How close is the Swordfish 24F compiler?

[normnet]

Quote:
Proton is working on a PIC24 basic version

Norm do you know this for sure?
I would not hold my breath on that.

As follows:
Norm

15th December 2009 04:06
Les
Proton Compiler Developer

Trying to place words in my mouth or force a happening is not going to be succesful.

Workloads have delayed the PIC24 version of the compiler, I have to make a living, and it's not through Proton.

The PIC24 version is still in the offing, but will be sometime next year. They have matured now to a point where they are a viable replacement for standard 8-bit microcontrollers.

There are no plans for a PIC32 version as of yet.
Regards
Les

26th August 2008 03:28
Les
Proton Compiler Developer

I will be looking for Alpha and Beta testers, but they must do some homework on the 16-bit devices first, as I simply don't have the time to tutor on the hardware.

Microchip have extensive, and excellent, datasheets and app notes for the PIC24 and dsPIC devices, so I recommend downloading them now.

Don't get bogged down with the dsPIC DSP functions, as these are seperate to the language, and the PIC24 devices don't have accumulators etc. I would recommend looking at the PIC24 devices first, as these are more conventional microcontrollers, and closer to PICmicros (in operation only). And they now come in handy 28-pin DIP packages for prototyping. Free samples are available from the Microchip web site.

I will let you all know when it's closer to the time of release.
Regards
Les

26th August 2008 02:36
Les
Proton Compiler Developer

Just for the record (Fanie).

The dsPIC and PIC24 instructions and architecture are completely, and utterly, different to the 8-bit micro's. The only things that have stayed the same are SFR names, but even their operation is different.

When Microchip state that they are backwardly compatible, They mean when you use their C compiler and do not perform any code with peripherals.

I am currently working on the Proton24 compiler, that will support dsPICs and PIC24 devices, but even these are different to each other, as there are two flavours of dsPIC, dsPIC30 and dsPIC33. Each 16-bit device has different instruction timings and SFR setups. I'm hoping to have an Alpha version for testing within the next 6 weeks, and the work is progressing nicely for that target date. It will be a seperate compiler and not an upgrade or an add-on.

I cannot, and will not, say anymore at this moment, except that I'm knackered. Late nights coding and early mornings going to work don't make for a relaxing lifestyle at the moment, but I'm sure it's worth it, as the results so far are promising, and the 16-bit devices have some nice features. It's taken quite a time to get familiar with the 16-bit architecture at a low level, but once you get used to it, it actually a pleasure to code in.
Regards
Les

[CharlieM]

Yea and here it is 2010 and there is not a full release of PDS. I don't think a proton version for pic 24 will be released soon.This is just my opinion.

[normnet]

I don't think a proton version for pic 24 will be released soon.

Probably not.

Norm

[Francis]

If/when the Proton one comes out do you think it'll be in the (dare i say it) slightly 'old fashioned' structure of Proton?

[JWinters]

I have to make a living, and it's not through Proton.

How is it that no one can make a living from selling PIC compilers? The market must be so much smaller than I thought.

[Raistlin]

Problem is market to an extent and the peanuts prices take HiTEC C , the lite version is free ok but for the full version its $1495 (£968) for 1 year of updates (updates are extra after that).

Then look at what Dave is charging £99 with no yearly fees. that's nearly a tenth of the price of hitec c , PDS same £99 and no yearly fees.
Take into account the dongle is probably £10 ,, no wonder they make no real cash

[Raistlin]

Oh then you got the aurdrinou assholle makezine titheads eating into that using lousy Atmel

[Francis]

Blimey, this topic gets more response than a technical query about SF coding...
Not sure what that means

[normnet]

If I may steal a little bandwidth a alternative to C for ARM is Armaide.
It's Oberon-07 code is Pascal and is very similar to Delphi (windows apps).

Direct technical support from the author.

Norm

[liak]

I have to make a living, and it's not through Proton.

How is it that no one can make a living from selling PIC compilers? The market must be so much smaller than I thought.

Yes!

- Captain Spock -:

I am now member of an endangered species!

More proudly, we are members of superior lifeforms!

regards,
liak

[rmteo]

....The market must be so much smaller than I thought.

Catering to what is essentially a hobby market is severely limited. Consider that the total number of members of the 3 most popular PIC BASIC forums is only about 13,000 (mikroE, including C, PASCAL and hardware products - 9,258, PicBASIC - 3,399, SF- 394) you get the idea. Even if you assume that every member actually owns/purchased the products (which is highly doubtful), the total dollar amount is still pretty small. In the case of SF, how long will $98,500 (394 x $250) - and probably much less after reseller discounts and the fact not everyone here is a buying member - last?

[Francis]

I must admit that I am genuinely surprised that (apparently) there are less than 400 SF users worldwide.

Is it down to lack of advertising?

For sure , the range of listed 'supported devices' is a bit limited. I realise that some experts here could roll their own and increase this, but people considering buying a compiler may like to get coding on day 1.


And I think the 'Help' could do with a tweak.
If 'hobby level' is a large part of the target audience then more comprehensive Help is required - like Proton.
I can think of a couple of examples where Forum queries were met with 'go look at the libraries and search for yourself' - true, but an extra line or two in Help would have, er..well, helped. Pity the poor newbie.

If the semi-pro or pro market is the target then they'll want more devices supported as default. Smaller Js for example.
Yes, the experienced people can generate the libararies, but I would have thought a pro has more important things to do time-wise i.e. get on with the job they are being paid to do.

But, on the whole, the 'open-ness' of SF is excellent. Whilst I don't pretend to be in the same league as others here, I have been able to do things quickly in SF that would have left me scratching my bonce in Proton or M.E.

I reckon with a bit of polishing and marketing it could be really successful.
Unfortunately it seems to have been abandoned. The only other Lunch-break I know of that has been longer is in Hitch Hiker's Guide

Well, all this is easy for me to say , I can just wave a Magic Wand and don't know the complexities.

[gramo]

I am driven by the word from Steve a while back that Dave will soon be back. Though this is definitely an extended break of official support.

I could not put a number on how many posts/comments/articles I have read where someone has recommended the use of SF over other compilers.

Perhaps the SE limitations allow end users to much flexibility? Consider PDS (50 lines of code) or MikroBasic (2K compiled limit). Such approaches lead end users to commit to product purchasing far sooner then a trial version of SF.

Either way, SF is an extremely solid platform that has made its mark to a lot of talented people - which can only do good in the long run.

Keep sharing and inspiring!

[CharlieM]

Well Here is where I am at. I am looking at CCS C. The compiler is expensive,but it is a full featured compiler that supports a wide varity of pics.even pic 24 and dspics. I am looking at the hardware debugger too. So I can develop and simulate code. I am looking at spending $600 dollars US for this.Support Is great and the release is the full release not in beta or "pre release" .

[rmteo]

CharlieM, check your PM.