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 85875

Article: 85875
Subject: Re: Xilinx Spartan 3 DCI Power Consumption
From: "Brian Davis" <brimdavis@aol.com>
Date: 17 Jun 2005 05:14:05 -0700
Links: << >>  << T >>  << A >>
James Morrison wrote:
>
> Does anyone have any good numbers for the power consumption
> used by the DCI circuitry on the I/O's?  The data sheet says,
> and I quote, "more power".  I kid you not--look it up.
>

 I've posted a few links below to old posts on this subject for
both V2 and S3.

 It's particularly bad for the parallel input terminators, which
have a +/-20% accuracy for the 2R pullup/pulldowns.

 In V2, I measured static DCI power at about 200 mW per bank
and 100 mW per LVDS_25_DCI pair (2.5V, 50 ohm VRP/N), much
greater than the 25 mW and 62.5 mW numbers from Xilinx.

 Back in early 2003, I opened webcases and filed CR's for the
various tool and documentation problems relating to the greatly
under-reported static DCI power in V2.

 After a three month webcase, and countless phone calls and emails,
the best Xilinx could do was give me an estimate, because they had
never properly characterized DCI power after implementing the
FreezeDCI bitgen patch.

 After six months, the documentation and tools still were wrong, so
I posted a summary of the various V2 DCI problems here in Fall 2003.

 If you look today, 2.5 years post-Webcase:

  Answer Record #11661 :
     ignores +/- 20% parallel terminator error, skips LVDS_25_DCI

  Answer Record #15633 :
     still says 62.5 mW per LVDS_25_DCI pair

  Answer Record #11794 :
     ignores stall-internal-resistor-at-R/2 case for VRP/VRN overhead

  WebPower now includes LVDS_25_DCI for V2 and S3, but still
under-reports power.

old DCI power posts:

http://groups-beta.google.com/group/comp.arch.fpga/msg/5daa6eea44f16213?hl=en

http://groups-beta.google.com/group/comp.arch.fpga/msg/4a7fa8984b3395db?hl=en

http://groups-beta.google.com/group/comp.arch.fpga/msg/ea29413b0b200f5c?hl=en

Brian


Article: 85876
Subject: Re: comp.arch.fpga.<mfr>
From: "Gabor" <gabor@alacron.com>
Date: 17 Jun 2005 05:18:54 -0700
Links: << >>  << T >>  << A >>


Unbeliever wrote:
> "Jon Beniston" <jon@beniston.com> wrote in message
> news:1119008115.592497.325850@g14g2000cwa.googlegroups.com...
> > comp.arch.fpga.cpu might be a good idea. Somewhere for all the NIOS &
> > MicroBlaze queries.
> >
> > Cheers,
> > Jon
> >
>
> Which newsgroup would I post my question relating to my implementation of a
> 6805 core in an fpga?  The .cpu group or the parent group or both.  At an
> average of 10 topics per day, I'd prefer to keep it in one place, like Mike.
>
> Cheers,
> Alf

I agree.  ~50 posts a day is not ~50 threads a day.  I use the google
groups site to access the newsgroup and almost never need to click
the "older topics" to find a thread that's still active.  I enjoy
seeing
threads related to vendors of fpga's I'm not currently using.  I also
enjoy seeing topics not related to my current work.  I would miss most
of this if I had to browse through multiple groups to find them.

Just my 2 cents.
Gabor


Article: 85877
Subject: Xlinix configuration: DONE pin too early?
From: "gw" <guenter.wolpert@orsys.de>
Date: 17 Jun 2005 05:30:42 -0700
Links: << >>  << T >>  << A >>
Hello,

We use different Xilinx FPGAs (Spartan3, Virtex2, etc.).
When downloading the code, everything works fine except the
fact that the DONE pine is going high before the last data is
written (tested on XC2V250fg456 or XC3S50vc100: DONE goes high
after writing the N-13th byte).
If we check the ODNE pin after loading ALL bytes, everyhing is
still fine, but we would like to check the INIT pin during the
configuration process in order to determine CRC errors.
Has anyone experince with this, thus is it possible to determine
when the DONE pins must go high and is it possible to define
a point at the end of the configuration process where the INIT
pin can be checked for CRC errors?

thanks for any comments....
Guenter


Article: 85878
Subject: Re: Idea exploration - Image stabilization by means of software.
From: CBFalconer <cbfalconer@yahoo.com>
Date: Fri, 17 Jun 2005 12:34:29 GMT
Links: << >>  << T >>  << A >>
Anton Erasmus wrote:
> <jon_harrisTIGER@hotmail.com> wrote:
> 
>> Are you taking pictures of stationary things from a stationary
>> camera?  That's what the astronomy case involves.  But since you
>> are concerned with a moving camera, I assumed you were talking
>> about pictures taken with a hand-held camera. In that case, the
>> reason the shake creates blur is that the exposure is too long. 
>> And the reason for the long exposure is lack of light (due to
>> lens limitations, ambient light conditions, or both).  If you
>> were to shorted the exposure, you would end up with a picture
>> that is too dark.
> 
> [Stuff snipped]
> 
> I have read a bit about the astronomical image processing mention
> in one of the other posts. One thing that came up quite often, is
> that CCD sensors are linear. i.e. double the exposure time gives
> double the energy recieved. (They do not mention CMOS, but it is
> probably the same.)
>
> I agree that blurring occurs when taking a single photo when the
> camaera is moved and the exposure is to long. If one can say take
> in stead of 1x 2sec exposure, take 20x 0.1 sec exposures, and
> stack them into a single picture in software, one should end up
> close to the exposure level of the 2sec picture, but without the
> blur. In the astronomical case, the "short" exposure pictures are
> in minutes for a total exposure time of hours. Of course the
> subject must not move during the "short" exposure time.
>
> If one scaled this down so that the "short" exposure time is in
> milli-seconds or even micro-seconds depending on what the sensors
> can do, then one can do the same thing just at a much faster
> scale. What is the minimum exposure time for a current CMOS sensor
> before one just see the inherant sensor noise ?

I think, at any particular time, you have an accumulated picture
and a picture increment to combine.  Around any particular point
you have a matrix of adjacent points, described by the indices -1,
0, +1 in each direction.  To do the combination, and arrive at a
new accumulated picture, you 'just' have to form the 9 value matrix
and combine all points accordingly.  Now you can get a new picture
increment, and recompute the matrix.  You combine the old into the
new, so you never have to worry about more than one increment of
motion.

Forming the matrix will depend on a herd of heuristics, such as
'where is the CG of the input points', 'where is the minimum',
'where is the maximum'.

This is pure speculation on my part.  I have no knowledge of real
world algorithms here.  It's how I would go about it.  I might want
a buffer for picture increment, so that the combining can take
place in parallel with the input of the next increment.

-- 
Chuck F (cbfalconer@yahoo.com) (cbfalconer@worldnet.att.net)
   Available for consulting/temporary embedded and systems.
   <http://cbfalconer.home.att.net>  USE worldnet address!


Article: 85879
Subject: Re: Good FPGA introduction book ?
From: "Gabor" <gabor@alacron.com>
Date: 17 Jun 2005 05:41:39 -0700
Links: << >>  << T >>  << A >>


Hal Murray wrote:
> >  I'm starting to learn FPGA programming (using a Xilinx Spartan II
> >200K). I will use VHDL and already have bought VHDL books, but I think
> >I also need a general introduction to FPGA so I plan to buy a book on
> >FPGA. I found this one:
>
> Have you checked the data sheet?

For Xilinx parts like Spartan2 there are both datasheets and user
guides.  They are not on the same web page so you need to look for
the user guide.  The user guide has  lots of detailed information
that isn't always easy to grok in the datasheet.

Also on the Xilinx site are tons of app notes and a plethora
(if plethora is the word I want to use) of tech tips.

>
> It's got all the info you need.  You will probably have to read it
> at least 6 times - more will make sense each pass.
>
> You can learn a lot of details and what's behind them if you
> follow this newsgroup.

Look through some older threads on board products and you'll find
links to interesting sites like:
http://www.fpga4fun.com/

Good Luck,
Gabor


Article: 85880
Subject: Re: Xilinx MAP problem (>1 External Macro Output Pin on single net)
From: "Gabor" <gabor@alacron.com>
Date: 17 Jun 2005 05:46:30 -0700
Links: << >>  << T >>  << A >>


Ian wrote:
> Hi Everyone,
>
> I would be really for any help or advice you can offer me on the
> following.
>
> I have created a simple tri-state bus as a macro using xdl. The design
> consists of two TBUFs driving a single long line.
>
> (Diagram at http://www.comms.scitech.susx.ac.uk/~ian/files/tbuf.gif)
>
> <img src="http://www.comms.scitech.susx.ac.uk/~ian/files/tbuf.gif">
>
> I attach external macro pins to Out, Enable and In of each TBUF.
> However, when I try to include the macro in a design, the DRC in the
> map phase complains that the Out pin is being driven by two sources.
> MAP Error Message:
>
>
> ERROR:MapLib:22 - Bus M0_DATA_LEFT_O_OBUF driven by bm_instance and
> bm_instance has multiple active drivers.
>
>
> This is not correct, as the O pins are external macro outputs! Is there
> anyway to prevent this?
>
> Xilinx don't list any help for this Error.
>
> /Ian.

What device are you targetting?  Newer FPGA's don't have TBUF's
anymore.  CPLD's never did...


Article: 85881
Subject: Re: IPIF LogiCore?
From: "Ben Jones" <ben.jones@xilinx.com>
Date: Fri, 17 Jun 2005 13:46:51 +0100
Links: << >>  << T >>  << A >>
Hi Michael,

> I'm trying to build a new core that will use a IPIF PLB interface to
> connect to the PLB. Most of my logic is ready to roll and tested
> independantly, but when I came to add the IPIF interface I fell over at
> the first hurdle :(

Since you're talking about PLBs and things, I'm guessing that you must be
using the EDK somewhere along the line. The best place to look for
information about what you're trying to do is in the Embedded System Tools
Reference Manual, in the chapter entitled "Create and Import Peripheral
Wizard".

