Searching \ for '[PIC] Anybody using Mikro Pascal for PIC?' in subject line. ()
Make payments with PayPal - it's fast, free and secure! Help us get a faster server
FAQ page: techref.massmind.org/techref/microchip/devices.htm?key=pic
Search entire site for: 'Anybody using Mikro Pascal for PIC?'.

Exact match. Not showing close matches.
PICList Thread
'[PIC] Anybody using Mikro Pascal for PIC?'
2006\08\25@105918 by Bob Axtell

face picon face
I came across Mikro Pascal (it has C as well). Anybody using it?
I read the manual last night, and I was very impressed. Its only
$125 for the pro version.

--Bob

2006\08\25@123124 by stef mientki

flavicon
face
Bob Axtell wrote:
> I came across Mikro Pascal (it has C as well). Anybody using it?
> I read the manual last night, and I was very impressed. Its only
> $125 for the pro version.
>  
No, but did you look at JAL ?
> --Bob
>
>  

2006\08\25@125735 by Bob Axtell

face picon face
stef mientki wrote:
> Bob Axtell wrote:
>  
>> I came across Mikro Pascal (it has C as well). Anybody using it?
>> I read the manual last night, and I was very impressed. Its only
>> $125 for the pro version.
>>  
>>    
> No, but did you look at JAL ?
>  
>> --Bob
>>
>>  
>>    
>
>  
I missed it. I have used Wouter's JAL a couple of years ago. That it?

--Bob

2006\08\25@133915 by Wouter van Ooijen

face picon face
> I missed it. I have used Wouter's JAL a couple of years ago. That it?

That's the grandfather. There is a new child now, you could call it
Kyle's Jal, Stef Mientki is the godfather.

Wouter van Ooijen

-- -------------------------------------------
Van Ooijen Technische Informatica: http://www.voti.nl
consultancy, development, PICmicro products
docent Hogeschool van Utrecht: http://www.voti.nl/hvu


2006\08\25@141204 by Timothy Weber

face picon face
Bob Axtell wrote:
> I came across Mikro Pascal (it has C as well). Anybody using it?
> I read the manual last night, and I was very impressed. Its only
> $125 for the pro version.

I used it for a couple projects.  Seemed nice, and I normally prefer
Pascal, but I've settled on BoostC now.  My observations on the mikro*
family were:

- Simulator very limited compared to MPSIM

- Not compatible with MPLAB - you have to use it in its own IDE as far
as I recall

- Generated code is very large compared to other compilers (on a small,
biased sample of code)

This was as of October of last year.
--
Timothy J. Weber
http://timothyweber.org

2006\08\25@142933 by Jack Smith

picon face


Bob Axtell wrote:
> I came across Mikro Pascal (it has C as well). Anybody using it?
> I read the manual last night, and I was very impressed. Its only
> $125 for the pro version.
>
> --Bob
>  

Bob:

I purchased the Pascal compiler late last year and tried to make it play
with my 18F4620-based Z90 panadapter project
(http://www.cliftonlaboratories.com) and abandoned it because it was too buggy
to use. I love Pascal and had great hopes for the compiler, but the
Mikro folks seemed too focused on new features and not enough on fixing
serious bugs in the current releases.

I have not downloaded an update for it in some months, so perhaps some
of the worst bugs have been sorted out.

I instead wrote the code in Mecanique's BASIC compiler, Swordfish. (This
is not your grandfather's BASIC, it's really an excellent product with
about all the bells and whistles one could want.) I found that  
Swordfish (a) produces faster and smaller code than Mikro Pascal did in
my head-to-head comparisons and (b) even when I was alpha testing
Swordfish, it was at least as stable (better, actually) than the
supposed 5th series commercial release version of Mikro Pascal.
Swordfish info at http://www.sfcompiler.co.uk/swordfish/

[My original idea was to parallel develop Swordfish and Mikro Pascal
versions of the code, as a backup in the event that Swordfish had
problems, as at that point it was only an early alpha test release.
After several weeks of parallel development, I found that the Swordfish
compiled code consistently worked, or if it it didn't, the bug was fixed
within a day or two, and that the supposed stable commercial Pascal
compiler was seriously broken and that it was not being fixed. Hence, I
dumped the Pascal plans and went 100% Swordfish, which turns out to have
been an excellent decision.]

