A new MISC forum?

[. .]

I'm looking to start up some groups for my own commercial projects
including future Misc ones.    As there does not seem to be a forum
for Misc development,  I can include a subforum for misc development
and other simple misc like processors you guys are doing.   I like
this misc idea and think it is promising.  I'm also interested in the
open os thing I mentioned before, and open hardware projects.

Starting a group on Usenet seems like a complex mess. But I've spent
a number of hours reading through Google group information, and it
seems to have a lot of what is needed.  I can start up a group and put
in sub categories as subgroups.  It can be setup as a non moderated
email list, which should minimise any legal responsibility, reduce
time looking after it and keeps conversations off the web.   It can be
non public invitation only, which would eliminate passer bys that seem
to be a hassle. I could expand it into a group web forum eventually.

I would prefer a freeform website, blog and forums, but as Google does
not seem to support that integration, it would be too much non project
downtime to look after a custom site for now.

I see many people have been on misc in the past, but it has died off,
so a crowd might be possible.

I would also be happy if anyone did summary reports back here every
now and then on developments, apart from anything specific I might
post back about.

What do people think? A Usenet subforum, MiscProcessorLike, could be
possible instead, but that would take time and I don't want to be
caught up in any legal/moderation hassle, if there is any.

I can discuss some of my vision if you like, it is pretty helping the
world through design and business.


Thanks again.


Steve.

[Paul Rubin]

". ." <nosp...@gmail.com> writes:
> I'm looking to start up some groups for my own commercial projects
> including future Misc ones.    As there does not seem to be a forum
> for Misc development,  I can include a subforum for misc development
> and other simple misc like processors you guys are doing. 

If it's MISC stuff in general, I think it's fine to put it here in CLF
unless the volume gets large enough to seriously cramp the other stuff.

If it's specifically for your own projects, I think it's still fine to
put it on CLF as long as the volume stays relatively low (occasional
announcements and discussion).

I prefer Usenet to web forums. I dislike mailing lists and hate Google
Groups. If you don't use Usenet, I think your best choice is a web
forum. I'm happier to use privately hosted ones but I can understand if
you don't want to run one, since they involve some hassle and
expenditure (the costs can be pretty low though). There are of course
also various no-cost hosted ones with various types of obnoxious
advertising and tracking.

I suppose you could even start a subreddit (reddit.com).

[Steve]

On Apr 28, 1:21 am, Paul Rubin <no.em...@nospam.invalid> wrote:

> ". ." <nospam...@gmail.com> writes:
>
> I prefer Usenet to web forums.  I dislike mailing lists and hate Google
> Groups.  If you don't use Usenet, I think your best choice is a web
> forum.  I'm happier to use privately hosted ones but I can understand if
> you don't want to run one, since they involve some hassle and
> expenditure (the costs can be pretty low though).  There are of course
> also various no-cost hosted ones with various types of obnoxious
> advertising and tracking.
>
> I suppose you could even start a subreddit (reddit.com).

Well, if that is all the interest there is, maybe it is a bit
pointless. There should be hundreds of posts or participants a week
about misc here, from past numbers in the archive and GA development
board sales, but 5 people at the moment and admittedly it just seems a
slow period. Having things washed in among other Forth topics is not
a way to grow it either from a market point of view. Very surprised
Green Array does not have a forum for their developers, or at least
setup a Usenet sub group for misc and point everybody there.

Paul, I agree about groups, seems extensive enough over Usenet, but
typical Google style, took hours to read bits and picees that could
have been simply done and implemented in 2 web pages, one for use and
one for administration, and no way to archive your data out easily.
However, I see how a mailing list could work like Usenet, just open
an email for that group, set to pop3 if in gmail, use the emai client
of choice, receive everything in conversation mode, and it breaks down
like Usenet topic lists.

But yes, a thought did occur to me about hosting services you
mentioned. Groups does this to an extent, but something that just
hosted full blown site, blog and board functionality, that you could
receive and reply to from email, would be great. Maybe something
that just hosted a structure of boards, that you could just insert
your own into with the running done by the hosting service. Are there
any you might be aware of Paul.

Also, you mentioned a guy that is interested in doing web browsers etc
on misc in another thread, who would that be? I asked about it and
mentioned a few other things, but you might have missed as I have
been on webforums a lot and combine responses.

I think I will start another thread on a misc development discussion,
to gauge interest in Misc development, Im starting to see something,
if people can let others know? For me to support Misc requires
significant resources, and community interaction on the open products
side. If there is no community wanting something it is not worth
doing as much. I`ll ask about that too.


Thanks.

[Paul Rubin]

Steve <nosp...@gmail.com> writes:
> Groups does this to an extent, but something that just
> hosted full blown site, blog and board functionality, that you could
> receive and reply to from email, would be great. Maybe something
> that just hosted a structure of boards, that you could just insert
> your own into with the running done by the hosting service. Are there
> any you might be aware of Paul.

I don't know of places that specialize in board hosting (they exist, I
just don't pay attention to them). I'm generally not too interested in
using those sites (even though they're usually free) since their main
purpose is to capture viewer information and sell advertising. I'd
rather pay a small amount for something whose contents are under my own
control.

There are a huge number of places that will sell you very cheap
full-fledged virtual servers on which you can run blogs or whatever else
you want. lowendbox.com is a good place to look for those, and you can
get a nice one for 1 or 2 dollars per month. But, you need a bit of
knowledge to run such a server, and the very cheap services don't offer
a lot of support (that's why they are so cheap). So you have to decide
if you're ready to deal with that.

> Also, you mentioned a guy that is interested in doing web browsers

That was sort of a joke. There is a guy who keeps asking about that
but discussions with him never seem to go anywhere.

> For me to support Misc requires significant resources, and community
> interaction on the open products side. If there is no community
> wanting something it is not worth doing as much.

I think a bunch of us have looked at the MISC stuff just out of interest
in what Chuck has been up to, but the capabilities of the current chip
are so constrained that it's quite hard to find practical target
applications. Maybe you are the one who will come up with the great
idea that the rest of us have missed. Or maybe a future GA chip will be
easier to develop with and that's when good applications will appear.
Or maybe, as you say, the whole thing will never take off.

[Hugh Aguilar]

On Apr 27, 10:07 am, ". ." <nospam...@gmail.com> wrote:

> I see many people have been on misc in the past, but it has died off,
> so a crowd might be possible.

That is only logical! Because everybody left the misc forum, that
crowd is now ready to jump on your discussion group, because it is
about the same topic.

> I can discuss some of my vision if you like, it is pretty helping the
> world through design and business.

"Business" is altruism???

> Steve.

Aren't you a sock-puppet of Rickman, the notorious troll of
comp.lang.forth?

[Steve]

Nope. Nice picture though. :-)

I don't know if you are being sincere or not above Hugh. Are you
saying there was a misc forum and its a good idea to start a new one?
Im working on the principle of better something than nothing.

Yes, altruism through business, lots of this stuff starting to
happen. I've run out of time and have to try to answer Paul post and
go, but Ill give a quick answer. You run a business to make your
income but also to spend some of the profit to helping out society.
But you can make that income on making products that really benefit
society. Also your business conduct and policy can affect the
market, especially if you are the size of Apple. Even if you are
small, people know what you are about and are worth supporting, and it
can encourage better conduct in other businesses, even if only a few
shall ones. Helping out society can mean a number of things, charity
is one, but you can start up other helpful businesses, sponsor
research and social or economic reform, even sponsor a politician you
know is going to benefit the way things are done in the community, in
partnership. There is enough money, it is just where it goes to.
There are lots of ideals, but this is the reality you can take baby
steps towards it.


Thanks


Steve.

[Steve]

Sorry Paul, I winter have much time. My software crashed and lost the
new thread I mentioned, and have used up my time trying to find an
article I remember on how Nvidia is topping out power savings because
of leakage on its chips. Pointing to Misc as an architectual solution
in the new thread.

On Apr 28, 4:04 am, Paul Rubin <no.em...@nospam.invalid> wrote:
> Steve <nospam...@gmail.com> writes:

/Hosting board site blog solutions.

>
> I don't know of places that specialize in board hosting (they exist, I
> just don't pay attention to them).  I'm generally not too interested in
> using those sites (even though they're usually free) since their main
> purpose is to capture viewer information and sell advertising.  I'd
> rather pay a small amount for something whose contents are under my own
> control.

Yes, that is my main interest too.

> There are a huge number of places that will sell you very cheap
> full-fledged virtual servers on which you can run blogs or whatever else
> you want.  lowendbox.com is a good place to look for those, and you can
> get a nice one for 1 or 2 dollars per month.  But, you need a bit of
> knowledge to run such a server, and the very cheap services don't offer
> a lot of support (that's why they are so cheap).  So you have to decide
> if you're ready to deal with that.

Yep, with all Im doing Id prefer to pay for something plus and play
and hosted.

> > Also, you mentioned a guy that is interested in doing web browsers
>
> That was sort of a joke.  There is a guy who keeps asking about that
> but discussions with him never seem to go anywhere.

Still worth checking out?

> > For me to support Misc requires significant resources, and community
> > interaction on the open products side.  If there is no community
> > wanting something it is not worth doing as much.
>
> I think a bunch of us have looked at the MISC stuff just out of interest
> in what Chuck has been up to, but the capabilities of the current chip
> are so constrained that it's quite hard to find practical target
> applications.  Maybe you are the one who will come up with the great
> idea that the rest of us have missed.  Or maybe a future GA chip will be
> easier to develop with and that's when good applications will appear.
> Or maybe, as you say, the whole thing will never take off.