The "Wizard" in question is (fortunately) not one of those
two-clever-by-half wizards, but more of a does-the-scut-work-for-you sort of
guy. You run it once ("create") and answer some questions, and it spits out
a VHDL (or Verilog) skeleton, to which you can attach your custom code.
Then, you run it again ("import") and it picks up your design and rolls it
up into an EDK "pcore" which you can then instantiate in your EDK design.

I'm sure it's possible to achieve all the above manually without the wizard,
but I haven't seen any documentation describing that. Give us a shout if you
need any more pointers.

Cheers,

        -Ben-



Article: 85882
Subject: Re: Xlinix configuration: DONE pin too early?
From: "John Adair" <removethisthenleavejea@replacewithcompanyname.co.uk>
Date: Fri, 17 Jun 2005 14:09:49 +0100
Links: << >>  << T >>  << A >>
Have you got the "DRIVE DONE PIN HIGH" option selected for one of your 
designs in the Generate Programming File stage of ISE?

John Adair
Enterpoint Ltd. - Home of MINI-CAN. PCI and CAN Development Board.
http://www.enterpoint.co.uk


"gw" <guenter.wolpert@orsys.de> wrote in message 
news:1119011442.346145.82770@o13g2000cwo.googlegroups.com...
> Hello,
>
> We use different Xilinx FPGAs (Spartan3, Virtex2, etc.).
> When downloading the code, everything works fine except the
> fact that the DONE pine is going high before the last data is
> written (tested on XC2V250fg456 or XC3S50vc100: DONE goes high
> after writing the N-13th byte).
> If we check the ODNE pin after loading ALL bytes, everyhing is
> still fine, but we would like to check the INIT pin during the
> configuration process in order to determine CRC errors.
> Has anyone experince with this, thus is it possible to determine
> when the DONE pins must go high and is it possible to define
> a point at the end of the configuration process where the INIT
> pin can be checked for CRC errors?
>
> thanks for any comments....
> Guenter
> 



Article: 85883
Subject: Re: AbusivepPricing information in marketing publications
From: David Brown <david@westcontrol.removethisbit.com>
Date: Fri, 17 Jun 2005 15:11:58 +0200
Links: << >>  << T >>  << A >>
Antti Lukats wrote:
> ROTFL
> ROTFL
> 
> this time really
> 
> TOP 5 listed !!
> 
> "Jim Granville" <no.spam@designtools.co.nz> schrieb im Newsbeitrag
> news:42b292ea$1@clear.net.nz...
> 

<snip>

>>
>>  Looking at
>>
> 
> http://www.altera.com/corporate/news_room/releases/products/nr-ultimate-products.html
> 
>>The headline claims :
>>" Altera's HardCopy II Structured ASICs and MAX II CPLDs Named Ultimate
>>Products by eeProductCenter "
>>
>>Hmmm, not my choice for Ultimate Products, but let's read on....
>>
>>"Both AlteraŽ products were ranked in the top five in the logic and
>>programmable logic category."
>>
>>So "Ultimate"  [1.Furthest or highest in degree or order; utmost or
>>extreme; 2.Being the last or concluding element of a series]
>>has suddenly spin-morphed to actually meaning - "Hey, we made the top
> 
> FIVE!"
> 
>>Now let's see, top five in the logic and programmable logic category ??
>>
>>So that could be, as an example that fits that claim :
>>1. Atmel
>>2. Actel
>>3. Lattice
>>4. Xilinx
>>5. Altera
> 
> 
> BIG ROTFL !!!
> 
> 
>>If it was any better, surely they would have said the top three, or top
>>two ?
>>

The "Ultimate" report is at 
http://www.eeproductcenter.com/ultimate/default.html .

And (of no surprise to anyone who thought about it) the reason Altera 
said "top five" is that it includes *both* the Hardcopy and Max II.  The 
rankings are:

1. Orange tree (C to VHDL tool)
2. Lattice (downloadable synthesis software)
3. Altera (Hardcopy II)
4. Cypress (PSOC 8-bit micro with configurable blocks)
5. Altera (Max II)
6. Actel (fpga starter kit)
7. Fairchild (tiny logic chips)
8. Celoxica (tools for image and video processing)
9. Exar (PCI bus UART)
10. Philips (tiny logic chips)

Of course, such a ranking is purely subjective and most people are going 
to disagree about it - especially the word "ultimate".  But I don't 
think Altera marketing was doing anything unreasonable here.

>>And further on, we see
>>...described MAX II CPLDs as "as the industry's lowest-cost CPLD
>>available now," and "ideal for volume-driven, price-sensitive
> 
> applications."
> 
>>Reality check time - quick look at their own web site
>>
> 
> http://www.buyaltera.com/scripts/partsearch.dll/showfilter?lookup=1,30,3074
> 
>>shows the Cheapest MAX II is listed at $6.00 - lowest cost ? - hmm,
>>there are over 43 other CPLDs from Altera that clearly have
>>lower prices {and many over at Xilinx too... ) !?.
>>
>>So what can "lowest cost CPLD" actually mean ?
>>
>>-jg
>>
> 

My guess is "lowest cost CPLD" means "lowest cost per macrocell".  I 
haven't checked the prices (except to note that the smallest MaxII is 
both too big and too expensive to replace a Max3000 we are using), but 
it is standard (though misleading) practice to talk about "lower cost" 
when you mean "better value for money".

