Parts of the Propeller Tool were written with proprietary tools which Parallax isn't allowed to release. Apparently the Propeller Tool was written before the big push to make things like this open source.
IIRC, Parallax (Ken) has expressed regret about not having complete control over the source code.
You bet we have regrets about not being able to release the entire source code.
It has one licensed component for the edit control which I chased down for over a year and was unable to get approval from the new corporate owner of the source code, even though it's well over 15 years old.
Fred, what is it you want to see done with the Propeller Tool?
Luckily we don't need the Propeller Tool source code any more.
It I understand correctly the original Spin compiler used by the Propeller Tool was written by Chip in x86 assembler. God only knows why (It seems Chip is such an extreme propeller head that he can churn out such code in assembler faster than using a high level language ) That code was given to Roy Eltham who created a Spin compiler in C++ from it. The source for that compiler is here: http://code.google.com/p/open-source-spin-compiler/
Then we have the PropellerIDE, a Propeller Tool in C++ by Jazzed. That code is now in the charge of lamestation and is available here: https://github.com/lamestation/PropellerIDE
Thanks to all these great efforts we now have Spin compiler and IDEs that run anywhere you can compile them, Linux, Mac, Windows, RaspberryPi even the MIPS processors used in cheap WIFI routers.
Thanks for your explanations. I just thought that everything Propeller has been open sourced. If I recall correctly, the Propeller Tool has been made in Delphi; one of my friends has been doing Delphi programming for 8 years so just by curiosity I wanted to take a peek at the source code with him. I still use the Propeller Tool a lot on my Windows machines... takes less than 1 minute to install and works really great.
I feel something is wrong with the swarm of different IDE for the Prop, and it is probably confusing for a newcomer : BST, SimpleIDE, PropellerIDE, PZST, msrobots Editor17, blockly prop, programming from ipad, programming from chromebooks...
On Parallax website, there is little or no mention of the others IDE. Still, you can find the Propeller Tool under the support tab, or on the various Propeller products download page. You really have to know whats happening on these forums to get the full picture.
Is Parallax going to give more visibility to the PropellerIDE? Will they support it? Are we just waiting for v1.0?
Interesting idea pik33. The Propeller Tool sources could be released sans the proprietary component.
Is it of any thing other than of historical interest though, a curiosity for code nerds?
I would hazard a guess that the missing component would be the editor.
The actual compiler is written in x86 assembler.
So as a cross-platform open source code base all we have is an empty shell of a Lazarus application with some menus and buttons on it. No use to man nor beast.
On the other hand, speaking of code nerds, I have no interest in the Pascal stuff but I'd love to see how Chip wrote the compiler in assembler.
I feel something is wrong with the swarm of different IDE for the Prop, and it is probably confusing for a newcomer : BST, SimpleIDE, PropellerIDE, PZST, msrobots Editor17, blockly prop, programming from ipad, programming from chromebooks...
On Parallax website, there is little or no mention of the others IDE
What could possibly be wrong with your customers putting in lot's of work to find new and interesting ways of working with your product? Time to worry is when nobody bothers.
Admittedly some of those efforts, BST, HomeSpun, were spurred on because the Propeller Tool was not open source. That's history now and has been debated to death already.
I don't know how much confusion there may be. BST is not on the radar much now a days as it is apparently abandoned and unsupported. PZST never was much on the radar. Editor17 is clearly a very early experiment. A A Chrome Book tool does not really exist as yet.
There is something wrong that as far as the Parallax web site is concerned SimpleIDE and PropellerIDE don't exist. They do not appear on the downloads page and the search box does not find them. A google search will get you so simpleide on learn.parallax.com but that is the C world not Spin.
It seems to be taking forever for an open source tool to hit the front page. SimpleIDE has been working for ages now.
Interesting. Looks like a Direct Digital Synthesis signal generator.
What platform are you on? Windows? Because the assembler that Free Pascal generates itself is in the format I have above. Which it would seeing as FPC uses the GCC assembler on Linux.
That is a signal generator written in asm because fpc was simply too slow.
About the asm syntax in Lazarus, compiler options -> assembler style. Then you can select Intel or AT&T. Works in every x86/x64 platform
I am not the Pascal guy in the company. Whenever I compile our Free Pascal applications I do it from the command line. But that might come in useful one day.