Wait for the new thread, an idea did occur to me, an literal opening
in the market. But I agree, it seems that the thing does not even
execute from it's external memory bus, mind boggeling, from what is
essentially a 2001 design. Just that alone would allow enough primary
performance and functionality to increase the number of applications
it is suitable for at least ten times. The 32 bit design looks the
most promising though, and that is more suitable for what is going to
be in the other thread.

[Steve]

On Apr 28, 5:51 am, Hugh Aguilar <hughaguila...@yahoo.com> wrote:

Ah, I see, he was the one that wanted two of my rings on the old
thread. I was actually thinking of sending him two if it ever
happened, but that was more a though bubble to discuss and maybe make
into a project. I am interested in doing something myself, but more
expensive.

[Steve]

On Apr 28, 3:37 am, Steve <nospam...@gmail.com> wrote:
> On Apr 28, 1:21 am, Paul Rubin <no.em...@nospam.invalid> wrote:
>
> > ". ." <nospam...@gmail.com> writes:
>
> > I dislike mailing lists and hate Google Groups.

Paul, when you say you dislike groups, do you mean just for usenet
use, because you can set them up very much like hosted bulletin boards
with a few new features?

/For everyone.

Tell you what, if anyone can find 20 eager people, I would like to
start a mailing list, if more than 20 than a group/board might be
possible with a eventual target of at least 100 active members. If we
can keep membership moderated and the discussion list private, it
should keep a better level of professional discussion with out passer
bys I can see in the history here.

Thanks again.

[Paul Rubin]

Steve <nosp...@gmail.com> writes:
>> > I dislike mailing lists and hate Google Groups.
> Paul, when you say you dislike groups, do you mean just for usenet
> use,

No I mean in general. I prefer either traditional (NNTP) Usenet or else
privately operated web forums.

> Tell you what, if anyone can find 20 eager people

I think on this group there are a dozen or so people who were interested
enough in the GA144 to spend a while looking at its docs and so on, and
a smaller group who tried to think up applications for it, but that's a
bit short of "eager".

I guess there are some also Forth cpu's like the b16 that you might or
might not count as MISC.

Overall I think "let's create a new XYZ group to drum up interest in
XYZ" has been tried many times and never works. The time to create an
XYZ group is when discussion of XYZ occupies a big enough chunk of the
traffic of a group on some related topic, that it helps everyone
involved to split off the XYZ stuff. So I think it's ok to keep MISC
stuff here on CLF until/unless traffic grows enough to want to split
it off. That's the MISC spirit, anyway: solve the problem you have.

[Steve]

On Apr 28, 8:03 am, Paul Rubin <no.em...@nospam.invalid> wrote:

> Steve <nospam...@gmail.com> writes:
>
> > Tell you what, if anyone can find 20 eager people
>
> I think on this group there are a dozen or so people who were interested
> enough in the GA144 to spend a while looking at its docs and so on, and
> a smaller group who tried to think up applications for it, but that's a
> bit short of "eager".

Eager to post, but including people not here but in the GA sphere.

> I guess there are some also Forth cpu's like the b16 that you might or
> might not count as MISC.

Including all those misc like & fpga etc. Might as well make it any
forth CPU if not enough people turn up.

comp.language.forth.processor Is that valid use on Usenet
comp.lang ? Because people are dispersed from here, it is unlikely I
could convince the Usenet big 8 committee to approve a sub group
anyway. But from history it appears numbers of misc threads here goes
up and down over time, a new misc processor should drive up numbers
again.

I could have proceeded this week except for the uncertainty as to
hosting and numbers. Maybe I should reach out further to get more
people?

> Overall I think "let's create a new XYZ group to drum up interest in
> XYZ" has been tried many times and never works.  The time to create an
> XYZ group is when discussion of XYZ occupies a big enough chunk of the
> traffic of a group on some related topic, that it helps everyone
> involved to split off the XYZ stuff.  So I think it's ok to keep MISC
> stuff here on CLF until/unless traffic grows enough to want to split
> it off.  That's the MISC spirit, anyway: solve the problem you have.

I agree, the traffic is just too low, but there must be many more out
there. Having XYZ in ABC might discourage interest too. When I
eventually release my open projects it might lead to people producing
a lot of threads, especially if I can attach real hardware products to
program.


/Everybody

If anybody knows of good hosting/servicing for full custom web, and
forum and blog, please let me know. I will have a look around and at
Paul options, but likely not to come up with the best options. Maybe
setting something up can improve demand to go to the big 8 committee
eventually.

If anybody knows of people out there that maybe interested in a group,
or the project proposals I'm going to list on the other threads,
please let them know.


Thanks.

[Paul Rubin]

Steve <nosp...@gmail.com> writes:
> comp.language.forth.processor Is that valid use on Usenet
> comp.lang ?

I think it's hard to justify this regardless of whether it's a "type
error" to have a hardware subgroup under comp.lang. An alt group would
be fine in principle but seems like unnecessary fragmentation to me.

> a new misc processor should drive up numbers again.

I'd wait til the numbers actually go up, before trying to do anything
about it.

> Maybe I should reach out further to get more people?

I don't think they are out there. Aren't there some forums devoted to
the old Seaforth stuff? They are very quiet.

> I agree, the traffic is just too low, but there must be many more out
> there.

That is possible, but I would want to see evidence before believing it.

> Having XYZ in ABC might discourage interest too.

Yeah, that's what I mean. Across the history of Usenet, people have had
that idea over and over, and it's difficult to refute by pure logic.
It's just my observation from experience that despite many tries, it
never works. Splitting off comp.lang.haskell from comp.lang.functional
was a recent example. Instead of increasing total traffic, it just
split the existing trickle from c.l.f. so now there are two almost-dead
groups. This is of course just my opinion, worth the $0.00 that you
paid for it, but I think the time to split a group is when the traffic
from the subtopic becomes unmanageably high, not when it's too small to
see.

> When I eventually release my open projects it might lead to people
> producing a lot of threads, especially if I can attach real hardware
> products to program.

That will be great, but until it happens I think we're getting ahead of
ourselves. When I make a million dollars I might get a fancy car, but
there's no point to hanging around dealerships until the money is in the
bank.

> If anybody knows of good hosting/servicing for full custom web, and
> forum and blog, please let me know.

Maybe this helps:
http://www.lowendbox.com/blog/wordpress-cheap-vps-lowendscript/
http://www.lowendbox.com/blog/yes-you-can-run-18-static-sites-on-a-64mb-link-1-vps/

[rickman]

On 4/28/2013 10:24 AM, Steve wrote:
> On Apr 28, 8:03 am, Paul Rubin<no.em...@nospam.invalid> wrote:
>> Steve<nospam...@gmail.com> writes:
>>
>>> Tell you what, if anyone can find 20 eager people
>>
>> I think on this group there are a dozen or so people who were interested
>> enough in the GA144 to spend a while looking at its docs and so on, and
>> a smaller group who tried to think up applications for it, but that's a
>> bit short of "eager".
>
> Eager to post, but including people not here but in the GA sphere.

You are looking for a way to talk about MISC processors but people are
already doing that here an to a lesser extent in sci.electronics.design.
I expect this topic would be welcome in comp.arch.embedded too.

Why try to re-invent the wheel in this regard?

As to using different medium than newsgroups, I wouldn't expect to find
much support for that in a newsgroup. People who are here have made
some effort to use this medium and so must feel it is more useful that
others.

Why not open a thread or two in some of the above mentioned groups and
see where they lead?

--

Rick

[rickman]

On 4/28/2013 2:55 AM, Steve wrote:
> On Apr 28, 5:51 am, Hugh Aguilar<hughaguila...@yahoo.com> wrote:
>> On Apr 27, 10:07 am, ". ."<nospam...@gmail.com> wrote:
>>
>>> I see many people have been on misc in the past, but it has died off,
>>> so a crowd might be possible.
>>
>> That is only logical! Because everybody left the misc forum, that
>> crowd is now ready to jump on your discussion group, because it is
>> about the same topic.
>>
>>> I can discuss some of my vision if you like, it is pretty helping the
>>> world through design and business.
>>
>> "Business" is altruism???
>>
>>> Steve.
>>
>> Aren't you a sock-puppet of Rickman, the notorious troll of
>> comp.lang.forth?
>
> Nope. Nice picture though. :-)
>
> I don't know if you are being sincere or not above Hugh. Are you
> saying there was a misc forum and its a good idea to start a new one?
> Im working on the principle of better something than nothing.

I don't really understand his post either. Is he being sarcastic or
just irrational?

BTW, I am not a sock puppeteer. But then if I were, that is exactly
what I would say isn't it? lol

Once when I wanted to get Chuck Moore's attention I had noticed that he
had on one of his blog pages that he didn't often respond to emails from
strangers because the world is full of predators. So I sent an email
with the subject "I am not a predator"... but that's exactly what a
predator *would* write... so I was a little surprised when he responded.
lol

--

Rick

[rickman]

Hey, I would be very interested in working with you if you are both
serious and capable. Why not start a discussion on your computer ideas
rather than worrying about where to discuss it?