> 
> all marketing BS
> 
> MAX2 is nice non volatile version of Xilinx XC2K
> 
> nothing in common what is considered CPLD (as Complex PLD)
> 
> its really nice, but it doesnt at all compete in the low range PLD market
> 
> the low range is 1USD and sub 1USD devices available from many vendors
> 
> Antti
> 
> 

Article: 85884
Subject: Re: Xlinix configuration: DONE pin too early?
From: "gw" <guenter.wolpert@orsys.de>
Date: 17 Jun 2005 06:21:57 -0700
Links: << >>  << T >>  << A >>
At least there is an external pull-up on the DONE pin.
Anyway, even if the DONE pin is an open collector only,
it shouldn't stop pulling low before the end of the data?

The only thing with the open collector option is that DONE
may come a little bit later than expected.

One additional thing we observed is that the FPGA started to
drive its outputs one clock before the DONE appeared. Therefore
it seems that the configuration process is already over before all
data is written to the FPGA.
 
Guenter


Article: 85885
Subject: Re: question regarding "Add I/O buffers" option - SOS
From: "Vladislav Muravin" <muravinv@advantech.ca>
Date: Fri, 17 Jun 2005 09:22:50 -0400
Links: << >>  << T >>  << A >>
Dear Gabor,

Thank you very much for your response.
Actually, what you've said was the first thing i have asked before agreeing 
to accept this kind of project.
But this never was met with enthusiasm... Managers stuff...

I am going to try manual instantiation of ~300 I/O buffers... Good luck to 
me.

Vladislav

"Gabor" <gabor@alacron.com> wrote in message 
news:1118956538.511166.57090@g44g2000cwa.googlegroups.com...
>
>
> Vladislav Muravin wrote:
>> Hi ppl,
>>
>> I previously asked something similar. The thing is that I am integrating 
>> a
>> design of somebody
>> who's been using schematic design capture and his design is delivered in 
>> a
>> generated VHDL code (or NGC as well).
>>
>> Now, the delivered design has already IBUFs / OBUFs inside, whereas my
>> high-level code (the top module) does not.
>> Consequently, I cannot do AND between two dsesigns, one with and one 
>> without
>> IBUFs / OBUFs.
>> If I turn on "Add I/O Buffers" option, I get an error, as the buffers are
>> trying to be driven by another buffers inferred by the tool.
>> If I turn off "Add I/O Buffers" option, all the logic is removed form
>> design, as LOC constraints on the pins cannot be applied,
>> because no buffers are inferred.
>>
>> As far as I understand the situation I have two choices:
>> (*) I design this core by myself.
>> (*) I manually add IBUFs / OBUFs in the top module and try synthesizing 
>> with
>> "Add I/O Buffers" option turned off.
>>
>> Is there a way to enable / disable the option "Add I/O Buffers" in Xilinx
>> per module / block?
>
> No.  The Add I/O buffers only really affects the top level module
> anyway.
> If you need more control, turn it off and instantiate all your I/O
> buffers (IBUF, IBUFG, OBUF, ughh...).
>
>> Is there a way to overcome this problem?
>>
>
> Yes.
> When you want to use another design as a black box, it should be built
> without IO buffers.  Only the top level design should have IO buffers.
> Therefore you need two projects, one to build the black box (in your
> case a schematic project) with add I/O buffers turned off.  You don't
> need to translate / map / place / route this project, just "synthesize"
> (yes there's HDL under the hood of ECS schematics) to create the
> ngc file.
>
> The second is your top level project (HDL in your case) with Add I/O
> Buffers turned on.  In this project you create an empty HDL module with
> the i/o list from the schematic.  If the schematic module connects to
> pins, you'll need to add the connections from this module to the pins
> in your HDL top-level code.
>
> Then just move the ngc file from the schematic project to the synthesis
> directory (project directory) of the HDL project.  Make sure the base
> filename is the same as the module name in your HDL.  XST will not try
> to synthesize your empty HDL module, but will insert the compiled ngc
> into the top level design.
>
>> Thank you all for your time and attention
>>
>> Sincerely,
>> Vladislav
> 



Article: 85886
Subject: Re: Xlinix configuration: DONE pin too early?
From: "John Adair" <removethisthenleavejea@replacewithcompanyname.co.uk>
Date: Fri, 17 Jun 2005 14:32:52 +0100
Links: << >>  << T >>  << A >>
The point is that DONE is not always an open drain drive. Have a look at 
this Xilinx answer 
http://www.xilinx.com/xlnx/xil_ans_display.jsp?iLanguageID=1&iCountryID=1&getPagePath=5865 
and you might get a clearer understanding.

