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 82675

Article: 82675
Subject: Re: salary ballpark please guys
From: "Bob" <nimby1_notspamm_@earthlink.net>
Date: Sat, 16 Apr 2005 04:55:42 GMT
Links: << >>  << T >>  << A >>
Dave,

I'd estimate at least $700K/yr for Atlanta, and about 20% more for San
Jose -- especially since this person will need to do technical
documentation/marketing material.

Bob


"Dave" <no@spam.com> wrote in message
news:d3oo69$1ft$00$1@news.t-online.com...
> Hi to all,
>
> Just wondering if you could give me an idea of a current salary ballpark
for
> the following details/person:
>
> Person:
> PhD in DSP on FPGA (3.5 years of RTL VHDL mainly for Xilinx and C++
> development)
> 1st class honours degree in Software Development/Electronic & Electrical
> Engineering
> 1 years industry experience (+ some during PhD)
> sales training/experience
> extensive presentation writing/giving experience
>
> Job description:
> DSP on FPGA design in RTL VHDL on Xilinx/Altera/Lattice/Actel etc.
> C++ development
> Customer visits/application support
> Technical documentation/marketing material
> Presentations to customers/conferences/seminars
>
> Job Location:
> Please give an idea for Atlanta and San Jose
>
> Thanks very much in advance for your time
>
>



Article: 82676
Subject: Re: LUT in fpga
From: Philip Freidin <philip@fliptronics.com>
Date: Sat, 16 Apr 2005 05:10:33 GMT
Links: << >>  << T >>  << A >>
On 13 Apr 2005 06:21:03 -0700, "Mohammed A Khader" <am.imak@gmail.com> wrote:
>Hi all,
>
> I am new to FPGA architecture.

Welcome.

>To know the what does LUT means I came
>across few  good articles posted in this group.
>But still not completely clear about it.
>
>A 4-bit input LUT in spartan 2 can be implemented as Function
>Generator or as 16-bit ROM ... accepted .

You need to accept it more fully :-)   Think of a ROM the following way:

16 flipflops (storage) each one either set or clear, and no clock
to the flipflops. Take the output of the 16 FFs, and connect them
to a 16 to 1 mux. The mux will need 4 control lines. When the control
lines are 0000, the output of the mux is the value in FF_0 , when it
is 1111, the mux output is the value in FF_15.

So really, there is no difference between calling it a LUT or a ROM.
It holds 16 bits, and via the mux, you can select to look at any 1 bit.

>But I can't imagine it as a 16-bit RAM or a shift Register.

Let me help your imagination, in stages.

First, lets take the 16 FFs I described above (without clocks, but
somehow preset to either 0 or 1), and lets add an async set and clear
line to each FF. We can now force the value of each FF via these new
signals. This is equivalent (but not actually) the way the FPGA uses
the configuration data to initialize the FFs that make up the ROM/LUT.

Now, lets add a common D and common clock input to each FF, and also
a clock enable to each FF (16 of them). Add a 4 line to 16 line decoder
and connect it to the 16 clock enable lines. The 4 inputs to the decoder
are connected to the 4 control lines of the output selection mux. So
when these 4 lines are 0000, the output mux shows us the content of FF_0,
and if we put a 0 or 1 on the D input line (that goes to all FFs), and
apply a rising clock edge (that goes to all FFs), then only FF_0 will be
loaded with the data on D, because the decoder is only enabling the
clock on FF_0. What we now have is a RAM.

>If it is a 16-bit Ram / Shift Register  then where is the input port.

There are 3 extra inputs, Clock, Write Enable (WE), and Data (D)

To turn the 16 FFs into a shiftregister, we need to add some additional
logic so that rather than only clocking 1 FF at a time (RAM mode), we
enable clocking of all 16 FFs, and rather that the 16 D inputs all
comming from a common D signal, the D signal only connects to FF_0. The
D of FF_n connects to the output of the prior FF, Q_n-1. Thus we have
a shiftregister.

>I think input port to LUT is only available during configuration of
>FPGA but not to the  designer.

This is true of the early Xilinx FPGA products, the XC2000 and XC3000
families. All Xilinx FPGA products since the XC4000 family (1990) have
had the RAM capability, and the SRL capability became available in the
Virtex-II product line (2001).

> Also please suggest me some good books to know the architecture of
>FPGA.

