G'day all
 A couple of (theoretically) simple questions, which I haven't found
definitive answers to elsewhere;
 - Normally, the instruction speed is 1/4 that of the crystal (ie.
5MHz with a 20MHz crystal), correct?; with PLL enabled, does that then
mean that the instructions run at the same speed as the crystal?  If
so, any disadvantages?
 - Seems there are limits to the above;  for the chip I am currently
playing with (PIC18F2620)
(http://www.microchip.com/wwwproducts/Devices.aspx?dDocName=en010284),
[1] 
 it seems to say that you can only use PLL for " frequencies up to 10
MHz. A PLL then multiplies the oscillator output frequency by 4 to
produce an internal clock frequency up to 40 MHz"; so does that mean
if a 10MHz crystal is used with PLL, the internal clock runs at 40MHz,
but the 1/4 instruction cycle means the instructions run @ 10MHz?  So
a 10MHz crystal with PLL gives an instruction speed of 10MHz, as
opposed to the 2.5MHz without PLL?
 - MIPS.  A newer, cheaper chip that I am looking at with similarish
specs, the PIC24FJ64GA104
(http://www.microchip.com/wwwproducts/Devices.aspx?dDocName=en536114),
[2]
 says it can run @ "Up to 16 MIPS Operation @ 32 MHz"; how are these
figures figured out?
 With the 18F2620 it says up to 10MIPS, which I can relate to 10MHz
crystal with PLL (4x 10 = 40, 40/4 = 10), are the 24F chips running at
one instruction every _2_ cycles instead of every 4 maybe?
 As per my previous posts, I'm attempting to play basic sound from
internal memory, it's working OK but I think I need to squeeze some
more speed out for room to breath, especially if I'm going to be doing
any other calculations at the same time (sound is 8bit 16kHz mono, so
interrupting and processing every 62.5us; actually interrupting every
31.2us and doing the routine every second interrupt, closest method I
could find to get the right speed).
 So instead of using a 20MHz straight crystal, am I better off with a
10MHz combined with PLL?
 Oh, and for those playing at home, an issue I was having writing
values via variable to PWM turned out to be a compiler error; it was
putting in the same value no matter what the variable held.  Fixed
with a compiler update.
 Thanks again
 Lee
 

Links:
------
[1]
http://www.microchip.com/wwwproducts/Devices.aspx?dDocName=en010284),
[2]
http://www.microchip.com/wwwproducts/Devices.aspx?dDocName=en536114),
-- 
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist