Is anyone aware of a simple development board for the GA144 chip? No
fluff, extra processors, etc like the (expensive) GreenArrays GA144
development boards, just the bare minimum to get the chip going and
write code for it. Ideally an open design, i.e. download Gerber files
and make a board or two. If there are none, I will probably design one
myself but didn't want to reinvent the wheel.

Thanks,
-Dave McGuire

--
Dave McGuire, President/Curator
Large Scale Systems Museum
New Kensington, PA

On Thursday, July 20, 2023 at 11:05:11 PM UTC-4, Dave McGuire wrote:
> Is anyone aware of a simple development board for the GA144 chip? No
> fluff, extra processors, etc like the (expensive) GreenArrays GA144
> development boards, just the bare minimum to get the chip going and
> write code for it. Ideally an open design, i.e. download Gerber files
> and make a board or two. If there are none, I will probably design one
> myself but didn't want to reinvent the wheel.

I don't recall seeing anything like that. There was a small board by one of the well known hobbyist companies, but it was really just a pin out adapter board, not even having decoupling caps.

I don't think anything like this ever materialized, because there really is no market. Someone selling this might sell a dozen up to maybe 100.

The real issue, is the development software. It was not a product of a single mind or even influence. It was a bit of a hodgepodge of various software to handle various functions that may or may not be needed.

--

Rick C.

- Get 1,000 miles of free Supercharging
- Tesla referral code - https://ts.la/richard11209

In article <2737c21a-be14-4af7-8105-a302c81f31f3n@googlegroups.com>,
Lorem Ipsum <gnuarm.deletethisbit@gmail.com> wrote:
>
>The real issue, is the development software. It was not a product of a
>single mind or even influence. It was a bit of a hodgepodge of various
>software to handle various functions that may or may not be needed.

Parallel processing: You couldn't configure a hypercube,not even a cube.
We (Dutch Forth) did a demo program. Not only required this a patch from
the seller the system software, in the next release the
program no longer worked.
Intriguing as the chip is, you're well advised to not waste any time
with it.

> Rick C.

Groetjes Albert
--
Don't praise the day before the evening. One swallow doesn't make spring.
You must not say "hey" before you have crossed the bridge. Don't sell the
hide of the bear until you shot it. Better one bird in the hand than ten in
the air. First gain is a cat spinning. - the Wise from Antrim -

On 7/20/23 23:29, Lorem Ipsum wrote:
>> Is anyone aware of a simple development board for the GA144 chip? No
>> fluff, extra processors, etc like the (expensive) GreenArrays GA144
>> development boards, just the bare minimum to get the chip going and
>> write code for it. Ideally an open design, i.e. download Gerber files
>> and make a board or two. If there are none, I will probably design one
>> myself but didn't want to reinvent the wheel.
>
> I don't recall seeing anything like that. There was a small board by one of the well known hobbyist companies, but it was really just a pin out adapter board, not even having decoupling caps.

Yeah, I've seen that one, Schmartboard. Not particularly useful.

> I don't think anything like this ever materialized, because there really is no market. Someone selling this might sell a dozen up to maybe 100.

I'm not talking about a commercial product, I'm talking about someone
having done a design and released it, as is common in the rest of the world.

> The real issue, is the development software. It was not a product of a single mind or even influence. It was a bit of a hodgepodge of various software to handle various functions that may or may not be needed.

There's at least one non-vendor toolchain for the GA144. Spotting
that is what (re)sparked my interest.

-Dave

--
Dave McGuire, President/Curator
Large Scale Systems Museum
New Kensington, PA

On Friday, July 21, 2023 at 8:39:18 AM UTC-4, Dave McGuire wrote:
> On 7/20/23 23:29, Lorem Ipsum wrote:
> >> Is anyone aware of a simple development board for the GA144 chip? No
> >> fluff, extra processors, etc like the (expensive) GreenArrays GA144
> >> development boards, just the bare minimum to get the chip going and
> >> write code for it. Ideally an open design, i.e. download Gerber files
> >> and make a board or two. If there are none, I will probably design one
> >> myself but didn't want to reinvent the wheel.
> >
> > I don't recall seeing anything like that. There was a small board by one of the well known hobbyist companies, but it was really just a pin out adapter board, not even having decoupling caps.
> Yeah, I've seen that one, Schmartboard. Not particularly useful.
> > I don't think anything like this ever materialized, because there really is no market. Someone selling this might sell a dozen up to maybe 100.
> I'm not talking about a commercial product, I'm talking about someone
> having done a design and released it, as is common in the rest of the world.

I know what you are talking about. I'm trying to explain there is no commercial support and the hobbyist market it vanishingly small.

> > The real issue, is the development software. It was not a product of a single mind or even influence. It was a bit of a hodgepodge of various software to handle various functions that may or may not be needed.
> There's at least one non-vendor toolchain for the GA144. Spotting
> that is what (re)sparked my interest.

Why don't you develop an SBC for the GA144? Then you can share with the rest of us. Well, if you can find anyone interested.

What are you interested in doing with the part? As Albert has said, the chip and tools are largely a disappointment in many ways. I don't know of any commercial designs using the GA144, nor even any projects that actually made effective use of the many processors.

I looked at using for an audio spectrum analyzer once. But the comms were a huge limitation. People think the comms are a special feature of the GA144, but they are actually a major limitation.

--

Rick C.

+ Get 1,000 miles of free Supercharging
+ Tesla referral code - https://ts.la/richard11209

I find the idea of the GA144 irresistible, but the limited memory makes it impractical.
What makes a lot more sense to me is to do one on an FPGA. The J1 comes in at 160 luts. Half that if you skip the barrel shifter. To do 144 of them would take up 21K luts, even if you double that for networking, a very reasonable amount. Probably cheaper than the green arrays chip as well. Although I am sure that their prices are negotiable.

For my master's thesis I was very interested in building a green array of J1 cpus, but since no one seems interested, I am currently planning on targeting 8 J1's each running cordic, kind of a competitor to the Parallax Propeller. I know that for real time control, it makes life much simpler to have one cpu for each motor. No need for interrupts.and all of the complexity of responding quickly.

Of course the real question is what do you plan to build? An obvious application for large arrays of cpus is image processing, but at first glance, that appears to be a very heavily occupied market. Very hard to get a toe hold.

I expect to do a talk tomorrow at the SVFIG meeting on "A Review of Forth Processors".
Christopher Lozinski

On 7/21/23 13:37, Lorem Ipsum wrote:
>>> I don't recall seeing anything like that. There was a small board by one of the well known hobbyist companies, but it was really just a pin out adapter board, not even having decoupling caps.
>> Yeah, I've seen that one, Schmartboard. Not particularly useful.
>>> I don't think anything like this ever materialized, because there really is no market. Someone selling this might sell a dozen up to maybe 100.
>> I'm not talking about a commercial product, I'm talking about someone
>> having done a design and released it, as is common in the rest of the world.
>
> I know what you are talking about. I'm trying to explain there is no commercial support and the hobbyist market it vanishingly small.

Then you most certainly *don't* know what I'm talking about. I will
explain again.

I'm not the least bit interested in commercial support. I'm not
interested in the hobbyist market, or any other kind of market, for this
chip or any board that it might get soldered to. I want to hack on this
rather neat little chip for my own enjoyment, and to do that, I need it
on a board.

>>> The real issue, is the development software. It was not a product of a single mind or even influence. It was a bit of a hodgepodge of various software to handle various functions that may or may not be needed.
>> There's at least one non-vendor toolchain for the GA144. Spotting
>> that is what (re)sparked my interest.
>
> Why don't you develop an SBC for the GA144? Then you can share with the rest of us. Well, if you can find anyone interested.

I'm thinking about designing a board, but given the attitude here,
I'm doubt I'll bother mentioning it here.

> What are you interested in doing with the part? As Albert has said, the chip and tools are largely a disappointment in many ways. I don't know of any commercial designs using the GA144, nor even any projects that actually made effective use of the many processors.

Hacking on it. Learning it and having some fun with it. I am a tech
guy, I have been for decades. I'm a commercial electronic designer, but
I also do a lot of things for my own enjoyment and education.

> I looked at using for an audio spectrum analyzer once. But the comms were a huge limitation. People think the comms are a special feature of the GA144, but they are actually a major limitation.

I understand that you hate this chip. None of this even comes close
to answering my initial query. If I'd known my post would've been a
trigger for a deluge of GA144-hater vitriol I'd have kept my damn mouth
shut and designed a board.

Wow.

-Dave

--
Dave McGuire, President/Curator
Large Scale Systems Museum
New Kensington, PA

Dave McGuire <mcguire@lssmuseum.org> writes:

> I'm thinking about designing a board, but given the attitude here,
> I'm doubt I'll bother mentioning it here.

Please do mention it.

--
Take Back Control! -- Mesh The Planet!
I do not play Nethack, I do play GNUS! o;-)
Solid facts do not need 1001 pictures.

On Friday, July 21, 2023 at 3:02:07 PM UTC-4, Christopher Lozinski wrote:
> I find the idea of the GA144 irresistible, but the limited memory makes it impractical.

Yeah, interesting, but rather incapable of doing much useful. It's really hard to get your head wrapped around the memory and also the comms limitations.

> What makes a lot more sense to me is to do one on an FPGA.

"One" what?

> The J1 comes in at 160 luts. Half that if you skip the barrel shifter.

Are you sure of this LUT count? Maybe 6 input LUTs, but even that's a stretch.

> To do 144 of them would take up 21K luts, even if you double that for networking, a very reasonable amount. Probably cheaper than the green arrays chip as well. Although I am sure that their prices are negotiable.
>
> For my master's thesis I was very interested in building a green array of J1 cpus, but since no one seems interested, I am currently planning on targeting 8 J1's each running cordic, kind of a competitor to the Parallax Propeller. I know that for real time control, it makes life much simpler to have one cpu for each motor. No need for interrupts.and all of the complexity of responding quickly.

So, you are focusing on ideas that get you excited, but with no particular purpose, no application?

> Of course the real question is what do you plan to build? An obvious application for large arrays of cpus is image processing, but at first glance, that appears to be a very heavily occupied market. Very hard to get a toe hold.
>
> I expect to do a talk tomorrow at the SVFIG meeting on "A Review of Forth Processors".

I would suggest you start with a definition of "forth processor". I claim that none exist. Everyone seems to have their own opinion of what a "forth" processor is.

--

Rick C.

-- Get 1,000 miles of free Supercharging
-- Tesla referral code - https://ts.la/richard11209

On Friday, July 21, 2023 at 7:22:48 PM UTC-4, Dave McGuire wrote:
> On 7/21/23 13:37, Lorem Ipsum wrote:
> >>> I don't recall seeing anything like that. There was a small board by one of the well known hobbyist companies, but it was really just a pin out adapter board, not even having decoupling caps.
> >> Yeah, I've seen that one, Schmartboard. Not particularly useful.
> >>> I don't think anything like this ever materialized, because there really is no market. Someone selling this might sell a dozen up to maybe 100.
> >> I'm not talking about a commercial product, I'm talking about someone
> >> having done a design and released it, as is common in the rest of the world.
> >
> > I know what you are talking about. I'm trying to explain there is no commercial support and the hobbyist market it vanishingly small.
> Then you most certainly *don't* know what I'm talking about. I will
> explain again.
>
> I'm not the least bit interested in commercial support. I'm not
> interested in the hobbyist market, or any other kind of market, for this
> chip or any board that it might get soldered to. I want to hack on this
> rather neat little chip for my own enjoyment, and to do that, I need it
> on a board.

What makes you think I don't understand this? I'm simply explaining to you that you are very, very unlikely to find such a board. If you want to design your own and share it with others, you are unlikely to find anyone who is interested, other than possibly a few people here. These days, pretty much everyone who might be interested, has seen the chip, learned enough about it to know there's no point. You will be traveling this road alone, pretty much from scratch.

> >>> The real issue, is the development software. It was not a product of a single mind or even influence. It was a bit of a hodgepodge of various software to handle various functions that may or may not be needed.
> >> There's at least one non-vendor toolchain for the GA144. Spotting
> >> that is what (re)sparked my interest.
> >
> > Why don't you develop an SBC for the GA144? Then you can share with the rest of us. Well, if you can find anyone interested.
> I'm thinking about designing a board, but given the attitude here,
> I'm doubt I'll bother mentioning it here.

So, you don't even want to share what you develop? Ok, that's not at all unusual.

> > What are you interested in doing with the part? As Albert has said, the chip and tools are largely a disappointment in many ways. I don't know of any commercial designs using the GA144, nor even any projects that actually made effective use of the many processors.
> Hacking on it. Learning it and having some fun with it. I am a tech
> guy, I have been for decades. I'm a commercial electronic designer, but
> I also do a lot of things for my own enjoyment and education.
> > I looked at using for an audio spectrum analyzer once. But the comms were a huge limitation. People think the comms are a special feature of the GA144, but they are actually a major limitation.
> I understand that you hate this chip.

It's not at all a matter of "hate". I've looked at the chip seriously. I've written code for it and hand assembled it counting execution time in a spreadsheet. I had become very familiar with it and now see past the "glitz" of its uniqueness and see how limited it really is.

> None of this even comes close
> to answering my initial query.

Actually, everything discussed was directly addressing your initial post where you ask about resources, including designing your own. I think I mentioned the support software, which you did not seem to have asked about.

> If I'd known my post would've been a
> trigger for a deluge of GA144-hater vitriol I'd have kept my damn mouth
> shut and designed a board.

I'm sorry my replies offend you. This is not "hating". I think everything I've posted has simply been relating facts. You can disagree with opinion, but not so much with facts.

I recall a poster here who was talking about some real pie-in-the-sky ideas.. I don't recall the details, but he wanted to design 32 bit, multiprocessor chips. When I tried to explain to him how his ideas were not realistic, he accused me of "hating" or something similar. I wasn't hating, I was just explaining how he would be unlikely to achieve any of his goals, because of their lofty nature. He even claimed he could not have a discussion because of "all the hate". He was messing with his own head. No one was "hating" on him, and he was at all times free to ignore any posts he didn't like.

I'm just sharing my experience. Please ignore me if I make you unhappy.

--

Rick C.

-+ Get 1,000 miles of free Supercharging
-+ Tesla referral code - https://ts.la/richard11209

In article <6c69c287-05c7-4a09-b9c0-93e230399310n@googlegroups.com>,
Lorem Ipsum <gnuarm.deletethisbit@gmail.com> wrote:
>On Friday, July 21, 2023 at 3:02:07 PM UTC-4, Christopher Lozinski wrote:
<SNIP>
>> I expect to do a talk tomorrow at the SVFIG meeting on "A Review of
>Forth Processors".
>
>I would suggest you start with a definition of "forth processor". I
>claim that none exist. Everyone seems to have their own opinion of what
>a "forth" processor is.

I'm content with the definition:
A Forth processor is a processor with an instruction set inspired by
or related to Forth instructions.

This is vague and what one considers a Forth processor, someone else
doesnot. So be it.

> Rick C.
--
Don't praise the day before the evening. One swallow doesn't make spring.
You must not say "hey" before you have crossed the bridge. Don't sell the
hide of the bear until you shot it. Better one bird in the hand than ten in
the air. First gain is a cat spinning. - the Wise from Antrim -

On 7/21/23 20:19, Lorem Ipsum wrote:
>>> I know what you are talking about. I'm trying to explain there is no commercial support and the hobbyist market it vanishingly small.
>> Then you most certainly *don't* know what I'm talking about. I will
>> explain again.
>>
>> I'm not the least bit interested in commercial support. I'm not
>> interested in the hobbyist market, or any other kind of market, for this
>> chip or any board that it might get soldered to. I want to hack on this
>> rather neat little chip for my own enjoyment, and to do that, I need it
>> on a board.
>
> What makes you think I don't understand this? I'm simply explaining to you that you are very, very unlikely to find such a board.

Ok, perhaps I misunderstood you. Usually when people talk about
"markets" they're talking about selling things to other people, or other
people adopting things. I don't care about that. This is for ME and me
alone, to satisfy a personal interest. I am a commercial embedded
systems designer; I design other things about which I'm interested in
sales and adoption. Not this.

> If you want to design your own and share it with others, you are unlikely to find anyone who is interested, other than possibly a few people here.

Apparently you think I should care about that. I don't.

> These days, pretty much everyone who might be interested, has seen the chip, learned enough about it to know there's no point. You will be traveling this road alone, pretty much from scratch.