The US patent office makes available all patents on-line, for free.

You "just" need to search for the appropriate patent.

start here:  http://patft.uspto.gov/netahtml/search-bool.html


Try search terms like "RAM" and "XILINX".  Here let me help you.
try 5566123 , it includes complete detailed schematics, including
something you didn't ask about, Dual-Port.

> Thanks .
>-- Mohammed A Khader.


Unfortunately the folling has several errors.

On Wed, 13 Apr 2005 16:19:28 +0200, Laurent Gauch <laurent.gauch@DELETEALLCAPSamontec.com> wrote:
>DO NOT CONFUSE LUT and CLB

Mohammed never mentioned CLBs

>Sorry but a LUT is : Logic Unit Table -> a LUT can only do combinatory works (AND OR 
>XOR ...).

No. It is Look Up Table.

>CLB is: Configurated Logic Block -> a CLB include one LUT followed by 
>one Flip-Flop in minimum.

True enough, but irrelevant to the question.

>For some Xilinx FPGA the CLB can be use as RAM or as SRL (not the LUT 
>itself ;-) )

No. The LUT  IS   what is used to implement the RAM or SRL.

>After that advice, please open a FPGA datasheet for understanding it 
>specific architecture !

Sure.


Happy learning,
Philip




===================
Philip Freidin
philip.freidin@fpga-faq.org
Host for WWW.FPGA-FAQ.ORG

Article: 82677
Subject: Re: Hobby or job? (FPGA User's groups anyone?)
From: "Simon Peacock" <nowhere@to.be.found>
Date: Sat, 16 Apr 2005 17:24:40 +1200
Links: << >>  << T >>  << A >>
this Simon.. is a professional :-)

but thoughts and ideas expressed by Simon may not be the same as all Simons.


"Simon" <news@gornall.net> wrote in message
news:rrOdnXY7xaVu9P3fRVn-qg@comcast.com...
> JJ wrote:
> > I'd imagine most people here are employed, consultants, small business,
> > occasional blue chip EEs etc beside the FPGA vendor reps. The traffic
> > reflects the niggling details that make simple things harder than 1st
> > seems and a little ragging on the vendors to boot. Not sure if open
> > source is of that much interest here except to the entry level or for
> > educational purposes.
> >
> > The free semi free tools are a great way for vendors to rope in new
> > paying customers perhaps a future woz or 2 will give them larger order,
> > no OSS agenda here.
> >
> > Since FPGAs are heading up into low end ASIC territory, not sure if
> > hobbyists will become a major deal here, mostly serious hobyists are
> > perhaps buddying engineers. It may seem as if FPGAs are a little
> > reminiscant of the 70s computer clubs but I don't expect to see
> > homebrew clubs popping up.
>
> Well, I'm a hobbyist. It makes a change from coding software for a
> living, at least when I can fit it in :-)
>
> The problems for hobbyists aren't really the vendors, or the tools, or
> even the large quantities of relatively domain-specific knowledge
> required. It's getting to grips with the teeny-weeny-pin soldering
> issues (at least it is for me). That's when you can get the parts in
> home-solderable form at all, eg: largest S3 is a PQ208. Whoosh. By the
> time you've added memory and a few peripherals, you're seriously out of
> pins :-( *And* it's a bitch to solder...
>
> Yes, I know there are good reasons (trust me - this is me being
> understanding :-) You really don't want to hear me whining about
> something :-)) Question though - would pin-grid array have the same
> signal integrity issues as BGA ? Or is it a function of lead-length
> rather than pin geography ? At least PGAs are solderable...
>
> I don't really expect A,X,whoever to produce parts that help the
> hobbyist, but it's immensely frustrating to get a quote of $350/board
> (inc. assembly/x-ray testing) when the part is "under 10$", well ok, $50
> in my quantities...
>
> It's actually cheaper to buy a development board for deployment (eg:
> AVNET virtex-4 FX-12 board) as well as development, which is a bit weird
> really!
>
> Simon.



Article: 82678
Subject: Technical Journals on FPGAs
From: JoeG <JoeG@yahoo.com>
Date: Sat, 16 Apr 2005 05:25:29 GMT
Links: << >>  << T >>  << A >>
Can anyone provide a recommendation on which journal or transaction 
series one ought to subscribe to in the IEEE family which provides the 
most coverage on FPGAs?

