
process, including purchasing, board fab, stuffing, everything, which
will distract you from your FPGA design :)

Of course, if no off-the-shelf board meets your needs, then your
choice is clear.

Besides, a product that costs only three times the bill-of-materials
is kinda cheap :)

-a

Article: 69196
Subject: Re: Data transfer for real time analysis
From: "Pini" <pinit@ee.bgu.ac.il>
Date: Thu, 29 Apr 2004 22:52:00 +0200
Links: << >>  << T >>  << A >>

I've found some boards with 32 LVDS channels. The problem is that all the 32
channels have a common clock. I would like to connect a different sub unit
to each pin of the 32 channels. How can I distribute the clock to all sub
units?


þþ"Hal Murray" <hmurray@suespammers.org> ëúá áäåãòä
news:9tSdnVZp843-mRLdRVn-sA@megapath.net...
> >We are planning a data acquisition system with 4-12 sub units. Each sub
unit
> >produce data of 25MBit/sec. The distance between the boxes and computer
is
> >about 2 meters.
> >I'm willing to transfer and analyze the data in real-time. I'm willing to
> >close a control loop in 1-5msec (I. wish 100usec ......).
> >What is the recommended serial protocol?
> >The sub units will be in-house made the host computer for data analysis
and
> >storage will be a PC (probably with raid HD).
> >What communication HW/protocol should I use in my sub units?
> >What communication HW/protocol should I use in the PC?
>
> One straw man would be 100 megabit ethernet.  That lets you use
> off-the-shelf gear on the PC end.  For 12 sub units, you probably
> need a gigabit ethernet card on the PC and a 12 port switch
> with a gigabit uplink.  (It also gives you lots of flexibility
> on the length which might be handy some day.)
>
> If you want to build your own PCI card, look at LVDS.  25 megabits
> should be easy at 2 meters.  Manchester encoding costs a factor
> of 2 in link bandwidth (which you have lots of) but it's real
> simple to decode.
>
>
> -- 
> The suespammers.org mail server is located in California.  So are all my
> other mailboxes.  Please do not send unsolicited bulk e-mail or
unsolicited
> commercial e-mail to my suespammers.org address or any of my other
addresses.
> These are my opinions, not necessarily my employer's.  I hate spam.
>



Article: 69197
Subject: Re: best machine setup for ISE ??
From: "Tim" <tim@rockylogic.com.nooospam.com>
Date: Thu, 29 Apr 2004 22:16:37 +0100
Links: << >>  << T >>  << A >>
Nicholas C. Weaver wrote:
> HT P4s are all the 400 MHz DDR bus, which is what you want.

Is that 200MHz?



Article: 69198
Subject: Re: Altera ByteBlaster II schematic
From: Rene Tschaggelar <none@none.net>
Date: Thu, 29 Apr 2004 23:58:09 +0200
Links: << >>  << T >>  << A >>
Leon Heller wrote:

> "Leon Heller" <leon_heller@hotmail.com> wrote in message
> news:408cffe8$0$19420$cc9e4d1f@news-text.dial.pipex.com...
> 
>>"Rene Tschaggelar" <none@none.net> wrote in message
>>news:408cd362$0$703$5402220f@news.sunrise.ch...
>>
>>>Leon Heller wrote:
>>>
>>>
>>>> I just came across this schematic for the ByteBlaster II:
>>>>
>>>>http://www.mcu.cz/atm/index.php?direction=0&order=&directory=EDA
>>>>
>>>>Someone has probably worked out the circuit by tracing the connections
>>
>>on a
>>
>>>>unit as Altera hasn't published it, AFAIK.
>>>
>>>And does it work ?
>>
>>I mean to try it, when I get my Cyclone board finished.
> 
> 
> A problem with this is that the BF550 and BF570 transistors used are
> unavailable - 20 weeks delivery according to Arrow. 8-(
> 
> I'll see if I can find equivalent devices.

Since this is not a precision analog device, I guess any bipolar does 
the job. Basically two inverters to boost the voltage plus an emitter
follower to lower the output impedance.

Rene
-- 
Ing.Buero R.Tschaggelar - http://www.ibrtses.com
& commercial newsgroups - http://www.talkto.net

Article: 69199
Subject: turning off clock for parts of design
From: tomtorfs@village.uunet.be (Tom Torfs)
Date: 29 Apr 2004 15:26:23 -0700
Links: << >>  << T >>  << A >>
Hello everyone,

I was wondering what the standard design practice is for turning off
the clock for some parts of an FPGA design while keeping other parts
running. I would assume simply gating the clock (followed by a global
clock buffer) introduces unacceptable amounts of clock skew. Perhaps
this method could be combined with use of a delay-locked loop as
present in the Xilinx FPGAs to keep the gated clock synchronous to the
rest of the system? Of course the gating circuit would also have to
take care of switching the gated clock glitchless, for which perhaps a
circuit could be used such as the one on
http://www.eedesign.com/features/exclusive/OEG20030626S0035 (referred
to by FAQ answer "Switching clocks without glitches"). Or am I
overlooking a simpler solution?

A similar problem exists when instead of turning the clock off it is
switched to a divided version (produced by counter) of itself, while
other parts keep running at the full speed or a different division
factor.

I would prefer to keep the whole system synchronous (e.g. all switched
or divided versions of the same clock with minimal skew) because else
the different parts would need anti-metastability sychronizers such as
2 or 3 flipflops in series on all their communication lines...

greetings,
Tom



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