View previous topic :: View next topic |
Author |
Message |
guy
Joined: 21 Oct 2005 Posts: 297
|
MPLAB-X IDE vs CCS PCWH IDE |
Posted: Fri Jul 29, 2016 4:54 am |
|
|
[if this has been discussed recently please refer me to the right thread - I couldn't find one]
It seems I would have to move on to new chips unsupported by MPLAB 8.92 and at some point we will all...
Of course I'm staying with CCS compiler but I have to choose between the heavy, possibly unstable MPLAB-X (PLEASE correct me if I'm wrong!) and CCS IDE which I know nothing about. And I'd like to use my PICKit3. Where do things stand at the moment between the two competing IDEs?
I like a light, stable IDE for simple debugging and to quickly get the project out the door. No whistles and bells please. |
|
|
drolleman
Joined: 03 Feb 2011 Posts: 116
|
|
Posted: Sat Jul 30, 2016 1:56 pm |
|
|
With ccs ide and mc programmers / debugging, it doesn't work well. I've tried using icd3 and real-ice with poor results or none at all. I use their programmer when using ccs ide with much better results. As far as X is concerned, if the current management of mplab x stays in place there will be no improvement. They keep trying to fix something that cannot be fixed. The people that promoted the new system need to save face, and the ones that really knew what they are doing are no longer there. Mplab x is clunky at best, and will always be.
ccs ide has been slowly getting better, so purchasing their programmer is probably your best option for new chips. I still use 8.92 when I can. |
|
|
temtronic
Joined: 01 Jul 2010 Posts: 9246 Location: Greensville,Ontario
|
|
Posted: Sat Jul 30, 2016 2:48 pm |
|
|
Perhaps something to consider is why use 'new' PICs ? I'm a diehard fan of 'if it ain't broke, don't fix it' ! Have a good look and see if there is a current PIC that CCS / MPLAB does like. JUst because a PIC is 'new and improved' dosn't mean that an 'older' one won't do the job, unless there's a new peripheral...
Just another way to look at the world.
I'm still using MPLAB 8.86 to pgm PICs using CCS C compilers.
Jay |
|
|
gjs_rsdi
Joined: 06 Feb 2006 Posts: 468 Location: Bali
|
|
Posted: Sun Jul 31, 2016 3:36 am |
|
|
Guy
Can get an example of PIC not supported by MPLAB v8.92?
Best wishes
Joe |
|
|
guy
Joined: 21 Oct 2005 Posts: 297
|
|
Posted: Sun Jul 31, 2016 3:58 am |
|
|
To answer temtronic and gjs_rsdi, the PIC16F18854..PIC16F18877 family offer an amazing collection of peripherals for an exceptionally good price. Aside from all of the standard new & old peripherals there are many small features such as selection of TTL/CMOS inputs per pin*, constant current output for pins (connect an LED without a resistor! or use it as short-circuit-proof output), a more independent ADC (ADC-square) and 35 analog inputs in a 40-pin device (24 for the 28-pin version). These can also be used as touch-sensing inputs.
The '18877 has 56KB of prog. memory and costs as low as $1.27 in 5K qty. for all this. My reaction was WOW!
* the reason I need to migrate is because I have a 16F887-based design and need more prog. memory. Most of the pin-to-pin alternatives changed the TTL input to CMOS and therefore are unsuitable for my design.
If you're looking for a new single chip to use for a variety of designs, I vote for this family. |
|
|
Ttelmah
Joined: 11 Mar 2010 Posts: 19553
|
|
Posted: Sun Jul 31, 2016 4:06 am |
|
|
Big caveat.....
You are in for a nasty surprise, if you think you can drive LED's without a resistor using the current limited drive.....
Read the data sheet:
"Special precautions must be taken when enabling the
current-controlled output drivers. An external resistor
must be inserted in series with the load to dissipate
most of the power. If an external resistor is not used
then the pin circuity will dissipate all the power, thereby
exceeding the maximum power specifications for the
pin and the device."
It's designed to give limited _slew rates_ by limiting the current during switching, not to give a limited current drive, for things like LED's, without a resistor. |
|
|
temtronic
Joined: 01 Jul 2010 Posts: 9246 Location: Greensville,Ontario
|
|
Posted: Sun Jul 31, 2016 5:26 am |
|
|
yup, the 'Devil's in the details'.... so it pays to read all 600+ pages of the datasheet..... what looks good on the first page... sigh..
Have you considered the 18F46K22 ? 64K of mem, lots of peripherals like 2 HW UARTS, 2 I2C. it's been my 'goto' PIC for years. Overkill on 99% of the projects but a 'Swiss army knife' PIC. Though it may cost more per PIC than a 'new', smaller PIC R&D time DOES cost $$$ ! MPLAB8.86 and PICKIT like it BTW!