Apparently you think I should care about that too. I don't.

You've turned "No, I haven't seen such a thing" to a whole shitload
of "Oh no, kid, you don't want to do that, oh no, nobody else will want
it, oh no, that chip sucks, oh no, there's no support" blah blah blah.

I have a GA144 chip that I got as a sample when they first shipped;
it has been sitting in my lab in its box since then. I've always been
intrigued by the design, but have never had an opportunity to do
anything with it. Though Chuck Moore is a bit "out there" I respect the
man and I'm interested to see what he's designed.

I'm getting older, I'm starting to notice it, I'm hot off the high of
having gotten a big(ish) Transputer-based cluster up and running at a
museum, and that reminded me of that GA144 sitting in a drawer. So I
said to myself, "Self, it's time do something with that chip, but GA's
dev boards are overpriced and overcomplicated."

I don't think I should have to explain all of this when the query was
a simple "Hey, has anyone laid out a board?"

>> I'm thinking about designing a board, but given the attitude here,
>> I'm doubt I'll bother mentioning it here.
>
> So, you don't even want to share what you develop? Ok, that's not at all unusual.

I think you missed my point.

> It's not at all a matter of "hate". I've looked at the chip seriously. I've written code for it and hand assembled it counting execution time in a spreadsheet. I had become very familiar with it and now see past the "glitz" of its uniqueness and see how limited it really is.

I understand that. I can read datasheets and user manuals as well as
the next guy. I usually design with ARM chips and I write firmware in C
and assembler. I know this will be very different, and that's part of
why I'm interested in it.

>> None of this even comes close
>> to answering my initial query.
>
> Actually, everything discussed was directly addressing your initial post where you ask about resources, including designing your own. I think I mentioned the support software, which you did not seem to have asked about.

I didn't ask about software because I had already found what I
thought I'd need. I certainly didn't ask for a critique of the chip.

>> If I'd known my post would've been a
>> trigger for a deluge of GA144-hater vitriol I'd have kept my damn mouth
>> shut and designed a board.
>
> I'm sorry my replies offend you. This is not "hating". I think everything I've posted has simply been relating facts. You can disagree with opinion, but not so much with facts.

I didn't ask for those facts. I can read documentation just fine. I
appreciate your opinion and your experience with these chips, but that
doesn't quell my curiosity about them.

> I recall a poster here who was talking about some real pie-in-the-sky ideas. I don't recall the details, but he wanted to design 32 bit, multiprocessor chips. When I tried to explain to him how his ideas were not realistic, he accused me of "hating" or something similar.

So this happens to you a lot.

> I wasn't hating, I was just explaining how he would be unlikely to achieve any of his goals, because of their lofty nature. He even claimed he could not have a discussion because of "all the hate". He was messing with his own head. No one was "hating" on him, and he was at all times free to ignore any posts he didn't like.

I suppose it's good to see Usenet hasn't really changed.

I have the same motivation, to warn people (usually kids) away from
bad ideas with "there be dragons here" kind of statements. I'm a
graybeard too.

Several years ago, a teenager popped up on a mailing list talking
about chasing an IBM z890 mainframe that was being auctioned off on
govdeals. I have one of those machines; it weighs just over a ton and
is much larger than it tends to look in pictures. The kid was very
excited about it, and I tried to warn him away from it, explaining that
if a PDP-11 falls on you, you'll end up in the hospital, but if a z890
falls on you, you're pretty much done. I move big iron all the time and
I know how dangerous something like this can be; I was concerned for his
safety.

Well, the kid ignored my advice and he got the z890. It was quite a
saga, involving his grandfather excavating land around his basement door
to get it into his house, etc, and he eventually got it up and running,
and he didn't get killed in the process. This caught the attention of
IBM, who immediately hired him and gave him a fantastic job working on
the development of cutting-edge hardware. He's basically set for life now.

If he'd taken my advice, which was motivated purely from the
standpoint of concern for his safety because I know that machine well,
he would've missed out on that great opportunity which changed his life.
Thankfully, he has forgiven me.

Us graybeards need to share our experience, sure. That's an
important part of being a member of a society. But there are times when
we need to let the wide-eyed kids chase crazy dreams. Sometimes they
will surprise us.

-Dave

--
Dave McGuire, President/Curator
Large Scale Systems Museum
New Kensington, PA

On Friday, 21 July 2023 at 04:05:11 UTC+1, Dave McGuire wrote:
> Is anyone aware of a simple development board for the GA144 chip? No
> fluff, extra processors, etc like the (expensive) GreenArrays GA144
> development boards, just the bare minimum to get the chip going and
> write code for it. Ideally an open design, i.e. download Gerber files
> and make a board or two. If there are none, I will probably design one
> myself but didn't want to reinvent the wheel.
>
> Thanks,
> -Dave McGuire
>
> --
> Dave McGuire, President/Curator
> Large Scale Systems Museum
> New Kensington, PA

Thank you very much for leaving the link to your organisation.
https://www.mact.io/about_us

This theme GA144 has been discussed for probably 10 years here and in other places now.

It is a shame, that no simple way for playing with these chips exists.
They are very unusual, but the low interest probably demonstrates,
that it is not really worth trying, except if you just want to for fun.

Unfortunately the entry cost at Green Arrays is at least $200 for 10 chips - SMT unsoldered so you have to do all of the work.
To my knowledge, there is not even a proven circuit diagram that you can use to build your own board.
https://www.greenarraychips.com/home/products/index.php

When I was more into Forth documentation,
I asked Greg at Greenarrays, as I was very interested,
if we could do a clean room implementation of a basic GA144 on FPGA - his feedback was extremely negative.
So I gave up and did other things.

I find it very unusual, that Chuck Moore is involved at GA,
but seems not really to be pushing it either.

To me it seems to be self limiting the sales,
as they would run out of chips soon,
and I wonder, if the fab who did these chips 10 to 15 years ago
still exists now and could do more if needed..

If you ask a question here on CLF,
you have to be aware of the negativity that is quite normal here.

I would suggest you ceck with one of the Forth facebook groups.
There you can find probably more help and in a friendy way.
https://www.facebook.com/groups/PROGRAMMINGFORTH

On Saturday, 22 July 2023 at 15:23:07 UTC+1, Jurgen Pitaske wrote:
> On Friday, 21 July 2023 at 04:05:11 UTC+1, Dave McGuire wrote:
> > Is anyone aware of a simple development board for the GA144 chip? No
> > fluff, extra processors, etc like the (expensive) GreenArrays GA144
> > development boards, just the bare minimum to get the chip going and
> > write code for it. Ideally an open design, i.e. download Gerber files
> > and make a board or two. If there are none, I will probably design one
> > myself but didn't want to reinvent the wheel.
> >
> > Thanks,
> > -Dave McGuire
> >
> > --
> > Dave McGuire, President/Curator
> > Large Scale Systems Museum
> > New Kensington, PA
> Thank you very much for leaving the link to your organisation.
> https://www.mact.io/about_us
>
> This theme GA144 has been discussed for probably 10 years here and in other places now.
>
> It is a shame, that no simple way for playing with these chips exists.
> They are very unusual, but the low interest probably demonstrates,
> that it is not really worth trying, except if you just want to for fun.
>
> Unfortunately the entry cost at Green Arrays is at least $200 for 10 chips - SMT unsoldered so you have to do all of the work.
> To my knowledge, there is not even a proven circuit diagram that you can use to build your own board.
> https://www.greenarraychips.com/home/products/index.php
>
> When I was more into Forth documentation,
> I asked Greg at Greenarrays, as I was very interested,
> if we could do a clean room implementation of a basic GA144 on FPGA - his feedback was extremely negative.
> So I gave up and did other things.
>
> I find it very unusual, that Chuck Moore is involved at GA,
> but seems not really to be pushing it either.
>
> To me it seems to be self limiting the sales,
> as they would run out of chips soon,
> and I wonder, if the fab who did these chips 10 to 15 years ago
> still exists now and could do more if needed..
>
> If you ask a question here on CLF,
> you have to be aware of the negativity that is quite normal here.
>
> I would suggest you ceck with one of the Forth facebook groups.
> There you can find probably more help and in a friendy way.
> https://www.facebook.com/groups/PROGRAMMINGFORTH

Another option might be the Minimalist group
https://www.facebook.com/groups/minimalistcomputing

On Saturday, July 22, 2023 at 6:34:49 AM UTC-4, none albert wrote:
> In article <6c69c287-05c7-4a09...@googlegroups.com>,
> Lorem Ipsum <gnuarm.del...@gmail.com> wrote:
> >On Friday, July 21, 2023 at 3:02:07 PM UTC-4, Christopher Lozinski wrote:
> <SNIP>
> >> I expect to do a talk tomorrow at the SVFIG meeting on "A Review of
> >Forth Processors".
> >
> >I would suggest you start with a definition of "forth processor". I
> >claim that none exist. Everyone seems to have their own opinion of what
> >a "forth" processor is.
> I'm content with the definition:
> A Forth processor is a processor with an instruction set inspired by
> or related to Forth instructions.
>
> This is vague and what one considers a Forth processor, someone else
> doesnot. So be it.