John Adair
Enterpoint Ltd. - Home of Broaddown2. The Ultimate Spartan3 Development 
Board.
http://www.enterpoint.co.uk


"gw" <guenter.wolpert@orsys.de> wrote in message 
news:1119014517.226639.8750@g47g2000cwa.googlegroups.com...
> At least there is an external pull-up on the DONE pin.
> Anyway, even if the DONE pin is an open collector only,
> it shouldn't stop pulling low before the end of the data?
>
> The only thing with the open collector option is that DONE
> may come a little bit later than expected.
>
> One additional thing we observed is that the FPGA started to
> drive its outputs one clock before the DONE appeared. Therefore
> it seems that the configuration process is already over before all
> data is written to the FPGA.
>
> Guenter
> 



Article: 85887
Subject: Re: AbusivepPricing information in marketing publications
From: John_H <johnhandwork@mail.com>
Date: Fri, 17 Jun 2005 13:36:03 GMT
Links: << >>  << T >>  << A >>
The text ALWAYS clearly states what the pricing timeframe and quantity 
are for the price.

Engineers looking to design in a new part in large quantities are 
typically looking for production a little ways out.  I would prefer to 
see 1H2006 pricing, but still... they make it clear.

The pricing method is used by too many vendors so why should Xilinx say 
this snazzy new part is available for $XX now in small quantities when 
others are advertizing the "mature volume" pricing?

If you know what the "typical markup" is for early adoption of a device 
or for small quantities of the part as purchased by your company, you 
can get a ballpark to the pricing without having to call for specifics.

I'm surprised that the UWG makes illegal the practice of giving a price 
where the price is clearly marked with a note and that note supplies the 
timeframe and quantitiy.  This is misleading why?  We do have market 
realities to consider, after all.


Kolja Sulimma wrote:
> Xilinx is doing it again.
> 
> http://www.xilinx.com/xlnx/xil_prodcat_product.jsp?title=s3e_overview
> "Spartan-3E FPGA devices with 100K system gates are available for under
> US$2.00*"
> Later in the text we learn that the pricing is for 2H2006. The current
> price is higher.
> 
> There are three correct formulations possible for these facts:
> - The devices will be be available for under US$2.
> - The devices are not available for under US$2.
> - The devices are available for <insert real price>
> 
> The formulation chosen by Xilinx marketing definetly is illegal in
> Germany under UWG. (Last time these text were distributed in Germany by
> distributors). And it probably is illegal in the US.
> 
> I do not understand why a big company with a good product again and
> againg ressorts to  unfair and illegal marketing that is designed to
> confuse customers?
> This really is bad style.
> 
> Kolja Sulimma
> (Otherwise a happy Xilinx user)

Article: 85888
Subject: Re: BGA Rework/Prototype Placement Anyone?
From: Mike Harrison <mike@whitewing.co.uk>
Date: Fri, 17 Jun 2005 13:37:17 GMT
Links: << >>  << T >>  << A >>
On Fri, 17 Jun 2005 11:04:01 +0100, "John Adair"
<removethisthenleavejea@replacewithcompanyname.co.uk> wrote:

>We have a product coming to assist in this area. Based on Spartan-3 in FG456 
>it consists of a PGA style board with the Spartan-3 in the middle. This 
>module is aimed at hobby or small run board builders that don't want the 
>setup charges of BGA lines.

Sounds like a really good idea - having started playing with the S3 devkit I keep having ideas of
other fun, low-volume or 1-off things that I could use an FPGA for if it were more easily
connect-to-able.

Do you have a rough spec yet ? Number of pins, size, price etc. ?

You may have already thought of some/all of these, but a few suggestions...

Would be nice if the pinout is arranged such that you can easily trade off required number of layers
needed on the host PCB against pins used, e.g useable on a 2-layer PCB if you don't need a lot of IO
pins. 

Having a power connector on the board would be useful, to reduce the amount of big tracks you'd need
(and allow a 2-layer host PCB for less demanding applications). Perhaps have JTAG and power
connection pins along an edge with some perforations, so that can be snapped off if you want minimum
module size, or left on to make it easier to use with simpler host boards. 
1.5 and 2.5V regulators on board ? 

Any chance the footprint can be the same as one of the various PC CPUs, so sockets can be found
cheaply ?    

Probably don't need seperately accessible connections to all of the IO bank supplies, but at least
one seperately connectable bank would be nice for when you want a few 'odd' standards that need a
different supply, e.g. lvds.  Maybe have some solder-bridge type link options for this?

Footprint for a user-fittable  SMD oscillator module or two  so high-speed clocks can stay on the
module. 


Article: 85889
Subject: Re: comp.arch.fpga.<mfr>
From: John_H <johnhandwork@mail.com>
Date: Fri, 17 Jun 2005 13:44:30 GMT
Links: << >>  << T >>  << A >>
I would prefer to keep the group together.  If one of the "little guys" 
comes up with a very marketable device, the buzz here will tend to get 
people interested.  Knowing some of the issues in other devices helps to 
shape expectations when switching to that vendor's part.

