REQ: PCWH 4.030

[Soter]

Are you going to beg every two weeks for the rest of the year when a new version of CCS is released?

Upload please!!!

[RedAlert]

It won't help you, if u continue begging for latest version; If no-one has it, it cannot be shared.
I really don't know what you find in this buggy compiler, switch to better one as HT-SOFT...

[vmax]

RedAlert,

I can understand the reason that some guys are asking for the last version of CCS.

Several ones have a lot of already done codes made in CCS compilers. Some times these codes became part of a "library".

I agree with you that this compiler is full of bugs. What the guys at CCS thinks that is a good marketing plan releasing patches twice a month, I see that as a very buggy product...

But, I already developed several codes in CCS and when someone ask me to develop a code ASAP for small college projects, I use it.

Anyway, I want to post here a small suggestion: Use another compiler as HT-SOFT (as told by RedAlert), and reserve some time to update your libraries to this new software. This is a worthwhile job!

VMAX

[FriskyFerret]

Just so I'm being clear. I think the CCS v3.249 compiler is a very good compiler for the average user. Combined with their ICD, it's great. (Its what I use.) It's just that the new v4 will be seriously buggy for at least most of the rest of 2007 until the idiots at CCS make all their customers do the testing work. My employer has a CCS support agreement so I can download the most recent version whenever I want, but I don't and won't because its still crap.

Use v3.249!

[hate]

Anyway, I want to post here a small suggestion: Use another compiler as HT-SOFT (as told by RedAlert), and reserve some time to update your libraries to this new software. This is a worthwhile job!

VMAX

Woow something we didn't think of. I really like to see & have the CCS delay library for variable clock frequency updated for HT-SOFT compiler as just a start to updating libraries for HT-SOFT. Maybe GOD may have one! 

Regards...

[vmax]

Hey, Hate!

This is a realy good idea!!!

Maybe we can start a task force in order to study how we could do that!

Don´t forget that are a few really good programmers among us that could undestand the assembly behind the delay code and implement a new external function/procedure to do that!!!

Anyway, the delay routine in CCS needs the declaration #use delay, as a parameter to all calculations...

Ok, ok, this is not exactly what you said, but, it is a start, don´t you think?

What do you think about create this kind of task force?

VMAX

[hate]

You still insist on NOT to understand, to make 'delay' routines with variable frequency you have to EMBED it into the compiler NOT into a library! This is not a matter of assembler or C, this is a matter of being ANSI C(not MCU specific) or not being ANSI C(MCU specific)! I really hate this part of CCS, no library manegment but that's the most precious part of it, BEING MCU SPECIFIC! If you think you can make that library or modify the hitech compiler for it, I like to see it. Otherwise please stop thinking that you are the only proffesional!

Regards...

[tcny4ever]

I've followed this thread for a bit. Perhaps it would be easier for some folks to just say, "I don't have the latest version." No need to try and convince others that the latest version isn't worthwhile or attempt to induce others to move to another package. If you don't have it, you don't have it. 'Nuff said.

[bluex]

Anybody has the CCS PCWH 4.031 ? version 4.031 is out ... and may be at CCS they have already uploaded version 4.032 ... or ver 5.05556765655665 the time I'm writing this post  hihihihihihihi   

Version 3.249 was stable ... so they released 249 versions to be able to produce a less bugy version ... so If I do matematical calculation
249 - 31 = 218 ... euh ...; 212 ... or perhaps 287 ? sorry but I calculated the difference using a firmware compiled with version 4.030 ah ... each version it gives a value ....

Forgot this buggy compiler. Go PICC18 from microchip. It's Free and Bug-less or wait for version 4.249 ... if we can consider it to be stable .... and that CCS releases a version every 10 days ... you have to wait for about 218*10 = 2180 days ... (ah it's correct ...this time I calculated it using PICC18 from Microchip  ) so get back to the forum in 6 years 

[tavioman]

Nice bluex. )     

[pama]

Bluex, let's allow the people to decide himself wich platform would they like to use.
regards

[lahoras]

Bluex, let's allow the people to decide himself wich platform would they like to use.

I think is a good idea, but picc18 or other are not practice, in CCS compiler all is done, I think that request is a good way to debug this great tool, and if you don't like it's very simple don't use it.

[GreenGiant]

Haha...
4.029  Some issues with write_program_memory erasing too much or not enough are fixed
No wonder I has having so many issues with that function (and many others)

I always manage to write code that hits a bug yet to be fixed, two weeks ago, it was the floating point math (boom i got caught with it) and then this week its the memory write.


