Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > comp.lang.forth > #8403 > unrolled thread

Forth for STM32F4 Discovery eval board

Back to article view | Back to comp.lang.forth

Contents

  Forth for STM32F4 Discovery eval board Reg Beardsley <user@nowhere.org> - 2011-12-28 16:50 -0600
    Re: Forth for STM32F4 Discovery eval board "Elizabeth D. Rather" <erather@forth.com> - 2011-12-28 09:14 -1000
    Re: Forth for STM32F4 Discovery eval board Brad <hwfwguy@gmail.com> - 2011-12-29 08:20 -0800
      Re: Forth for STM32F4 Discovery eval board rickman <gnuarm@gmail.com> - 2011-12-29 16:40 -0800
        Re: Forth for STM32F4 Discovery eval board "Elizabeth D. Rather" <erather@forth.com> - 2011-12-29 15:03 -1000
          Re: Forth for STM32F4 Discovery eval board Paul Rubin <no.email@nospam.invalid> - 2011-12-30 00:39 -0800
            Re: Forth for STM32F4 Discovery eval board rickman <gnuarm@gmail.com> - 2011-12-30 15:22 -0800
              Re: Forth for STM32F4 Discovery eval board Paul Rubin <no.email@nospam.invalid> - 2011-12-30 17:32 -0800
              Re: Forth for STM32F4 Discovery eval board Albert van der Horst <albert@spenarnc.xs4all.nl> - 2012-01-01 13:25 +0000
                2012 - the year of the GA144 Howerd <howerdo@yahoo.co.uk> - 2012-01-01 10:24 -0800
                  Re: 2012 - the year of the GA144 rickman <gnuarm@gmail.com> - 2012-01-02 17:20 -0800
                    Re: 2012 - the year of the GA144 Jan Coombs <jan_2011-02@murray-microft.co.uk> - 2012-01-03 07:41 +0000
                      Re: 2012 - the year of the GA144 rickman <gnuarm@gmail.com> - 2012-01-03 01:27 -0800
          Re: Forth for STM32F4 Discovery eval board Brad <hwfwguy@gmail.com> - 2011-12-30 08:27 -0800
    Re: Forth for STM32F4 Discovery eval board John Smith <user@example.net> - 2011-12-31 05:04 -0600
      Re: Forth for STM32F4 Discovery eval board "Elizabeth D. Rather" <erather@forth.com> - 2011-12-30 12:28 -1000
        Re: Forth for STM32F4 Discovery eval board David Schultz <abuse@127.0.0.1> - 2011-12-30 21:04 -0600
          Re: Forth for STM32F4 Discovery eval board MikeS <mike.smart@talktalk.net> - 2011-12-31 10:30 -0800
      Re: Forth for STM32F4 Discovery eval board David Schultz <abuse@127.0.0.1> - 2012-01-19 18:32 -0600
    Re: Forth for STM32F4 Discovery eval board StephenPelc <stephen@mpeforth.com> - 2012-01-03 17:04 -0800
      Re: Forth for STM32F4 Discovery eval board Albert van der Horst <albert@spenarnc.xs4all.nl> - 2012-01-04 17:18 +0000

Page 1 of 2  [1] 2  Next page →

#8403 — Forth for STM32F4 Discovery eval board

I'm looking for a very basic Forth to use w/ the <$20US Discovery  board 
to develop a software defined radio.

Requirements:

  configurable RS-232 link (address, baud rate, etc)
  assembler for basic ARM Cortex M4 instruction set
  block interface to flash memory
  run from RAM
  load RAM from flash at boot
  commented source code

I've found riscy pygness & some old work at eddiem.com.  Is there 
anything else I've missed?  I'm "retired" so commercial offerings at 
10-20x the cost of the board are out of the question.

I'd prefer a reliable minimal features Forth to a buggy, full of 
features that don't work right implementation.