Yes, it certainly is vague, and unknowable. If the creator doesn't say, how do you know what inspired the "instruction" set? What does "related" mean? Is the X86 line a Forth processor, as the instruction set includes many instructions corresponding to Forth "instructions" (which I assume means "primitives"), so "related to".

Sounds like you don't really care about defining the term. So I guess you don't actually use it?

--

Rick C.

+- Get 1,000 miles of free Supercharging
+- Tesla referral code - https://ts.la/richard11209

On Saturday, July 22, 2023 at 10:19:36 AM UTC-4, Dave McGuire wrote:
> On 7/21/23 20:19, Lorem Ipsum wrote:
> >>> I know what you are talking about. I'm trying to explain there is no commercial support and the hobbyist market it vanishingly small.
> >> Then you most certainly *don't* know what I'm talking about. I will
> >> explain again.
> >>
> >> I'm not the least bit interested in commercial support. I'm not
> >> interested in the hobbyist market, or any other kind of market, for this
> >> chip or any board that it might get soldered to. I want to hack on this
> >> rather neat little chip for my own enjoyment, and to do that, I need it
> >> on a board.
> >
> > What makes you think I don't understand this? I'm simply explaining to you that you are very, very unlikely to find such a board.
> Ok, perhaps I misunderstood you. Usually when people talk about
> "markets" they're talking about selling things to other people, or other
> people adopting things.

Yes, exactly. If you want to buy something, someone has to be selling it. That's not very likely when there is no market for such a thing.

> I don't care about that. This is for ME and me
> alone, to satisfy a personal interest. I am a commercial embedded
> systems designer; I design other things about which I'm interested in
> sales and adoption. Not this.

I get that. There is nothing unclear about what you are saying. But... if you design a board, it still needs a ton of software to support it. If you make the board available to others, they might share the drudgery of creating some of the perfunctory elements.

If you have no interest in this at all, then no need to even respond to it.

> > If you want to design your own and share it with others, you are unlikely to find anyone who is interested, other than possibly a few people here.
> Apparently you think I should care about that. I don't.

Ok, enough said. You no longer need to respond to anything related to this comment.

> > These days, pretty much everyone who might be interested, has seen the chip, learned enough about it to know there's no point. You will be traveling this road alone, pretty much from scratch.
> Apparently you think I should care about that too. I don't.

You do a lot of reading and replying about things you don't care about.

> You've turned "No, I haven't seen such a thing" to a whole shitload
> of "Oh no, kid, you don't want to do that, oh no, nobody else will want
> it, oh no, that chip sucks, oh no, there's no support" blah blah blah.

You are reading words I never wrote. That's a bad habit. It makes communication difficult.

> I have a GA144 chip that I got as a sample when they first shipped;
> it has been sitting in my lab in its box since then. I've always been
> intrigued by the design, but have never had an opportunity to do
> anything with it. Though Chuck Moore is a bit "out there" I respect the
> man and I'm interested to see what he's designed.

I know you don't want to hear this, but you can learn a lot by reading all the info, then writing a program using the F18A assembly language. Make it a program you might actually want to use, something practical. This will teach you about the "issues" of using the GA144 without building anything. That's why I used a spread sheet to analyze code. I learned a lot more, in a shorter time, than building hardware and getting up to speed on all the "unique" software required to make it work.

> I'm getting older, I'm starting to notice it, I'm hot off the high of
> having gotten a big(ish) Transputer-based cluster up and running at a
> museum, and that reminded me of that GA144 sitting in a drawer. So I
> said to myself, "Self, it's time do something with that chip, but GA's
> dev boards are overpriced and overcomplicated."

Yeah, you will find very, very little in common between the Transputer and the GA144, very little.

> I don't think I should have to explain all of this when the query was
> a simple "Hey, has anyone laid out a board?"

Why do you feel a need to explain yourself? I've offered advice, directly related to your original post and your subsequent posts. You mischaracterize your original post, which was actually more of a wide open question. It literally asked nothing like, "Hey, has anyone laid out a board?" Perhaps you should reread it.

> >> I'm thinking about designing a board, but given the attitude here,
> >> I'm doubt I'll bother mentioning it here.
> >
> > So, you don't even want to share what you develop? Ok, that's not at all unusual.
> I think you missed my point.

Which you have not clarified.

> > It's not at all a matter of "hate". I've looked at the chip seriously. I've written code for it and hand assembled it counting execution time in a spreadsheet. I had become very familiar with it and now see past the "glitz" of its uniqueness and see how limited it really is.
> I understand that. I can read datasheets and user manuals as well as
> the next guy. I usually design with ARM chips and I write firmware in C
> and assembler. I know this will be very different, and that's part of
> why I'm interested in it.

"Different" is not the issue. Lots of things are "different". The GA144 is not very good. It has a very few shining features, then the massive shortcomings get in the way of using those features. It can be used for designs, but it is not well suited for many. DSP is one area where it might find some utility, with the massive parallelism and arrayed comms. But the limited memory (a huge liability) and the limited comms makes it hard to use for anything other than perhaps systolic processing.

I looked at using it to implement a frequency transform. Trying to map an FFT across the nodes was uniquely difficult with the reverse addressing of the coefficients. A DFT was perhaps practical, but limited by the memory. I suppose I could have tried generating the coefficients on the fly with a CORDIC algorithm in each node. I didn't think of that at the time... Hmmmm

> >> None of this even comes close
> >> to answering my initial query.
> >
> > Actually, everything discussed was directly addressing your initial post where you ask about resources, including designing your own. I think I mentioned the support software, which you did not seem to have asked about.
> I didn't ask about software because I had already found what I
> thought I'd need. I certainly didn't ask for a critique of the chip.

And yet you continue to discuss it.

> >> If I'd known my post would've been a
> >> trigger for a deluge of GA144-hater vitriol I'd have kept my damn mouth
> >> shut and designed a board.
> >
> > I'm sorry my replies offend you. This is not "hating". I think everything I've posted has simply been relating facts. You can disagree with opinion, but not so much with facts.
> I didn't ask for those facts. I can read documentation just fine. I
> appreciate your opinion and your experience with these chips, but that
> doesn't quell my curiosity about them.

Ok, so where is the problem? You talk like I'm being rude by talking about the GA144 when you didn't ask, and then you say you appreciate my opinion. Which is it?

> > I recall a poster here who was talking about some real pie-in-the-sky ideas. I don't recall the details, but he wanted to design 32 bit, multiprocessor chips. When I tried to explain to him how his ideas were not realistic, he accused me of "hating" or something similar.
> So this happens to you a lot.

Only with people who don't know what they want. Like you. You continue to complain about my posts, yet, you reply to them. Who is the one with the problem?

> > I wasn't hating, I was just explaining how he would be unlikely to achieve any of his goals, because of their lofty nature. He even claimed he could not have a discussion because of "all the hate". He was messing with his own head. No one was "hating" on him, and he was at all times free to ignore any posts he didn't like.
> I suppose it's good to see Usenet hasn't really changed.
>
> I have the same motivation, to warn people (usually kids) away from
> bad ideas with "there be dragons here" kind of statements. I'm a
> graybeard too.

Just to be clear, I'm not warning anyone away from the GA144. I'm simply pointing out that there are ways to get a handle on the device without building or buying hardware. It's usually the younger ones who want to get their hands on something physical, rather than spending time working on the computer to learn.

