View Single Post
  #13   Spotlight this post!  
Unread 23-06-2015, 00:30
AustinSchuh AustinSchuh is offline
Registered User
FRC #0971 (Spartan Robotics) #254 (The Cheesy Poofs)
Team Role: Engineer
 
Join Date: Feb 2005
Rookie Year: 1999
Location: Los Altos, CA
Posts: 803
AustinSchuh has a reputation beyond reputeAustinSchuh has a reputation beyond reputeAustinSchuh has a reputation beyond reputeAustinSchuh has a reputation beyond reputeAustinSchuh has a reputation beyond reputeAustinSchuh has a reputation beyond reputeAustinSchuh has a reputation beyond reputeAustinSchuh has a reputation beyond reputeAustinSchuh has a reputation beyond reputeAustinSchuh has a reputation beyond reputeAustinSchuh has a reputation beyond repute
Re: Talons without libraries?

Quote:
Originally Posted by asid61 View Post
Runs at 100MHz each for 4 cores, but I need to do port read/write tests. It should only take one clock tick per reading, but now I'm wondering how I would report results without losing some functionality. I could just add another XMOS and run encoder readings on 4 cores.
Really? That's a pretty fancy chip.

On the Cortex-M3, it takes 3 cycles to access the AHB bus on the fast GPIO, and on older generation chips, slow GPIO would take north of 10 cycles (if I remember right).

Peripheral accesses tend to be pretty slow, even on embedded chips. CAN receiving code that I have used on a i7 over PCI takes a very long time. The CPU busy waits while that is happening. The result is that it takes about 1 core to read CAN messages (!).