Have Fun!
Reg

[toc] | [next] | [standalone]

#8404

On 12/28/11 12:50 PM, Reg Beardsley wrote:
>
> I'm looking for a very basic Forth to use w/ the <$20US Discovery board
> to develop a software defined radio.
>
> Requirements:
>
> configurable RS-232 link (address, baud rate, etc)
> assembler for basic ARM Cortex M4 instruction set
> block interface to flash memory
> run from RAM
> load RAM from flash at boot
> commented source code
>
> I've found riscy pygness & some old work at eddiem.com. Is there
> anything else I've missed? I'm "retired" so commercial offerings at
> 10-20x the cost of the board are out of the question.
>
> I'd prefer a reliable minimal features Forth to a buggy, full of
> features that don't work right implementation.

SwiftX from FORTH, Inc www.forth.com supports the ARM with a very 
full-functioned interactive cross-compiler.  It's a mature, commercil 
product with good support and not "buggy".

Cheers,
Elizabeth

-- 
==================================================
Elizabeth D. Rather   (US & Canada)   800-55-FORTH
FORTH Inc.                         +1 310.999.6784
5959 West Century Blvd. Suite 700
Los Angeles, CA 90045
http://www.forth.com

"Forth-based products and Services for real-time
applications since 1973."
==================================================

[toc] | [prev] | [next] | [standalone]

#8426

On Dec 28, 3:50 pm, Reg Beardsley <u...@nowhere.org> wrote:
> I've found riscy pygness & some old work at eddiem.com.  Is there
> anything else I've missed?  I'm "retired" so commercial offerings at
> 10-20x the cost of the board are out of the question.
>
I ran across a blog about using MPE's ARM cross compiler on mbed:
http://toddbot.blogspot.com/2010/04/mbed-cortex-m3-and-mpe-forth.html

The author says the port was easy.

Since you have more time than money, you should see if MPE will give
you the tools in exchange for useful code.

-Brad

[toc] | [prev] | [next] | [standalone]

#8454

On Dec 29, 11:20 am, Brad <hwfw...@gmail.com> wrote:
> On Dec 28, 3:50 pm, Reg Beardsley <u...@nowhere.org> wrote:> I've found riscy pygness & some old work at eddiem.com.  Is there
> > anything else I've missed?  I'm "retired" so commercial offerings at
> > 10-20x the cost of the board are out of the question.
>
> I ran across a blog about using MPE's ARM cross compiler on mbed:http://toddbot.blogspot.com/2010/04/mbed-cortex-m3-and-mpe-forth.html
>
> The author says the port was easy.
>
> Since you have more time than money, you should see if MPE will give
> you the tools in exchange for useful code.
>
> -Brad

I think I tried that route once.  I seem to recall that we didn't come
to terms.  But if you think about it I can see why.  A stranger says
he will add onto your house in exchange for living in your basement.
Are you likely to either expect his work to be good or to want him in
your house?

Rick

[toc] | [prev] | [next] | [standalone]

#8455

On 12/29/11 2:40 PM, rickman wrote:
> On Dec 29, 11:20 am, Brad<hwfw...@gmail.com>  wrote:
...
>> Since you have more time than money, you should see if MPE will give
>> you the tools in exchange for useful code.
>>
>> -Brad
>
> I think I tried that route once.  I seem to recall that we didn't come
> to terms.  But if you think about it I can see why.  A stranger says
> he will add onto your house in exchange for living in your basement.
> Are you likely to either expect his work to be good or to want him in
> your house?

There's also the fact that those of us who make a living from Forth need 
to be able to buy groceries, make house payments, and meet payroll 
because our staff like groceries, etc., too.

Infrequently FORTH, Inc. has exchanged software for services. A couple 
of times it sort-of worked out. I expect MPE has had a similar experience.

Cheers,
Elizabeth

-- 
==================================================
Elizabeth D. Rather   (US & Canada)   800-55-FORTH
FORTH Inc.                         +1 310.999.6784
5959 West Century Blvd. Suite 700
Los Angeles, CA 90045
http://www.forth.com

"Forth-based products and Services for real-time
applications since 1973."
==================================================

[toc] | [prev] | [next] | [standalone]

#8458

"Elizabeth D. Rather" <erather@forth.com> writes:
> Infrequently FORTH, Inc. has exchanged software for services. A couple
> of times it sort-of worked out. I expect MPE has had a similar
> experience.

