 |
 |
| View previous topic :: View next topic |
| Author |
Message |
MikeW
Joined: 15 Sep 2003
Posts: 184
Location: Warrington UK
|
|
 Posted: Tue Aug 15, 2006 1:56 pm |
 |
|
I agree with PCM.
I have used the CCS compiler for years, and knew that V4 would be pain until much much later.
I am a "beta tester", but it even crashes my dual core laptop.
no other application has managed to lock up my dual core before.
dont get me wrong, I still think that the CSS compiler is the best value, and MOST IMPORTANT OF ALL, has the most active community.
I will continue to use 3.249 for the moment, and maybe run it throughV4 just to see.
I think it is criminal that CCS has let V4 into the wild at this stage.
didnt they learn anything from the v2 to v3 outrage ?
Mike |
|
 |
Adrianf
Joined: 02 Oct 2003
Posts: 2
Location: North Wales (UK)
|
|
| Posted: Wed Aug 16, 2006 7:12 am |
|
|
I feel gutted that I could be so stupid. As a user for the last couple of years I've been very pleased with the product, content you might even say.
When I heard about a new version I thought "great! - improvements and advances - and surely it'll be sorted after the beta trials".
Like a fool (not an angel) I rushed in and installed the released 4.001 with excitement - and promptly suffered from all the problems listed above.
I feel let down, betrayed and now impatient - how long am I going to have to wait 'til I get what I was expecting?
If only I'd read this thread before .......
Come on CCS - this is no way to treat your loyal customers! |
|
|
Ttelmah
Guest
|
|
| Posted: Wed Aug 16, 2006 8:48 am |
|
|
It is 'nice', to have the file on the website, since it allows me to download it (hurrah), but it really should have been flagged as a 'beta' release. Most people 'in the know' about the way that the CCS versions work, will be fully aware that the latest version really is a beta, but to anyone without the experience to know this, the version may well be fairly 'depressing'...
As it currently stands, the code to me is a fairly low 'alpha', possibly just pushing 'beta' status.
Best Wishes |
|
|
rwyoung
Joined: 12 Nov 2003
Posts: 563
Location: Lawrence, KS USA
|
|
| Posted: Wed Aug 16, 2006 8:56 am |
|
|
Question about compiler vs ide performance.
Browsing through this sticky topic, I notice most of the complains are in relation to the download manager and ide.
How have you found the actual compiler output to be, especially if you are using MPLAB as your editor?
_________________
Rob Young
The Screw-Up Fairy may just visit you but he has crashed on my couch for the last month! |
|
|
Ttelmah
Guest
|
|
| Posted: Wed Aug 16, 2006 10:22 am |
|
|
The new version is better.
The earlier release had some fairly bad problems. The first for me, was that it generated an output file, in several cases, that failed to program. It appeared to write two sets of values to the same area of memory, which the programmer then complained about. This seemed to be fixed, before I could raise it. The next was that it gave an error 'expecting a comma', where ever I used the CCS 'shortcut', of giving a constant string to a function expecting an integer. This now appears fixed.
The current version, has compiled two projects for me, that failed on the older V4 release, and the code generated, looks reasonable. The code is larger than from the V3 compiler (not much, but a little - I am currently comparing the listings to find out why).
I'd actually say that the publication on the website, about co-incides, with the version just about becoming a reasonable 'beta'. It actually appears to 'work' for basic compilations now, so the next test will be to play with the new features, and see which of these actually work...
Best Wishes |
|
|
Dinesh
Joined: 15 Jul 2005
Posts: 22
Location: UK
|
|
| Posted: Wed Aug 16, 2006 12:11 pm |
|
|
I have compiled my current project that is on V3 to V4. It compliled with zero errors and produced helpful and meaningful warnings - an improvement on V3.
I have tested the target product and it appears to be fine.
Tried the Document->Generate Documentation and managed to produce a document based on default template.
This is much welcomed as it will help speed up formal documentation.
Tried a few of keywords and they seem to work.
One problem that I encountered was multi-line comments lost cr/lf and ended up as one confusing paragraph. It needs to maintain text formating.
Has any one experienced this 
I look forward to fixes for this - its a productive utility 
_________________
W2k sp4 512MB (Motherboard:ASUS A7S8X)
Last edited by Dinesh on Wed Aug 16, 2006 5:08 pm; edited 1 time in total |
|
|
treitmey
Joined: 23 Jan 2004
Posts: 1094
Location: Appleton,WI USA
|
|
| Posted: Wed Aug 16, 2006 3:08 pm |
|
|
Looks like the sym table construction has changed
project compiled in 3.249 but run out of ram in V4
I can't get 16bit pointers
| Code: | #include <16f877.h>
#device *=16
#use delay(clock=16000000,restart_wdt)
#fuses hs,nowdt,noprotect,nolvp//we have a wdt now
#use rs232(baud=19200,xmit=PIN_B3,invert,stream=DEBUG)
#case
#zero_ram
int8 A0[16];
int8 A1[16];
int8 A2[16];
int8 A3[16];
int8 A4[16];
int8 A5[16];
int8 A6[16];
int8 A7[16];
int8 A8[16];
int8 A9[16];
int8 A10[16];
void main(void)
{
setup_adc_ports(NO_ANALOGS);set_tris_a(0);set_tris_b(0);set_tris_c(0);set_tris_d(0);set_tris_e(0);
restart_wdt();
fprintf(DEBUG,"Starting U11\n\r");
fprintf(DEBUG,"Done U11\n\r");
while(1);
} |
from the sym file
| Quote: | Compiler Settings:
Processor: PIC16F877
Pointer Size: 8
ADC Range: 0-255
Opt Level: 9
Short,Int,Long: 1,8,16
Case Sensitive: Yes
|
UPDATE: 16 bit pointers (#device *=16) has now been fixed in Ver4.004
Last edited by treitmey on Mon Aug 21, 2006 9:31 am; edited 1 time in total |
|
|
Dinesh
Joined: 15 Jul 2005
Posts: 22
Location: UK
|
|
| Posted: Wed Aug 16, 2006 6:14 pm |
|
|
| Dinesh wrote: | Tried the Document->Generate Documentation and managed to produce a document based on default template.