> Several years ago, a teenager popped up on a mailing list talking
> about chasing an IBM z890 mainframe that was being auctioned off on
> govdeals. I have one of those machines; it weighs just over a ton and
> is much larger than it tends to look in pictures. The kid was very
> excited about it, and I tried to warn him away from it, explaining that
> if a PDP-11 falls on you, you'll end up in the hospital, but if a z890
> falls on you, you're pretty much done. I move big iron all the time and
> I know how dangerous something like this can be; I was concerned for his
> safety.
>
> Well, the kid ignored my advice and he got the z890. It was quite a
> saga, involving his grandfather excavating land around his basement door
> to get it into his house, etc, and he eventually got it up and running,
> and he didn't get killed in the process. This caught the attention of
> IBM, who immediately hired him and gave him a fantastic job working on
> the development of cutting-edge hardware. He's basically set for life now..
>
> If he'd taken my advice, which was motivated purely from the
> standpoint of concern for his safety because I know that machine well,
> he would've missed out on that great opportunity which changed his life.
> Thankfully, he has forgiven me.
>
> Us graybeards need to share our experience, sure. That's an
> important part of being a member of a society. But there are times when
> we need to let the wide-eyed kids chase crazy dreams. Sometimes they
> will surprise us.

In article <c0600e4e-63d0-473f-b27b-a56def1027e6n@googlegroups.com>,
Lorem Ipsum <gnuarm.deletethisbit@gmail.com> wrote:
<SNIP>
>Yes, it certainly is vague, and unknowable. If the creator doesn't say,
>how do you know what inspired the "instruction" set? What does
>"related" mean? Is the X86 line a Forth processor, as the instruction
>set includes many instructions corresponding to Forth "instructions"
>(which I assume means "primitives"), so "related to".
>
>Sounds like you don't really care about defining the term. So I guess
>you don't actually use it?

At the time I have the only capable computer around the Dutch Forth
we designed a Forth processor, called FIETS (Forth Implementation
by Enhanced Translator and Systems.)
(You guessed it the acronym came first. It is the Dutch word for
push bike.)
It run on an emulator on cp/m and it worked nicely.
Then Chuck Moore come along and designed the NOVIX and we never
pushed it to FPGA. I can read 5" floppies though, so it
might be published some time. Anyway that is the circumstance
I would used the term Forth processor.

I recently bought a HP (20 kg) with a capable GPU, in order to
do CUDA. And here I find myself struggling with an Clojure
implementation challenge...

> Rick C.

Groetjes Albert
--
Don't praise the day before the evening. One swallow doesn't make spring.
You must not say "hey" before you have crossed the bridge. Don't sell the
hide of the bear until you shot it. Better one bird in the hand than ten in
the air. First gain is a cat spinning. - the Wise from Antrim -

On Saturday, July 22, 2023 at 10:23:07 AM UTC-4, Jurgen Pitaske wrote:
> On Friday, 21 July 2023 at 04:05:11 UTC+1, Dave McGuire wrote:
> > Is anyone aware of a simple development board for the GA144 chip? No
> > fluff, extra processors, etc like the (expensive) GreenArrays GA144
> > development boards, just the bare minimum to get the chip going and
> > write code for it. Ideally an open design, i.e. download Gerber files
> > and make a board or two. If there are none, I will probably design one
> > myself but didn't want to reinvent the wheel.
> >
> > Thanks,
> > -Dave McGuire
> >
> > --
> > Dave McGuire, President/Curator
> > Large Scale Systems Museum
> > New Kensington, PA
> Thank you very much for leaving the link to your organisation.
> https://www.mact.io/about_us
>
> This theme GA144 has been discussed for probably 10 years here and in other places now.
>
> It is a shame, that no simple way for playing with these chips exists.
> They are very unusual, but the low interest probably demonstrates,
> that it is not really worth trying, except if you just want to for fun.
>
> Unfortunately the entry cost at Green Arrays is at least $200 for 10 chips - SMT unsoldered so you have to do all of the work.
> To my knowledge, there is not even a proven circuit diagram that you can use to build your own board.
> https://www.greenarraychips.com/home/products/index.php

They did one thing right for the hobbyist, the package is a QFN (no lead), with just one row of pins. This pin count range is often a smaller package with a two row arrangement, which is much harder to lay out.

> When I was more into Forth documentation,
> I asked Greg at Greenarrays, as I was very interested,
> if we could do a clean room implementation of a basic GA144 on FPGA - his feedback was extremely negative.
> So I gave up and did other things.

I had a similar conversation. I was told they could not stop me from implementing the instruction set, but they would not look kindly on the use of their tools to write the code for it. That's when I started working with paper and pencil (figuratively) and learned what I really needed to know.

I asked about the memory interface, as I wanted to make it work with a fast DRAM (100-133 MHz). This would have required three internal nodes (minimum) and the use of the comms channels, of course. To run at full speed the DRAMs have to be timed carefully, so I needed timing data on the comms. They would not share that with me, claiming I could use it to reverse engineer the design. Really? You have a chip, no one is buying, but you are worried about people reverse engineering it?

I did some more work with paper and pencil to prove this simply could not work at speed and gave up. The memory interface is one of the lesser discussed failures. To make up for the limited on chip memory, they added the external memory interface. But they didn't do enough work to assure it would be compatible with DRAM (without slowing it to half speed), meanwhile SRAM continued to get more and more expensive, and power hungry, and is now very hard to find.

> I find it very unusual, that Chuck Moore is involved at GA,
> but seems not really to be pushing it either.

Chuck Moore was never active in promoting the GA144 and is now retired. The last I heard, he was working on using the comms for general purpose data transfer, rather than having to code each design from scratch. I never heard that it got anywhere.

> To me it seems to be self limiting the sales,
> as they would run out of chips soon,

If they were selling any, maybe.

> and I wonder, if the fab who did these chips 10 to 15 years ago
> still exists now and could do more if needed..

It was built at a fab that specializes in old processes. I expect the process is still available.

> If you ask a question here on CLF,
> you have to be aware of the negativity that is quite normal here.

If, by negativity, you mean relating facts that may be "inconvenient", then yes, that happens a lot here.

> I would suggest you ceck with one of the Forth facebook groups.
> There you can find probably more help and in a friendy way.
> https://www.facebook.com/groups/PROGRAMMINGFORTH

Just be careful which group you go with. Some are more friendly than others. A guy who goes by Peter Forth (not his real name) has banned several people from his group.

--

Rick C.

--- Get 1,000 miles of free Supercharging
--- Tesla referral code - https://ts.la/richard11209

