Forbidden Technologies
Forbidden Technologies
The cancellation of the Risc PC 2 is inconvenient for me at Eidos
because we were planning an on-line version of Optima for RPC2.
As it happens, I have this new company with some cash in it, but
not enough on its own to launch RPC2 which would need UKP 2m-3m
in working capital (to complete development and build stock).
I wondered what people thought of these alternatives:
1) raise a few million quickly from Acorn users eg on this NG to
complete and launch RPC2 next year (1999);
2) move the Acorn community on to an ARM-based Unix (eg Linux) box
and spend the money on tools to assist this process;
3) build a new machine running eg Linux with say eight ARM10s
( see http://www.techweb.com/se/directlink.cgi?EET19980803S0011 )
- such a new machine would have hardware FP, PMT, PCI for graphics
cards and SCSI, standard Unix tools eg C++, and would be launched
perhaps in a year from now.
In cases 2 and 3, we would create a RISC OS user-interface.
Moving to Linux would remove the cost of keeping RISC OS up to date
and would allow support for the more modern 32 modes in ARM chips.
I'm off to Italy for a few days, but will look through this thread
when I get back to see if anyone is interested in these ideas.
--
Stephen B Streater
Forbidden Technologies Limited
Steven Purchase
> As it happens, I have this new company with some cash in it, but
> not enough on its own to launch RPC2 which would need UKP 2m-3m
> in working capital (to complete development and build stock).
>
> I wondered what people thought of these alternatives:
>
> 1) raise a few million quickly from Acorn users eg on this NG to
> complete and launch RPC2 next year (1999);
A nice idea, but would end up with a more out-dated mahine. Still, it be
very nice to see the Phoebe make it through to production, and I still
don't see why they don't try to make some of their R&D back by selling a
couple of thousand...?
> 2) move the Acorn community on to an ARM-based Unix (eg Linux) box
> and spend the money on tools to assist this process;
Probably a good idea, but the existing ARM-based machines aren't anywhere
near as powerful as Phoebe could have been. So...
> 3) build a new machine running eg Linux with say eight ARM10s
> ( see http://www.techweb.com/se/directlink.cgi?EET19980803S0011 )
> - such a new machine would have hardware FP, PMT, PCI for graphics
> cards and SCSI, standard Unix tools eg C++, and would be launched
> perhaps in a year from now.
>
> In cases 2 and 3, we would create a RISC OS user-interface.
I think that might be a good way forward. Collaborating with the two
existing ARM-based machines that already exist would be a way to reduce
the time and money it takes to get a working machine available.
As for software, I feel there would be a bright, if slow future for
development of software on a new OS.
> Moving to Linux would remove the cost of keeping RISC OS up to date
> and would allow support for the more modern 32 modes in ARM chips.
...And open the door to many more technologies that already exist for
Linux/Unix systems and will be ported in the future... I see a far better
future for a machine like this than I did for RISC-OS machines.
Please, if anyone has the money/power/will to make a go of a project like
this, go for it. I'm sure you'll be able to count on support for a very
large percentage of the Acorn community.
--
Steven Purchase of Technium220 st...@technium.demon_DOT_co_DOT_uk
Get Doh! and WimpBar2 for free at...
http://www.technium.demon.co.uk/
Flying is the art of throwing yourself at the ground... and missing.
Dave Roberts
<URL:mailto:Forb...@surprise.demon.co.uk> wrote:
> 2) move the Acorn community on to an ARM-based Unix (eg Linux) box
> and spend the money on tools to assist this process;
>
> 3) build a new machine running eg Linux with say eight ARM10s
> ( see http://www.techweb.com/se/directlink.cgi?EET19980803S0011 )
> - such a new machine would have hardware FP, PMT, PCI for graphics
> cards and SCSI, standard Unix tools eg C++, and would be launched
> perhaps in a year from now.
>
> In cases 2 and 3, we would create a RISC OS user-interface.
These sound like good ideas, and some in .advocacy would be quick to point
out they mentioned them first! I can see only one real downside, and that
being, what would up and coming software like Avante do? It would surely be
impossible without a *major* rewrite to make these work? I would sincerely
like to be told wrong on this point as maybe this *is* light at the end of
a torturously long tunnel.