Usual disclaimer -- my relationship with Swordfish's developer is that
I've been an alpha and beta tester of the Swordfish compiler and have
been highly impressed with what I've seen. No financial links, etc.

Jack

2006\08\25@144921 by Bob Axtell

face picon face
Timothy Weber wrote:
{Quote hidden}

That's the kind of info I like to hear.

--Bob

2006\08\25@145013 by Bob Axtell

face picon face
Jack Smith wrote:
{Quote hidden}

Thanks so much for the note.

--Bob

{Quote hidden}

2006\08\25@145144 by Bob Axtell

face picon face
stef mientki wrote:
> Bob Axtell wrote:
>  
>> I came across Mikro Pascal (it has C as well). Anybody using it?
>> I read the manual last night, and I was very impressed. Its only
>> $125 for the pro version.
>>  
>>    
> No, but did you look at JAL ?
>  
>> --Bob
>>
>>  
>>    
>
>  
Can you point me to a link for JAL? I have two applications where I'd
like to try it,
PIC12F629 and PIC16F88.

--Bob

2006\08\25@151044 by Wouter van Ooijen

face picon face
> Can you point me to a link for JAL? I have two applications where I'd
> like to try it, PIC12F629 and PIC16F88.

http://pic.flappie.nl/ check jal v2

Wouter van Ooijen

-- -------------------------------------------
Van Ooijen Technische Informatica: http://www.voti.nl
consultancy, development, PICmicro products
docent Hogeschool van Utrecht: http://www.voti.nl/hvu


2006\08\25@152357 by Byron A Jeff

face picon face
On Fri, Aug 25, 2006 at 09:12:42PM +0200, Wouter van Ooijen wrote:
> > Can you point me to a link for JAL? I have two applications where I'd
> > like to try it, PIC12F629 and PIC16F88.
>
> http://pic.flappie.nl/ check jal v2

Is Source available? I notice there isn't a Linux package AFAICT.

BAJ

2006\08\25@152746 by Tamas Rudnai

face picon face
> Can you point me to a link for JAL? I have two applications where I'd
> like to try it,

http://www.voti.nl/jal/
http://pic.flappie.nl/
http://jal.sourceforge.net/

Basically follow the jal webring or google on it -- there couple of very
interesting projects and examples based on jal.

Tamas


On 25/08/06, Bob Axtell <spam_OUTengineerTakeThisOuTspamneomailbox.com> wrote:
{Quote hidden}

> -

2006\08\25@160525 by Timothy Weber

face picon face
Jack Smith wrote:
>
> I love Pascal and had great hopes for the compiler, but the
> Mikro folks seemed too focused on new features and not enough on fixing
> serious bugs in the current releases.

Oh yeah - I remember that too!

I think if all the features they already have worked really well, it
would be a terrific suite.
--
Timothy J. Weber
http://timothyweber.org

2006\08\25@160656 by Xiaofan Chen

face picon face
On 8/25/06, Byron A Jeff <.....byronKILLspamspam@spam@cc.gatech.edu> wrote:
> On Fri, Aug 25, 2006 at 09:12:42PM +0200, Wouter van Ooijen wrote:
> > > Can you point me to a link for JAL? I have two applications where I'd
> > > like to try it, PIC12F629 and PIC16F88.
> >
> > http://pic.flappie.nl/ check jal v2
>
> Is Source available? I notice there isn't a Linux package AFAICT.
>

Yes the source is available.
http://oase.uci.kun.nl/~mientki/data_www/pic/jal_v2_release.html

The Source (freeBSD license) is available at:
http://www.casadeyork.com/jalv2src.zip

I believe you can build the binary under Linux since it is the main
development environment for the author. I remember I built it
under Ubuntu 5.10 with some minor tweaks for the Makefile
but I have no access to Linux right now (under training in US).