I have a rule in life: Never write software in a language whose name begins with "P": Perl, PHP, Python, Pascal, Prolog, PL/M, Postscript, Processing... there are many more. For some reason all the "P" languages are miserable
Pascal, as it is today, is good - you can program what you want in the way you want so I don't want any more (The Spin is even better - maybe we can do a spin compiler for Windows, where if you want the procedure in a new thread, you simply write cognew(procedure)
Pascal, as it is today, is good - you can program what you want in the way you want so I don't want any more (The Spin is even better - maybe we can do a spin compiler for Windows, where if you want the procedure in a new thread, you simply write cognew(procedure)
You sort of already have a Spin compiler for Windows. You can use spin2cpp to translate Spin code to C++ and then compile that with a Windows C++ compiler.
I too would love to see the x86 pasm assembler code. I wrote a mini emulator in 486 assembler specifically targeting the fast 486 instructions. But that was >20 years ago. Remember PharLap anyone?
FredBlais
Posts: 370
Comments
Parts of the Propeller Tool were written with proprietary tools which Parallax isn't allowed to release. Apparently the Propeller Tool was written before the big push to make things like this open source.
IIRC, Parallax (Ken) has expressed regret about not having complete control over the source code.
It has one licensed component for the edit control which I chased down for over a year and was unable to get approval from the new corporate owner of the source code, even though it's well over 15 years old.
Fred, what is it you want to see done with the Propeller Tool?
Thanks,
Ken Gracey
It I understand correctly the original Spin compiler used by the Propeller Tool was written by Chip in x86 assembler. God only knows why (It seems Chip is such an extreme propeller head that he can churn out such code in assembler faster than using a high level language
Then we have the PropellerIDE, a Propeller Tool in C++ by Jazzed. That code is now in the charge of lamestation and is available here: https://github.com/lamestation/PropellerIDE
Thanks to all these great efforts we now have Spin compiler and IDEs that run anywhere you can compile them, Linux, Mac, Windows, RaspberryPi even the MIPS processors used in cheap WIFI routers.
Now, what is it you wanted to do?
I feel something is wrong with the swarm of different IDE for the Prop, and it is probably confusing for a newcomer : BST, SimpleIDE, PropellerIDE, PZST, msrobots Editor17, blockly prop, programming from ipad, programming from chromebooks...
On Parallax website, there is little or no mention of the others IDE. Still, you can find the Propeller Tool under the support tab, or on the various Propeller products download page. You really have to know whats happening on these forums to get the full picture.
Is Parallax going to give more visibility to the PropellerIDE? Will they support it? Are we just waiting for v1.0?
Is it of any thing other than of historical interest though, a curiosity for code nerds?
I would hazard a guess that the missing component would be the editor.
The actual compiler is written in x86 assembler.
So as a cross-platform open source code base all we have is an empty shell of a Lazarus application with some menus and buttons on it. No use to man nor beast.
On the other hand, speaking of code nerds, I have no interest in the Pascal stuff but I'd love to see how Chip wrote the compiler in assembler.
So how about it Ken? Spill the source!
Admittedly some of those efforts, BST, HomeSpun, were spurred on because the Propeller Tool was not open source. That's history now and has been debated to death already.
I don't know how much confusion there may be. BST is not on the radar much now a days as it is apparently abandoned and unsupported. PZST never was much on the radar. Editor17 is clearly a very early experiment. A A Chrome Book tool does not really exist as yet.
There is something wrong that as far as the Parallax web site is concerned SimpleIDE and PropellerIDE don't exist. They do not appear on the downloads page and the search box does not find them. A google search will get you so simpleide on learn.parallax.com but that is the C world not Spin.
It seems to be taking forever for an open source tool to hit the front page. SimpleIDE has been working for ages now.
And now the Lazarus has a very good code editor component...
So it is time to rewrite it in x64 assembler
(they often ask me if x64 processors CAN be programmed in assembler... so I always answer, yes, they can...)
.file "test.c" .text .globl main .type main, @function main: .LFB0: .cfi_startproc pushq %rbp .cfi_def_cfa_offset 16 .cfi_offset 6, -16 movq %rsp, %rbp .cfi_def_cfa_register 6 movl %edi, -20(%rbp) movq %rsi, -32(%rbp) movl $200, -4(%rbp) movl $100, -8(%rbp) movl -8(%rbp), %eax movl -4(%rbp), %edx addl %edx, %eax popq %rbp .cfi_def_cfa 7, 8 ret .cfi_endproc .LFE0: .size main, .-main .ident "GCC: (Debian 4.7.2-5) 4.7.2" .section .note.GNU-stack,"",@progbitsLooks like we are all set to go.