Also with respect to 'needing' more memory, don't take this wrong, but ever since memory got cheap , it's led to a lot of bloatware or sloppy programming. In the 'good old days' you HAD to be careful about coding. My remote energy control system is in 8KB, handles 512 remotes over miles of copper,printer AND a floppy disk unit as well. I know of a complete, WYSIWYG word processor that was cut into 2KB of code!
If you've got a lot of 'common' routines, perhaps you can 'tweak' here, slim down 'there' and reduce memory requirements?
just food for thought.
Jay |
|
|
guy
Joined: 21 Oct 2005 Posts: 297
|
|
Posted: Sun Jul 31, 2016 6:25 am |
|
|
Couldn't agree with you more Jay. I started with PIC assembly and for years wouldn't upgrade to C. Until I did and hardly ever went back
56KB of memory is enough for a cellular data logger and I can save the external EEPROM, or whatever other applications that require external memory (fonts, bitmaps, audio etc.)
That's what's exciting about this family of PICs - it opens up new possibilities, not just 'more of the same', nor peripherals like the CLC or NCO that seem like they were invented for a specific project and will rarely be used. It seems Microchip could put as many transistors as they wanted regardless of size and $$, so they just load more and more, and we pay less. |
|
|
gjs_rsdi
Joined: 06 Feb 2006 Posts: 468 Location: Bali
|
|
Posted: Sun Jul 31, 2016 4:04 pm |
|
|
Jay is right in my opinion.
I had enough hard time to move from 16F648A to 16F1847 and from 18F2520 to 18F26K22, to learn the new data sheets, the new functions, even they are pin to pin replacements.
And as Jay had said:
Quote: | the 'Devil's in the details'.... |
Best wishes
Joe |
|
|
guy
Joined: 21 Oct 2005 Posts: 297
|
|
Posted: Tue Aug 02, 2016 2:55 pm |
|
|
Is anyone using CCS IDE (PCW) with PIC kit3 ? Is it comfortable? Stable?
No need to respond if you don't use PCW with PICKit3. |
|
|
TimothyCarter
Joined: 08 Aug 2016 Posts: 22
|
PICkit3 with CCS IDE |
Posted: Mon Aug 08, 2016 2:18 pm |
|
|
I've tried to use the PICkit3 with the CCS IDE - short answer it doesn't work. The CCS IDE gives a memory fault error in windows.
I do use MPLAB X v3.35 - very stable in my opinion (for what I'm doing), with CCS v5.061 and the PICkit3. I'm using the PIC24FJ256GB406, which is also a "new" chip (not according to Microchip, but according to CCS) - and CCS has very little support for it, and I have had to contact support to obtain Beta versions just to make the CCS programming tools even program it (let's not even mention Debug).
Sometimes it's nice using "cutting edge" stuff, other times it can be a pain in the a**.
guy wrote: | Is anyone using CCS IDE (PCW) with PIC kit3 ? Is it comfortable? Stable?
No need to respond if you don't use PCW with PICKit3. |
_________________ “Computer science education cannot make anybody an expert programmer any more than studying brushes and pigment can make somebody an expert painter.”
- Eric S. Raymond |
|
|
guy
Joined: 21 Oct 2005 Posts: 297
|
|
Posted: Tue Aug 09, 2016 1:17 am |
|
|
Wow! Just downloading MPLAB-X 3.35, more than half GB! This has to be good
Lucky I upgraded my PC recently. I will report back if I get over the objection stage and start playing around with it... |
|
|
TimothyCarter
Joined: 08 Aug 2016 Posts: 22
|
|
Posted: Thu Sep 29, 2016 12:07 pm |
|
|
I actually had to use MPLABX as the particular part I am using is not available with MPLAB 8 ... the PIC24FJ256GB406. _________________ “Computer science education cannot make anybody an expert programmer any more than studying brushes and pigment can make somebody an expert painter.”
- Eric S. Raymond |
|
|
guy
Joined: 21 Oct 2005 Posts: 297
|
|
Posted: Thu Sep 29, 2016 2:01 pm |
|
|
Reopening the thread, I tried to make the shift to MPLAB-X, went over the tutorials, no use... Debugging icons disappearing, features from MPLAB missing... I went back to good old 8.92, optimized the code and stayed with the old part I used before.
Not sure how long we can keep this attitude... Maybe push Microchip to go back to supporting MPLAB? |
|
|
gjs_rsdi
Joined: 06 Feb 2006 Posts: 468 Location: Bali
|
|
Posted: Thu Sep 29, 2016 9:59 pm |
|
|
Tried MPLABX and was a disaster.
Hope I will never need to change from MPLAB v8.92.
Best wishes
Joe |
|
|
|