Greg Bailey of Green Arrays made an interesting proposition in his Forth
Day talk:

   http://www.forth.org/svfig/videos/fd2011/bailey.ogm 

fast forward to 20 minutes into the video.

[toc] | [prev] | [next] | [standalone]

#8506

On Dec 30, 3:39 am, Paul Rubin <no.em...@nospam.invalid> wrote:
> "Elizabeth D. Rather" <erat...@forth.com> writes:
>
> > Infrequently FORTH, Inc. has exchanged software for services. A couple
> > of times it sort-of worked out. I expect MPE has had a similar
> > experience.
>
> Greg Bailey of Green Arrays made an interesting proposition in his Forth
> Day talk:
>
>    http://www.forth.org/svfig/videos/fd2011/bailey.ogm
>
> fast forward to 20 minutes into the video.

I listened to this.  They are willing to let you become part of their
"slave labor" and write an app note in exchange for an Eval board.
But then I guess they are already giving away the software so what
else could they offer?

Interesting that Greg Baily understands that the company will live and
die by the app notes.  I sent him an email a couple of weeks ago
asking if or when they will be publishing app notes on adding USB or
Ethernet support without adding dedicated chips for this.  Most MCUs
have versions with all sorts of interface protocol support.  The GA144
should be able to do much of this in software like USB up to 12 Mbps
and Ethernet at 10 Mbps.  But I don't want to have to roll my own.  I
haven't received a reply yet, but it is the holidays.

The GA chips have a lot of raw potential, in many ways they are
amazing devices.  But the company needs to provide more info, a lot
more info, on how to do useful stuff with them.  As it currently
stands I am looking at designing a board with a GA144 along with a
SiBlue chip as a level converter and possibly an ARM MCU as a
interface device for Ethernet and USB.  I think this shows some of the
current limitations of the GA144.

Rick

[toc] | [prev] | [next] | [standalone]

#8508

rickman <gnuarm@gmail.com> writes:
> I listened to this.  They are willing to let you become part of their
> "slave labor" and write an app note in exchange for an Eval board.
> But then I guess they are already giving away the software so what
> else could they offer?

The most interesting part of the offer, I thought, was the opportunity
to go up and work with the GA guys for a week or two.  Plus I didn't get
the impression that they wanted to make any claim to what you develop
other than publishing the app note.  I don't feel all that likely to
take them up on the offer (since I'm not much of a Forther) but it's
attractive in some ways, and I have some ideas that this might be an
interesting way to develop if I have the time for it.

[toc] | [prev] | [next] | [standalone]

#8553

In article <40d03fae-b016-49e9-b927-fcc6fcc944e1@h3g2000yqa.googlegroups.com>,
rickman  <gnuarm@gmail.com> wrote:
<SNIP>
>
>The GA chips have a lot of raw potential, in many ways they are
>amazing devices.  But the company needs to provide more info, a lot
>more info, on how to do useful stuff with them.  As it currently
>stands I am looking at designing a board with a GA144 along with a
>SiBlue chip as a level converter and possibly an ARM MCU as a
>interface device for Ethernet and USB.  I think this shows some of the
>current limitations of the GA144.

It remains to be seen whether Lecrone will end up behind bars.
If not, it will spell the end of GA144 anyway.
With a firm hold of the patents, he will be instructed by his
clients to finish off GreenArrays.

USB should indeed be high on the priority list. The crystal application
note should long have been finished. colorforth as a source for
applications should be abandoned. I don't see an enticement
for clever developers. I don't see how they can beat Lecrone.

What is worse, I don't see much insight in above matters.
I don't see a future for GreenArrays.
(I did purchase a couple of chips, while it lasts.)

>
>Rick

Groetjes Albert

--
-- 
Albert van der Horst, UTRECHT,THE NETHERLANDS
Economic growth -- being exponential -- ultimately falters.
albert@spe&ar&c.xs4all.nl &=n http://home.hccnet.nl/a.w.m.van.der.horst

[toc] | [prev] | [next] | [standalone]

#8566 — 2012 - the year of the GA144

Hi Albert,

I have just invested in a GA144 evaluation board, to see if I can use
the chip to upgrade two 8051-based products, and for new development,
so I find your comments worth a reply :

> It remains to be seen whether Lecrone will end up behind bars.
This may be what justice requires, and maybe what Chuck and friends
would wish, but my understanding is that what all parties want is a
"fair" distribution of money, where the definition of "fair" is
currently being decided. In this context, the threat of a jail
sentence is designed to get everyone to abide by the rules. For Mr.
Lecrone to end up in jail would require that he and his lawyers are
very careless.