Come on Ken where's the code?
function synteza:Tsample; var a:pointer; b:integer; begin a:=@tabl; asm push rax push rbx push rcx push rdx push rbp push rsi push rdi mov rbp,qword ptr a mov rsi,rbp add rsi,4*17000 // 65536: last sin; mov ecx,256 mov edi,0 mov ebx,[rsi] //phase @p001: mov eax,[rsi+12] // freq ramp cmp eax,0 jne @p012 // zero -> new freq in 4 mov eax,[rsi+4] jmp @p013 @p012: add eax,[rsi+4] // add ramp mov [rsi+4],eax // save ramped freq dec dword ptr [rsi+8] // dec time jne @p013 mov dword ptr[rsi+12],0 @p013: add ebx,eax mov [rsi],ebx // new phase to tabl add eax,[rsi+8] //freq ramp mov [rsi+4],eax //save ramped freq mov edx,[rsi+24] // vol ramp cmp edx,0 jne @p002 // zero -> new volume in 16 mov edx,[rsi+16] // volume jmp @p003 @p002: add edx,[rsi+16] // ramp mov [rsi+16],edx // save vol dec dword ptr [rsi+20] jne @p003 mov dword ptr [rsi+24],0 @p003: and rbx,$0000000000FFFFFF // 24-bit phase shr rbx,10 mov eax,[rbp+rbx*4] // sinus imul edx //vol*sin add esi,32 add edi,edx mov ebx,[rsi] //phase dec ecx jnz @p001 @p021: sar edi,15 mov [rbp+4*16385],edi // second channel mov ecx,256 mov edi,0 @p101: mov eax,[rsi+12] // freq ramp cmp eax,0 jne @p112 // zero -> new freq in 4 mov eax,[rsi+4] jmp @p113 @p112: add eax,[rsi+4] // add ramp mov [rsi+4],eax // save ramped freq dec dword ptr [rsi+8] // dec time jne @p113 mov dword ptr[rsi+12],0 @p113: add ebx,eax mov [rsi],ebx // new phase to tabl add eax,[rsi+8] //freq ramp mov [rsi+4],eax //save ramped freq mov edx,[rsi+24] // vol ramp cmp edx,0 jne @p102 // zero -> new volume in 16 mov edx,[rsi+16] // volume jmp @p103 @p102: add edx,[rsi+16] // ramp mov [rsi+16],edx // save vol dec dword ptr [rsi+20] jne @p103 mov dword ptr [rsi+24],0 @p103: and rbx,$0000000000FFFFFF // 24-bit phase shr rbx,10 mov eax,[rbp+rbx*4] // sinus imul edx //vol*sin add esi,32 add edi,edx mov ebx,[rsi] //phase dec ecx jnz @p101 @p121: sar edi,15 mov [rbp+4*16386],edi // popad pop rdi pop rsi pop rbp pop rdx pop rcx pop rbx pop rax end; synteza[0]:=tabl[16385]; synteza[1]:=tabl[16386]; end;Interesting. Looks like a Direct Digital Synthesis signal generator.
What platform are you on? Windows? Because the assembler that Free Pascal generates itself is in the format I have above. Which it would seeing as FPC uses the GCC assembler on Linux.
About the asm syntax in Lazarus, compiler options -> assembler style. Then you can select Intel or AT&T. Works in every x86/x64 platform
Thanks for the tip re: assembler style.
I am not the Pascal guy in the company. Whenever I compile our Free Pascal applications I do it from the command line. But that might come in useful one day.
I have a rule in life: Never write software in a language whose name begins with "P": Perl, PHP, Python, Pascal, Prolog, PL/M, Postscript, Processing... there are many more. For some reason all the "P" languages are miserable
Enjoy!
Mike
There is this "PASM": http://en.wikipedia.org/wiki/Parrot_assembly_language which of course begins with a "P" and is miserable.
@mindrobots,
"PHORTH"
I made a promise to the forum to no longer comment on stack based, syntax free and confusing languages:)
@Clusso, Yes. We have to keep nagging at Ken and Chip.