Computer Aided Design of Integrated Circuits and Systems, IEEE (TCAD)
Very Large Scale Integration Systems (TVLSI)
Solid-State Circuits (J-SSC)


Thanks in advance ....

Article: 82679
Subject: Re: salary ballpark please guys
From: "Simon Peacock" <nowhere@to.be.found>
Date: Sat, 16 Apr 2005 17:26:07 +1200
Links: << >>  << T >>  << A >>
depends...

Do you normally add "Would you like fries with that?"


Simon



"Dave" <no@spam.com> wrote in message
news:d3oo69$1ft$00$1@news.t-online.com...
> Hi to all,
>
> Just wondering if you could give me an idea of a current salary ballpark
for
> the following details/person:
>
> Person:
> PhD in DSP on FPGA (3.5 years of RTL VHDL mainly for Xilinx and C++
> development)
> 1st class honours degree in Software Development/Electronic & Electrical
> Engineering
> 1 years industry experience (+ some during PhD)
> sales training/experience
> extensive presentation writing/giving experience
>
> Job description:
> DSP on FPGA design in RTL VHDL on Xilinx/Altera/Lattice/Actel etc.
> C++ development
> Customer visits/application support
> Technical documentation/marketing material
> Presentations to customers/conferences/seminars
>
> Job Location:
> Please give an idea for Atlanta and San Jose
>
> Thanks very much in advance for your time
>
>



Article: 82680
Subject: Re: Hobby or job? (FPGA User's groups anyone?)
From: Mac <foo@bar.net>
Date: Sat, 16 Apr 2005 05:26:45 GMT
Links: << >>  << T >>  << A >>
On Fri, 15 Apr 2005 18:27:10 -0700, Simon wrote:

> JJ wrote:
>> I'd imagine most people here are employed, consultants, small business,
>> occasional blue chip EEs etc beside the FPGA vendor reps. The traffic
>> reflects the niggling details that make simple things harder than 1st
>> seems and a little ragging on the vendors to boot. Not sure if open
>> source is of that much interest here except to the entry level or for
>> educational purposes.
>> 
>> The free semi free tools are a great way for vendors to rope in new
>> paying customers perhaps a future woz or 2 will give them larger order,
>> no OSS agenda here.
>> 
>> Since FPGAs are heading up into low end ASIC territory, not sure if
>> hobbyists will become a major deal here, mostly serious hobyists are
>> perhaps buddying engineers. It may seem as if FPGAs are a little
>> reminiscant of the 70s computer clubs but I don't expect to see
>> homebrew clubs popping up.
> 
> Well, I'm a hobbyist. It makes a change from coding software for a 
> living, at least when I can fit it in :-)
> 
> The problems for hobbyists aren't really the vendors, or the tools, or 
> even the large quantities of relatively domain-specific knowledge 
> required. It's getting to grips with the teeny-weeny-pin soldering 
> issues (at least it is for me). That's when you can get the parts in 
> home-solderable form at all, eg: largest S3 is a PQ208. Whoosh. By the 
> time you've added memory and a few peripherals, you're seriously out of 
> pins :-( *And* it's a bitch to solder...
> 
> Yes, I know there are good reasons (trust me - this is me being 
> understanding :-) You really don't want to hear me whining about 
> something :-)) Question though - would pin-grid array have the same 
> signal integrity issues as BGA ? Or is it a function of lead-length 
> rather than pin geography ? At least PGAs are solderable...
> 
> I don't really expect A,X,whoever to produce parts that help the 
> hobbyist, but it's immensely frustrating to get a quote of $350/board 
> (inc. assembly/x-ray testing) when the part is "under 10$", well ok, $50 
> in my quantities...
> 
> It's actually cheaper to buy a development board for deployment (eg: 
> AVNET virtex-4 FX-12 board) as well as development, which is a bit weird 
> really!
> 
> Simon.

Nowadays people are soldering surface mount parts themselves in toaster
ovens. I think you can even do BGA's this way.

Have you checked out the PCBexpress website? If not, have a look.
www.pcbexpress.com. They make circuit boards, but they also have a lot of
information about hobbyist / low-volume production techniques.

Mind you, I haven't done any of this stuff myself, yet, but I would much
rather cook a board in the oven than try to solder it, even if it doesn't
have any BGA's on it. Of course, with BGA's, hand-soldering is not an
option.