Regards,
Xiaofan

2006\08\25@160751 by Stef Mientki

flavicon
face

> Is Source available? I notice there isn't a Linux package AFAICT.
>
>  
http://www.wattystuff.net/tiki/tiki-index.php?page=JAL+V2+PRESS+RELEASE&PHPSESSID=bcaf097144de3f1bc584a1be49df8690

(hope this link doesn't break)
Stef

2006\08\25@161238 by Wouter van Ooijen

face picon face
> Is Source available?

AFAIK yes.

> I notice there isn't a Linux package AFAICT.

dunno. but ask on the jallist!

Wouter van Ooijen

-- -------------------------------------------
Van Ooijen Technische Informatica: http://www.voti.nl
consultancy, development, PICmicro products
docent Hogeschool van Utrecht: http://www.voti.nl/hvu


2006\08\25@163725 by Wouter van Ooijen

face picon face
> > I notice there isn't a Linux package AFAICT.
> dunno. but ask on the jallist!

I obviously understimated the alertness of some people...

Wouter van Ooijen

-- -------------------------------------------
Van Ooijen Technische Informatica: http://www.voti.nl
consultancy, development, PICmicro products
docent Hogeschool van Utrecht: http://www.voti.nl/hvu


2006\08\28@122916 by Padu

face picon face
From: "Bob Axtell"
> Timothy Weber wrote:
>> I used it for a couple projects.  Seemed nice, and I normally prefer
>> Pascal, but I've settled on BoostC now.  My observations on the mikro*
>> family were:
>>
>> - Simulator very limited compared to MPSIM
>>
>> - Not compatible with MPLAB - you have to use it in its own IDE as far
>> as I recall
>>
>> - Generated code is very large compared to other compilers (on a small,
>> biased sample of code)
>>
>> This was as of October of last year.
>>

> That's the kind of info I like to hear.


I've been using it since version 2. It was very buggy in the earlier
versions, and I've chosen mp because of my familiarity with Delphi and
because I'm a software engineer instead of an EE. After using delphi, java
and c# I avoid at all costs having to return to asm (although I had to a few
times). Most of the bugs are gone now, unless you are using a newly
developed feature or library. I agree that the hex file may not be optimal
in performance and/or size, but for the things I am doing it is good enough.

I'm happy with it, and the time it saved me with the libraries and the
shorter learning curve from my pascal background far outweights its
shortcomings.

Cheers

Padu

2006\08\28@125447 by Bob Axtell

face picon face
Padu wrote:
{Quote hidden}

It appeals to me, but apparently it was very buggy before, and NOBODY
except you likes it.

Like you, I love Delphi and Borland Pascal. Pascal is like writing a
love letter; it flows easily from
the heart. I may give it a shot, it only costs $125 or so.

--Bob

2006\08\28@130912 by Jack Smith

picon face

>>  
>>    
> It appeals to me, but apparently it was very buggy before, and NOBODY
> except you likes it.
>
> Like you, I love Delphi and Borland Pascal. Pascal is like writing a
> love letter; it flows easily from
> the heart. I may give it a shot, it only costs $125 or so.
>
> --Bob
>  

Bob:

If you do, please give us a report on what you find, and in particular
whether you find it too buggy for serious use.

The free version (I assume it's still available) will generate up to 2K
code and may give you a sense for how good/bad it is.

As I said, like you, I love Turbo Pascal and really wanted to program
18F-series PICs with Mikro Pascal, but, I also wanted the code to work
and not require me to wonder why a := b*c sometimes yields different
results than a := c*b, depending on the relative values of b and c.

Jack

2006\08\28@150723 by Timothy Weber

face picon face
Padu wrote:
> I've been using it since version 2. It was very buggy in the earlier
> versions, and I've chosen mp because of my familiarity with Delphi and
> because I'm a software engineer instead of an EE. After using delphi, java
> and c# I avoid at all costs having to return to asm (although I had to a few
> times). Most of the bugs are gone now, unless you are using a newly
> developed feature or library. I agree that the hex file may not be optimal
> in performance and/or size, but for the things I am doing it is good enough.

This is great to hear - I'm another Delphi cultist so perhaps I'll give
it another look when I get a chance.