One of the main disadvantages of Linux (listening only to what other
people have said) is the lack of decent commercial software. Maybe we have
what they are looking for?
--
Dave Roberts
Da...@pharpech.demon.co.uk
mrp...@leeds.ac.uk
Rob Hemmings
>
>
> The cancellation of the Risc PC 2 is inconvenient for me at Eidos
> because we were planning an on-line version of Optima for RPC2.
>
> not enough on its own to launch RPC2 which would need UKP 2m-3m
> in working capital (to complete development and build stock).
>
> I wondered what people thought of these alternatives:
<snip>
My interest is in continuing to use the Risc OS user interface and my
existing Risc OS based software with which I am very productive. I'm
certainly interested in any options to achieve this, including some
investment of money.
The first option sounds immediately attractive although perhaps the others
are more viable in the long term. What I would want to know for options 2
or 3 is what would happen to existing Risc OS software?
--
-------------------------------------------------------------
Rob Hemmings Southport
Tel: +44 (0)1704 573210 ro...@argonet.co.uk
Darrell Taylor
[snip]
> 1) raise a few million quickly from Acorn users eg on this NG to
> complete and launch RPC2 next year (1999);
Interesting thought, but is it really possible to raise that amount of
cash this way? maybe if everyone sent their 500UKP depsit to
Eidos/Forbidden Technologies instead of Acorn this would work?
Are their 6000 people who would put a 500UKP deposit down? How many
have actually placed orders for Pheobe already, my guess (though it may
be completely wrong) is < 500.
Companies who use RiscPCs in commercial/industrial applications may be
interested in investing in this venture.
Now then, if this did work would it be an Eidos badged workstation?
If it was an Eidos badged workstation it would have immediate exposure
just because it has the name Eidos attached to it. I think it would
certainly get the attention of the computing industry.
> 2) move the Acorn community on to an ARM-based Unix (eg Linux) box
> and spend the money on tools to assist this process;
>
> 3) build a new machine running eg Linux with say eight ARM10s
> ( see http://www.techweb.com/se/directlink.cgi?EET19980803S0011 )
> - such a new machine would have hardware FP, PMT, PCI for graphics
> cards and SCSI, standard Unix tools eg C++, and would be launched
> perhaps in a year from now.
This would introduce an amount of stability to the future of RiscOS
(though it wouldn't actually be RiscOS but something that looks and
feels like it) because I don't thing Linux will be dissappearing in
a hurry, nor will ARM so it seems to be the logical choice. The Chalice
ARM motherboard thingy would fit the bill for this wouldn't it?
> In cases 2 and 3, we would create a RISC OS user-interface.
Hasn't this already been batted around before? ISTR seeing something
tucked away in a corner at either Wakefield or AW97.
> Moving to Linux would remove the cost of keeping RISC OS up to date
> and would allow support for the more modern 32 modes in ARM chips.
I think the most important thing to remember if a RiscOS like window
manager was to be developed for Linux, is that it shouldn't lose any
of that RiscOS 'feel' in the transfer.
> I'm off to Italy for a few days, but will look through this thread
> when I get back to see if anyone is interested in these ideas.
Your ideas have certainly provoked some thoughts, and I too will be
interested to see what develops over the coming weeks.
--
Darrell
Dmitry Petrov
>
> The cancellation of the Risc PC 2 is inconvenient for me at Eidos
> because we were planning an on-line version of Optima for RPC2.
>
> As it happens, I have this new company with some cash in it, but
> not enough on its own to launch RPC2 which would need UKP 2m-3m
> in working capital (to complete development and build stock).
>
> I wondered what people thought of these alternatives:
>
> complete and launch RPC2 next year (1999);
>
> and spend the money on tools to assist this process;
>
> 3) build a new machine running eg Linux with say eight ARM10s
> ( see http://www.techweb.com/se/directlink.cgi?EET19980803S0011 )
> - such a new machine would have hardware FP, PMT, PCI for graphics
> cards and SCSI, standard Unix tools eg C++, and would be launched
> perhaps in a year from now.
register and vote for these alternatives.