--Mac


Article: 82681
Subject: Re: salary ballpark please guys
From: Mac <foo@bar.net>
Date: Sat, 16 Apr 2005 05:36:03 GMT
Links: << >>  << T >>  << A >>
On Sat, 16 Apr 2005 04:55:42 +0000, Bob wrote:

> Dave,
> 
> I'd estimate at least $700K/yr for Atlanta, and about 20% more for San
> Jose -- especially since this person will need to do technical
> documentation/marketing material.
> 
> Bob
> 

Do normal people really make 700k a year purely in salary? My first
reaction is that you are making a joke.

I would have guessed 150-200k for a PhD with only a little relevant
experience, but that would have been nothing but a guess. That is, I don't
know any PhD's (in EE).

How long does it take to get a PhD? If it's only five or six years, it may
actually be worth it! ;-)

--Mac

> 
> "Dave" <no@spam.com> wrote in message
> news:d3oo69$1ft$00$1@news.t-online.com...
>> Hi to all,
>>
>> Just wondering if you could give me an idea of a current salary ballpark
> for
>> the following details/person:
>>
>> Person:
>> PhD in DSP on FPGA (3.5 years of RTL VHDL mainly for Xilinx and C++
>> development)
>> 1st class honours degree in Software Development/Electronic & Electrical
>> Engineering
>> 1 years industry experience (+ some during PhD)
>> sales training/experience
>> extensive presentation writing/giving experience
>>
>> Job description:
>> DSP on FPGA design in RTL VHDL on Xilinx/Altera/Lattice/Actel etc.
>> C++ development
>> Customer visits/application support
>> Technical documentation/marketing material
>> Presentations to customers/conferences/seminars
>>
>> Job Location:
>> Please give an idea for Atlanta and San Jose
>>
>> Thanks very much in advance for your time
>>
>>



Article: 82682
Subject: Re: Hobby or job? (FPGA User's groups anyone?)
From: Quiet Desperation <nospam@nospam.com>
Date: Fri, 15 Apr 2005 22:38:48 -0700
Links: << >>  << T >>  << A >>
In article <d3pr78127f2@enews2.newsguy.com>, Phil Tomson
<ptkwt@aracnet.com> wrote:

> Well, much cheaper tools anyway ;-)  I can't afford ModelSim, but I can 
> download and compile Icarus or GHDL (both open source) for all of my HDL 
> simulation needs.  Are they as slick as ModelSim?  No, but they're 
> actually pretty capable and I get things done with them.

Eh. I just log into work over the VPN link, and use Modelsim for my own
projects. And Synplicity. And the full Xilinx tools. All from my laptop
wirelessly from my couch. :)

Yes, you may throw things at me now.

Bosees don't mind as long as its after hours and the Suns are just
sitting there counting out time.

Article: 82683
Subject: hdl designer cvs in remote repository
From: "badduck" <nachiket@gmail.com>
Date: 15 Apr 2005 22:50:09 -0700
Links: << >>  << T >>  << A >>
has anyone figured out how to do a cvs operation using HDL Designer? i
use cvsroot as :extssh:name@domain:/cvsroot_path and get the error
"Error: cvs.exe add: unknown method in CVSroot: .....
cvs.exe [add aborted]: Bad CVSROOT.


Article: 82684
Subject: Xilinx tools on Linux
From: David <dmsbox2000-news1@yahoo.com>
Date: Sat, 16 Apr 2005 06:56:31 GMT
Links: << >>  << T >>  << A >>


There are ISE Foundation and ISE Webpack versions for Red Hat Enterprise 
Linux 3.0  How strict is this port?  Has anybody managed to run these 
tools on Fedora or Mandrake Linux (now Mandriva), or any other Linux 
variant?  If so, how well did it work?


Article: 82685
Subject: Re: salary ballpark please guys
From: ptkwt@aracnet.com (Phil Tomson)
Date: 16 Apr 2005 07:08:20 GMT
Links: << >>  << T >>  << A >>
In article <iV08e.7071$yq6.1107@newsread3.news.pas.earthlink.net>,
Bob <nimby1_notspamm_@earthlink.net> wrote:
>Dave,
>
>I'd estimate at least $700K/yr for Atlanta, and about 20% more for San
>Jose -- especially since this person will need to do technical
>documentation/marketing material.
>

$700K/yr!  Wow, where do I sign up?!