Though - The hex file was, according to my notes, *twice* as large as
one generated from substantially similar code (except for what needed to
be changed for porting) by BoostC.  Again, that was only a single code
sample.  So I hope they've done some optimization too - or that that
sample was unusual.
--
Timothy J. Weber
http://timothyweber.org

2006\08\28@151518 by Victor Fraenckel

flavicon
face
*I agree with Padu. I began using mP at version 2.1 on my first PIC
application ever. There were bugs in that version, but the development
team, with the help of active help forum members,  has been constantly
eliminating them and adding new features. The current version is 6.0 and
it is mainly in the new features where bugs occur. mE also had a mikroC
and a mikroBasic version and it is _my sense_ that the code generator
for all compilers is the same.
*

*I don't think you can go wrong with mP, especially if you test drive
the free version to get your feet wet!
*

*I am  a  long  time  Pascal  user going back to Pascal on a DEC PD1170
and the original Turbo Pascal.
*

*Vic
*

*
____________________________________________________________________________________________*

*Victor Fraenckel
KC2GUI
victorf ATSIGN windreader DOT com**
*


Padu wrote:
{Quote hidden}

2006\08\28@155545 by Bob Axtell

face picon face
Victor Fraenckel wrote:
> *I agree with Padu. I began using mP at version 2.1 on my first PIC
> application ever. There were bugs in that version, but the development
> team, with the help of active help forum members,  has been constantly
> eliminating them and adding new features. The current version is 6.0 and
> it is mainly in the new features where bugs occur. mE also had a mikroC
> and a mikroBasic version and it is _my sense_ that the code generator
> for all compilers is the same.
> *
>
> *I don't think you can go wrong with mP, especially if you test drive
> the free version to get your feet wet!
>  
I am going to try it, and see how it does with a small F88 project.

> *
>
> *I am  a  long  time  Pascal  user going back to Pascal on a DEC PD1170
> and the original Turbo Pascal.
>  
Lotsa people never knew... did you know that Borland sold 3,000,000
copies of TP3.0 in Europe?
That's a LOT of users... consequently, there is a lot of PASCAL CODE in
Europe, especially
in Germany.

--Bob


{Quote hidden}

2006\08\28@161015 by Bob Blick

face picon face

> Though - The hex file was, according to my notes, *twice* as large as
> one generated from substantially similar code (except for what needed to
> be changed for porting) by BoostC.

Perhaps there was different padding in the hex file - was the actual
rom/ram usage double?

Also, if it was a small enough test, the differences in startup routines
could account for it, yet be insignificant in a larger program.

It'd be interesting to look at disassembled code from different compilers.

Cheerful regards,

Bob


2006\08\28@161053 by stef mientki

flavicon
face
Padu wrote:
{Quote hidden}

wouldn't it be nice if we had some benchmark programs to compare the
different compilers,
but I guess it's (almost) impossible.
And now that's the first advantage I see for C.
(btw for those who didn't know I'm a Delphi/JAL fan(atic) )

Stef

2006\08\28@162640 by Victor Fraenckel

flavicon
face
Bob,

Let us know how you make out using mP with a F88

Vic

*____________________________________________________________________________________________*

*Victor Fraenckel
KC2GUI
victorf ATSIGN windreader DOT com**
*

Bob Axtell wrote:
{Quote hidden}

2006\08\28@164224 by Wouter van Ooijen

face picon face
> Lotsa people never knew... did you know that Borland sold 3,000,000
> copies of TP3.0 in Europe?
> That's a LOT of users... consequently, there is a lot of
> PASCAL CODE in
> Europe, especially
> in Germany.

I used it a lot (TP2, TP3, TP4, TP5). The WISP.EXE tool is in TP (I
think TP6).

Wouter van Ooijen

-- -------------------------------------------
Van Ooijen Technische Informatica: http://www.voti.nl
consultancy, development, PICmicro products
docent Hogeschool van Utrecht: http://www.voti.nl/hvu


2006\08\28@173651 by Jan-Erik Soderholm

face picon face
> did you know that Borland sold 3,000,000
> copies of TP3.0 in Europe?

And it was written by Anders Hejlsberg
from Denmark... :-) He later joined Borland
and was "the architect for all versions of the
Turbo Pascal compiler and the first three
versions of Borland Delphi", according to
Wikipedia...