I was put in contact once with an entrepreneur who had an idea for an
industrial lighting control that could save a small but significant
percentage of the power consumed. He had been burnt a couple of times
be consulting engineers who weren't able to meet his expectations. His
desired working arrangement was so onerous and the conditions for
payment so vague I couldn't work with him. A couple of years later a
friend was out of work so I tried to put them in touch. The guy had
gotten *nowhere* with his idea but still had the same onerous working
arrangement and so we still could not work together. The moral: a great
idea that you never do anything with is not so great...

--

Rick

[Hugh Aguilar]

On Apr 27, 11:55 pm, Steve <nospam...@gmail.com> wrote:
> On Apr 28, 5:51 am, Hugh Aguilar <hughaguila...@yahoo.com> wrote:
> > On Apr 27, 10:07 am, ". ." <nospam...@gmail.com> wrote:
>
> > > I see many people have been on misc in the past, but it has died off,
> > > so a crowd might be possible.
>
> > That is only logical! Because everybody left the misc forum, that
> > crowd is now ready to jump on your discussion group, because it is
> > about the same topic.
>
> > > I can discuss some of my vision if you like, it is pretty helping the
> > > world through design and business.
>
> > "Business" is altruism???
>
> > > Steve.
>
> > Aren't you a sock-puppet of Rickman, the notorious troll of
> > comp.lang.forth?
>
> Nope.  Nice picture though. :-)
>
> I don't know if you are being sincere or not above Hugh.  Are you
> saying there was a misc forum and its a good idea to start a new one?
> Im working on the principle of better something than nothing.

I'm saying that if the previous forum died out, another one is going
to die out too. There is no interest in that kind of thing. I was
assuming that you were one of Rickman's sock-puppets --- his troll-
platform is to be an expert on stack-based FPGA processors --- if
you're not Rickman though, then I'll give you the benefit of the doubt
and assume that you know something about the subject, and that you
aren't here just to taunt people and get an emotional response out of
them.

I have experience with what you are describing. In the mid 1990s, I
worked for a company called Testra. I wrote MFX, which was an
assembler/simulator/compiler for their MiniForth processor (now called
the RACE). It was built on a Lattice isp1048 PLD. The company was only
interested in selling motion-control boards based on the MiniForth ---
but the motion-control boards are the product --- the MiniForth itself
is not a product, and there is no interest anywhere outside of Testra
in using it. The fact that it is a Forth processor, is not interesting
to anybody -- there is a gee-whiz aspect to it being Forth --- but
nobody spends money on gee-whiz.

I doubt that you are going to come up with anything more powerful than
the MiniForth, and the MiniForth is almost 20-year-old technology now
--- if nobody cared about this in the 1990s, when the mainstream micro-
controller was the Dallas 80c320, why would anybody care nowadays,
when the mainstream micro-controller is the ARM? In those days, it was
possible for a homebrew processor to outperform the mainstream
technology, but that is not realistic nowadays.

> Yes, altruism through business, lots of this stuff starting to
> happen.  I've run out of time and have to try to answer Paul post and
> go, but Ill give a quick answer.   You run a business to make your
> income but also to spend some of the profit to helping out society.
> But you can make that income on making products that really benefit
> society.   Also your business conduct and policy can affect the
> market, especially if you are the size of Apple.  Even if you are
> small, people know what you are about and are worth supporting, and it
> can encourage better conduct in other businesses, even if only a few
> shall ones.  Helping out society can mean a number of things, charity
> is one, but you can start up other helpful businesses, sponsor
> research and social or economic reform, even sponsor a politician you
> know is going to benefit the way things are done in the community, in
> partnership.  There is enough money, it is just where it goes to.
> There are lots of ideals, but this is the reality you can take baby
> steps towards it.

I am very turned off by the "altruism through business" idea. Almost
everybody who makes that kind of speech, is gouging the heck out of
their customers, and their employees, and anybody else who crosses
their path. Everybody has some hobby-horse that they use to "prove"
that they are of so much higher moral caliber than everybody around
them. For example, at Testra, the owner John Hart was strongly opposed
to abortion (he was Catholic), and that qualified him as being holier-
than-thou, compared to slobs like me. Of course, the Women's Rights
folks think that he is the devil --- it is all a matter of perspective
--- one person's saint is another's devil. For the most part, people
want to be holier-than-thou, in order to have justification for
cheating thou out of money --- otherwise, how would they differentiate
themselves from the common sociopath?

[Richard Owlett]

rickman wrote:
> [ *snip* ]
> People
> who are here have made some effort to use this medium...
>

I'm here 'cause it required LEAST effort! ;}

Are u using a ?$ product?

[Steve]

Paul Rubin wrote:
> Steve <nosp...@gmail.com> writes:
> > comp.language.forth.processor Is that valid use on Usenet
> > comp.lang ?
>
> I think it's hard to justify this regardless of whether it's a "type
> error" to have a hardware subgroup under comp.lang. An alt group would
> be fine in principle but seems like unnecessary fragmentation to me.

That's one of the issues I was seeing. It is about programming a
forth language processors, but also about the hardware development for
it, and about the processor. So, obviously being in either place
crosses over. comp.lang.forth.hardware sounds good, but to be more
specific comp.lang.hardware.processor is just too long.
comp.lang.hard.proc still a litthe convoluted, but that is the nature
of nested knowledge structures.

> > a new misc processor should drive up numbers again.
>
> I'd wait til the numbers actually go up, before trying to do anything
> about it.

Aim to open groups in preparation for product and development anyway,
so this fits in. But seeing the scattering you mentioned, it would
be interesting to see whose out there. comp.lang.functional and
comp.lang.haskell is that a spell checker typo?

> > If anybody knows of good hosting/servicing for full custom web, and
> > forum and blog, please let me know.
>
> Maybe this helps:
> http://www.lowendbox.com/blog/wordpress-cheap-vps-lowendscript/
> http://www.lowendbox.com/blog/yes-you-can-run-18-static-sites-on-a-64mb-link-1-vps/

Thanks again, I was going to research the lowendbox site you posted
from before.



I've neglected a few things, so I've got to knock off looking into
this for now until I can on top of everything.

[Paul Rubin]

Steve <nosp...@gmail.com> writes:
> comp.lang.forth.hardware sounds good

Yes, that would be a good choice if there was enough traffic to need a
new group. I don't think there is that much traffic right now. There
was one person grumbling about hardware posts very recently, but it
hasn't been a problem in general as far as I know. Forth CPU's and
stack hardware are a significant motivator for the Forth language these
days, IMHO. So I consider Forth hardware to be on-topic for CLF until
there's 100's of posts per day about it for sustained periods. I do
realize others may see it differently.

> Aim to open groups in preparation for product and development anyway,

I agree with Rick: do the product first, worry about the venue later.

> comp.lang.functional and comp.lang.haskell is that a spell checker
> typo?

Not as far as I can see. Comp.lang.functional was a fairly quiet group
about functional-programming languages such as Haskell, ML, Erlang, and
so on. One day someone got the bright idea that splitting off Haskell
into its own group would increase traffic. Instead it decreased
traffic, so now instead of one fairly quiet group, there are two dead
groups.

[Steve]

rickman wrote:
> On 4/28/2013 10:24 AM, Steve wrote:
> > Eager to post, but including people not here but in the GA sphere.
>
> You are looking for a way to talk about MISC processors but people are
> already doing that here an to a lesser extent in sci.electronics.design.
> I expect this topic would be welcome in comp.arch.embedded too.
>
> Why try to re-invent the wheel in this regard?

I will be starting my own groups, so this slots in alongside easily.

> As to using different medium than newsgroups, I wouldn't expect to find
> much support for that in a newsgroup. People who are here have made
> some effort to use this medium and so must feel it is more useful that
> others.

I actually would prefer a newsgroup or mailing list where things are
kept off air, but it is difficult to get a newsgroup approved, and I
still need to do a site and blog, and a webforum offers everybody a
lot of extra functionality. I actually remember putting a suggestion
into Google many years ago for something that functions like groups
with main pages that you can email post like Usenet. If I don't find
a deal I like, I might try groups for that reason. I actually don't
like webforum as much, because of the mentality of some of the people
that turn up and the things they do with the technology. Nothing pure
about it, but emotional in conversations rather than intelligent. But
the mentality is not as altruistic as in professional Usenet groups.

>
> Why not open a thread or two in some of the above mentioned groups and
> see where they lead?

Yep, that is why I replied, that is the thread I mentioned I am
preparing in previous posts. I already mentioned a ring computer in
the $1 computer thread ($10+ retail) as a spin idea on all these ultra
cheap arm and x86 boards out there in something simple that misc is
ideally suited for, except for the consumer wireless, as complex to
design as a zx81 (which had most of the parts integrated over the
x80). Like a zx81 really useless compared to an PC or C64, but really
affordable, fun and sellable in the short term until something better
comes. If not for bt or wifi support, it would be ultra simple.
What you are left with is a micro embedded platform in a ring that
could be used in other products. If one insists, they can add usb,
micro/mini sd or include display support, but is not needed.

Steve.