Then, with collected information, it will be much easier to decide what
would be the right thing to do.
I, personally, was going to get RPC2 (as well as some my friends in
Russia and USA).
Now, I may think about CATS offered by ChalTech
(http://www.chaltech.com).
Regards,
Dmitry Petrov
Liam Gretton
> 1) raise a few million quickly from Acorn users eg on this NG to
> complete and launch RPC2 next year (1999);
Nice idea, but sounds very risky for the casual investor (such as myself).
The experiences of the poor Amiga community spring to mind, waitng for years
for their boat to come in.
> 2) move the Acorn community on to an ARM-based Unix (eg Linux) box
> and spend the money on tools to assist this process;
Bleugh. Much as I love Unix, if I wnated a unix box I'd get a PC or a cheap
Sun or Alpha.
> 3) build a new machine running eg Linux with say eight ARM10s
> ( see http://www.techweb.com/se/directlink.cgi?EET19980803S0011 )
> - such a new machine would have hardware FP, PMT, PCI for graphics
> cards and SCSI, standard Unix tools eg C++, and would be launched
> perhaps in a year from now.
Better but still sounds a bit pie-in-the-sky.
IMHO those of us who want an up to date RISC OS box want Phoebe; the best
thing about it was its PCI architecture and faster bus.
--
Liam Gretton
li...@binliner.demon.co.uk
l...@star.le.ac.uk
John Rees
dar...@rmtwenty.demon.co.uk (Darrell Taylor) wrote:
> If it was an Eidos badged workstation it would have immediate exposure
> just because it has the name Eidos attached to it. I think it would
> certainly get the attention of the computing industry.
Two words spring to mind, TOMB RAIDER. The publicity a machine built
by the people who brought the world Lara Croft would grab the attention of
not just the computing industry, but a large chunk of the general public.
One of Acorn's problems has been it is associated with schools and is seen
as fuddy duddy. The kudos of being connected to one of the leading lights
in the games world would do any new RISC OS machine a power of good.
--
John Campbell Rees
<jw...@argonet.co.uk> http://www.argonet.co.uk/users/jwcr/index.htm
"Just like swatting flies with a laser cannon. The aims a bit tricky, but it sure takes care of the flies" - Lord Miles Vorkosigan from "Komarr" by Lois McMaster Bujold
Ross Tierney
<URL:mailto:dar...@rmtwenty.demon.co.uk> wrote:
Although I work for Stephen, don't assume I know anything about what
he's planning - Okay? :)
> > 1) raise a few million quickly from Acorn users eg on this NG to
> > complete and launch RPC2 next year (1999);
>
> Interesting thought, but is it really possible to raise that amount of
> cash this way? maybe if everyone sent their 500UKP depsit to
> Eidos/Forbidden Technologies instead of Acorn this would work?
> Are their 6000 people who would put a 500UKP deposit down? How many
> have actually placed orders for Pheobe already, my guess (though it may
> be completely wrong) is < 500.
I heard a figure of less than 150 (146ish) had payed their deposits to
Acorn for an RPC2.
> Companies who use RiscPCs in commercial/industrial applications may be
> interested in investing in this venture.
>
> Now then, if this did work would it be an Eidos badged workstation?
> If it was an Eidos badged workstation it would have immediate exposure
> just because it has the name Eidos attached to it. I think it would
> certainly get the attention of the computing industry.
Stephen may not be able to reply until next week, but i thought I'd
better clear this up here once and for all.
It would NOT be an Eidos product, or in fact related to Eidos plc in
any way. It would be a product of "Forbidden Technologies Ltd." in
which Stephen has a directing hand.
The company exists now, but isn't doing anything yet. This 'box' would
pretty much be it's first product. It has nothing to do with Eidos
other than the fact Stephen is connecting them.
> This would introduce an amount of stability to the future of RiscOS
> (though it wouldn't actually be RiscOS but something that looks and
> feels like it) because I don't thing Linux will be dissappearing in
> a hurry, nor will ARM so it seems to be the logical choice. The Chalice
> ARM motherboard thingy would fit the bill for this wouldn't it?
