EmbDev.net

Forum: FPGA, VHDL & Verilog Some basic fpga questions


Author: James Yunker (Company: Naval Undersea Warfare Center) (newport_j)
Posted on:

Rate this post
0 useful
not useful
Let me ask some really basic questions here.

What is the difference between an fpga development board and just an 
fpga board?

Also, since most of my code that I want to speed up has been ported to 
64 bits years ago what is there if any difference between 32 bit abd 64 
bit linux aspertains to an fpga?

I know these may seem elementary to some, but to someone who is very new 
to fpgas they are not.

Any help appreciated.

Thanks in advance.


Respectfully,


Newport_j

Author: Grendel (Guest)
Posted on:

Rate this post
0 useful
not useful
Hi,

to 64 Bit:
This only matters for PROGRAMS that run on a processor inside the FPGA - 
like Cortex-A9 Cores (32 Bit) in Zynq.


In the FPGA fabric for actual data processing, you can use ANY 
resolution you want. Also 62 bits, or e.g. 144 bits if you need it. Just 
a question of # of ressources you want to throw at the problem.


> What is the difference between an fpga development board
> and just an fpga board?

Development is - obviously - for development purposes.
Other Boards might be intended to go into customer products or be used 
directly in a productive environment (such as a PCIe accelerator Card).

Author: Klaus Falser (Guest)
Posted on:

Rate this post
0 useful
not useful
James Yunker wrote:
> Also, since most of my code that I want to speed up has been ported to
> 64 bits years ago what is there if any difference between 32 bit abd 64
> bit linux aspertains to an fpga?

FPGAs are completely different to microprocessors, so speeding up an 
algorithm by simply bringing the code to FPGA will not work.
A FPGA can implement a microprocessor which can than run Linux, but this 
processor and Linux wll be much slower than a high end microprocessor,

An FPGA brings you only an advantage if implement part of the algorithm 
in hardware.
An example would be a cryptographic operation:
Instead of having a sequence of xor and shift operations executed by the 
processor you have a hardware unit, which is loaded with 1024 bit and 
all the shifting, xor-ing is done in a few clock cylces by hardware.
But even in that case you have to ponder the overhead time needed to 
laod the hardware unit.

Reply

Entering an e-mail address is optional. If you want to receive reply notifications by e-mail, please log in.

Rules — please read before posting

  • Post long source code as attachment, not in the text
  • Posting advertisements is forbidden.

Formatting options

  • [c]C code[/c]
  • [avrasm]AVR assembler code[/avrasm]
  • [vhdl]VHDL code[/vhdl]
  • [code]code in other languages, ASCII drawings[/code]
  • [math]formula (LaTeX syntax)[/math]




Bild automatisch verkleinern, falls nötig
Note: the original post is older than 6 months. Please don't ask any new questions in this thread, but start a new one.