[Steve]

rickman wrote:
> On 4/28/2013 3:23 AM, Steve wrote:
> > On Apr 28, 5:51 am, Hugh Aguilar<hughaguila...@yahoo.com> wrote:
> >> On Apr 27, 10:07 am, ". ."<nospam...@gmail.com> wrote:
> >
> > Ah, I see, he was the one that wanted two of my rings on the old
> > thread. I was actually thinking of sending him two if it ever
> > happened, but that was more a though bubble to discuss and maybe make
> > into a project. I am interested in doing something myself, but more
> > expensive.
>
> Hey, I would be very interested in working with you if you are both
> serious and capable. Why not start a discussion on your computer ideas
> rather than worrying about where to discuss it?

Hmm, actually it was this second post I was meaning to reply to. Yep,
preparing a thread, spent 6-10 hours on it (mainly due to trying to
locate one article link) just got to proofread it after I do some
business today.

It is on the table Rick, but I have a kickstarter project to start and
a top engineer to be involved. Misc will have to come second place
this year, but aim to do some preporations. At the moment I would
like to discuss community open projects, like the open hardware Ring,
that would bring a small embedded platform with it we could all use.
I actually want to do something commercially much better one day than
what I'm proposing. If I can make money I can finance people to do
open projects in the future

Steve.

[Steve]

Hugh Aguilar wrote:
> On Apr 27, 11:55 pm, Steve <nospam...@gmail.com> wrote:
> > On Apr 28, 5:51 am, Hugh Aguilar <hughaguila...@yahoo.com> wrote:
> > > On Apr 27, 10:07 am, ". ." <nospam...@gmail.com> wrote:
> > I don't know if you are being sincere or not above Hugh.  Are you
> > saying there was a misc forum and its a good idea to start a new one?
> > Im working on the principle of better something than nothing.

> I'm saying that if the previous forum died out, another one is going

Which previous forum are people mentioning?

> to die out too. There is no interest in that kind of thing. I was
> assuming that you were one of Rickman's sock-puppets --- his troll-
> platform is to be an expert on stack-based FPGA processors --- if
> you're not Rickman though, then I'll give you the benefit of the doubt
> and assume that you know something about the subject, and that you
> aren't here just to taunt people and get an emotional response out of
> them.

Nope. Might joke sometimes, but that is obvious.

/Race miniforth.

Thanks, I'll have to look that up.

> to anybody -- there is a gee-whiz aspect to it being Forth --- but
> nobody spends money on gee-whiz.

Yes, it has to offer something much better,


> > altruism through business,

>
> I am very turned off by the "altruism through business" idea. Almost
> everybody who makes that kind of speech, is gouging the heck out of
> their customers, and their employees, and anybody else who crosses
> their path. Everybody has some hobby-horse that they use to "prove"

True altruism is looking after your employees and customers, and the
community aswell. I see where you are coming from, but Im not
interested in that sort of thing, the only thing I am interested in is
the Open community aspect to get started to make money towards
altruistic purposes. Normally I would steer away from such things,
but as it would help the community achieve long term products and
goals they want, and it helps me establish against big players who are
just interested in getting into a bigger mansion or pent house, so I
can put money strategically towards making life in general better for
people, I see it as a win win situation for eveybody concerned. Like
linux people, mainly in Europe, do this..

.. But I am not ready to launch my own open projects and open license,
so I am saying the community can have a go itself. It doesn't worry
me, a little competition is good and it would get things started, and
we can all work together in the end. I'm very for the aspect of
commercial viability in open licenses, so even in mine I am planning
to have clauses that allow for exelusive use for a period before
something is folded into the free aspect, and ways to separate out the
use of long term exclusive code so that it is never included, plug-ins
for example. Without these, there is little incentive to pay to
develope a lot of code commercially for an unlimited amount of
competitors to get it for free and send you out of business.
Commercial developers will still share things because it is best to
have them in the base standard code base, and for interpolility. But
there is nothing to stop somebody seeing something and doing their own
version in the code base. With things under short term exclusiveness
clauses, you just wait the 6 months to a year or so for it to be
folded back in rather than do your own version. For long term
exclusive code, people might make their own version for the codebase
and see which is better. I've designed it so companies can get a
season of exclusiveness and dynamically compete season to season. In
the end, some will just do a quality job and have exclusive code after
the codebase has settled down in their product's sector, so they will
end up on top, but for the rest of us we can get the less refined
mature codebase version. They win because people will pay extra for
quality and service. The codes win, because not only do they get
reputation and experience, but are also consultants on their code and
hirable to recode the code for companies. They can even setup their
own said support lines. Their are a few other innovations I'm
planning to add to give coders chance for financial reward. I'm not
very happy with certain aspects of copyright and software patents, so
am building a alternative ego system into the license that better all
around, fir those that want to use it. Once established as a
business practice, common law in court will start to support it making
it a standard of practice.

So, that is a picture of the things I'm about Hugh. As you can
imagine, I get some issues from people about this at times.

What is important to you Hugh, in respect to altruism?

[Steve]

Paul Rubin wrote:
> Steve <nosp...@gmail.com> writes:
> > comp.lang.forth.hardware sounds good
>
> Yes, that would be a good choice if there was enough traffic to need a
> new group. I don't think there is that much traffic right now. There
> was one person grumbling about hardware posts very recently, but it
> hasn't been a problem in general as far as I know. Forth CPU's and
> stack hardware are a significant motivator for the Forth language these
> days, IMHO. So I consider Forth hardware to be on-topic for CLF until
> there's 100's of posts per day about it for sustained periods. I do
> realize others may see it differently.

But if forth hardware people are spread around different forums, this
would give them a central place to come. My main concern is to get
away from non hardware oriented people crashing the
conversation .hardware isn't specific and might attract some bleed
across. But yes, I agree, it seems a good strategy, and.
hardware.processor.misc, could always come afterwards if needed. The
only questions are, do you think inviting these people is enough of a
crowd, and wherever to make it .hardware or .hard considering it
could be expanded to .proc.misc one day? The other thing we have not
considered, is that we want to include all forth/misc like processors,
but there is no comp.lang.ForthLike around. Hmm, what about this,
comp.lang.hard.proc.ForthLike as a future expansion path (if I have to
go through the trouble or registering, I might as well try to argue
for pre-approvals for future expansion names at the same time)?

I'm interested in a forum that is a nice friendly melting pot of
helpful people.

> > Aim to open groups in preparation for product and development anyway,
>
> I agree with Rick: do the product first, worry about the venue later.
>
>

> Not as far as I can see. Comp.lang.functional was a fairly quiet group
> about functional-programming languages such as Haskell, ML, Erlang, and
> so on. One day someone got the bright idea that splitting off Haskell
> into its own group would increase traffic. Instead it decreased
> traffic, so now instead of one fairly quiet group, there are two dead
> groups.

Paul sorry, you mentioned clf, I thought I was missing something.

I think we are starting to think the opposite, to bring proper from
all different forth and misc like processors into one place.

[Steve]

On Apr 29, 5:56 am, Steve <nospam...@gmail.com> wrote:
> Paul Rubin wrote:
> > Steve <nospam...@gmail.com> writes:
> > > comp.lang.forth.hardware sounds good

> comp.lang.hard.proc.ForthLike as a future expansion path (if I have to
>

Sorry, that should have been, comp.lang.forth.hard.proc.ForthLike

[Steve]

About: A discussion on a new forum for Forth Hardware, Misc or
ForthLike processors. On Usenet, the web or Google groups, or through
a maiing list.

I will leave this discussion open for the foreseeable future, to
anybody that wants to support or talk about having such a forum. It
maybe that more people will turn up who would like to be in one, that
haven't been around for a while.

[Paul Rubin]

Steve <nosp...@gmail.com> writes:
> But if forth hardware people are spread around different forums, this
> would give them a central place to come.

http://xkcd.com/927/

> My main concern is to get away from non hardware oriented people
> crashing the conversation

That's not a problem, really, there was an unusual occurrence last week
when someone who's not really a regular griped, but don't worry about
it. The Forth software crowd tends to be interested in hardware as well.

>do you think inviting these people is enough of a crowd,

I think the invitations are not likely to attract significant interest
no matter who you invite, unless there is already technical substance at
the new place. So I still advise making the technical announcements
first, and worrying about venues and invitations later.

> I think we are starting to think the opposite, to bring proper from
> all different forth and misc like processors into one place.

Perhaps so. I think we are going around in circles with this
metadiscussion. If you want to start a forum or blog or whatever, good
luck with it. I can only speak for myself but I don't feel likely to
enroll or participate unless there's something very very interesting
there, because of the hassle of dealing with yet another forum.

[Steve]

Paul Rubin wrote:
> Steve <nosp...@gmail.com> writes:
> > But if forth hardware people are spread around different forums, this
> > would give them a central place to come.
>
> http://xkcd.com/927/

:-)

Paul, sometimes you just have to say stuff it, and do it.

> Perhaps so. I think we are going around in circles with this
> metadiscussion. If you want to start a forum or blog or whatever, good
> luck with it. I can only speak for myself but I don't feel likely to
> enroll or participate unless there's something very very interesting
> there, because of the hassle of dealing with yet another forum.

Nope, I think we have come full circle and discussed most everything
we need. That's why I put the last post in to wait and see if anybody
else is eager to come before doing anything.

I think the problem is that we are all getting along and not so eager
to do things like forth rings etc as young guys are. But hopefully we
have the wisdom to see the worth of an idea, and when something is a
really bad idea. I'm thinking for now the forum thing falls between
good and bad, unless people turn up. What we need is a drive in
Forth in general, to get new blood, a website that promotes the
beneficial parts of the history of Forth (practical benefits brought),
and it's and Misc's benefits now, and education. To pass onto
students, and quote to other groups. Plenty of Forth sites, does
anybody have one like that out there (a question to readers in
general). Otherwise the Forth crowd is aging.

