CCS vs IAR choice with TI RF module

[SuperDave]

I'm debating using a PIC16F1828 to control a TI CC2500. I know and love MPLAB/PIC. Or, I could get the TI CC2510 which integrates the TI CC2500 with the TI MSP430, then get the IAR IDE and learn a whole new world but presumably IAR's IDE is set up for the combination.

Is it easier to integrate the 16F1828/CC2500 or learn the IAR? Board space is not a problem. The development kits of course assume IAR.

Has anyone fought this or a similar battle? One forum indicated that the CC2500 had problems. Guidance appreciated.

Dave

[temtronic]

It all depends on what your time is worth, and how soon the project is due.

The 'learning curve' for a new processor, language, IDE can be very, very steep. Maybe you can afford to blow a couple of months getting 'kinda familiar' with the new chip....not me.

I've been with PICs for 25 years, since the 68HC11 debacle, and have never been disappointed with PICs or the CCS compilers.
Getting too old for the 'new fangled' chips, heck I stay with 3-5 different PICs as I can do all that I need to with just them!

You've acquired a wealth of onhands knowledge about PICs why 'jump ship' especially if after 2,3 months you have to abandon it and come back to what you KNOW !!?

[PCM programmer]

This is really a Google question. Search for this:

[bkamen]

Hey Tem,

What would you consider the HC11 debacle?

I'm curious... (PM me if you'd like since it's OT)

-Ben
_________________
Dazed and confused? I don't think so. Just "plain lost" will do. :D

[asmallri]

[temtronic]

Ok...I'll post here as it is WHY I've been using the CCS compiler since V2.542.
Short story... back in mid80's, I designed a new product (hackerproof remote control), decided HC11 (then new and THE micro) was was to go. Got a sample, books, etc. Needed few for R&D real tests, tried to order them, was told by disti that I wasn't on 'the list'. Unless I was buying in 10,000 piece lots. I was NEVER going to be on 'the list'. Saw an ad about Microchip seminar <1 hour from me. Great, I went, was very impressed, ASKED about availability 16C84 and 16C71 were 'new' then!! They supplied me with samples, books, etc. Sometime later came across CCS ad in Circuit Cellar Ink, phoned them, asked couple of questions, gave them what was left on my CC. I've NEVER looked at another family of micro since then. Frankly it's the dynamic duo. PICs are small and mighty, easy to understand, even the assembler AND the CCS C compiler with examples, this forum and their support staff allowed this 'old dog' to learn something new!

Jay
aka Temtronic

[Ttelmah]

I suspected it was going to be the compiler debacle. I was another caught by this, I was using several other processors for embedded work, and assembler at this point for the PIC. Tried CCS, and found it so simple for basic jobs that despite also using several other chips still, I will always think of CCS and a PIC as the easiest solution for fast small jobs.

Best Wishes

[bkamen]

Speaking of compilers...

I know CCS isn't perfect (what compiler is?)....


But I had to do a project with XC8 and it's been horrible.

I've found 4 bugs in the compiler and then experienced another that can't be replicated with a simple example because it's based on code-size that I submitted to Microchip to solve.

They have failed miserably.

Their new model is "pay for yearly updates and maybe the bug will be fixed".

And their concept of "pay for optimization" is burning them. They are so busy coding into the compiler when or when not to encode trash instructions, that it's simply not the best the compiler can be.

One of the bugs I found was a direct result of their stupid optimization model. (inserting garbage/duplicate instructions) The bug was a missing MOVLB instruction that ended up modifying the wrong variable in the wrong page of RAM.

Yea. Fun.
_________________
Dazed and confused? I don't think so. Just "plain lost" will do. :D