Phil

Article: 82686
Subject: Re: free-ip
From: "Daniel Florin" <florin@primelec.ch>
Date: Sat, 16 Apr 2005 09:09:27 +0200
Links: << >>  << T >>  << A >>
Just downloaded, thanks very much!
Dani

"Antti Lukats" <antti@openchip.org> schrieb im Newsbeitrag 
news:d3ml4e$9da$05$1@news.t-online.com...
> ok I wasnt lazy the freelib1 is now downloadable
> http://gforge.openchip.org/projects/wwwfreeipcom/
>
> antti
>
> "Daniel Florin" <florin@primelec.ch> schrieb im Newsbeitrag
> news:d3ma49$pm4$1@news.hispeed.ch...
>> Over the last couple of weeks I have tried accessing the web site
>> www.free-ip.com and gotten an error message. Is it mirrored someplace 
>> else
>> on the web?
>>
>> I am looking for Free-LIB1, a library of NCO, DAC and PWM parameterized
>> cores in VHDL, which was available at www.free-ip.com.
>>
>> Any information is very appreciated.
>>
>>
>
> 



Article: 82687
Subject: Differential timing specification in Xilinx FPGA
From: Marc Le Roy <nospam@please.com>
Date: Sat, 16 Apr 2005 09:09:48 +0200
Links: << >>  << T >>  << A >>
Hello,

I am searching how to specify a relative timing constraint (and not a 
placement constraint) that would be enforced either by the VHDL 
synthesis tool or the place and route tool in the following case 
(decoding of a serial link data/strobe like the one depicted in figure 5 
of http://products.sel.sony.com/semi/PDF/wpIEEE1394.pdf ) :

* the serial link is made of one data line D and one strobe line S

* the clock CLK at half bit rate is regenerated by an XOR between D and S

* the data signal D is sampled on both the rising edge and the falling 
edge of CLK.

The problem is that when the clock edge is caused by a change on the D 
line, the setup time is very dependant on the routing of D, S and CLK. 
Manual placement constraints allow to solve the problem, but it is 
rather tricky. I would prefer to find a way to specify a relative timing 
contraint between D and CLK.

Marc

Article: 82688
Subject: Re: salary ballpark please guys
From: Simon <news@gornall.net>
Date: Sat, 16 Apr 2005 00:15:45 -0700
Links: << >>  << T >>  << A >>
Mac wrote:
> On Sat, 16 Apr 2005 04:55:42 +0000, Bob wrote:
> 
> 
>>Dave,
>>
>>I'd estimate at least $700K/yr for Atlanta, and about 20% more for San
>>Jose -- especially since this person will need to do technical
>>documentation/marketing material.
>>
>>Bob
>>
> 
> 
> Do normal people really make 700k a year purely in salary? My first
> reaction is that you are making a joke.
> 
> I would have guessed 150-200k for a PhD with only a little relevant
> experience, but that would have been nothing but a guess. That is, I don't
> know any PhD's (in EE).
> 
> How long does it take to get a PhD? If it's only five or six years, it may
> actually be worth it! ;-)

Well, to join two threads together, if a 'professional' in this industry 
makes $700k/year, I'm about to become a professional! And I already have 
a PhD :-) [it takes 3 years in the UK, on average btw]

I'm sure that was just a reaction post to an open-ended question though. 
Please tell me I'm right, or I'll have wasted the last 10 years (!)

ATB,
	Simon (the unprofessional one. Oh, wait!)



Article: 82689
Subject: Re: Hobby or job? (FPGA User's groups anyone?)
From: Simon <news@gornall.net>
Date: Sat, 16 Apr 2005 00:25:49 -0700
Links: << >>  << T >>  << A >>
Mac wrote:
> 
> Nowadays people are soldering surface mount parts themselves in toaster
> ovens. I think you can even do BGA's this way.
> 
> Have you checked out the PCBexpress website? If not, have a look.
> www.pcbexpress.com. They make circuit boards, but they also have a lot of
> information about hobbyist / low-volume production techniques.

Yep, and pad2pad (who are cheaper, in general). PCBExpress (or rather, 
screaming circuits, their partner) want ~$650 for 1 board I was thinking 
of (10 IC's, 1 BGA, a few through-hole components). Pad2pad were $350 
for the same. It makes 'hobbyist' a bit expensive...