Yeah. I'd say we need to make a move on this soon - I wanna do it, but
then I've got an ulterior motive: It'll keep me in a job.
What *I'd* like to manage also is to get the focus of the developer
community directed this way so they might be able to keep their jobs
too. If Avante, Ovation Pro, Photodesk and Top Model to list but a
very few of the great titles we currently have could be directed
towards this solution it may be a saving grace for the whole
community.
> I think the most important thing to remember if a RiscOS like window
> manager was to be developed for Linux, is that it shouldn't lose any
> of that RiscOS 'feel' in the transfer.
For me personally, I agree. If as much of the RISC OS 'Look and Feel'
can be retained under Linux, I'd be delighted. And as a bonus, we all
get an OS that's stable as hell and all the Pre-Emptive Multi-tasking
you could wish for. Also, theoretically a distriution could be made to
work on other platforms other than ARM to tempt some of the other 4
million Linux users to the RISC OS GUI.
Cheers,
Ross.
--
Ross Tierney.
r...@eidos.co.uk "...Breathing in... Breathing out..."
kra...@argonet.co.uk -An Old Friend
Ross Tierney
<URL:mailto:dpe...@acorn.ru> wrote:
> Forbidden Technologies wrote:
> May be it is a good idea to create a web page where people could
> register and vote for these alternatives.
We have the web *site* already. But there's nothing on it at the mo.
We've been very busy investigating all of Stephen's options already
(Thurs) and we havn't had time to do anything for the web site.
Because of this I doubt we'll have the chance to do anything soon
either. I'll get in contact with some friends and see if I can't get
something whipped-up in my spare-time.
Ross Tierney
<URL:mailto:rhem...@argonet.co.uk> wrote:
> The first option sounds immediately attractive although perhaps the
> others are more viable in the long term. What I would want to know
> for options 2 or 3 is what would happen to existing Risc OS software?
There are ways and means. Initial estimates say it would take about
six months for us to get some system that would run a large proportion
of code designed for RISC OS under Linux. Most of it would actually be
fairly straight-forward, but time-consuming.
We need to start something fairly soon, because the future of Optima
is centered around this hypothetical product and if we're going to do
a lot of work on it, we'll need to start fairly soon.
The more interest we have, the more likely we are to do it.
Cheers,
Ross.
(Not speaking in any official capacity what-so-ever!)
Ross Tierney
<URL:mailto:li...@binliner.demon.co.uk> wrote:
> > 3) build a new machine running eg Linux with say eight ARM10s
> > ( see http://www.techweb.com/se/directlink.cgi?EET19980803S0011 )
> > - such a new machine would have hardware FP, PMT, PCI for graphics
> > cards and SCSI, standard Unix tools eg C++, and would be launched
> > perhaps in a year from now.
>
> Better but still sounds a bit pie-in-the-sky.
In fact it's fairly straight forward. Our solution is actually simple
and elegant. After some enquiries I made yesterday (Thurs), we could
have such a system ready (and waiting for the ARM 10's) by Q3 of next
year.
> IMHO those of us who want an up to date RISC OS box want Phoebe; the
> best thing about it was its PCI architecture and faster bus.
(Hypothetical bit)
I said simple and elegant :-
* Main Processor Bus 32bit @ 66Mhz - Same as RPC2;
* 3 or 4 slot Standard PCI bus - Same as RPC2;
* Support for Graphics Cards in the OS. RISC OS was found to be too
heavily tied to the VIDC architecture to make this an
impossibility on the RPC2). Allows for more than 4Mb V-RAM;
[1}
* Up to 512Mb S-DRAM - Same as RPC2;
* Nothing to do with Micro$oft;
* Stable, PMT Operating System. 4 million fellow users instead of
40,000; [2]
* Big-name software developer backed platform - Corel, Oracle etc;
* RISC OS front-end providing familiar 'Look and Feel';
* DMA IDE/ATA Interface (4 devices);
* USB support;
* IRDA support;
* Very Fast SCSI and Ethernet Capability via PCI or perhaps on the
motherboard itself - freeing up PCI slot.