On Saturday, 22 July 2023 at 20:01:32 UTC+1, Lorem Ipsum wrote:
> On Saturday, July 22, 2023 at 10:23:07 AM UTC-4, Jurgen Pitaske wrote:
> > On Friday, 21 July 2023 at 04:05:11 UTC+1, Dave McGuire wrote:
> > > Is anyone aware of a simple development board for the GA144 chip? No
> > > fluff, extra processors, etc like the (expensive) GreenArrays GA144
> > > development boards, just the bare minimum to get the chip going and
> > > write code for it. Ideally an open design, i.e. download Gerber files
> > > and make a board or two. If there are none, I will probably design one
> > > myself but didn't want to reinvent the wheel.
> > >
> > > Thanks,
> > > -Dave McGuire
> > >
> > > --
> > > Dave McGuire, President/Curator
> > > Large Scale Systems Museum
> > > New Kensington, PA
> > Thank you very much for leaving the link to your organisation.
> > https://www.mact.io/about_us
> >
> > This theme GA144 has been discussed for probably 10 years here and in other places now.
> >
> > It is a shame, that no simple way for playing with these chips exists.
> > They are very unusual, but the low interest probably demonstrates,
> > that it is not really worth trying, except if you just want to for fun.
> >
> > Unfortunately the entry cost at Green Arrays is at least $200 for 10 chips - SMT unsoldered so you have to do all of the work.
> > To my knowledge, there is not even a proven circuit diagram that you can use to build your own board.
> > https://www.greenarraychips.com/home/products/index.php
> They did one thing right for the hobbyist, the package is a QFN (no lead), with just one row of pins. This pin count range is often a smaller package with a two row arrangement, which is much harder to lay out.
> > When I was more into Forth documentation,
> > I asked Greg at Greenarrays, as I was very interested,
> > if we could do a clean room implementation of a basic GA144 on FPGA - his feedback was extremely negative.
> > So I gave up and did other things.
> I had a similar conversation. I was told they could not stop me from implementing the instruction set, but they would not look kindly on the use of their tools to write the code for it. That's when I started working with paper and pencil (figuratively) and learned what I really needed to know.
>
> I asked about the memory interface, as I wanted to make it work with a fast DRAM (100-133 MHz). This would have required three internal nodes (minimum) and the use of the comms channels, of course. To run at full speed the DRAMs have to be timed carefully, so I needed timing data on the comms. They would not share that with me, claiming I could use it to reverse engineer the design. Really? You have a chip, no one is buying, but you are worried about people reverse engineering it?
>
> I did some more work with paper and pencil to prove this simply could not work at speed and gave up. The memory interface is one of the lesser discussed failures. To make up for the limited on chip memory, they added the external memory interface. But they didn't do enough work to assure it would be compatible with DRAM (without slowing it to half speed), meanwhile SRAM continued to get more and more expensive, and power hungry, and is now very hard to find.
> > I find it very unusual, that Chuck Moore is involved at GA,
> > but seems not really to be pushing it either.
> Chuck Moore was never active in promoting the GA144 and is now retired. The last I heard, he was working on using the comms for general purpose data transfer, rather than having to code each design from scratch. I never heard that it got anywhere.
> > To me it seems to be self limiting the sales,
> > as they would run out of chips soon,
> If they were selling any, maybe.
> > and I wonder, if the fab who did these chips 10 to 15 years ago
> > still exists now and could do more if needed..
> It was built at a fab that specializes in old processes. I expect the process is still available.
> > If you ask a question here on CLF,
> > you have to be aware of the negativity that is quite normal here.
> If, by negativity, you mean relating facts that may be "inconvenient", then yes, that happens a lot here.
> > I would suggest you ceck with one of the Forth facebook groups.
> > There you can find probably more help and in a friendy way.
> > https://www.facebook.com/groups/PROGRAMMINGFORTH
> Just be careful which group you go with. Some are more friendly than others. A guy who goes by Peter Forth (not his real name) has banned several people from his group.
>
> --
>
> Rick C.
>
> --- Get 1,000 miles of free Supercharging
> --- Tesla referral code - https://ts.la/richard11209

Forth2020 I think is the one you mean, it is to my knowledge the restricted Peter Fucking group - not the one I mentioned.
I have never been there so I do not know them and actually I do not want to know.

On Saturday, July 22, 2023 at 4:04:49 PM UTC-4, Jurgen Pitaske wrote:
> On Saturday, 22 July 2023 at 20:01:32 UTC+1, Lorem Ipsum wrote:
> > On Saturday, July 22, 2023 at 10:23:07 AM UTC-4, Jurgen Pitaske wrote:
> > > On Friday, 21 July 2023 at 04:05:11 UTC+1, Dave McGuire wrote:
> > > > Is anyone aware of a simple development board for the GA144 chip? No
> > > > fluff, extra processors, etc like the (expensive) GreenArrays GA144
> > > > development boards, just the bare minimum to get the chip going and
> > > > write code for it. Ideally an open design, i.e. download Gerber files
> > > > and make a board or two. If there are none, I will probably design one
> > > > myself but didn't want to reinvent the wheel.
> > > >
> > > > Thanks,
> > > > -Dave McGuire
> > > >
> > > > --
> > > > Dave McGuire, President/Curator
> > > > Large Scale Systems Museum
> > > > New Kensington, PA
> > > Thank you very much for leaving the link to your organisation.
> > > https://www.mact.io/about_us
> > >
> > > This theme GA144 has been discussed for probably 10 years here and in other places now.
> > >
> > > It is a shame, that no simple way for playing with these chips exists..
> > > They are very unusual, but the low interest probably demonstrates,
> > > that it is not really worth trying, except if you just want to for fun.
> > >
> > > Unfortunately the entry cost at Green Arrays is at least $200 for 10 chips - SMT unsoldered so you have to do all of the work.
> > > To my knowledge, there is not even a proven circuit diagram that you can use to build your own board.
> > > https://www.greenarraychips.com/home/products/index.php
> > They did one thing right for the hobbyist, the package is a QFN (no lead), with just one row of pins. This pin count range is often a smaller package with a two row arrangement, which is much harder to lay out.
> > > When I was more into Forth documentation,
> > > I asked Greg at Greenarrays, as I was very interested,
> > > if we could do a clean room implementation of a basic GA144 on FPGA - his feedback was extremely negative.
> > > So I gave up and did other things.
> > I had a similar conversation. I was told they could not stop me from implementing the instruction set, but they would not look kindly on the use of their tools to write the code for it. That's when I started working with paper and pencil (figuratively) and learned what I really needed to know.
> >
> > I asked about the memory interface, as I wanted to make it work with a fast DRAM (100-133 MHz). This would have required three internal nodes (minimum) and the use of the comms channels, of course. To run at full speed the DRAMs have to be timed carefully, so I needed timing data on the comms. They would not share that with me, claiming I could use it to reverse engineer the design. Really? You have a chip, no one is buying, but you are worried about people reverse engineering it?
> >
> > I did some more work with paper and pencil to prove this simply could not work at speed and gave up. The memory interface is one of the lesser discussed failures. To make up for the limited on chip memory, they added the external memory interface. But they didn't do enough work to assure it would be compatible with DRAM (without slowing it to half speed), meanwhile SRAM continued to get more and more expensive, and power hungry, and is now very hard to find.
> > > I find it very unusual, that Chuck Moore is involved at GA,
> > > but seems not really to be pushing it either.
> > Chuck Moore was never active in promoting the GA144 and is now retired. The last I heard, he was working on using the comms for general purpose data transfer, rather than having to code each design from scratch. I never heard that it got anywhere.
> > > To me it seems to be self limiting the sales,
> > > as they would run out of chips soon,
> > If they were selling any, maybe.
> > > and I wonder, if the fab who did these chips 10 to 15 years ago
> > > still exists now and could do more if needed..
> > It was built at a fab that specializes in old processes. I expect the process is still available.
> > > If you ask a question here on CLF,
> > > you have to be aware of the negativity that is quite normal here.
> > If, by negativity, you mean relating facts that may be "inconvenient", then yes, that happens a lot here.
> > > I would suggest you ceck with one of the Forth facebook groups.
> > > There you can find probably more help and in a friendy way.
> > > https://www.facebook.com/groups/PROGRAMMINGFORTH
> > Just be careful which group you go with. Some are more friendly than others. A guy who goes by Peter Forth (not his real name) has banned several people from his group.
> >
> > --
> >
> > Rick C.
> >
> > --- Get 1,000 miles of free Supercharging
> > --- Tesla referral code - https://ts.la/richard11209
> Forth2020 I think is the one you mean, it is to my knowledge the restricted Peter Fucking group - not the one I mentioned.
> I have never been there so I do not know them and actually I do not want to know.

One of the few things we seem to agree on.

I believe the 2020 group is the one that has, in the name of maintaining Win32Forth, rewritten it to suit his purposes. Win32Forth is a useful tool, but has lost all support. It used to have a Yahoo group, with lots of archived info, but Verizon seems to have ended group support. I believe I started a similar group at groups.io, but it gets very little activity.

--

Rick C.

--+ Get 1,000 miles of free Supercharging
--+ Tesla referral code - https://ts.la/richard11209

On 7/22/23 14:43, Lorem Ipsum wrote:
>>> What makes you think I don't understand this? I'm simply explaining to you that you are very, very unlikely to find such a board.
>> Ok, perhaps I misunderstood you. Usually when people talk about
>> "markets" they're talking about selling things to other people, or other
>> people adopting things.
>
> Yes, exactly. If you want to buy something, someone has to be selling it. That's not very likely when there is no market for such a thing.

Again you've missed my point. I don't want to buy something. I was
never looking to buy something.

>> I don't care about that. This is for ME and me
>> alone, to satisfy a personal interest. I am a commercial embedded
>> systems designer; I design other things about which I'm interested in
>> sales and adoption. Not this.
>
> I get that. There is nothing unclear about what you are saying. But... if you design a board, it still needs a ton of software to support it. If you make the board available to others, they might share the drudgery of creating some of the perfunctory elements.