> Mind you, I haven't done any of this stuff myself, yet, but I would much
> rather cook a board in the oven than try to solder it, even if it doesn't
> have any BGA's on it. Of course, with BGA's, hand-soldering is not an
> option.

Like you, I've heard of it, but not actually done it. I intend to 
though. I've read some of the articles and I'll be purchasing one of 
these 'toaster' ovens (a strange term!) RSN(tm). I'm off to Vegas for a 
week (trade show) so it'll have to wait a bit longer...

The PQ208's aren't actually that bad - it takes a bit of practice, which 
is itself expensive! But once you've the knack, you can knock out a 
board in a few hours - and that includes the inspection with a 
magnifying glass to find the solder joins...

I have heard of people using vias to mate with the balls of the chip and 
soldering from below, filling the vias like little ink-wells. Sounds 
like a good way to destroy a chip to me :-)


ATB,
	Simon

Article: 82690
Subject: Re: salary ballpark please guys
From: "Thomas Entner" <aon.912710880@aon.at>
Date: Sat, 16 Apr 2005 10:19:05 +0200
Links: << >>  << T >>  << A >>
>> Just wondering if you could give me an idea of a current salary ballpark
> for
>> the following details/person:
>>
>> Person:
>> PhD in DSP on FPGA (3.5 years of RTL VHDL mainly for Xilinx and C++
>> development)
>> 1st class honours degree in Software Development/Electronic & Electrical
>> Engineering
>> 1 years industry experience (+ some during PhD)
>> sales training/experience
>> extensive presentation writing/giving experience
>>
>> Job description:
>> DSP on FPGA design in RTL VHDL on Xilinx/Altera/Lattice/Actel etc.
>> C++ development
>> Customer visits/application support
>> Technical documentation/marketing material
>> Presentations to customers/conferences/seminars
>>
>> Job Location:
>> Please give an idea for Atlanta and San Jose

> I'd estimate at least $700K/yr for Atlanta, and about 20% more for San
> Jose -- especially since this person will need to do technical
> documentation/marketing material.


I think, there is a '0' too much (if it was not a joke alltogether ;-), 
otherwise I'll buy a ticket to the Bay Area... (I am from Europe)

Thomas




Article: 82691
Subject: FPGA Design Introduction
From: "Clemens Hagen" <ernte23@gmx.at>
Date: Sat, 16 Apr 2005 10:27:21 +0200
Links: << >>  << T >>  << A >>
Hi

I wonder if I could find online a good introduction to start with FPGA and 
hardware design. So the document should say something about the trade of 
between performance and size, how the critical path determines the system 
performance and so on.

Does anybody know a good online ressourse for this basis stuff?

thanks 



Article: 82692
Subject: Re: Xilinx tools on Linux
From: buchty@atbode100.lrr.in.tum.de (Rainer Buchty)
Date: Sat, 16 Apr 2005 09:17:40 +0000 (UTC)
Links: << >>  << T >>  << A >>
In article <zG28e.2191$J12.2162@newssvr14.news.prodigy.com>,
 David <dmsbox2000-news1@yahoo.com> writes:
|> There are ISE Foundation and ISE Webpack versions for Red Hat Enterprise 
|> Linux 3.0  How strict is this port?  Has anybody managed to run these 
|> tools on Fedora or Mandrake Linux (now Mandriva), or any other Linux 
|> variant?  If so, how well did it work?

Debian and SuSE work nicely. Just the usual environment variable hassles.

Rainer


Article: 82693
Subject: Re: sharing a common resource... potential problems...
From: Philip Freidin <philip@fliptronics.com>
Date: Sat, 16 Apr 2005 09:17:44 GMT
Links: << >>  << T >>  << A >>
On 15 Apr 2005 06:41:48 -0700, Puneetsingh81@gmail.com (CODE_IS_BUG) wrote:
>if i am having a set of modules which reference some common control
>registers and each of these modules can read or write to these
>registers then there is a severe problem of 'bus' first of all and
>secondaly the 'mode' of the control registers(INOUT??). Is there any
>design reference or document that may be of help to me?

The first thing I would do is re-think the design approach. Having
multiple writers of a resource is usually just a bad idea. If the
situation is that different modules write to mutually exclusive
bits of the registers, then you just need to break these registers
up into separate registers.