Jan-Erik.




2006\08\28@184612 by Padu

face picon face
Then he left borland (as most of the really smart guys from the golden age
of delphi) and went to microsoft and he was the main guy behind .net and
c#....
IMHO, the guy is the best software engineer/architect in the world.


{Original Message removed}

2006\08\28@191322 by Timothy Weber

face picon face
Bob Blick wrote:
>> Though - The hex file was, according to my notes, *twice* as large as
>> one generated from substantially similar code (except for what needed to
>> be changed for porting) by BoostC.
>
> Perhaps there was different padding in the hex file - was the actual
> rom/ram usage double?

I think so.  The note from my log says "Ported LedDriver to mikroC.  Wow
- uses twice as much ROM!"

The descendant of this program - pretty similar, I think - uses 925
words in BoostC.  As I recall, it was also around 1K in PICC Lite, and
around 2K in mikroC.

I assumed at the time that the code generator was common between mikroC
and Pascal.

> It'd be interesting to look at disassembled code from different compilers.

I should have done a more detailed analysis, but I needed to move ahead.
 I remember glancing at the source and thinking "Wow, THAT's wasteful!"
but I can't remember what 'that' was.
--
Timothy J. Weber
http://timothyweber.org

2006\08\28@194450 by Bob Blick

face picon face

> The descendant of this program - pretty similar, I think - uses 925
> words in BoostC.  As I recall, it was also around 1K in PICC Lite, and
> around 2K in mikroC.

Sounds like BoostC and HiTech are both about as efficient as you can get,
without user intervention.

When I put LCDTerm on the web I wanted to make sure it would compile with
PICCLITE, and noticed that it used about 5% more ROM than the paid-for
version of HiTech C. Ram usage was the same between the two.

Cheers,

Bob


2006\08\28@202853 by Timothy Weber

face picon face
Bob Blick wrote:
>> The descendant of this program - pretty similar, I think - uses 925
>> words in BoostC.  As I recall, it was also around 1K in PICC Lite, and
>> around 2K in mikroC.
>
> Sounds like BoostC and HiTech are both about as efficient as you can get,
> without user intervention.

That's been my experience with BoostC - I can usually hand-tune critical
pieces in assembler and gain a bit, but in general it looks like they
have a good optimizer.

> When I put LCDTerm on the web I wanted to make sure it would compile with
> PICCLITE, and noticed that it used about 5% more ROM than the paid-for
> version of HiTech C. Ram usage was the same between the two.

Interesting!  Not the way I'd slice those features if I were looking to
hook people with the free version, but maybe there was a good reason.
--
Timothy J. Weber
http://timothyweber.org

2006\08\28@212149 by Bob Axtell

face picon face
Jan-Erik Soderholm wrote:
>> did you know that Borland sold 3,000,000
>> copies of TP3.0 in Europe?
>>    
>
> And it was written by Anders Hejlsberg
> from Denmark... :-) He later joined Borland
> and was "the architect for all versions of the
> Turbo Pascal compiler and the first three
> versions of Borland Delphi", according to
> Wikipedia...
>
> Jan-Erik.
>
>
>
>
>  
Well, I remember TP3 with a tear in my eye. It was a truly GREAT
program.

--Bob

2006\08\28@234708 by dbwood

picon face
I've been using HI-TECH PICC for a long time, also, and I find that it
does generate very nice code. My two biggest complaints with it has
been:

1) The documentation has had errors in it that have caused some
confusion. The errors that I found I reported to HI-TECH and
[hopefully] the corrections have made it into the manual by now.

2) The compiler insists on maintaining pointers in "C". That is, there
is no way for the programmer to specify to the compiler (preferably
with a #pragma) that it should use the FSR register for the pointer
storage only. Instead, the compiler always generates code that
maintains a "C" version of the pointer (and storage FOR that pointer),
and loading the FSR register every time. This means that you HAVE to
hand code FSR usage in ISRs, either with linked in assembly code or in-
line, if you want tight code for your ISR (or other code).

Aside from this, it has served me well.

Douglas Wood

{Original Message removed}

More... (looser matching)
- Last day of these posts
- In 2006 , 2007 only
- Today
- New search...