I *do* like the idea of moving the CPU stuff.  There has been a huge 
amount of processor related traffic which doesn't really relate to the 
hardware but to the "core" used.  I don't use any processors in my FPGA 
designs yet so the threads just clutter my FPGA experience here.  I read 
comp.lang.verilog but don't bother with comp.lang.vhdl and subscribe to 
both.  There's no reason someone designing with CPUs couldn't do both. 
Also, cross posting when a topic covers FPGA hardware AND the CPU is 
reasonable.

I like to see the occasional posts here on DSP implementation when they 
refer to hardware yet wouldn't enjoy a diatribe on how to adapt 
coefficients.  Hardware related DSP can be cross posted for those as well.

I'd hate to see the vendor schism.


Adam Megacz wrote:

> I see a lot of posts here that are specific to a single vendor's
> devices or that device-vendor's tools.  I think it would make things a
> bit more organized if we had a few subgroups beneath comp.arch.fpga
> for the specific vendors (just like comp.os), and kept comp.arch.fpga
> itself for vendor-agnostic (or vendor-comparative) discussions.
> Traffic has been pretty high lately (~50 posts/day).

<snip>

Article: 85890
Subject: Re: AbusivepPricing information in marketing publications
From: David Brown <david@westcontrol.removethisbit.com>
Date: Fri, 17 Jun 2005 16:02:38 +0200
Links: << >>  << T >>  << A >>
John_H wrote:
> The text ALWAYS clearly states what the pricing timeframe and quantity 
> are for the price.
> 
> Engineers looking to design in a new part in large quantities are 
> typically looking for production a little ways out.  I would prefer to 
> see 1H2006 pricing, but still... they make it clear.
> 
> The pricing method is used by too many vendors so why should Xilinx say 
> this snazzy new part is available for $XX now in small quantities when 
> others are advertizing the "mature volume" pricing?
> 
> If you know what the "typical markup" is for early adoption of a device 
> or for small quantities of the part as purchased by your company, you 
> can get a ballpark to the pricing without having to call for specifics.
> 
> I'm surprised that the UWG makes illegal the practice of giving a price 
> where the price is clearly marked with a note and that note supplies the 
> timeframe and quantitiy.  This is misleading why?  We do have market 
> realities to consider, after all.
> 

It is clearly misleading to say " *are* available" for a particular 
price, when they are most definitely not available.  Giving a 500k price 
for a year in the future is stretching "mature volume pricing" quite a 
bit, although it would not be unreasonable if the date and quantity were 
quoted in the text, rather than as a small-print footnote.  I don't 
think anyone would consider the Altera Cyclone II press release to be 
misleading or illegal:

<quote from Altera press release>
Pricing and Availability

The first member of the Cyclone II device family, the EP2C35 device, 
will be available in February 2005. Volume pricing for the EP2C35 will 
be $22 in 250,000 unit volumes. The web edition of Quartus II version 
4.1 software supports the entire Cyclone II family and can be downloaded 
for free on www.altera.com/q2webedition.
</quote>

Now do you see the difference?  I'm not claiming Altera's marketing 
people are perfect, and they are as happy as anyone else to omit useful 
information (there were no prices given in the Stratix II 180 press 
release!), but they are clear on what their price information actually is.

mvh.,

David



> 
> Kolja Sulimma wrote:
> 
>> Xilinx is doing it again.
>>
>> http://www.xilinx.com/xlnx/xil_prodcat_product.jsp?title=s3e_overview
>> "Spartan-3E FPGA devices with 100K system gates are available for under
>> US$2.00*"
>> Later in the text we learn that the pricing is for 2H2006. The current
>> price is higher.
>>
>> There are three correct formulations possible for these facts:
>> - The devices will be be available for under US$2.
>> - The devices are not available for under US$2.
>> - The devices are available for <insert real price>
>>
>> The formulation chosen by Xilinx marketing definetly is illegal in
>> Germany under UWG. (Last time these text were distributed in Germany by
>> distributors). And it probably is illegal in the US.
>>
>> I do not understand why a big company with a good product again and
>> againg ressorts to  unfair and illegal marketing that is designed to
>> confuse customers?
>> This really is bad style.
>>
>> Kolja Sulimma
>> (Otherwise a happy Xilinx user)

Article: 85891
Subject: Re: Idea exploration - Image stabilization by means of software.
From: Jonathan Bromley <jonathan.bromley@doulos.com>
Date: Fri, 17 Jun 2005 15:03:47 +0100
Links: << >>  << T >>  << A >>
On Fri, 17 Jun 2005 11:36:40 +0200, Anton Erasmus
<nobody@spam.prevent.net> wrote:

>On Thu, 16 Jun 2005 22:08:13 -0700, "Jon Harris"
><jon_harrisTIGER@hotmail.com> wrote:
>
>>Are you taking pictures of stationary things from a stationary camera?  That's
>>what the astronomy case involves.  But since you are concerned with a moving
>>camera, I assumed you were talking about pictures taken with a hand-held camera.