Once again, the software is available. I didn't ask about that here
because I found it already and have judged it to be sufficient for my needs.

>>> These days, pretty much everyone who might be interested, has seen the chip, learned enough about it to know there's no point. You will be traveling this road alone, pretty much from scratch.
>> Apparently you think I should care about that too. I don't.
>
> You do a lot of reading and replying about things you don't care about.

Wrong.

>> You've turned "No, I haven't seen such a thing" to a whole shitload
>> of "Oh no, kid, you don't want to do that, oh no, nobody else will want
>> it, oh no, that chip sucks, oh no, there's no support" blah blah blah.
>
> You are reading words I never wrote. That's a bad habit. It makes communication difficult.

You wrote every word. No, not verbatim, but you said, in essence all
of it. I'm not going to go pull the quotes out of your previous posts;
you can do that yourself if you really don't remember.

>> I have a GA144 chip that I got as a sample when they first shipped;
>> it has been sitting in my lab in its box since then. I've always been
>> intrigued by the design, but have never had an opportunity to do
>> anything with it. Though Chuck Moore is a bit "out there" I respect the
>> man and I'm interested to see what he's designed.
>
> I know you don't want to hear this, but you can learn a lot by reading all the info, then writing a program using the F18A assembly language. Make it a program you might actually want to use, something practical. This will teach you about the "issues" of using the GA144 without building anything. That's why I used a spread sheet to analyze code. I learned a lot more, in a shorter time, than building hardware and getting up to speed on all the "unique" software required to make it work.

I've read that. What I want to do is BUILD A DAMN BOARD. What
language am I speaking here?

My goal is not to "get up to speed". This is not a commercial
endeavor. It is not about efficiency. I work in this field, but my
interest in this chip is not commercial.

I realize that you're not saying what you said above because you
actually believe it. You're a net.troll, and you say these things to
get a rise out of people. Unfortunately I often cannot resist the bait.

>> I'm getting older, I'm starting to notice it, I'm hot off the high of
>> having gotten a big(ish) Transputer-based cluster up and running at a
>> museum, and that reminded me of that GA144 sitting in a drawer. So I
>> said to myself, "Self, it's time do something with that chip, but GA's
>> dev boards are overpriced and overcomplicated."
>
> Yeah, you will find very, very little in common between the Transputer and the GA144, very little.

I typed and deleted a few responses here. But the most appropriate
seems to be:

"NO SHIT, SHERLOCK."

>> I don't think I should have to explain all of this when the query was
>> a simple "Hey, has anyone laid out a board?"
>
> Why do you feel a need to explain yourself? I've offered advice, directly related to your original post and your subsequent posts. You mischaracterize your original post, which was actually more of a wide open question. It literally asked nothing like, "Hey, has anyone laid out a board?" Perhaps you should reread it.

I'll refer you to my original post. That's exactly what I wrote.
Again not verbatim, but that was my request.

>>>> I'm thinking about designing a board, but given the attitude here,
>>>> I'm doubt I'll bother mentioning it here.
>>>
>>> So, you don't even want to share what you develop? Ok, that's not at all unusual.
>> I think you missed my point.
>
> Which you have not clarified.

Wow.

>>> It's not at all a matter of "hate". I've looked at the chip seriously. I've written code for it and hand assembled it counting execution time in a spreadsheet. I had become very familiar with it and now see past the "glitz" of its uniqueness and see how limited it really is.
>> I understand that. I can read datasheets and user manuals as well as
>> the next guy. I usually design with ARM chips and I write firmware in C
>> and assembler. I know this will be very different, and that's part of
>> why I'm interested in it.
>
> "Different" is not the issue. Lots of things are "different". The GA144 is not very good.

In your opinion. I stopped caring about your opinion when I realized
that you're just a troll, not any sort of knowledgeable person whose
advice I should value.

I'll point out yet again, that I did not ask you (or anyone else) if
they thought the GA144 is any good. I've been in this business a long
time; I will come to my own conclusions about that.

>> I didn't ask about software because I had already found what I
>> thought I'd need. I certainly didn't ask for a critique of the chip.
>
> And yet you continue to discuss it.

Because you're a highly skilled troll and I lack self control.

> Ok, so where is the problem? You talk like I'm being rude by talking about the GA144 when you didn't ask, and then you say you appreciate my opinion. Which is it?

I was being generous. You seem to need it.

>>> I recall a poster here who was talking about some real pie-in-the-sky ideas. I don't recall the details, but he wanted to design 32 bit, multiprocessor chips. When I tried to explain to him how his ideas were not realistic, he accused me of "hating" or something similar.
>> So this happens to you a lot.
>
> Only with people who don't know what they want. Like you. You continue to complain about my posts, yet, you reply to them. Who is the one with the problem?

I know exactly what I want. I clearly stated it in my first post in
this thread. I want a GA144 on a simple board with no fluff so I can
explore it. I've explained this multiple times.

>>> I wasn't hating, I was just explaining how he would be unlikely to achieve any of his goals, because of their lofty nature. He even claimed he could not have a discussion because of "all the hate". He was messing with his own head. No one was "hating" on him, and he was at all times free to ignore any posts he didn't like.
>> I suppose it's good to see Usenet hasn't really changed.
>>
>> I have the same motivation, to warn people (usually kids) away from
>> bad ideas with "there be dragons here" kind of statements. I'm a
>> graybeard too.
>
> Just to be clear, I'm not warning anyone away from the GA144. I'm simply pointing out that there are ways to get a handle on the device without building or buying hardware. It's usually the younger ones who want to get their hands on something physical, rather than spending time working on the computer to learn.

Yes but that's like an inflatable doll. It will get the job done,
but it's usually not what you really want. I want HARDWARE. I've
already read the documentation.

Why am I explaining this yet again? Of course: I Have Been Trolled.

I came to comp.sys.forth with my guard down thinking that there would
be civilized people here. I regret that I was mistaken.

-Dave

--
Dave McGuire, President/Curator
Large Scale Systems Museum
New Kensington, PA

On 7/22/23 15:01, Lorem Ipsum wrote:
>> I would suggest you ceck with one of the Forth facebook groups.
>> There you can find probably more help and in a friendy way.
>> https://www.facebook.com/groups/PROGRAMMINGFORTH
>
> Just be careful which group you go with. Some are more friendly than others. A guy who goes by Peter Forth (not his real name) has banned several people from his group.

Interesting! Peter Forth (whatever his real name may be) has never
been anything other than nice and friendly to me, where YOU have never
been anything other than a dick.

Every newsgroup has their armchair expert, wannabe authority troll.
I'm glad I found this one's early before making any friends here.

*PLONK*

-Dave

--
Dave McGuire, President/Curator
Large Scale Systems Museum
New Kensington, PA

On 23/07/2023 10:37 am, Lorem Ipsum wrote:
> On Saturday, July 22, 2023 at 4:04:49 PM UTC-4, Jurgen Pitaske wrote:
>> Forth2020 I think is the one you mean, it is to my knowledge the restricted Peter Fucking group - not the one I mentioned.
>> I have never been there so I do not know them and actually I do not want to know.
>
> One of the few things we seem to agree on.
>
> I believe the 2020 group is the one that has, in the name of maintaining Win32Forth, rewritten it to suit his purposes. Win32Forth is a useful tool, but has lost all support. It used to have a Yahoo group, with lots of archived info, but Verizon seems to have ended group support. I believe I started a similar group at groups.io, but it gets very little activity.
>

IIRC the forth promoted was based on SP-Forth and closed-source. Can't seem
to find any current links so may no longer be promoted. Main attraction of the
group today appears to be their Zoom meetings and presentations often by well-
known Forth personalities.

On Saturday, July 22, 2023 at 9:49:04 PM UTC-4, Dave McGuire wrote:
> On 7/22/23 14:43, Lorem Ipsum wrote:
> >>> What makes you think I don't understand this? I'm simply explaining to you that you are very, very unlikely to find such a board.
> >> Ok, perhaps I misunderstood you. Usually when people talk about
> >> "markets" they're talking about selling things to other people, or other
> >> people adopting things.
> >
> > Yes, exactly. If you want to buy something, someone has to be selling it. That's not very likely when there is no market for such a thing.
> Again you've missed my point. I don't want to buy something. I was
> never looking to buy something.

So, "Is anyone aware of a simple development board for the GA144 chip?" is not asking for something to buy? Sorry, I thought that was the point of your original post. My bad.