> If not, it will spell the end of GA144 anyway.
This is saying that the GA144 is doomed either way - is this what you
intended? I can see no connection between Mr. Lecrone's fate and that
of the GA144.

> With a firm hold of the patents, he will be instructed by his clients to finish off GreenArrays.
OK, so if Mr. Lecrone remains out of jail, and still has a "firm hold
of the patents", he could sue GreenArrays and force them to stop
trading.
What this is implying is that the patent and legal system is a very
powerful bully - they are able to steal ownership of ideas and
physical chips that are clearly created by Chuck Moore and his
colleagues.
If this happens, and I remain a strong optimist that it will not, I
will have to use another chip.

Obviously it would be nice if the GA144 (or plug-compatible
replacement) would still be available in five years time...

> USB should indeed be high on the priority list.
USB is very complicated, and will take time to get it to work on the
GA144.

> The crystal application note should long have been finished.
Meanwhile Chuck is driving a crystal to create video sync - this
should be added to the app note too. It is clearly possible to do
this...

> colorforth as a source for applications should be abandoned.
colorForth source needs a wrapper if it is to exist outside the
rarefied atmosphere of the colorForth environment - a bit like humans
trying to live on Pandora.
As a design concept however, the use of pre-parsed "coloured" tokens
is very effective.
I would like to make the interface easier...

> I don't see an enticement for clever developers.
Nothing about the GA144 or colorForth is going to entice any developer
who is used to mainstream tools.
However some people are curious to see how anything this simple can do
anything useful.

> I don't see how they can beat Lecrone.
Don't be so defeatist! Demonisation is not helpful either. This is
just rich people playing rich-people's games.
I will still provide colorForth distros because I think its important
to follow what is simple, elegant and beautiful in life, both within
and outside of work and technology.

...
> (I did purchase a couple of chips, while it lasts.)
Cool. Thats $40 to a worthwhile cause ;-)

> What is worse, I don't see much insight in above matters.
Its not insight, its speculation.

> I don't see a future for GreenArrays.
I do.

A few days ago I watched the film "Himalaya". A buddhist monk said
that if there is a choice of two ways, you should always choose the
more difficult one....
My own view is that you should follow the path of maximum fun -
difficulty is irrelevant if you enjoy what you do...

Best wishes and a Happy New Year,

Howerd  :-)



On Jan 1, 2:25 pm, Albert van der Horst <alb...@spenarnc.xs4all.nl>
wrote:
> In article <40d03fae-b016-49e9-b927-fcc6fcc94...@h3g2000yqa.googlegroups.com>,rickman  <gnu...@gmail.com> wrote:
>
> <SNIP>
>
>
>
> >The GA chips have a lot of raw potential, in many ways they are
> >amazing devices.  But the company needs to provide more info, a lot
> >more info, on how to do useful stuff with them.  As it currently
> >stands I am looking at designing a board with a GA144 along with a
> >SiBlue chip as a level converter and possibly an ARM MCU as a
> >interface device for Ethernet and USB.  I think this shows some of the
> >current limitations of the GA144.
>
> It remains to be seen whether Lecrone will end up behind bars.
> If not, it will spell the end of GA144 anyway.
> With a firm hold of the patents, he will be instructed by his
> clients to finish off GreenArrays.
>
> USB should indeed be high on the priority list. The crystal application
> note should long have been finished. colorforth as a source for
> applications should be abandoned. I don't see an enticement
> for clever developers. I don't see how they can beat Lecrone.
>
> What is worse, I don't see much insight in above matters.
> I don't see a future for GreenArrays.
> (I did purchase a couple of chips, while it lasts.)
>
>
>
> >Rick
>
> Groetjes Albert
>
> --
> --
> Albert van der Horst, UTRECHT,THE NETHERLANDS
> Economic growth -- being exponential -- ultimately falters.
> albert@spe&ar&c.xs4all.nl &=nhttp://home.hccnet.nl/a.w.m.van.der.horst

