Which FPGA brand is industry standard for defense and radio/radar market?

Why only those two? What about Microsemi/Microchip for example? They 
have their market in space with radiation hardened FPGA.
If you are set to Xilinx or Intel, I can't help, but if not, I would not 
only focus on those two.

Federico Massimi wrote:
> Is this my impression true?
I can't tell about Italy, but I'm an FPGA developer in germany and we 
use Xilinx and Microchip (latter because of the space qualification).
Xilinx has/had a revenue of over 2 billion dollars and that's not 
because of some niches.
Personally, I liked Intel/Altera more because I found their tools to be 
more intuitive, but on the other hand it seems like you have to pay to 
get full support of VHDL (you have to pay to fully use the standard 
language).
But I don't have knowledge about FPGA development in defense.

(Btw. I wonder, how any war can happen? I only see defense companies, I 
have never heard about an offense company...)

Federico Massimi wrote:
> Greetings,
>      I have a small company that works in the defense and
> telecommunications sector.

most probably a VERY small company, in terms you yourself

Answering your questions:

Of course there is no FPGA-Type standard. There are functional standards 
for med end defense and flight systems, like eg do 254.

THIS is, what you have to know.

The other issue is, that several vendors do not have appropriate FPGAs 
to handle the requirments given in these documents or if a particular 
device is too expensive or does not have qualfied parameters.

For example: I changed a Xilinx device againts another one recently just 
because of SEE / SEU issues in terms of vulnerability against particle 
beams. Altera had a device too, but they wanted to go with Xilinx and 
the cores.

In another example, I changed from Altera to Lattice, because the device 
is cheaper and the little smaller size was no issue.

So gain: There is no "Standard Vendor".


> (at  least in my country, Italy) everyone, absolutely everyone,
> uses Altera.

This is wrong. 2 FPGA developers from Bergamo and Cassino working for my 
client use Xilinx more then Altera, and 70% of their business is in 
Italy. Moreover some suppliers I know working for FIAT and the City of 
Milano (Energy Dissipation) create some of their devices with Lattice 
FPGAs. I have the specs on my desk. Another supplier for power plant 
control uses Xilinx next to Altera.

Von Gerhard H,
wegen Überforderung der BBS-Software heute vermutlich als Gast


Developper wrote:
> Federico Massimi wrote:
>> Greetings,
>>      I have a small company that works in the defense and
>> telecommunications sector.
>
> most probably a VERY small company, in terms you yourself

> The other issue is, that several vendors do not have appropriate FPGAs
> to handle the requirments given in these documents or if a particular
> device is too expensive or does not have qualfied parameters.
>
> For example: I changed a Xilinx device againts another one recently just
> because of SEE / SEU issues in terms of vulnerability against particle
> beams. Altera had a device too, but they wanted to go with Xilinx and
> the cores.

There is no SEU issue in Xilinx that cannot be handled by cyclic
scrubbing of the configuration RAM. BTDT for Virtex. The FPGA can even
handle that itself if implemented in triple module redundancy. It just
must be done often enough so that SEUs cannot accumulate. TMR is most
impressive. You can inject masses of errors and the system just runs on,
just like you would try to shoot a terminator with a pump gun.
After each clock cycle, the world is OK again.

Picoblazes and the like need a small update since their registers are
really windows into the configuration RAM. Resetting the registers
to the power-up values every few seconds annoys the programmers.

I'd be happy to rewrite that and the TMR library in contemporary
VHDL. The 2nd implementation is usually much cleaner.  :-)
Like the accident with the Fortran punched card deck.
The joys of being a freelancer. They use us because we know things.

And no, there has never been a Führer or a Führungsoffizier, just
normal space technology like ISS.

Gerhard

Federico Massimi wrote:
>  defense and
> telecommunications sector. In the coming months we have to train some of
> our engineers to develop on FPGA. The question is Xilinx or Altera?
>   How is the industry standard?

You are confusing things, yes there are standards, but those standards 
are not met by simply using 'standard devices' or a 'preferred brand'. 
It is your job to find out if and how your hardware implementation meets 
customer requirements and legal standards. Some of the standards you 
need to be concerned with include:

https://en.wikipedia.org/wiki/DO-254
https://en.wikipedia.org/wiki/STANAG_3910
https://en.wikipedia.org/wiki/STANAG_4586
https://en.wikipedia.org/wiki/STANAG_5066
https://en.wikipedia.org/wiki/International_Traffic_in_Arms_Regulations
https://en.wikipedia.org/wiki/Export_Administration_Regulations

...


> Since I have to spend quite lot of money on training and software, I
> would like to make a conscious choice, Do you advise me to invest in
> training and software for Altera Intel FPGA or Xilinx FPGA?

Well, in my humble opinion, it's not worth hiring an engineer who relies 
on outside training to get all the know-how to do his job. An engineer 
must be able to B. Draw a "decision matrix" to choose between different 
devices and fill in the blanks after researching the available 
documentation.