* Nothing to do with Int*l (optional: SA=Intel, ARM10=Not);
* Future upgrade-path clear:
(Now) StrongARM -> (1999) ARM10+FP/SA2 -> (2000+)ARM11/SA3.
* Multiple Processor only.
(Now) 2 or 4 x SA110 upgradable in later release to 8 or 10
ARM10s or SA2 or SA3's (whichever is fastest);
* True 32bit Mode ARM support in OS. RISC OS used 26bit which will be
dropped in the ARM 10. RISC OS >could< never run on an
ARM10 for the exact same reason it doesn't run on the
SA1500 super-processor now;
* Hardware Floating Point: ARM10 will provide this for the first
time;
Hopefully, we would also be able to write some software to make the
porting of RISC OS software written in C or C++ much easier to
transfer across so we might get your favourites: Ovation Pro, Top
Model, Photodesk, DataPower etc. Hey, maybe even !Zap could be
ported...
Cheers,
Ross.
---===---
[1] Because we would use standard PCI Graphics card, we remove the
cost and problem of tying the OS to something like VIDC. Reduces
complexity too.
[2] By using Linux, we remove the requirement of such things as IOMD2,
removing even more cost and complexity.
Peter Naulls
> My interest is in continuing to use the Risc OS user interface and my
> existing Risc OS based software with which I am very productive. I'm
> certainly interested in any options to achieve this, including some
> investment of money.
>
<aol>
Me too
</aol>
Peter
--
+--------------------------------------------+------------------------------+
| Peter Naulls | Phone 64-7-854-9478 |
| pna...@usa.net | Java and JVM Consultant |
| http://chocky.home.ml.org/ | Technical Author |
| Java for Risc OS: chocky.home.ml.org/java/ | Program performance analyst |
+--------------------------------------------+------------------------------+
Unsolicted email received will be dealt with under the full extent of the law
Dave Roberts
<URL:mailto:kra...@argonet.co.uk> wrote:
[snip poss machine spec]
> Hopefully, we would also be able to write some software to make the
> porting of RISC OS software written in C or C++ much easier to
> transfer across so we might get your favourites: Ovation Pro, Top
> Model, Photodesk, DataPower etc. Hey, maybe even !Zap could be
> ported...
If you can pull this off, and especially if current developers can move
their products over to it, it would have my money! What it would seem to
need is some sort of mass meeting! All developers of software and hardware
need to get together. To find out what can be salvaged from the ashes and
what would need to be done to make any new venture work. To find out
whether the will still exists to make this happen! With a concerted effort
I strongly believe that something good and better(!) may come out of this.
Arjan Bokx
<URL:mailto:Da...@pharpech.demon.co.uk> wrote:
...which would be called the Phoenix, of course.
(see, already people are rallying around and thinking about the future!)
Arjan Bokx
Andreas Dehmel
I don't understand why you all go on about Linux. Think more broadly,
think Unix in general. The best OS I know is Unix. The best GUI I
know is the Wimp. Write a standard lib + window manager that runs
on Unix (or X should be OK too, but keep away from any high-level
libs that force their dubious philosophy on you), and make that
_any_ Unix. Wouldn't it be grand if you could just buy an Alpha
box, get the Wimp environment for it and enjoy the best of both
worlds? Although most people would probably use Linux PCs for it
I think it'd be a very bad move to restrict yourself to just one
Unix variant. The additional effort in keeping it portable across
Unix flavours would be relatively small, the benefits for people
willing to get some serious equipment (or install it on their
workstation at work, even) would be massive.
IMHO going Unix is the best solution. The best hardware is there and
will probably always be there so we'll never have to worry about
slow processors or missing hardware FP again. The OS is unrivalled
in stability and power. It's just the current GUI that's completely
and utterly pathetic, and projects like KDE who're just trying to
emulate the latest Windows version are going in the wrong direction
entirely too.
Also I don't like the suggestions about using ARM Unix machines
only, it's even worse than restricting yourself to x86 machines.