[toc] | [prev] | [next] | [standalone]

#8614 — Re: 2012 - the year of the GA144

On Jan 1, 1:24 pm, Howerd <howe...@yahoo.co.uk> wrote:
> Hi Albert,
>
> I have just invested in a GA144 evaluation board, to see if I can use
> the chip to upgrade two 8051-based products, and for new development,
> so I find your comments worth a reply :
>
> > It remains to be seen whether Lecrone will end up behind bars.
>
> This may be what justice requires, and maybe what Chuck and friends
> would wish, but my understanding is that what all parties want is a
> "fair" distribution of money, where the definition of "fair" is
> currently being decided. In this context, the threat of a jail
> sentence is designed to get everyone to abide by the rules. For Mr.
> Lecrone to end up in jail would require that he and his lawyers are
> very careless.

I have to say I am not familiar enough with the "Lecrone" matter to
understand any of this.  I heard about a lawyer who had allegedly
taken the money from the patents but I didn't realize he actually had
taken the patents.  Are details available anywhere or is this all by
word of mouth?


> > If not, it will spell the end of GA144 anyway.
>
> This is saying that the GA144 is doomed either way - is this what you
> intended? I can see no connection between Mr. Lecrone's fate and that
> of the GA144.
>
> > With a firm hold of the patents, he will be instructed by his clients to finish off GreenArrays.
>
> OK, so if Mr. Lecrone remains out of jail, and still has a "firm hold
> of the patents", he could sue GreenArrays and force them to stop
> trading.
> What this is implying is that the patent and legal system is a very
> powerful bully - they are able to steal ownership of ideas and
> physical chips that are clearly created by Chuck Moore and his
> colleagues.
> If this happens, and I remain a strong optimist that it will not, I
> will have to use another chip.
>
> Obviously it would be nice if the GA144 (or plug-compatible
> replacement) would still be available in five years time...

If the rights to patents used within the GA144 are in question then it
would be foolish to design in these devices.  I don't want to believe
or spread rumor.  Where can I verify any of this?


> > USB should indeed be high on the priority list.
>
> USB is very complicated, and will take time to get it to work on the
> GA144.

It's not that bloody complex.  They have it running on 8051 processors
and nearly every MCU line has a USB version of the device.  Most have
both device and host software.  The memory limitations of the GA
devices my present an issue, but I can't imagine it would make it that
much harder using external memory.


> > The crystal application note should long have been finished.
>
> Meanwhile Chuck is driving a crystal to create video sync - this
> should be added to the app note too. It is clearly possible to do
> this...

This is actually a sore point with me.  I have not heard any details
of what has been done or how they do it.  In particular the last I
heard they had a 10 MHz crystal oscillating... some of the time
because there isn't enough resolution using a software timing loop for
that speed of operation.  Again, if there are any details I'd love to
hear them by other than the rumor mill.


> > colorforth as a source for applications should be abandoned.
>
> colorForth source needs a wrapper if it is to exist outside the
> rarefied atmosphere of the colorForth environment - a bit like humans
> trying to live on Pandora.
> As a design concept however, the use of pre-parsed "coloured" tokens
> is very effective.
> I would like to make the interface easier...

I have mixed feelings about color forth.  I don't mind using it
myself, but to many it will be a "no go" for using the chip.  There is
so much existing code written that would need to be ported for many
people to try using this device.


> > I don't see an enticement for clever developers.
>
> Nothing about the GA144 or colorForth is going to entice any developer
> who is used to mainstream tools.
> However some people are curious to see how anything this simple can do
> anything useful.

I'm still curious.  I'd like to see a full set of professional level
app notes explaining how to use the device to implement various
building blocks most users would want.