>If one scaled this down so that the "short" exposure time is in
>milli-seconds or even micro-seconds depending on what the sensors can
>do, then one can do the same thing just at a much faster scale.
>What is the minimum exposure time for a current CMOS sensor before one
>just see the inherant sensor noise ?

Get real.  You've got to read out the sensor, as well as 
merely exposing it.  Unless you have a highly integrated CMOS
imager with piles of memory on it, there is essentially only
one (*) serial data path out of the device and you need to 
squeeze all the pixels through that.  Shortening the exposure
of a CCD sensor is easy; it's done all the time to control
exposure time.  But it is *very* hard to get the image OUT
of a multi-megapixel CCD in less than a few milliseconds.

There is precisely one way in which you CAN do this: it's
called "Time Delay Integration" or TDI and it's been 
widely used for hi-res military sensors for ages.  If you
can arrange that the motion is at a constant speed along
the sensor's Y axis, then you can organise the sensor's
vertical shift clock so that it keeps in step with the
motion, and a fairly long exposure can be perfectly
sharp.  Brilliant for taking aerial reconnaissance
photos from a 'plane in flight.  But it needs lots of
smarts in the motion sensing and the camera readout
electronics.

If I read the OP's idea correctly, he wants to correct
the picture for motion blurring after it's been exposed.
This is perfectly possible if you know the exact behaviour
of the motion.  The OP should find out about "deconvolution".
To take a simpler example:  Suppose you have a stream of
data, and before you get to see that data you know that it 
has been processed by a moving-average filter.  How can
you reconstruct the original data points, if you know the
exact form of the moving-average filter function?

(*) Some sensors have multiple serial output paths - 
see Dalsa's offerings, for example.  The number of
paths is never more than a small handful, though.
Broadside readout of the lines into local on-chip
memory is the only hope.
-- 
Jonathan Bromley, Consultant

DOULOS - Developing Design Know-how
VHDL, Verilog, SystemC, Perl, Tcl/Tk, Verification, Project Services

Doulos Ltd. Church Hatch, 22 Market Place, Ringwood, BH24 1AW, UK
Tel: +44 (0)1425 471223          mail:jonathan.bromley@doulos.com
Fax: +44 (0)1425 471573                Web: http://www.doulos.com

The contents of this message may contain personal views which 
are not the views of Doulos Ltd., unless specifically stated.

Article: 85892
Subject: Re: Xlinix configuration: DONE pin too early?
From: "gw" <guenter.wolpert@orsys.de>
Date: 17 Jun 2005 07:29:51 -0700
Links: << >>  << T >>  << A >>


John Adair schrieb:
> The point is that DONE is not always an open drain drive. Have a look at
> this Xilinx answer and you might get a clearer understanding.
>

>>Answers Database
>>Virtex, Virtex-II Configuration - The DONE pin goes High, but the device does
>>not start up (I/Os are inactive/3-stated)

Well, this points to a different topic.
We don't have any problems after the done pin is high. We have the
problem (?)
that the DONE pin goes high earlier than expected. The loaded code
works very well.

Although I doubt that this is a problem, I'll go and check the
bitstream option and the lower pull-up value (we use more than 330
Ohms).

Can you confirm that the DONE only appears *after* all configuration
data has been sent?

Guenter


Article: 85893
Subject: Xilinx FFT
From: Javier Castillo <jcastillo@opensocdesign.com>
Date: Fri, 17 Jun 2005 16:33:55 +0200
Links: << >>  << T >>  << A >>
Hello,

  I am trying to connect a FFT module generated with coregen to a
microprocessor. The problem is that the microprocessor works with
floating data in IEEE754 format and the xilinx FFT module accepts,
fixed point and block-floating point. How can I convert the data from
the microprocessor to one the core understand.

Regards

Javier Castillo



Article: 85894
Subject: Re: Xlinix configuration: DONE pin too early?
From: "Berty" <wooster.berty@gmail.com>
Date: 17 Jun 2005 09:07:31 -0700
Links: << >>  << T >>  << A >>
Some time ago I wrote a loader to load the Xilinx Spartan2 but this
should be the same for the Virtex2 and Spartan3 as far as I know and
when I checked back than the signal behave the done should only go
active after the whole bitstream was loaded and the crc was ok.
You can control when it will go using the Startup Option (By default it
come before the output are enable and the internal reset is released).

If the reason you are concern about good CRC is that if it is bad you
want to try and reload or maybe load different bitstream than what I
did was once I ended loading the bitstream I let 16 clocks pass and
than checked the done.
I don't recall why I use 16 clock as the startup don't let you
delay for so long and probably count to 6 should be enough but ...

Have fun.


Article: 85895
Subject: Re: comp.arch.fpga.<mfr>
From: Bob Perlman <bobsrefusebin@hotmail.com>
Date: Fri, 17 Jun 2005 16:15:11 GMT
Links: << >>  << T >>  << A >>
Hi - 

If we're going to split up the newsgroup--and I strongly suggest that
we don't--here are some suggestions:

  .vendor-to-vendor-flames - FPGA vendor employees argue endlessly
about whose performance claims are more specious, whose parts are less
available, and whose manners are more uncouth.