Don't make the same mistake Acorn did when they developed their
OS/GUI: KEEP IT PORTABLE!!! Since we're only talking the GUI here
that's even a lot simpler than if it involved a whole new OS too.
So by all means, port the Wimp to Unix (I've been toying with the
idea myself and with yesterday's devastating news there's no way
to go for me but in that direction). Now that Acorn programmers
_have_ to get their act together something like this might even
be done faster than you'd think.
Also it shouldn't be too hard to write something like OSlib for Unix
to make programs written in high-level languages more portable.
Alas the days of ARM-hacking are probably gone forever, but that's
a little price to pay if the Wimp can be safed and Acorn developers
are willing to migrate to Unix/Wimp -- just imagine Ovation Pro on
an Alpha box... :-)
Andreas
--
--------------------------------------------------------------------
Andreas Dehmel When asked how his parrot had died
Orleansstr. 34 the mathematician replied:
D-81667 Muenchen ``Polynomial. Polygon.''
deh...@forwiss.tu-muenchen.de
Tel. 089 / 28095-218 http://www.forwiss.tu-muenchen.de/~dehmel
Henning Hansen
<URL:mailto:Forb...@surprise.demon.co.uk> wrote:
>
> because we were planning an on-line version of Optima for RPC2.
>
> As it happens, I have this new company with some cash in it, but
> not enough on its own to launch RPC2 which would need UKP 2m-3m
> in working capital (to complete development and build stock).
>
> I wondered what people thought of these alternatives:
>
> complete and launch RPC2 next year (1999);
Acorn dropped it, and in a year it will not be any better.
>
> 2) move the Acorn community on to an ARM-based Unix (eg Linux) box
> and spend the money on tools to assist this process;
Too boring. We want to be something different.
>
> 3) build a new machine running eg Linux with say eight ARM10s
> ( see http://www.techweb.com/se/directlink.cgi?EET19980803S0011 )
> - such a new machine would have hardware FP, PMT, PCI for graphics
> cards and SCSI, standard Unix tools eg C++, and would be launched
> perhaps in a year from now.
Yes, that's the vision I would go for. With lots of ARM's, you not only
move the Acorn community, but also grab a great deal of the Linux people
out there. Hopefully, some good RISC OS software could be ported, like
Avante, Photodesk, Ovation, etc. - and software developers could be
attracted from other platforms as well (Tomb Raider ?)
I'm concerned about the laptop issue - it would be a good idea to have
a software compatible Linux laptop in mind.
>
> In cases 2 and 3, we would create a RISC OS user-interface.
>
> Moving to Linux would remove the cost of keeping RISC OS up to date
> and would allow support for the more modern 32 modes in ARM chips.
>
> I'm off to Italy for a few days, but will look through this thread
> when I get back to see if anyone is interested in these ideas.
>
Henning Hansen
--
2M ELECTRONIC A/S, Holsbjergvej 41 B, DK-2620 Albertslund
http://www.2m.dk fax: +45 43620092 phone: +45 43624433
André van den Berg
> > I'm
> > certainly interested in any options to achieve this, including some
> > investment of money.
> >
>
> <aol>
> Me too
> </aol>
>
> Peter
AND WE!!!
Met vriendelijke groet / Kind regards,
Andre van den Berg
ECD Computers Automatisering / Detachering
Patrijsweg 16
2289 EX Rijswijk / The Netherlands
Tel. +31(0)70 319 4343
Fax: +31(0)70 319 4963
VATNo: NL009507590B01
Acorn E-mail : ec...@xs4all.nl
ECD E-mail : a...@ecd.nl
PC E-mail : in...@ecd.nl
Detachering E-mail: detac...@ecd.nl
Homepage : http://www.ecd.nl
Daniel Pead
>1) raise a few million quickly from Acorn users eg on this NG to
> complete and launch RPC2 next year (1999);
... provided you can find several thousand "enthusiasts" with a couple
of grand burning a hole in their pocket and a fetish for yellow objects
you might get your money back - although turnover isn't the same thing
as profit & Acorn would no doubt want their palms crossed with silver.