Just here to post the new thread.


Thanks


Steve.

[Jecel]

Steve,

you might enjoy looking at the archives for the previous mailing list:

http://www.ultratechnology.com/maillist.htm

I participated in that until it died out. A while later, Eric Laforest tried to create a discussion group on Google Plus (before there were communities there) but it didn't work out because most people he invited still used web browsers so old that they couldn't access Google Plus. Before that we were just Cc:ing all the participants and we continued to do so for a while, though that dicussion died out too.

-- Jecel

[rickman]

No, I meant they are here for a reason. Often did some work to find
newsgroups (you have to admit they aren't widely known). At a minimum
they found a news reader and a news provider, free or not and learned
how to use them... or they at a very minimum found Google Groups or one
of the other web interfaces and dug around until they found this little
corner of the cyber-world. So anyone here has done a bit more work than
just open Google and search for... well, whatever they were interested
in. Or am I wrong about that?

--

Rick

[Steve]

I've posted the new thread, "Misc Expansion.." yesterday, can you guys
see it? I just found out that groups was only displaying a few
threads.

I thought the whole group had only a few handful of members active. I
was wondering where these several hundred posts for April was
yesterday. My apologies, looks like the Forth group here is far from
dead.

Thanks.

Steve.

[Steve]

Jecel, good to know what has happened.

[rickman]

Obviously I have seen some of your posts, but I don't see a threat with
the subject "Misc Expansion".

--

Rick

[Steve]

Sorry Rick, the . . indicates it is a truncated version of the title.
Pretty excited about it, the more I think about it, the more I realise
I could do some of this stuff sooner rather than latter. But the
schedule is to do other things. This thread represents years of
experience and planning out there for the communities benefit, if they
choose it, I certainly couldn't afford to fund it myself any time
soon. Nobody is putting their hands up, or enthused yet. When I was
writing up the description of the OS, I thought, hmm this is pretty
significant functionality. and something I could do. Of course the
interfacing is where it can go obscure for being an easier job, and
without it is not so significant in functionality. Remote API
interfacing is basically writing different drivers. Mr Moore sees
things in terms of what was done, part experience, but what if it can
be done differently. What if we can get 90% of the true embedded
programming benefits of an simple OS in basically what is an extension
to firmware, a Firmware OS, in 256-512 words. I think that is
something he would love very much. The ring thing is starting to look
to me like when a drunk sees a pretty woman. If I am allowed to carry
over the GA engineering, instantly I have much of the non ring basics
for free. It must be in delusion, it can't be that simple (still a
tall order). A first edition can drop any more complex interface, to
proof it. The issue, where the external ram and flash memory, maybe
there is a combo chip that has most of what is wanted, maybe stick it
on that small RFduino mentioned in the $1 computer thread, as an
expansion board, maybe get the cheapest arm all in one combination
chip and put it on one side and misc on the other, or underneath.
Yes, that is probably the easiest solution, interface a low low end
all in one to a misc chip, as long as you can pull data off the chips
address bus to form a external memory for the misc chip, it will
probably be cheaper to use That part than to buy and use external
memory and wireless parts, and to write all drivers and interface
them. What do you think Rick?

[Hugh Aguilar]

On Apr 28, 10:29 pm, Steve <nospam...@gmail.com> wrote:
> /Race miniforth.
>
> Thanks, I'll have to look that up.

> ...

> What is important to you Hugh, in respect to altruism?

If you want to be altruistic, you could build a soft-core Forth
processor, with the VHDL in the public-domain so the users could
extend it with application-specific primitives.

The MiniForth was completed in 1996. The RACE is essentially the same
as the MiniForth, AFAIK. That was pretty cool back then (well, I
thought so; the rest of the world wasn't much interested). Nowadays, I
think something like that could be done fairly easily with our
improved modern technology --- not by me though, as I only know a very
little about VHDL. It is not cool enough to justify being proprietary.
Similarly, Unix originally was cool enough to justify being
proprietary, but after awhile, it had to go open-source in order to
hold the public's interest --- Forth chips are at that stage now.

I think there would be a huge boost in Forth's popularity if there was
a public-domain Forth processor available. Public-domain, in the sense
that users don't have to buy the physical chips one at a time from a
single-source supplier, but can build the processors themselves using
inexpensive FPGA chips. Also, as I said, the users need to be able to
add their own application-specific primitives --- there is really no
point to using a soft-core processor other than extensibility --- if
you aren't going to extend it, then you might as well just use an off-
the-shelf processor (off-the-shelf processors were pretty much the
only option available in the 1990s). Linux would never have become
popular if it had been proprietary, and the same is true for the
MiniForth or any other proprietary Forth chip --- it has to be public-
domain to become popular. People make money supporting Linux, and the
same could be true of a public-domain Forth chip. The ARM is popular
because it is widely second-sourced --- similarly, Forth would only
become popular if the Forth chip were widely second-sourced.

I don't think the world needs another proprietary processor. Using the
MiniForth, you are locked into Testra. Using the PIC24, you are locked
into MicroChip. What difference does it make? There is a little bit of
difference, in that with the MiniForth you can pay Testra to design
and build an application-specific custom version, whereas MicroChip
isn't going to do this unless you have a huge volume application and
deep pockets. It is not that much of a difference though --- in
neither case can you make the extensions yourself.

[Steve]

Hugh Aguilar wrote:
> On Apr 28, 10:29 pm, Steve <nospam...@gmail.com> wrote:
> > /Race miniforth.
> >
> > Thanks, I'll have to look that up.
> > ...
> > What is important to you Hugh, in respect to altruism?
>
> If you want to be altruistic, you could build a soft-core Forth
> processor, with the VHDL in the public-domain so the users could
> extend it with application-specific primitives.

If you mean for FPGA's they are very limited, compromises. If for
Asic, that is better again. If a OK CAD version then even better
efficencies. But giving the farm away leaves only a few stragglers to
tend it part time while they earn an income elsewhere. Isn't it much
better to have somebody that can earn a descent income from it as a
reward for good work, and to have more for better work tommorow? Even
better again if those people choose to put lots of the money back into
the community to positively guide it? It I is not the money but what
you do with it. While it would be more altruistic, so is giving all
your money away to the poor rather than starting a business to make
more money to not only give away but to rebuild the lives of the poor
and give some of them jobs, it wouid not be as efficient, and in
cases, less altrustic as a consequence.

It would be better for GA to do licensed fpga oriented designs, and
offer a licensed cell design and full chips in a commercially open
extendable way, then for us to do pd. If they get paid they have
incentive to do more and hire more, to do great work, if you give it
away the rich might not even appreciate it to use it, and if they did
would likely use it to make a higher profit margin off the public.
Even if they did try to undercut they could or would be forcing
competitors doing potentially better design work off the market, or
likely restricting the quality of work they could do, this quality
ultimately being for the altruistic benefit of Mankind. Effectiveness
is a consideration in altruism. Hugh, I can see you have been hurt
in the past, and I don't blame you, but I am not the one to make up
for all that, I can only do what I can do, hopefully with
affectiveness. These things here are for others to do something in
their own way, while I do the little of what I can do. A calling to
others.

Aren't there all ready pd forth chip designs, you could work with
those people to do something like what you want with them? I
encourage you.

Unix etc. The truth is that these were mediocre versions of
commercial systems of the 60's and 70's, and after the commercial
systems got their act together with OS/2, Windows and Mac, it became
mediocre on the desktop up till Next, but as Linux rermained mediocre
compared to next Mac and windows. Even in copying the work of better
companies and people, let alone producing as much new work as the
commercial industry, they have remained mediocre. We might have been
better off if the PC had been an extended CPM machine and had a Unix
PC a year or so after (I have an article).

But I still welcome your vhdl original/older technology alternative.
The simple reality is that we need a design with a full address bus to
feed an array. 16 bit word address for data banks and code banks is
perfectly acceptable, 18 bit even if you can find cheap matching
memory, 32 bit word address for data banks and code banks even better
again.

Misc is by no means mediocre in it's self concept, just under
developed I feel.

> I think there would be a huge boost in Forth's popularity if there was
> a public-domain Forth processor available. Public-domain, in the sense
> that users don't have to buy the physical chips one at a time from a
> single-source supplier, but can build the processors themselves using
> inexpensive FPGA chips. Also, as I said, the users need to be able to
> add their own application-specific primitives --- there is really no
> point to using a soft-core processor other than extensibility --- if
> you aren't going to extend it, then you might as well just use an off-
> the-shelf processor (off-the-shelf processors were pretty much the
> only option available in the 1990s). Linux would never have become
> popular if it had been proprietary, and the same is true for the
> MiniForth or any other proprietary Forth chip --- it has to be public-
> domain to become popular. People make money supporting Linux, and the
> same could be true of a public-domain Forth chip. The ARM is popular
> because it is widely second-sourced --- similarly, Forth would only
> become popular if the Forth chip were widely second-sourced.

