Przemek Klosowski wrote:
> On Wed, May 29, 2013 at 7:22 AM, Charles Steinkuehler <
> [email protected]> wrote:
>
>   
>> The PRU can do a great job with step/dir generation.  Anything that
>> has encoders should really be running with FPGA hardware and something
>> like the Mesa boards.
>>
>>     
>
> Why do you  say so? PRU is running at 200MHz and can read and write GPIO
> every 5ns. I think that's close to what the FPGA is capable of. Of course
> FPGA can then process these in  parallel using independent hardware units,
> but still.
>
>   
5 ns is only one instruction.  Do you think it can read, say, the 3 
signals from 4 encoders
and keep count, all with only one instruction?  Not possible.  It would 
probably take
20 -30 instructions for each encoder channel, and possibly quite a bit more.
Assuming one PRU was dedicated to encoders, that would mean sampling the
encoders at about a 1 MHz rate.  This is still fantastic (if my totally 
off-the cuff
guess at how many instructions it takes is anywhere near correct) and an
incredible improvement on reading encoders via the LinuxCNC encoder
HAL component.  I would not be astonished to find out that it actually
takes 200 instruction per encoder, reducing the rate to a 4 us sampling 
period.
Still, not bad at all, and will handle many hobby-level machines.

Jon


------------------------------------------------------------------------------
Introducing AppDynamics Lite, a free troubleshooting tool for Java/.NET
Get 100% visibility into your production application - at no cost.
Code-level diagnostics for performance bottlenecks with <2% overhead
Download for free and get started troubleshooting in minutes.
http://p.sf.net/sfu/appdyn_d2d_ap1
_______________________________________________
Emc-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/emc-users

Reply via email to