The only hope I can see is for some vertical market where you can ship a
RPC2 as part of a 5-figure turnkey package - maybe (e.g.) some of the
groups working on video editing could make this happen.
A safer bet for a buyout might be a cheap A7000 replacement for schools
- some sort of "IMac-killer" - rather than a geek's dream machine... or
a "modular" system which could be used to build either & sold as
components to home-brew enthusiasts and niche-market "turnkey"
manufacturers.
Otherwise, you can get too much tasty PC or Mac G3 hardware for the
price, and if you can't face making Billgatus $100 richer you can run
one of the free unixoid operating systems on it. Heck, you can even get
UNIX applications off the shelf in PC World now.
>
>2) move the Acorn community on to an ARM-based Unix (eg Linux) box
> and spend the money on tools to assist this process;
>
>
>3) build a new machine running eg Linux with say eight ARM10s
> ( see http://www.techweb.com/se/directlink.cgi?EET19980803S0011 )
> - such a new machine would have hardware FP, PMT, PCI for graphics
> cards and SCSI, standard Unix tools eg C++, and would be launched
> perhaps in a year from now.
Oh by the way - most of the exiting new 3D graphics cards and stuff are
now coming out for AGP, not PCI; Affordable scanners and storage devices
are probably more likely to be a mixture of parallel port, USB and IDE
than SCSI... so think carefully about that spec!
I'd say that the new RiscPC should be in the form of a CD-ROM that you
plug into your existing RiscPC, Intel, Alpha or G3 box. Much cheaper to
produce.
Get a UNIX distribution, strip out all the gunk (TeX, 15 different
Tetris clones, 8 different mailers etc), make sure all the GUI-based
admin tools actually work, set up an X-based GUI that *really is* a GUI
and not just a tool for running 8 copies of XTerm side by side, and
clean up the documentation so there's one source for everything instead
of the current motley collection of man pages, TeXInfo, HTML, Postscript
and plain text.
--
Daniel Pead
Email: d...@octpen.demon.co.uk WWW: http://www.octpen.demon.co.uk/
Mike Clarkson
<URL:mailto:li...@binliner.demon.co.uk> wrote:
> > 1) raise a few million quickly from Acorn users eg on this NG to
> > complete and launch RPC2 next year (1999);
>
myself).
> The experiences of the poor Amiga community spring to mind, waitng
for years
> for their boat to come in.
>
> > and spend the money on tools to assist this process;
>
a cheap
> Sun or Alpha.
>
> > ( see http://www.techweb.com/se/directlink.
cgi?EET19980803S0011 )
> > - such a new machine would have hardware FP, PMT, PCI for
graphics
> > cards and SCSI, standard Unix tools eg C++, and would be
launched
> > perhaps in a year from now.
>
>
the best
> thing about it was its PCI architecture and faster bus.
>
seems to me to be the inability to run on anything post SA1 - ie. not
on ARM10, SA2 etc., due to lack of 32-bit compliance in RiscOS. Any
thoughts on this one?
--
Mike Clarkson, Shrewsbury School.
Mike Clarkson
>
> > The cancellation of the Risc PC 2 is inconvenient for me at Eidos
> > because we were planning an on-line version of Optima for RPC2.
>
>
> > As it happens, I have this new company with some cash in it, but
> > not enough on its own to launch RPC2 which would need UKP 2m-3m
> > in working capital (to complete development and build stock).
>
> up front outlay of ~500 ukp necessary.
>
> > I wondered what people thought of these alternatives:
>
> > complete and launch RPC2 next year (1999);
>
> recoup their costs and make some sort of profit in any licensing deal
> but this may be feasible if there is enough interest. From the n/g
> postings, I doubt the orders for PRPC2 much exceeded a couple of
> hundred and, unfortunately, we've no way of knowing what the recent
> offer would have produced in orders.
>
> Should Acorn hold out for much more than their development costs then
> it's probably going to be too expensive.
>
> I can't help thinking that the timing of this alarming turn of events
> might not solely be down to accountancy control but could be due to
> some unforeseen and expensive problems with Phoebe. This path could be
> very risky; I might venture a thou or so (if the October crash doesn't
> come sooner than expected) but not without seeing some confidence
> boosting figures.