This is much welcomed as it will help speed up formal documentation.
Tried a few of keywords and they seem to work.
One problem that I encountered was multi-line comments lost cr/lf and ended up as one confusing paragraph. It needs to maintain text formating.
Has any one experienced this
I look forward to fixes for this - its a productive utility |
This was due to feature I missed that is specified the current CCS manual. e.g /**:CHANGES
Missing the : was the cause - my error.
However, the formatted text is striped of leading spaces which can mess up your multi-line comments.
Perhaps there should be an option to strip or not strip the leading spaces.
_________________
W2k sp4 512MB (Motherboard:ASUS A7S8X) |
|
|
Ttelmah
Guest
|
|
| Posted: Thu Aug 17, 2006 2:40 am |
|
|
The code you post Trietmey, compiles fine for me under V4.
It showed another error in the IDE though. If you select to manually create a project (which the 'pop up' help says can be used with an existing .c file, or will create one), it will not allow you to proceed unless the file exists. Normally in V3, I can simply select such a manual project, a dummy filename, and then paste the test code 'in' from the post on the group. With V4, I have to actually create the file first...
Gives:
| Code: |
ROM used: 164 words (2%)
Largest free fragment is 2048
RAM used: 24 (14%) at main() level
25 (14%) worst case
Stack: 1 locations
|
When compiled with V4.
Best Wishes |
|
|
wulffert
Joined: 17 Sep 2003
Posts: 6
Location: New York, NY
|
| My comments |
| Posted: Thu Aug 17, 2006 7:06 am |
|
|
All,
I will be blunt, I have used this product for 3 years, and I have to say V4.001 SUCKS!  It is loaded with bugs, forget about using the ICD-U40. The ICD debugger windows have all sorts of bugs, I gave up after a couple of hours. I am seeing the same problems that you all are. I went back to using 3.xx, my maintenance period is almost up, I hope they extend it for the 'Crapware" they produced. I paid good money, sure it is cheaper than the competitors, but I expected a working release. Not a Beta version, heck an Alpha version.
Rant Over.
Mark |
|
|
Guest
|
|
| Posted: Thu Aug 17, 2006 11:54 am |
|
|
My debugger window will not even open at all in v4.004. And now the debug toolbar
in my v3.249 shows up as a tiny sliver that cannot be resized. |
|
|
wulffert
Joined: 17 Sep 2003
Posts: 6
Location: New York, NY
|
|
| Posted: Thu Aug 17, 2006 12:02 pm |
|
|
| Anonymous wrote: | My debugger window will not even open at all in v4.004. And now the debug toolbar
in my v3.249 shows up as a tiny sliver that cannot be resized. |
Yeah, the window gets sized all the way to the right, you have to rezize that window every time you start the IDE. |
|
|
Guest
|
|
| Posted: Thu Aug 17, 2006 12:49 pm |
|
|
I've been WAITING for x64 ICD support and was told that it would work in V4.... now it seems ICD doesn't work even in 32bit systems so I'm STILL screwed.
Really looking into microchips compiler now, going to check an see if hitech supports ICD clones and dsPic
Why don't they hire another engineer!? The cost of a good engineer's salary has GOT to be way less then the cost of pissing off all your customers. |
|
|
Ttelmah
Guest
|
|
| Posted: Thu Aug 17, 2006 3:00 pm |
|
|
Unfortunately, 'publishing' V4, is definately not a good idea. At present, parts may just about reach a 'beta' standard, but most still rings as an 'alpha' level product...
4.004, seem for the first time to integrate properly with MPLAB.
It still retains the I/O error 32, if you compile a file, and then try to compile it again. Having to exit the ICE, every time you want to change anything in the code, is foul...
The 'fixed' arithmetic, does not support multiplication yet (let alone division...).
In quite a few cases, if you have a project 'in use', and then try to exit it, you will get a message that the "process cannot access the file, because it is in use by another process". The only program running is pcw itelf, and only the one copy...
The automatic indexed variable length constant strings, appears not to work at all. I have tried a simple 'sprintf' from an array of such strings, and get nothing.
Best Wishes |
|
|
Storic
Joined: 03 Dec 2005
Posts: 182
Location: Australia SA
|
|
| Posted: Fri Aug 18, 2006 6:15 pm |
|
|
Hi,
V4.004 it is good to see that the wizard is working, I have tried with the modbus lib however "modbus.c" is not available as yet. (looking forward to try it out when avilable) 
I also see you have included the PIC24F (limited) DSPIC33 in the wizard, I gues this means the compiler for the PIC24F and dsPIC is getting closer. 
Andrew
_________________
What has been learnt if you make the same mistake?  |
|
|
|
|
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum
|
Powered by phpBB © 2001, 2005 phpBB Group
|
FAQ 
Search
Register
Profile 
Log in to check your private messages 
Log in