Why give away better commercial work to work on something for
unaltruistic people to use that likely up to 1/100 of the people on
the planet benefit from compared to commercial work? It is not like
pd software that any user can download for free, at least on a misc
machine you could download a code pattern. You could at least start
up a business to fund development of top quality product. But I don't
discourage you from doing it othetwise, even if not top quality, it
encourages commercial companies to do better. In the end it is not
about a few private individuals and otherwise commercial entities
benefiting from your work, it I'd about how the products and designs
benefit the word and positively affect the conduct of their industries
isn't it?

> I don't think the world needs another proprietary processor. Using the
> MiniForth, you are locked into Testra. Using the PIC24, you are locked
> into MicroChip. What difference does it make? There is a little bit of
> difference, in that with the MiniForth you can pay Testra to design
> and build an application-specific custom version, whereas MicroChip
> isn't going to do this unless you have a huge volume application and
> deep pockets. It is not that much of a difference though --- in
> neither case can you make the extensions yourself.

It's who has the money and what they do with it, pd integrated
hardware tends to just hand free money in form of work done, to the
commercial sector to spend on themselves. Except for the prospect of
delivering a better misc way of doing things in products across the
industry, ultimately benefiting society, which admittedly is a slimmer
chance than a well funded commercial business archieving it, like Arm
is still progressively doing, it is a bit of a waste.


Thanks Hugh


Steve.

[Hugh Aguilar]

On May 1, 9:23 pm, Steve <nospam...@gmail.com> wrote:
> Hugh Aguilar wrote:
> > On Apr 28, 10:29 pm, Steve <nospam...@gmail.com> wrote:
> > > /Race miniforth.
>
> > > Thanks, I'll have to look that up.
> > > ...
> > > What is important to you Hugh, in respect to altruism?
>
> > If you want to be altruistic, you could build a soft-core Forth
> > processor, with the VHDL in the public-domain so the users could
> > extend it with application-specific primitives.
>
> If you mean for FPGA's they are very limited, compromises.  If for
> Asic, that is better again.

I was talking about FPGAs, as I mean something that could be done in
very small volume.

Why are soft-core FPGA processors so limited? You mean that they are
slower?

I don't know very much about hardware, as I just write software, so
I'm not familiar with the limitations involved.

[Steve]

I'm thought you were. I'm interested in a fair bit of benefit. You
could learn and get together with some of the many fpga softcore forth
people to do something better.

FPGA's were power hungry and slow, but the advantage was that you
could do things more directly, by passing code and in parralel, great
advantages, and response was very analogue. But in making a processor
things slow down with higher power consumption. The integrated
circuitry in them speeds things up to compensate, they make great
parts to do something custom on a circuite board. They have changed,
speeds increased and lower powered versions. But even today the same
problems probably still apply. But the truth is that Misc arrays have
gotten over much of it's advantages, finer, faster, lowered powered
parallel compared to Cisc competitors FPGA used to compete against,
just we don't have leading edge Misc on top processes to see the full
advantages.


Thanks


Steve.

[Brad Eckert]

On Wednesday, May 1, 2013 6:33:59 PM UTC-7, Hugh Aguilar wrote:
>
> > What is important to you Hugh, in respect to altruism?
>
> If you want to be altruistic, you could build a soft-core Forth
> processor, with the VHDL in the public-domain so the users could
> extend it with application-specific primitives.

Hard CPUs were missing from FPGAs for a long time, leaving a hole in the market for soft cores to fill. But that hole is now being filled. The real trick isn't solving a problem, it's finding the right problem to solve. Solving the wrong problem is a bigger problem than the actual problem.

[Steve]

Brad it is just Hugh's interest, and it is a credible option, I'm just
suggesting that he could go ahead and do it. It wouid be a benifit.


Forgot to post, there is a forth processor page at Forth.org that also
points to one at ultra technology, that lists pd firth soft cores, on
Dr Ting I think 32-64 bit versions. Would be a good way just to
develop instruction set ideas and proof with code, that some day could
go to silicon. He could make his dream come true. So, it does not
matter too much if the core is out competed, as long as he or somebody
uses it, it is still alive.

[rickman]

I think you are not really that familiar with FPGAs. There are large,
high power devices out there, but there are also smaller, low power
devices that can rival many processors in terms of speed and efficiency.

I won't say there are any FPGAs that can match the speed/power
consumption of the GA144, but I don't think there *are* any processors
that can do that.

But compared to a Cortex CM3, a softcore MISC processor can keep up and
possibly be lower power.

> The integrated
> circuitry in them speeds things up to compensate, they make great
> parts to do something custom on a circuite board. They have changed,
> speeds increased and lower powered versions. But even today the same
> problems probably still apply. But the truth is that Misc arrays have
> gotten over much of it's advantages, finer, faster, lowered powered
> parallel compared to Cisc competitors FPGA used to compete against,
> just we don't have leading edge Misc on top processes to see the full
> advantages.

I'm not sure what you are saying here. You seem to be saying that the
GA144 is somehow better than softcore FPGA processors. That all depends
on what you are using it for. I found the GA144 to be very difficult to
use for most apps. There may be a few which it is well suited for, but
it is *not* a general purpose CPU or MCU. It is a special beast all its
own.

I don't agree that the factor limiting the GA144 is that it isn't built
on "top processes". This chip is hobbled by the lack of decent
interfaces. They promote the use of soft interfaces which limit I/O to
slow protocols. They also don't provide flexible voltage I/Os so
anything other than 1.8 volt I/O requires messy level conversion. Try
driving a simple clock LCD with a GA144. You will need a number of
level conversion chips.

--

Rick

[Steve]

Well the last time I looked was around four years ago at the new low
powered generation, silicon bye, or something like that, a far cry
from the 6 or so mw per node of misc that could be under 1 mw on the
latest process.

> I won't say there are any FPGAs that can match the speed/power
> consumption of the GA144, but I don't think there *are* any processors
> that can do that.

Depends, but we are talking exclusively about softcore performance on
fpga aren't we? I expressed that for a direct parallel circuite it is
much better for fpga, but making a custom asic is better again, having
such a thing controlled by a processor is still better. But I suppose
if you if you compared most processor arrays to an fog a parallel
circuit the pridessir is going to pay in sie way, either to much
current drain at speed, or too much circuit to run at lower speed in
parralel. But then it might hinedtly depend, if you have a small fog
a up against a small speed processor, say an basic small arm converted
to good high speed low powered version, then there is only so much you
can fit into a equivalent die space (30k transistors, but less in that
areas on the fpga) to run against it, realistically.

> But compared to a Cortex CM3, a softcore MISC processor can keep up and
> possibly be lower power.

>

> I'm not sure what you are saying here. You seem to be saying that the
> GA144 is somehow better than softcore FPGA processors. That all depends
> on what you are using it for. I found the GA144 to be very difficult to
> use for most apps. There may be a few which it is well suited for, but
> it is *not* a general purpose CPU or MCU. It is a special beast all its
> own.

We are only talking performance and power consumption, not
programming, and I did specify that misc had not reached it's peak, by
any stretch, to prove that. Plus, if it was at this pleak, it would
be easier to program, having at least one core that had large program
pace for regular programming like a general purpose. Please those are
in reference my words we are talking about.

> I don't agree that the factor limiting the GA144 is that it isn't built
> on "top processes". This chip is hobbled by the lack of decent
> interfaces. They promote the use of soft interfaces which limit I/O to
> slow protocols. They also don't provide flexible voltage I/Os so
> anything other than 1.8 volt I/O requires messy level conversion. Try
> driving a simple clock LCD with a GA144. You will need a number of
> level conversion chips.
>

I agree, the design needs to improve, though at top process, the
artificial software buss maybe able to run at 1ghz, and the software
stream how many instructions of a word that can be run in a cycle in
that design. This peak also includes redesign, I'm sorry if I did
not mention it. The rest of the array for stream processing. But
now they are free of higher consideratiibs handled by the control
node, then maybe they can be easier to program and more efficiently
program. That would be very power full compared to now. What would
you agree Rick?

But many people miss things in what I write, and take it differently.

> --
>
> Rick


Thanks

Steve.

[rickman]

On 5/5/2013 3:40 AM, Steve wrote:
>
>
> rickman wrote:
>> On 5/2/2013 3:12 AM, Steve wrote:
>>>
>>> FPGA's were power hungry and slow, but the advantage was that you
>>> could do things more directly, by passing code and in parralel, great
>>> advantages, and response was very analogue. But in making a processor
>>> things slow down with higher power consumption.
>>
>> I think you are not really that familiar with FPGAs. There are large,
>> high power devices out there, but there are also smaller, low power
>> devices that can rival many processors in terms of speed and efficiency.
>
>
> Well the last time I looked was around four years ago at the new low
> powered generation, silicon bye, or something like that, a far cry
> from the 6 or so mw per node of misc that could be under 1 mw on the
> latest process.

You need to look again. Check out the iCE40 family from Lattice.
Xilinx and Altera are very focused on the networking and comms markets
where the large, power hungry devices are still in demand. The iCE40 is
from a startup bought by Lattice, Silicon Blue who understands there is
a market for small, low power devices.

>> I won't say there are any FPGAs that can match the speed/power
>> consumption of the GA144, but I don't think there *are* any processors
>> that can do that.
>
> Depends, but we are talking exclusively about softcore performance on
> fpga aren't we?

No, I am saying the GA144 has the highest speed/power tradeoff of any
processor I've seen, at least for powerful processors. Some of the
watch type processors might edge it out since they are designed just for
low power with speed just not a consideration... above 32 kHz anyway.