If the writers are trully writing to the same bits, and this is really
what you want (see the start of my reply), then you could do something
like have multiple registers, and then mix their outputs (OR, AND,
priority, ...). The problem really comes down to what happens if 2 or
more writers both try to write at the same time. How is this resolved.
A complex solution to this might be something like a separate module
for these annoying registers, that takes in a bus and a write request
from each of the writers. Inside this module is where you decide what
to do with simultaneous writes.

On the third had, re-think, and change your design approach.

Philip Freidin




Philip Freidin
Fliptronics

Article: 82694
Subject: Re: salary ballpark please guys
From: buchty@atbode100.lrr.in.tum.de (Rainer Buchty)
Date: Sat, 16 Apr 2005 09:23:56 +0000 (UTC)
Links: << >>  << T >>  << A >>
In article <4260ca75$0$32388$91cee783@newsreader02.highway.telekom.at>,
 "Thomas Entner" <aon.912710880@aon.at> writes:
|> I think, there is a '0' too much (if it was not a joke alltogether ;-), 
|> otherwise I'll buy a ticket to the Bay Area... (I am from Europe)

Consultant job in the NYC metro area as a freshly graduated PhD, ~100K/yr 

A Friend of mine is working at LLNL (after a Postdoc year at Cornell), ~120K/yr 

Thomas: In Europe they don't pay their PhD nearly as well. They just wonder
why everyone who has the chance to work in the US does it...

Rainer

Article: 82695
Subject: Re: Xilinx tools on Linux
From: General Schvantzkoph <schvantzkoph@yahoo.com>
Date: Sat, 16 Apr 2005 08:41:49 -0400
Links: << >>  << T >>  << A >>
On Sat, 16 Apr 2005 06:56:31 +0000, David wrote:

> 
> 
> There are ISE Foundation and ISE Webpack versions for Red Hat Enterprise 
> Linux 3.0  How strict is this port?  Has anybody managed to run these 
> tools on Fedora or Mandrake Linux (now Mandriva), or any other Linux 
> variant?  If so, how well did it work?

They run fine on Fedora Core 3 once you have them installed. The only
problem is the installer which is looking for an older version of the
libcurl library. You can do a soft link between the version in FC3 and the
required version.

ln -s /usr/lib/libcurl.so.3.0.0 /usr/local/lib/libcurl.so.2 

If that doesn't work for you then the other alternative is to use RHEL3
or a clone, I used Whitebox Linux, to do the install and then you can
rsync the directories to your other machines.


Article: 82696
Subject: Re: salary ballpark please guys
From: "Dave" <no@spam.com>
Date: Sat, 16 Apr 2005 15:34:33 +0200
Links: << >>  << T >>  << A >>
> Dave,
>
> I'd estimate at least $700K/yr for Atlanta, and about 20% more for San
> Jose -- especially since this person will need to do technical
> documentation/marketing material.

$70k right!?





Article: 82697
Subject: Re: salary ballpark please guys
From: "Dave" <no@spam.com>
Date: Sat, 16 Apr 2005 15:35:06 +0200
Links: << >>  << T >>  << A >>
> How long does it take to get a PhD? If it's only five or six years, it may
> actually be worth it! ;-)

In the UK it is usually 3/4 years.

More like 5/6 in Europe though I think.... 



Article: 82698
Subject: Re: salary ballpark please guys
From: "Dave" <no@spam.com>
Date: Sat, 16 Apr 2005 15:36:14 +0200
Links: << >>  << T >>  << A >>
> Well, to join two threads together, if a 'professional' in this industry 
> makes $700k/year, I'm about to become a professional! And I already have a 
> PhD :-) [it takes 3 years in the UK, on average btw]

Any salary insights then based on your experience?

Cheers

Dave



Article: 82699
Subject: Re: salary ballpark please guys
From: "Dave" <no@spam.com>
Date: Sat, 16 Apr 2005 15:37:18 +0200
Links: << >>  << T >>  << A >>
> Consultant job in the NYC metro area as a freshly graduated PhD, ~100K/yr
>
> A Friend of mine is working at LLNL (after a Postdoc year at Cornell), 
> ~120K/yr
>
> Thomas: In Europe they don't pay their PhD nearly as well. They just 
> wonder
> why everyone who has the chance to work in the US does it...

I expect living costs in the NYC metro area are higher than Atlanta for 
example?

Pretty good salaries - does put Europe in the shade a little.... 





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