> >> I don't care about that. This is for ME and me
> >> alone, to satisfy a personal interest. I am a commercial embedded
> >> systems designer; I design other things about which I'm interested in
> >> sales and adoption. Not this.
> >
> > I get that. There is nothing unclear about what you are saying. But... if you design a board, it still needs a ton of software to support it. If you make the board available to others, they might share the drudgery of creating some of the perfunctory elements.
> Once again, the software is available. I didn't ask about that here
> because I found it already and have judged it to be sufficient for my needs.

Please correct me if I'm wrong, as it has been a long time since I looked at the tools, but they require a GA144 to act as a boot device for another GA144, no? Maybe I'm not remembering the eval board accurately.

> >>> These days, pretty much everyone who might be interested, has seen the chip, learned enough about it to know there's no point. You will be traveling this road alone, pretty much from scratch.
> >> Apparently you think I should care about that too. I don't.
> >
> > You do a lot of reading and replying about things you don't care about.
> Wrong.

LOL This is a perfect example of you replying to things you keep saying you don't care about.

> >> You've turned "No, I haven't seen such a thing" to a whole shitload
> >> of "Oh no, kid, you don't want to do that, oh no, nobody else will want
> >> it, oh no, that chip sucks, oh no, there's no support" blah blah blah.
> >
> > You are reading words I never wrote. That's a bad habit. It makes communication difficult.
> You wrote every word. No, not verbatim, but you said, in essence all
> of it.

Ah, see, that's the problem. You are reading things I didn't write. You don't like what I'm saying, so you perceive it as hostile. It's not hostility. I'm just sharing the facts of what I found.

> I'm not going to go pull the quotes out of your previous posts;
> you can do that yourself if you really don't remember.

That's fine. I know I wrote nothing remotely like, "Oh no, kid, you don't want to do that, oh no, nobody else will want it, oh no, that chip sucks, oh no, there's no support". YOU have added the hostile tone, that I never wrote.

But the lack of support is pretty accurate, and there's virtually no one working with the chip these days. So that's also accurate.

As to you *wanting* to work with the chip. That's up to you. People play with all manner of things. All I'm trying to say is that the chip is not really any more suited to most MCU tasks than any of a thousand other MCUs. So, it has no purpose in life. No hostility intended. If you want to try it out, go for it.

> >> I have a GA144 chip that I got as a sample when they first shipped;
> >> it has been sitting in my lab in its box since then. I've always been
> >> intrigued by the design, but have never had an opportunity to do
> >> anything with it. Though Chuck Moore is a bit "out there" I respect the
> >> man and I'm interested to see what he's designed.
> >
> > I know you don't want to hear this, but you can learn a lot by reading all the info, then writing a program using the F18A assembly language. Make it a program you might actually want to use, something practical. This will teach you about the "issues" of using the GA144 without building anything.. That's why I used a spread sheet to analyze code. I learned a lot more, in a shorter time, than building hardware and getting up to speed on all the "unique" software required to make it work.
> I've read that. What I want to do is BUILD A DAMN BOARD. What
> language am I speaking here?
>
> My goal is not to "get up to speed". This is not a commercial
> endeavor. It is not about efficiency. I work in this field, but my
> interest in this chip is not commercial.

You keep repeating that. This also indicates that you think I'm saying something I'm not. If you want to build a board, go ahead. No one is stopping you.

> I realize that you're not saying what you said above because you
> actually believe it. You're a net.troll, and you say these things to
> get a rise out of people. Unfortunately I often cannot resist the bait.

Ok, so I'm a troll, because you don't like what I'm saying. Sorry, that's not what a troll is. I'm not deliberately trying to get a rise out of you, or to create confrontation. I'm trying to have a technical discussion, but you don't want to hear what I have to say. Instead of simply ignoring it, you turn hostile. So don't blame the hostility on me!

> >> I'm getting older, I'm starting to notice it, I'm hot off the high of
> >> having gotten a big(ish) Transputer-based cluster up and running at a
> >> museum, and that reminded me of that GA144 sitting in a drawer. So I
> >> said to myself, "Self, it's time do something with that chip, but GA's
> >> dev boards are overpriced and overcomplicated."
> >
> > Yeah, you will find very, very little in common between the Transputer and the GA144, very little.
> I typed and deleted a few responses here. But the most appropriate
> seems to be:
>
> "NO SHIT, SHERLOCK."

LOL I've written nothing to deserve a response like that. I think it is you who are out of line, not me.

> >> I don't think I should have to explain all of this when the query was
> >> a simple "Hey, has anyone laid out a board?"
> >
> > Why do you feel a need to explain yourself? I've offered advice, directly related to your original post and your subsequent posts. You mischaracterize your original post, which was actually more of a wide open question. It literally asked nothing like, "Hey, has anyone laid out a board?" Perhaps you should reread it.
> I'll refer you to my original post. That's exactly what I wrote.
> Again not verbatim, but that was my request.

I'm starting to think maybe you have some cognitive issues. Here is your OP.

"Is anyone aware of a simple development board for the GA144 chip? No
fluff, extra processors, etc like the (expensive) GreenArrays GA144
development boards, just the bare minimum to get the chip going and
write code for it. Ideally an open design, i.e. download Gerber files
and make a board or two. If there are none, I will probably design one
myself but didn't want to reinvent the wheel."

The very first sentence is asking if such a board exists. I realize now that you are fishing for an open source design you can copy, but that first sentence says to me you would like to buy a board. Then you go on to explain that you would build a board if someone else has designed it.

> >>>> I'm thinking about designing a board, but given the attitude here,
> >>>> I'm doubt I'll bother mentioning it here.
> >>>
> >>> So, you don't even want to share what you develop? Ok, that's not at all unusual.
> >> I think you missed my point.
> >
> > Which you have not clarified.
> Wow.

I can't help you if that is the extent of your replies, "wow".

> >>> It's not at all a matter of "hate". I've looked at the chip seriously.. I've written code for it and hand assembled it counting execution time in a spreadsheet. I had become very familiar with it and now see past the "glitz" of its uniqueness and see how limited it really is.
> >> I understand that. I can read datasheets and user manuals as well as
> >> the next guy. I usually design with ARM chips and I write firmware in C
> >> and assembler. I know this will be very different, and that's part of
> >> why I'm interested in it.
> >
> > "Different" is not the issue. Lots of things are "different". The GA144 is not very good.
> In your opinion. I stopped caring about your opinion when I realized
> that you're just a troll, not any sort of knowledgeable person whose
> advice I should value.
>
> I'll point out yet again, that I did not ask you (or anyone else) if
> they thought the GA144 is any good.

On Saturday, July 22, 2023 at 9:49:05 PM UTC-4, Dave McGuire wrote:
> On 7/22/23 15:01, Lorem Ipsum wrote:
> >> I would suggest you ceck with one of the Forth facebook groups.
> >> There you can find probably more help and in a friendy way.
> >> https://www.facebook.com/groups/PROGRAMMINGFORTH
> >
> > Just be careful which group you go with. Some are more friendly than others. A guy who goes by Peter Forth (not his real name) has banned several people from his group.
> Interesting! Peter Forth (whatever his real name may be) has never
> been anything other than nice and friendly to me, where YOU have never
> been anything other than a dick.
>
> Every newsgroup has their armchair expert, wannabe authority troll.
> I'm glad I found this one's early before making any friends here.

You can anyone you wish. But I'm not the only person who has had issues with Peter Forth. Juergen has already indicated his displeasure. Don't you find it odd that he uses a fake name?

I have done nothing to you. You have misinterpreted my posts as somehow hostile. I've only related information. Not one harsh word. Not one! Yet, you perceive me as a troll and an "armchair expert" which would imply I've done nothing in Forth or engineering otherwise. This is far from the truth.

Whatever. There are a lot of good people here. Many of them are happy to help. But if you can't handle honesty, I guess this is not the place for you. If you are willing to accept information from others, without taking it personally, then this is a great place to discuss Forth.

> *PLONK*

I always find it amusing that people feel the need to be so dramatic when they killfile someone. That's the sort of thing that teenagers do. Sorry you have so much trouble with all this.

--

Rick C.

-++ Get 1,000 miles of free Supercharging
-++ Tesla referral code - https://ts.la/richard11209