  .homework-problems - Repeated questions about traffic lights and
Coke machines, followed by "do it yourself" responses, "here's where
to start looking" responses, and "here's how to do it" responses, the
last of which guarantee that the stream of questions will never cease.

  .schematics-vs-hdl - Because we haven't quite beaten this one to
death yet.

  .i-need-an-orcad-symbol - Well, I mean, *I* don't need an Orcad
symbol.  It's for a friend.

  .recruiters - Offers of fabulous FPGA design jobs that require you
to relocate to the Falkland Islands or Faluja.  Sign up before
midnight tonight and get a free set of steak knives.

  .things-i-could-have-googled-for - Self-explanatory.  Or maybe not.

Bob Perlman
Cambrian Design Works


On Fri, 17 Jun 2005 02:34:48 -0700, Adam Megacz
<megacz@cs.berkeley.edu> wrote:

>
>I see a lot of posts here that are specific to a single vendor's
>devices or that device-vendor's tools.  I think it would make things a
>bit more organized if we had a few subgroups beneath comp.arch.fpga
>for the specific vendors (just like comp.os), and kept comp.arch.fpga
>itself for vendor-agnostic (or vendor-comparative) discussions.
>Traffic has been pretty high lately (~50 posts/day).
>
>If you support this (or if you don't), please post in this thread.
>Also let me know what vendors you'd like to see listed.  Off the top
>of my head, alphabetically,
>
>   Actel
>   Altera
>   Atmel
>   Cypress
>   Lattice
>   Quicklogic
>   Xilinx
>
>I think it would be wise to keep the list short by limiting it to
>vendors currently producing and selling chips, and which are available
>from distributors at the retail level (ie EOL'ed or a specialty
>device).
>
>If there is sufficient support, I will RFD the matter over on
>news.announce.newgroups and reference this thread as justification.
>The call for votes will be crossposted here.
>
>Thanks,
>
>  - a


Article: 85896
Subject: re:Problem for xilinx!!!
From: dejavu99@hotmail-dot-it.no-spam.invalid (damidar)
Date: Fri, 17 Jun 2005 11:17:47 -0500
Links: << >>  << T >>  << A >>
now I will try to make it

thanks too gabor


Article: 85897
Subject: implementing webserver application
From: kurapati77@gmail-dot-com.no-spam.invalid (kurapati)
Date: Fri, 17 Jun 2005 11:17:47 -0500
Links: << >>  << T >>  << A >>
Hi Ed,

Thanks for ur info of that gsrd reference design is available for
ml403. I have used it and run the treck tcp/IP stack. 

For me more interesting is running a simple webserver application
using powerpc. 

Can you tell me how can I start implementing it on ml403 board. 

I have seen one webserver application using lwIP for virtex2pro in
application notes 663 but how one can use the same for virtex 4 
device on ml403. what kind of chages required to run the webserver
application.

please give me some hints or notes which help me to use the trimode
ethernet MAC..

with regards
Rajesh


Article: 85898
Subject: found edk example
From: kurapati77@gmail-dot-com.no-spam.invalid (kurapati)
Date: Fri, 17 Jun 2005 11:17:48 -0500
Links: << >>  << T >>  << A >>
Hi Ed

I found echoserver application in EDK examples..
hope that will help me some extend to use the TMAC..

I come to u if I get any problems..

bye


Article: 85899
Subject: Re: Xlinix configuration: DONE pin too early?
From: "Dave Garnett" <dave.garnett@metapurple.co.uk>
Date: Fri, 17 Jun 2005 17:18:26 +0100
Links: << >>  << T >>  << A >>

"Berty" <wooster.berty@gmail.com> wrote in message 
news:1119024451.072046.217090@z14g2000cwz.googlegroups.com...
> Some time ago I wrote a loader to load the Xilinx Spartan2 but this
> should be the same for the Virtex2 and Spartan3 as far as I know and
> when I checked back than the signal behave the done should only go
> active after the whole bitstream was loaded and the crc was ok.
> You can control when it will go using the Startup Option (By default it
> come before the output are enable and the internal reset is released).
>
> If the reason you are concern about good CRC is that if it is bad you
> want to try and reload or maybe load different bitstream than what I
> did was once I ended loading the bitstream I let 16 clocks pass and
> than checked the done.
> I don't recall why I use 16 clock as the startup don't let you
> delay for so long and probably count to 6 should be enough but ...
>
> Have fun.
>
>

I too have written a loader for Spartan II, and I can confirm that DONE 
becomes active before the correct number of bits is loaded. At one point my 
load code stopped clocking when it saw DONE go high - but the fpga would 
then not operate. I now always send the correct number of clocks and then 
check DONE and everything is fine ...

Dave




 Posted Via Nuthinbutnews.Com Premium Usenet Newsgroup Services
----------------------------------------------------------
    ** SPEED ** RETENTION ** COMPLETION ** ANONYMITY **
----------------------------------------------------------        
                http://www.nuthinbutnews.com



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