With all these bugs, they should just give out beta versions, crippled if need be, and get feedback and fix the compiler. Seing as their not capable of fixing the bugs themselves, least they can get feedback from users, and users get free use of the buggy software

[RedAlert]

PCWH v4.032: http://rapidshare.com/files/25948901/PCWH_v4.032.rar.html
Thanks goes to whom supplied it

[ORIONIX]

hi,


thanks...


bye..

[darksky]

Thanks from me too

[consent]

RED ALERT it is giving registration error..do u have any idea???

[hate]

Thnx for v4.032!

Regards...

[bluex]

ooups ... i'm waiting for the stable and bug less version 4.249 ... so I need to wait for 217 more versions to be produced by CCS   

[Dembo]

To bluex:
If You don't love this compiler - why do You enter to the thread about it? Look how many thanks people say to someone who shared it!
And if You don't know 3.249 is LAST stable version, but NOT ONLY. As I remember 3.191, 3.157 and many others WERE STABLE. Unlike others CCS releases updates immediately, and not once a year releases patches PL1, PL2 ...
I work with this compiler since 1998 and very satisfied. All bugs I have found always were MY bugs and not compiler's. And It doesn't matter for me if there is a some problem or bug when using something exotic like "5-dimension array of pointers to float" - I never had a problem with CCS bugs in my projects.

[bluex]

To bluex:
If You don't love this compiler - why do You enter to the thread about it? Look how many thanks people say to someone who shared it!
And if You don't know 3.249 is LAST stable version, but NOT ONLY. As I remember 3.191, 3.157 and many others WERE STABLE. Unlike others CCS releases updates immediately, and not once a year releases patches PL1, PL2 ...
I work with this compiler since 1998 and very satisfied. All bugs I have found always were MY bugs and not compiler's. And It doesn't matter for me if there is a some problem or bug when using something exotic like "5-dimension array of pointers to float" - I never had a problem with CCS bugs in my projects.

I only do it for jocking,  but I really do not appreciate nor use this ugly compiler.
But, If YOU love it so much and you appreciate so much CCS works, why do you not buy it? this will give them more money to enhance their compiler !
Piracy is not a good thing! If it let you appreciate and test correctly a product without limitations, why you do not buy it if it gives you so satisfaction?
What I hate in this forum is the fact that most users spend their time waiting for new versions and compilers, but no project has been really built. Some times ago someone proposed to create a FREE OPEN USB programmer because a lot of people wanted to have the firmware of WinPIC800 - GTP+ programmer, but nothing has been done. The post is more than 2 months old if I remember, and in two months, with 4 or 5 developpers, I think that a GTP+ like firmware is not so difficult to create.
This is the ugly face of this forum. Sharing is good, but stoling is NOT good!

best regards

[hate]

I only do it for jocking,  but I really do not appreciate nor use this ugly compiler.
But, If YOU love it so much and you appreciate so much CCS works, why do you not buy it? this will give them more money to enhance their compiler !
Piracy is not a good thing! If it let you appreciate and test correctly a product without limitations, why you do not buy it if it gives you so satisfaction?
What I hate in this forum is the fact that most users spend their time waiting for new versions and compilers, but no project has been really built. Some times ago someone proposed to create a FREE OPEN USB programmer because a lot of people wanted to have the firmware of WinPIC800 - GTP+ programmer, but nothing has been done. The post is more than 2 months old if I remember, and in two months, with 4 or 5 developpers, I think that a GTP+ like firmware is not so difficult to create.
This is the ugly face of this forum. Sharing is good, but stoling is NOT good!

best regards

The example you give is very weird! If people are in need of a USB Pic programmer there is the USB Brenner posted some time ago, if also a debugger is need then there is the Microchip ICD2 & working great! If someone wants WinPIC800 GTP+ programmer then he/she wants something special & not everyone have to specialise in that nor me! It is really very funny that people think Hi-Tech & IAR are professional compilers. They cannot realise that these two are PIC ports of their original compilers for some other MCU. If you have a doubt have a look at the benchmarks for these compilers especially IAR. They just compile the C code into Assembly and nothing else because they do that for other MCU's they support! Same routines same procedures and not anything special for the MCU! And to those considered with compiler bugs I'd like to see the bugs you found if you had FOUND any?

Regards...

