PIC 18F not so popular yet?

On a PIC one cannot imagine it because it is impossible or close to impossible. On something like an AVR, I have done it, and it makes some applications much easier and more elegant. It is easy to make re-entrant interrupt routines. One also do not have to bend over backwards to not cause the compiler to use "complex" routines such as diviide or multiply or a switch statement having more than about 8 cases.

Regards Anton Erasmus

I'm sure the marketing dept at Microchip will repeat this refrain :) I believe any $5+ region single chip uC, is going to have to compete with other $5+ uC, and in 2004, that means the PIC18 and dsPICs are going to compete with each other, as well as the new ARM_uC's, plus the DSP_uC offerings from Motorola and TI, as well as the high performance Analog uC.

Price points mean it also comes up against the growing families of ARM_uC, and the 32 bit /100 MIPS DSPs from TI.

-jg

I never have and I cannot imagine I ever will use reenterant code in a PIC like microcontroller application.

Mike Harding

Absolutely not! For the most part, the 18F is just as easy as a 16F to configure (in fact many registers are designed for backwards compatibility) and you get to avoid a lot of bank switching.

There are a few electrical characteristics that might get you in trouble (the A/D conversion comes to mind) but otherwise it would be crazy for a beginner to learn the 16F and "upgrade" later.

--
Ben Jackson

http://www.ben.com/

If Microchip did that it would probably make the dsPIC redundant as well. Since Philips have released the LPC ARM processors, I think the dsPIC has pretty much missed the market window. Atmel are also not far behind Philips with low-cost ARM devices.

Persoanally I think the Microchip 16F series chips are only really suitable for assembly code programming. For high level language there are so many better choices.

regards, Johnny.

What's the multiply (or MAC) performance on the ARM7 core?

Best regards, Spehro Pefhany

--
"it's the network..."                          "The Journey is the reward"
speff@interlog.com             Info for manufacturers: http://www.trexon.com
Embedded software/hardware/analog  Info for designers:  http://www.speff.com

I made the exact same switch, necessary on a design for faster processing (5 MIPS PIC16 to 10 MIPS PIC18) (PIC18F452 can use a totally internal PLL circuit to multiply an on-board

10MHz to an on-chip 40MHz, back to 10MIPS at 4 clocks per cycle.) Pinout (PLCC) ISTR was identical, perhaps apart from the clock input pin?

Since using this, I've done a couple more designs with PIC18. I've not wanted to use a PIC16 again :-)

In two designs I used a sector of PIC18 Flash program memory for the PIC to write calibration values to. (PIC18 can erase and reprogram its program memory, I believe at any operating supply voltages, not just at 5V)

No PIC16 missing banksel bugs any more.

Existing PIC16 assembler code needs vey few changes to convert to PIC18.

Paul T