> > I don't see how they can beat Lecrone.
>
> Don't be so defeatist! Demonisation is not helpful either. This is
> just rich people playing rich-people's games.
> I will still provide colorForth distros because I think its important
> to follow what is simple, elegant and beautiful in life, both within
> and outside of work and technology.
>
> ...> (I did purchase a couple of chips, while it lasts.)
>
> Cool. Thats $40 to a worthwhile cause ;-)
>
> > What is worse, I don't see much insight in above matters.
>
> Its not insight, its speculation.
>
> > I don't see a future for GreenArrays.
>
> I do.
>
> A few days ago I watched the film "Himalaya". A buddhist monk said
> that if there is a choice of two ways, you should always choose the
> more difficult one....
> My own view is that you should follow the path of maximum fun -
> difficulty is irrelevant if you enjoy what you do...
>
> Best wishes and a Happy New Year,
>
> Howerd  :-)


I'm not looking to have fun.  I would like to use the GA144 device to
design a product.  But I don't want to develop a USB application on
the device on my own.  Just as with most MCUs I would like to use
building blocks to create my application.

Rick



> On Jan 1, 2:25 pm, Albert van der Horst <alb...@spenarnc.xs4all.nl>
> wrote:
>
> > In article <40d03fae-b016-49e9-b927-fcc6fcc94...@h3g2000yqa.googlegroups.com>,rickman  <gnu...@gmail.com> wrote:
>
> > <SNIP>
>
> > >The GA chips have a lot of raw potential, in many ways they are
> > >amazing devices.  But the company needs to provide more info, a lot
> > >more info, on how to do useful stuff with them.  As it currently
> > >stands I am looking at designing a board with a GA144 along with a
> > >SiBlue chip as a level converter and possibly an ARM MCU as a
> > >interface device for Ethernet and USB.  I think this shows some of the
> > >current limitations of the GA144.
>
> > It remains to be seen whether Lecrone will end up behind bars.
> > If not, it will spell the end of GA144 anyway.
> > With a firm hold of the patents, he will be instructed by his
> > clients to finish off GreenArrays.
>
> > USB should indeed be high on the priority list. The crystal application
> > note should long have been finished. colorforth as a source for
> > applications should be abandoned. I don't see an enticement
> > for clever developers. I don't see how they can beat Lecrone.
>
> > What is worse, I don't see much insight in above matters.
> > I don't see a future for GreenArrays.
> > (I did purchase a couple of chips, while it lasts.)
>
> > >Rick
>
> > Groetjes Albert

[toc] | [prev] | [next] | [standalone]

#8617 — Re: 2012 - the year of the GA144

On 03/01/12 01:20, rickman wrote:
    . . .
>
> I'm not looking to have fun.  I would like to use the GA144 device to
> design a product.  But I don't want to develop a USB application on
> the device on my own.  Just as with most MCUs I would like to use
> building blocks to create my application.

Perhaps you could help Bernd with the USB block for his b16, and 
combine your processor and other logic into one part?

   http://bernd-paysan.de/b16.html

Jan Coombs

[toc] | [prev] | [next] | [standalone]

#8619 — Re: 2012 - the year of the GA144

On Jan 3, 2:41 am, Jan Coombs <jan_2011...@murray-microft.co.uk>
wrote:
> On 03/01/12 01:20, rickman wrote:
>     . . .
>
> > I'm not looking to have fun.  I would like to use the GA144 device to
> > design a product.  But I don't want to develop a USB application on
> > the device on my own.  Just as with most MCUs I would like to use
> > building blocks to create my application.
>
> Perhaps you could help Bernd with the USB block for his b16, and
> combine your processor and other logic into one part?
>
>    http://bernd-paysan.de/b16.html
>
> Jan Coombs

Thanks for the suggestion.  I was not aware of the USB and Ethernet
interfaces he was working on.  I had rolled my own Forth CPU which
ended up not entirely unlike his B16.  Mine fetches one instruction at
a time and has an opcode format optimized for minimum memory usage.  I
never spent enough time on it to truly optimize it for anything only
using it for a small project once.  The GA144 is an amazing design in
many ways and I expect I would roll my own F18 for an FPGA.  GA seems
to be ok with that, but does ask for some compensation if I am going
to use their tools... fair enough.

I'll contact Bernd and see if there is anything in particular I could
offer help with.  One of the reasons I would like to see GA provide a
reference design for USB and Ethernet is because I have not done
either before, so I don't know if I will be of much help, but I'm
willing to learn and figure things out.