> I expressed that for a direct parallel circuite it is
> much better for fpga, but making a custom asic is better again, having
> such a thing controlled by a processor is still better. But I suppose
> if you if you compared most processor arrays to an fog a parallel
> circuit the pridessir is going to pay in sie way, either to much
> current drain at speed, or too much circuit to run at lower speed in
> parralel. But then it might hinedtly depend, if you have a small fog
> a up against a small speed processor, say an basic small arm converted
> to good high speed low powered version, then there is only so much you
> can fit into a equivalent die space (30k transistors, but less in that
> areas on the fpga) to run against it, realistically.

You have too many typos for me to understand your statements.

>> But compared to a Cortex CM3, a softcore MISC processor can keep up and
>> possibly be lower power.
>
>>
>> I'm not sure what you are saying here. You seem to be saying that the
>> GA144 is somehow better than softcore FPGA processors. That all depends
>> on what you are using it for. I found the GA144 to be very difficult to
>> use for most apps. There may be a few which it is well suited for, but
>> it is *not* a general purpose CPU or MCU. It is a special beast all its
>> own.
>
> We are only talking performance and power consumption, not
> programming, and I did specify that misc had not reached it's peak, by
> any stretch, to prove that. Plus, if it was at this pleak, it would
> be easier to program, having at least one core that had large program
> pace for regular programming like a general purpose. Please those are
> in reference my words we are talking about.

I am talking usability of the hardware. You are focusing on the CPU,
but the CPU has to live on a board and do useful stuff. The GA144 is
actually an MCU with a number of I/O pins and seems to be intended to be
embedded in products, not the core of a phone or tablet. Turns out it
sucks at this for several reasons, mostly because the I/O has
shortcomings, but also because you need to add so many support chips.
Products these days want to see integration. Put the Flash on the CPU,
put all the interfaces on the CPU, including the funky I/O voltages they
may need, like USB, etc. GA144 falls very short in this regard.

Even as a higher end processor it falls short by not having a memory
controller, but rather depending on three CPUs to provide this. This
ends up being a real bottle neck for processing.

>> I don't agree that the factor limiting the GA144 is that it isn't built
>> on "top processes". This chip is hobbled by the lack of decent
>> interfaces. They promote the use of soft interfaces which limit I/O to
>> slow protocols. They also don't provide flexible voltage I/Os so
>> anything other than 1.8 volt I/O requires messy level conversion. Try
>> driving a simple clock LCD with a GA144. You will need a number of
>> level conversion chips.
>>
>
> I agree, the design needs to improve, though at top process, the
> artificial software buss maybe able to run at 1ghz, and the software
> stream how many instructions of a word that can be run in a cycle in
> that design. This peak also includes redesign, I'm sorry if I did
> not mention it. The rest of the array for stream processing. But
> now they are free of higher consideratiibs handled by the control
> node, then maybe they can be easier to program and more efficiently
> program. That would be very power full compared to now. What would
> you agree Rick?
>
> But many people miss things in what I write, and take it differently.

Or they read exactly what you write...

--

Rick

[Steve]

rickman wrote:
> On 5/5/2013 3:40 AM, Steve wrote:
> >
> >
> > rickman wrote:
> >> On 5/2/2013 3:12 AM, Steve wrote:
> >>>
> >>> FPGA's were power hungry and slow, but the advantage was that you
> >>> could do things more directly, by passing code and in parralel, great
> >>> advantages, and response was very analogue. But in making a processor
> >>> things slow down with higher power consumption.
> >>
> >> I think you are not really that familiar with FPGAs. There are large,
> >> high power devices out there, but there are also smaller, low power
> >> devices that can rival many processors in terms of speed and efficiency.
> >
> >
> > Well the last time I looked was around four years ago at the new low
> > powered generation, silicon bye, or something like that, a far cry
> > from the 6 or so mw per node of misc that could be under 1 mw on the
> > latest process.
>
> You need to look again. Check out the iCE40 family from Lattice.
> Xilinx and Altera are very focused on the networking and comms markets
> where the large, power hungry devices are still in demand. The iCE40 is
> from a startup bought by Lattice, Silicon Blue who understands there is
> a market for small, low power devices.
>

Without going into all this, silicon blue low powered was my primary
point, in comparison to misc, so not even that with soft core beats
misc on an similar area use basis. It dies not matter about higher
powered, it gets worse for FPGA by area.

> >> I won't say there are any FPGAs that can match the speed/power
> >> consumption of the GA144, but I don't think there *are* any processors
> >> that can do that.
> >
> > Depends, but we are talking exclusively about softcore performance on
> > fpga aren't we?
>
> No, I am saying the GA144 has the highest speed/power tradeoff of any
> processor I've seen, at least for powerful processors. Some of the
> watch type processors might edge it out since they are designed just for
> low power with speed just not a consideration... above 32 kHz anyway.

But where we came from and what we were talking, about was about FPGA
softcore to misc comparison, jumping sideways does not invalidate
that, and is hardly fair or inconfusing, isn't it? My hlead can jump
around as fast as the next, Rick, but I keep it on track and pick
relavaent side lines, with some observations. If you want to talk
oranges when we are talking apples please talk oranges separately
instead of saying apples does not equal oranges and I am therefore
wrong, sort of thing. I could similarly next say something irrelevant
and claim that is more right, just does not make sense.

>
> > I expressed that for a direct parallel circuite it is
> > much better for fpga, but making a custom asic is better again, having
> > such a thing controlled by a processor is still better. But I suppose

> > if you if you compared most processor arrays to an fpga parallel
> > circuit the processor is going to pay in some way, either to much

> > current drain at speed, or too much circuit to run at lower speed in

> > parralel. But then it might inevitably depend, if you have a small fpga
> > up against a small speed processor, say a basic small arm converted
> > to a good high speed low powered chip process version, then there is only so much you
> > can fit into an equivalent die space (30k transistors, but on the same area on the FPGA less than that

>>) to run against it, realistically.
>
> You have too many typos for me to understand your statements.
>

Freakin auto spell correction, substitutes and splits into nonsense
words things it does not understand, and answering these posts in
depth is wearing my time out too much to keep up with correction.
Plus this keyboard has some issues. Sorry. I've corrected and
rearranged the little quote above for you to look at.

> >> But compared to a Cortex CM3, a softcore MISC processor can keep up and
> >> possibly be lower power.
> >
> >>
> >> I'm not sure what you are saying here. You seem to be saying that the
> >> GA144 is somehow better than softcore FPGA processors. That all depends
> >> on what you are using it for. I found the GA144 to be very difficult to
> >> use for most apps. There may be a few which it is well suited for, but
> >> it is *not* a general purpose CPU or MCU. It is a special beast all its
> >> own.
> >
> > We are only talking performance and power consumption, not
> > programming, and I did specify that misc had not reached it's peak, by
> > any stretch, to prove that. Plus, if it was at this pleak, it would
> > be easier to program, having at least one core that had large program
> > pace for regular programming like a general purpose. Please those are
> > in reference my words we are talking about.
>
> I am talking usability of the hardware. You are focusing on the CPU,

Again it was about what was talked about, and the topic/subtopic, if
you want to shift topic say, 'ok, whatever you think about the current
topic', then 'what about xyz different angle' for the next topic,
don't just call into question what I'm saying into relevance something
else. This just confuses it.

> but the CPU has to live on a board and do useful stuff. The GA144 is
> actually an MCU with a number of I/O pins and seems to be intended to be
> embedded in products, not the core of a phone or tablet. Turns out it
> sucks at this for several reasons, mostly because the I/O has
> shortcomings, but also because you need to add so many support chips.
> Products these days want to see integration. Put the Flash on the CPU,
> put all the interfaces on the CPU, including the funky I/O voltages they
> may need, like USB, etc. GA144 falls very short in this regard.

Yep, I hear you, that is why I am looking at attaching it to another
integrated processor, and just have it do processing, but the other
chip run everything else. Until it has a large execution space bus,
even processing will be slowed. A 32 bit with large execution space
you can do more. As it is, notice for the cheap version of the ring
computer I wanted only a few interfaces, like two or so, to keep it
very simple and basic. With the present version there is little you
can do. If it had a large execution bus, everything would be a bit
easier and you can do more, but even then, all the extra functionality
companion chips for cpu's would be drying up, because they are being
integrated with cpu's etc and are no longer needed.

>
> Even as a higher end processor it falls short by not having a memory
> controller, but rather depending on three CPUs to provide this. This
> ends up being a real bottle neck for processing.

Yep.

>
> >> I don't agree that the factor limiting the GA144 is that it isn't built
> >> on "top processes". This chip is hobbled by the lack of decent
> >> interfaces. They promote the use of soft interfaces which limit I/O to
> >> slow protocols. They also don't provide flexible voltage I/Os so
> >> anything other than 1.8 volt I/O requires messy level conversion. Try
> >> driving a simple clock LCD with a GA144. You will need a number of
> >> level conversion chips.
> >>
> >
> > I agree, the design needs to improve, though at top process, the
> > artificial software buss maybe able to run at 1ghz, and the software
> > stream how many instructions of a word that can be run in a cycle in
> > that design. This peak also includes redesign, I'm sorry if I did
> > not mention it. The rest of the array for stream processing. But
> > now they are free of higher consideratiibs handled by the control
> > node, then maybe they can be easier to program and more efficiently
> > program. That would be very power full compared to now. What would
> > you agree Rick?
> >
> > But many people miss things in what I write, and take it differently.
>
> Or they read exactly what you write...

