Install files for V3.2.5.5 Compiler, required please

[IanHemmingsen]

Do to rework required on an old piece of code, I find I need to install version 3.2.5.5 of the Compiler I used at the time
Does anyone have a copy of the files or able to point me to where I can download them.
FYI the reason I need to go back, is with current release version 3.5.0.6 "LookUp" is broken, and with current beta 3.5.2.7, "LookUp is fixed again, but I am having other code problems.
I have raised code problems with Les

regards
Ian

Hi, its great to see you visiting our forum. Why not try Proton Compiler for FREE? Download the FREE version of Proton Compiler, Its called Amicus18 and its available from HERE Already using proton Compiler??? Get rid of these pesky messages... get LICENSED USER STATUS

[John Drew]

Ian,
I have 3.5.2.6 for XP if that will help. 43MB
John

Edit: Just found 3.5.2.5 msi file if you can use that, it's 40 MB

[IanHemmingsen]

Hi John,
Need to go back further in time:-( Need V3.2.5.5
Your Ver 3.5.x.x is fairly recent
Have just found V3.2.5.2 here, and have got it running...I think... but not convinced yet
Might have to make another Virtual Machine or find an old computer somewhere to run it on
I suspect it was patched/replaced files to get to V3.2.5.5 from V3.2.5.2 back in those days

regards
Ian

[John Drew]

Sorry Ian,
I slipped a digit!
Good luck.
Cheers
John

[Les]

Keeping the original compiler version that an important project is compiled with should be a normal practice on all platforms and languages. It's simple to do with Proton, just zip up the ProtonIDE folder with a relevant name. Earlier versions need to zip the Crownhill and ProtonIDE folders.

Ian You say that you've bought up an issue with me, however, looking through the Beta section this is actually what I got from you:

"Short story:- source built with V3.5.2.6 works. Uninstall & install V3.5.2.7 and code almost behaves same.
Late Friday afternoon and time to go home, so I can't give any more clues as to what is going wrong yet.
Program is using conditional compiles [$ifdef] and .inc files. My gut feel lead at this stage
Nothing in compiler changes seems to apply
Its a big program, so will take a while to find whats up.
Looking at code and what's physically happening don't make sense "

Which, you must agree, is not much to go on, even if you had attached the offending program.

[IanHemmingsen]

>Which, you must agree, is not much to go on, even if you had attached the offending program.
I agree entirely, you would need the complete software, hardware, and sensors/simulators etc
But I would appreciate some ideas/thoughts on how to isolate this problem, as I am the only one with the above requirements

If you look at my history, I have uncovered a few obscure compiler issues over the years, so bare with me
This one, I can't pin down yet and give you the offending line(s) of code or cause

The problem seems to be compiler related, as changes to program size, change behaviour in this section of code, and compiles with V3.5.0.6 work correctly with this section. [But unfortuately elsewhere "LookUp" section of code broken]

The problem area of code who's behaviour depends on compiler version, debugging RS232 dumps being present (one example given here), total code size
That is, I could make problem come and go, by compiling with different amounts of debug dumps being present. I don't think the dumps are the cause as code doesn't work when no dumps compiled but are shifting the code around in memory location. The variables are not being flagged as crossing any page boundarys
Micro is 16F916. Total code size around 5-6K

;##Input:- DigFilterIn [Word]
;##Changes:- DigFilDiff [Word], DigFilterSign [Bit]
;##Output:- DigFilterAcc [Word]
DigFil: ;
;enter with DigFilterIn = last sample value (word form)
$ifdef DF1 ; These three lines are an example of RS232 debugging dumps put after each calculation line
Clear GIE : RSOut "\r\na",Dec DigFilterIn : If GIE_Flag = 1 Then Set GIE: ;Stop interrupts so RS232 works and re-enable if they were enabled. GIE_Flag is a bit variable
$endif
'DigFilDiff = DigFilterIn * 16 - DigFilterAcc ;Calc difference from current smothed value
DigFilDiff = DigFilterIn << 4 ;Calc difference from current smothed value (x16)
DigFilDiff = DigFilDiff - DigFilterAcc ;Calc difference from current smothed value
DigFilterSign = DigFilDiff_High.7 ;Get Sign of the diference (Topbit of word is signbit)
DigFilDiff = Abs DigFilDiff ;Magnitude the difference
'DigFilDiff = (DigFilDiff + 3 )/4 ;1/4 of the difference, or 1 or 0 (Dither adjust)
DigFilDiff = DigFilDiff + 3 ;1/4 of the difference, or 1 or 0 (Dither adjust)
DigFilDiff = DigFilDiff >> 2 ;1/4 of the difference, or 1 or 0 (Dither adjust)
If DigFilterSign = 0 Then ;+ve change
DigFilterAcc = DigFilterAcc + DigFilDiff ;Update smothed value
Else ;-Ve
DigFilterAcc = DigFilterAcc - DigFilDiff ;Update smothed value
EndIf
DigFilterOut = DigFilterAcc >> 4 ;/16 Generate new smoothed output value from filter
Return

The digital filter works (V3.2.5.5 & V3.5.0.6), but variable behaviour with current beta V3.5.2.7

[bill9399]

Ian.

Probably has no bearing on your problem whatsoever, but, have you tried using the latest version of the fuse configurator (V2?) to update these settings to comply with latest MCHip nomenclature changes?

There have been a few changes in these from the compiler version you seek to the latest version of the compiler.
I had problems with re-compiling an old previously working OK program of mine, with latest compiler, getting spurious errors, until I updated the config settings using the above.

[IanHemmingsen]

Hi bill.
Have started a thread in beta section with my code issue.
Good point. Made me realise I will need to go back to the assembler used at the time also.
Will see if new thread in beta area solves problem

[See_Mos]

Hi Ian,

Do you have an approximate date of the install?

I keep quite a few backups from my documents and downloads so there is a good chance that I will have a copy somewhere.

[roger]

Hi,

Does anybody have a copy of the compiler/ide for Protonbasic 3.4.0.7, it would have been released about October 2009?

My archived copy seems to be corrupted.

Would appreciate any help.

roger