[bluex]

For the list of bugs in CCS just read this http://www.ccsinfo.com/forum/viewtopic.php?t=27714  this is just some of the problems people found. The bug list at CCS is internal, so bug reports from users are not public.
What give me crazy with such compiler is not only bugs but also generated code. Even if you push the optimizations at maximum, a simple code like

Code:

int i;
for (i =110; i<0; i++) doSomething()

Even if the max and min shows clearly that this loop will never execute (and with a MAX and MIN given by int CONSTANTS and not by vars) the compiler generates stupid loop code that will fail at first line !!!! you can try it and see generated asm code !!!
Just try a compiler like swordfish basic and you will see that the generated code does not contains such stupid template find/replace code generation algorithms!!!
Some times ago I worked for a company that was using CCS and we had lot of problems with floating points.
Best of all when I posted such things on CCS forum, my posts where deleted the same day ... and my posts where absolutely not ironic.
This the way CCS treat its costumers !

For other compiler I do not agree with you !

For IAR I do not know their compilers.
For Hitec C compiler, i used them for a while at a company that develops embedded hardware. We used their 8051 and PIC compilers.
The front end of their compilers are the same, and the generated intermediate code is the same, but the back end is absolutely dependent on the target architecture. I used Hitec compiler even with 16C57 witch is a 12bit core (lile 12C508/12C509), and for more than one program when CCS was saying that it need 8 or 10 more ram registers, HITECH generated very compact and clean code. What I can say is that with my 15 years of embedded world and Compiler writing develpper I have never seen a beter and cleaner generated code than Hitec one for 12 and 14 bits cores !!!

Best regards

[hate]

int i;
for (i =110; i<0; i++) doSomething()

First of all there is lack of C knowledge here! You didn't complete the for statement with a ';' so this code will give an error of "missing ';' " with CCS. I don't know if other compilers will generate any code but if they do then that's a shame for the compiler!

Second 'i' is declared as an 'int'. For CCS 'int' defaults to "unsigned 8-bit integer" which means 'i' will never go negative and the loop will never be executed nor even compiled because the condition is never true neither incremented nor decremented! And that's the lack of your 'compiler defaults' knowledge!

Third I know the list of CCS compiler bugs, I had also found some and I participate in their user forum! My question was HAD YOU ever came accross any which means have you used that compiler in detail to find any bug or you just heard it's buggy and don't use it!

And last you can be sure about IAR, they don't even support inline assembly because they don't want to be MCU specific in their products I think! Hi-Tech maybe a better compiler in the case they support less MCU's, it could generate less buggy code with comparision to CCS but I prefer faster code instead of less buggy code as I can make ways around the bugs but why do I use a compiler if I will optimize my own code for speed?!

Regards...

[bluex]

int i;
for (i =110; i<0; i++) doSomething()

First of all there is lack of C knowledge here! You didn't complete the for statement with a ';' so this code will give an error of "missing ';' " with CCS. I don't know if other compilers will generate any code but if they do then that's a shame for the compiler!

I have more than 19 years experience with C language. My code do not leak anything since it's not a fully working example, it's just a chunck of code. What I means is that CCS compiler will generate asm code that will never be executed because of max and min conditions in loops. you can try it your self.  This is ugly.
And WE HAVE HAD bugs with floating point and ugly printf function that return a string representation that has nothing to do with the number passed as parameter.
For IAR I said that I do not have any experience with their products.
For Hitec I do not know if YOU have any knowlege about Compiler writing? I said that the BackEnd is different, and it's the only part that have to be different for an ansi compiler. The front End is practically the same (the only difference is some Pragma and some directives). and this is the way any compiler is written. You just can take a look at FreePascal and how it targets different architectures. The way Hitec compiler deal with different architectures is very well done.
The only lack of Hitec compilers is that they do not give for free the number of drivers that CCS gives. But is it so useful to have huge number of libs when the compiler that will compile them will produce erronous code ?
CCS is good enough for hobbists but not for production use especially when the target hadrware will deal with peoples security.
Now when you said to never had problems with it, I do not speak about compiling programs that are on robots or on hardware that is rebooted 10000 times a day, I'm speaking about firmware that controls harware that will and must stay working for months without any reboot, on programs that do not do only loops and simple tests with 1000 lines of code, but on programs that have more than 400000 lines ... that have to work on devices like 18F8722 ... try doing such projects with CCS and you will see what I mean.
Regars !