You must have, because you didn't disagree this time. :-)

Now, I've had enough issues, and as elsewhere today, I would like to
close off this conversation, even though it was ok.


Thank you Rick for the good conversation here.


Steve.

> --
>
> Rick

[rickman]

I'm not sure what your comparison point is. Are you designing a chip?
I didn't realize that.

>>>> I won't say there are any FPGAs that can match the speed/power
>>>> consumption of the GA144, but I don't think there *are* any processors
>>>> that can do that.
>>>
>>> Depends, but we are talking exclusively about softcore performance on
>>> fpga aren't we?
>>
>> No, I am saying the GA144 has the highest speed/power tradeoff of any
>> processor I've seen, at least for powerful processors. Some of the
>> watch type processors might edge it out since they are designed just for
>> low power with speed just not a consideration... above 32 kHz anyway.
>
> But where we came from and what we were talking, about was about FPGA
> softcore to misc comparison, jumping sideways does not invalidate
> that, and is hardly fair or inconfusing, isn't it? My hlead can jump
> around as fast as the next, Rick, but I keep it on track and pick
> relavaent side lines, with some observations. If you want to talk
> oranges when we are talking apples please talk oranges separately
> instead of saying apples does not equal oranges and I am therefore
> wrong, sort of thing. I could similarly next say something irrelevant
> and claim that is more right, just does not make sense.

I don't actually know what you are talking about. You said FPGAs were
big and power hungry and I pointed out that they all aren't. Then you
started talking about the power of MISC ASICs which you can't build to
the best of my knowledge. So what is your point? Yes, any ASIC built
on an appropriate process will be lower power than an FPGA built on a
similar process. But apples aren't oranges and there is only one off
the shelf MISC I am aware of, the GA144 which is very inappropriate for
many, many designs for many reasons. So for people who want to do
something *real*, the alternative is an FPGA MISC which is *real*, *now*
and can be made to work in many applications.

>>> I expressed that for a direct parallel circuite it is
>>> much better for fpga, but making a custom asic is better again, having
>>> such a thing controlled by a processor is still better. But I suppose
>>> if you if you compared most processor arrays to an fpga parallel
>>> circuit the processor is going to pay in some way, either to much
>>> current drain at speed, or too much circuit to run at lower speed in
>>> parralel. But then it might inevitably depend, if you have a small fpga
>>> up against a small speed processor, say a basic small arm converted
>>> to a good high speed low powered chip process version, then there is only so much you
>>> can fit into an equivalent die space (30k transistors, but on the same area on the FPGA less than that
>>> ) to run against it, realistically.
>>
>> You have too many typos for me to understand your statements.
>>
>
> Freakin auto spell correction, substitutes and splits into nonsense
> words things it does not understand, and answering these posts in
> depth is wearing my time out too much to keep up with correction.
> Plus this keyboard has some issues. Sorry. I've corrected and
> rearranged the little quote above for you to look at.

Why don't you turn autospell off?

Most of what you talk about is speculation, totally disconnected from
the real world. Who would be building the MISC processors you speak of?

>>>> But compared to a Cortex CM3, a softcore MISC processor can keep up and
>>>> possibly be lower power.
>>>
>>>>
>>>> I'm not sure what you are saying here. You seem to be saying that the
>>>> GA144 is somehow better than softcore FPGA processors. That all depends
>>>> on what you are using it for. I found the GA144 to be very difficult to
>>>> use for most apps. There may be a few which it is well suited for, but
>>>> it is *not* a general purpose CPU or MCU. It is a special beast all its
>>>> own.
>>>
>>> We are only talking performance and power consumption, not
>>> programming, and I did specify that misc had not reached it's peak, by
>>> any stretch, to prove that. Plus, if it was at this pleak, it would
>>> be easier to program, having at least one core that had large program
>>> pace for regular programming like a general purpose. Please those are
>>> in reference my words we are talking about.
>>
>> I am talking usability of the hardware. You are focusing on the CPU,
>
> Again it was about what was talked about, and the topic/subtopic, if
> you want to shift topic say, 'ok, whatever you think about the current
> topic', then 'what about xyz different angle' for the next topic,
> don't just call into question what I'm saying into relevance something
> else. This just confuses it.

I guess I am talking about the hardware because it is a bit more
relevant to what can be done. I actually have no idea what you want to
do, so I can't say what is relevant to your needs.

>> but the CPU has to live on a board and do useful stuff. The GA144 is
>> actually an MCU with a number of I/O pins and seems to be intended to be
>> embedded in products, not the core of a phone or tablet. Turns out it
>> sucks at this for several reasons, mostly because the I/O has
>> shortcomings, but also because you need to add so many support chips.
>> Products these days want to see integration. Put the Flash on the CPU,
>> put all the interfaces on the CPU, including the funky I/O voltages they
>> may need, like USB, etc. GA144 falls very short in this regard.
>
> Yep, I hear you, that is why I am looking at attaching it to another
> integrated processor, and just have it do processing, but the other
> chip run everything else. Until it has a large execution space bus,
> even processing will be slowed. A 32 bit with large execution space
> you can do more. As it is, notice for the cheap version of the ring
> computer I wanted only a few interfaces, like two or so, to keep it
> very simple and basic. With the present version there is little you
> can do. If it had a large execution bus, everything would be a bit
> easier and you can do more, but even then, all the extra functionality
> companion chips for cpu's would be drying up, because they are being
> integrated with cpu's etc and are no longer needed.

That is a loosing proposition. Treating the GA144 as a coprocesssor
doesn't work because it has so little I/O capability. Until GA decides
to join the rest of the world with standard interfaces, this will only
ever be an MCU and not really a great one at that. If you can find an
app that suits the various limitations and matches the strengths of the
GA144, then it can be a quite good device perhaps. But it will never be
a replacement for an ARM in a tablet or phone.

When you figure out exactly what you want to do with MISC I would be
happy to discuss it with you.

--

Rick

[hughag...@yahoo.com]

On Saturday, April 27, 2013 10:07:07 AM UTC-7, Steve wrote:
> I'm looking to start up some groups for my own commercial projects
> including future Misc ones.    As there does not seem to be a forum
> for Misc development,  I can include a subforum for misc development
> and other simple misc like processors you guys are doing.   I like
> this misc idea and think it is promising.  I'm also interested in the
> open os thing I mentioned before, and open hardware projects.
>
> Starting a group on Usenet seems like a complex mess. But I've spent
> a number of hours reading through Google group information, and it
> seems to have a lot of what is needed.  I can start up a group and put
> in sub categories as subgroups.  It can be setup as a non moderated
> email list, which should minimise any legal responsibility, reduce
> time looking after it and keeps conversations off the web.   It can be
> non public invitation only, which would eliminate passer bys that seem
> to be a hassle. I could expand it into a group web forum eventually.
>
> I would prefer a freeform website, blog and forums, but as Google does
> not seem to support that integration, it would be too much non project
> downtime to look after a custom site for now.

If you do start a group, it would be best to use something other than Google Groups. This new format is just horrible. The old format was pretty good, but Google made a huge step backward with their supposed improvement.

Now I'm stuck in this new format, and I don't know how to revert it to the old format. :-( This used to be an option in the tools menu, but it isn't there anymore --- maybe Google has carried out their threat of getting rid of the old format altogether.

[rickman]

I was in the same position a while back. The only solution I found was
to exit Google Groups and use a newsreader. I'm using Thunderbird and
www.eternal-september.org It is not the end-all be-all of newsgroups,
but it works... much better than Google Groups.

Another reason to not create a group in Google Groups is because it will
*only* be readable in Google Groups. A Google group isn't visible in
newsgroups.

--

Rick

[Steve]

I'm using old news group, because new does not work at all in my
browser. But everybody has a negative about every option, so Just to
just bite the bullet.

I like news, but I can't re-edit a first post to add updates, which in
my other thread would be great. I can't go back to correct typos and
mistakes I miss, which is a good reason to leave my faulty spell
checker on (I managed to switch to a slightly better spellchecker and
keyboard). I can't re-edit my posts and delete useless posts to
condense a thread. I know exactly what I am planning and have stated
exactly what things are about, but it gets lost in static that gets
louder than what was said.

Online hosting, people complain about tracking usage etc. Now
vbulletin, which is the best self hosting package I've seen so far
seems to track links.

Groups seems to have some strange interface my standards compliant
browser can't render. If it was actually better than old groups, I
would just use that. Now you tell me that the new groups is worse.
We will see, with Google I can host a website group, pages and
email. Being on the web rather than news, is not such an issue, for
my own groups, I would prefer to have them behind closed doors, as
they are community development. Somethings that happen I would not
want on the internet. Seen past mailing lists with infighting, and
when you have a large number of general people without moderation,
things can get pretty off.

Still, have a while to find an ideal solution for my own groups. A
comp.Lang.forth.hardware would be nice for us here instead.

[Steve]

I just went to the new group on another browser. Wow, you were
right. I can't evern open up threads in multiple windows. Trying to
even find the group was a bit of a labor. The first thing people do
is try to go to a news group. Searching posts instead makes one feel
like they are in azoo. being looked at.

All this dumbing down for dumb people, makes one, dumb.