Rick

[toc] | [prev] | [next] | [standalone]

#8480

On Dec 29, 6:03 pm, "Elizabeth D. Rather" <erat...@forth.com> wrote:
> There's also the fact that those of us who make a living from Forth need
> to be able to buy groceries, make house payments, and meet payroll
> because our staff like groceries, etc., too.

Compare the revenue of SwiftX licenses to that of consulting and
education services. What would be the impact of FOSSing SwiftX?

- Lose a small revenue stream from license sales
+ More design-ins for SwiftX
+ Consulting work maintaining these design-ins

The Forths people choose depend on their budget. If they don't have a
budget, they take what they can get. If they do have a budget, they go
with the best they can get, which for the ARM target is MPE. Given the
success of open source companies, this model may be worth looking at.

-Brad

[toc] | [prev] | [next] | [standalone]

#8499

The board arrived today :-) As soon as my USB<->RS-232 adaptor shows up 
I'm going to give Frank Sergeant's Riscy Pygness a shot.

The entire attraction of Forth is the self hosting development 
environment.  So crosscompilers aren't really appealing.

I want to be able to debug effectively on the target system.  gcc/gdb 
are more or less usable but somewhat underwhelming coming from a Solaris 
Studio 12 world.  Besides, this really feels like Forth territory.

This is a an entertainment project on a $20.70 computer.  I want to play 
w/ software defined radio.  $1200 for a crosscompiler that forces me to 
run Linux or Windows?  Huh?  High end commercial SDRs cost less. It 
would make far more sense to just buy an appliance.

A barebones native Forth for $100 that would run on a range of eval 
boards I'd do if it was well documented. However, I'd want to read the 
manual first, cover to cover.  I'd want to get a book, not a giant PDF.

My last adventure w/ Forth was on a C64.   The documentation was so bad 
that all I ever did in Forth was write a decompiler.  A short time later 
I was running VMS on a MicroVAX II and later just about every Unix 
workstation made so I never got around to Forth again.

At the moment I know nothing at all about the ARM other than it's a 
processor family.  I don't even have a manual for it yet in the 3+ feet
of processor manuals on my shelves.  But I've still got all my old Forth 
books w/ listings for Z80, 6502, etc.  ARM is just another set of 
instructions and features.

Have Fun!
Reg

[toc] | [prev] | [next] | [standalone]

#8501

On 12/31/11 1:04 AM, John Smith wrote:
> The board arrived today :-) As soon as my USB<->RS-232 adaptor shows up
> I'm going to give Frank Sergeant's Riscy Pygness a shot.
>
> The entire attraction of Forth is the self hosting development
> environment. So crosscompilers aren't really appealing.
>
> I want to be able to debug effectively on the target system. gcc/gdb are
> more or less usable but somewhat underwhelming coming from a Solaris
> Studio 12 world. Besides, this really feels like Forth territory.

What are you using for a terminal?  If you're using a PC with a terminal 
emulator, the setup is exactly what it would be using a cross-compiler, 
except you'd have a more powerful compiler and faster interaction, since 
you wouldn't have to download whole text files for your target to 
compile.  An interactive Forth cross compiler *does* support debugging 
effectively on the target.

> This is a an entertainment project on a $20.70 computer. I want to play
> w/ software defined radio. $1200 for a crosscompiler that forces me to
> run Linux or Windows? Huh? High end commercial SDRs cost less. It would
> make far more sense to just buy an appliance.

You don't have to run anything on your target except your application.

> A barebones native Forth for $100 that would run on a range of eval
> boards I'd do if it was well documented. However, I'd want to read the
> manual first, cover to cover. I'd want to get a book, not a giant PDF.

Try the free evaluation version of SwiftX. Check out its documentation 
and capabilities.

Cheers,
Elizabeth

-- 
==================================================
Elizabeth D. Rather   (US & Canada)   800-55-FORTH
FORTH Inc.                         +1 310.999.6784
5959 West Century Blvd. Suite 700
Los Angeles, CA 90045
http://www.forth.com

"Forth-based products and Services for real-time
applications since 1973."
==================================================

[toc] | [prev] | [next] | [standalone]

#8510

On 12/30/2011 04:28 PM, Elizabeth D. Rather wrote:
> Try the free evaluation version of SwiftX. Check out its documentation
> and capabilities.
> 
> Cheers,
> Elizabeth
> 

I purchased one of the STM32F4DISCOVERY boards recently as well but I
think this is bad advice. I don't see any support for the STM32F407 used
on the Discovery board or any STM32 product for that matter.

[toc] | [prev] | [next] | [standalone]

#8519

On Dec 31, 3:04 am, David Schultz <ab...@127.0.0.1> wrote:
> On 12/30/2011 04:28 PM, Elizabeth D. Rather wrote:
>
> > Try the free evaluation version of SwiftX. Check out its documentation
> > and capabilities.
>
> > Cheers,
> > Elizabeth
>
> I purchased one of the STM32F4DISCOVERY boards recently as well but I
> think this is bad advice. I don't see any support for the STM32F407 used
> on the Discovery board or any STM32 product for that matter.

I have recently been using the STM32LDiscovery board (ultra low power
variant, also <GBP10). I had MPE Forth up and running in a couple of
days - most time was spent reading the STM documentation and migration
guides.
There are some differences between the early F1 series of Cortex-M3
and the F2 and L15 variants. From the F4 data sheet it appears to be
pretty compatible with the F2 and L15.
I have the ADC and DMA functioning and the GPIO setup (the biggest
change from the F1).

Reg I can thoroughly recommend VFX cross-compiler - I haven't written
any ARM assembly in the 8 years I have been using it.
The native code runs about 5 times faster than a 'traditional'
threaded implementation. The 'Stamp' version should be adequate for
your requirements.
Note the Discovery boards and the STM support code (Windows) gives you
SWdebug and programming. The cross-compiler will generate a new image
in a click a key! It would appear that the F4 will execute code at
full speed from flash although MPE code also supports run from RAM and
'reflash' via serial port.
I understand Stephen is currently working on Cortex-M4 support.

Mike

[toc] | [prev] | [next] | [standalone]

#9057

On 12/31/2011 05:04 AM, John Smith wrote:
> The board arrived today :-) As soon as my USB<->RS-232 adaptor shows up
> I'm going to give Frank Sergeant's Riscy Pygness a shot.
>

Good luck with that.

I have spent some time starting the process and it is tedious because
of all the differences between the SMT32F407 and the part used in the
Olimex board targeted by that version.

But then I discovered that the tcl program that massages everything
cannot handle a code .org other than zero while 0x08000000 is required.
I quick look at riscy.tcl shows spots where you can set the flash size,
ram size, and ram start, but not the start of flash. I know next to
nothing about TCL and have little desire to dig into to it to see how to
fix it. I did fire off an email to Frank Sergeant asking if he had any
hints.

But while waiting for a reply I dusted off a copy of eFORTH for the 8086
and started the process of porting that. I am to the point now where I
have coded all the primitives but still need to clean up some details in
the boot process before trying it for the first time.

[toc] | [prev] | [next] | [standalone]

#8646

On Dec 29 2011, 9:50 am, Reg Beardsley <u...@nowhere.org> wrote:
> I'm looking for a very basic Forth to use w/ the <$20US Discovery  board
> to develop a software defined radio.
>
> Requirements:
>
>   configurable RS-232 link (address, baud rate, etc)
>   assembler for basic ARM Cortex M4 instruction set
>   block interface to flash memory
>   run from RAM
>   load RAM from flash at boot
>   commented source code
>
> I've found riscy pygness & some old work at eddiem.com.  Is there
> anything else I've missed?  I'm "retired" so commercial offerings at
> 10-20x the cost of the board are out of the question.

MPE has done an STM32F4 port. This will become part of the cross
compiler distribution. The STM32F4 and Kinetis parts will be included
in all versions of the cross compiler in the release at the end of
January 2012.

The Stamp version costs GBP75, say USD 118. It is limited to 120kb
code and 64k RAM. If you take the existing ARM Stamp code from the
LPC210x folder and port it to the STM32F4, you will get what you want.

Stephen

[toc] | [prev] | [next] | [standalone]

Page 1 of 2  [1] 2  Next page →

Back to top | Article view | comp.lang.forth

csiph-web