I think the timing was just that it was time to write the cheque for the first production batch. They decided not to.
>
> >
> > 2) move the Acorn community on to an ARM-based Unix (eg Linux) box
> > and spend the money on tools to assist this process;
> >
> > 3) build a new machine running eg Linux with say eight ARM10s
> > ( see http://www.techweb.com/se/directlink.
cgi?EET19980803S0011 )
> > - such a new machine would have hardware FP, PMT, PCI for
graphics
> > cards and SCSI, standard Unix tools eg C++, and would be
launched
> > perhaps in a year from now.
> >
> > In cases 2 and 3, we would create a RISC OS user-interface.
>
> Probably either of these is the best long-term solution. One point
here
> is that the Acorn (as was) or RISC OS (as it must now be thought of)
> following is not due just to the front end look and feel of the GUI
but
> to the ease with which moderate skilled users can create WIMP
> applications in Basic. I doubt this facility can easily be ported
into
> Unix.
>
> A far better option, if feasible, would be to forget about the Acorn
> machine's architecture and port a RISC OS front end onto a PC/Linux
> base. I've heard reasons why this can't be done but, hell, look at
the
> size of Windows 98 and see how kludgy that still is. RISC OS is tiny
> and fast in comparison - how much code would it take, and how much
> speed loss would ensue from initially inserting a level of
translation
> and gradually converting RISC OS modules from ARM code to the Intel
'86
> family?
>
> > Moving to Linux would remove the cost of keeping RISC OS up to
date
> > and would allow support for the more modern 32 modes in ARM chips.
>
> Let us down gently, do you mean?
>
> > I'm off to Italy for a few days, but will look through this thread
> > when I get back to see if anyone is interested in these ideas.
>
> Do the words 'straws' and 'clutching' suggest anything?
Bill Oldroyd
>
> The cancellation of the Risc PC 2 is inconvenient for me at Eidos
> because we were planning an on-line version of Optima for RPC2.
Sorry to hear this.
... snipped - sensible ideas on what to do next ...
ARM based Linux systems, with the familar and well-liked bits of the
RiscOS GUI, multi-processor support etc..
This is what Acorn should have been doing for some time (like since the
early 1990's), but were being lead in the wrong direction by RiscOS.
The only addition I would make to your proposals is to think carefully
about how existing users would transfer to this new platform,
maintaining their investment in existing software/hardware. It doesn't
mean transferring everything we have now to the new system, just making
a good link between out existing kit and new kit, and perhaps making it
easier to use Linux on our existing kit in parallel with RiscOS.
Bill
dav...@argonaut.com
> In fact it's fairly straight forward. Our solution is actually simple
> and elegant. After some enquiries I made yesterday (Thurs), we could
> have such a system ready (and waiting for the ARM 10's) by Q3 of next
> year.
After a mass of depressing articles a nice constructive solution was pleasant
to find. This idea seems well thought out and has many benefits (not least the
large user base that Linux already has). Also it would provide a focus for all
those developers that must be wondering what to do now.
> Hopefully, we would also be able to write some software to make the
> porting of RISC OS software written in C or C++ much easier to
> transfer across so we might get your favourites: Ovation Pro, Top
> Model, Photodesk, DataPower etc. Hey, maybe even !Zap could be
> ported...
If this can be achieved and is as clean and elegant as you describe you have
a very interested customer here. I have supported Acorn as a good customer
and as a developer of software and I would like to see this community move on
to something that provide all those features we have desired for years.
If software can be made easy to port then we have the prospect of a good
collection of software that due to the nature of Linux would be available to
the masses for the first time. Another benefit would be a decent C++ compiler
that would enable far more complex programs to be created *and* without
Acorn's tie to education could enter more interesting markets.
I hope other people can see this as a very bright prospect if it gets the
backing it deserves.
...David
-----== Posted via Deja News, The Leader in Internet Discussion ==-----
http://www.dejanews.com/rg_mkgrp.xp Create Your Own Free Member Forum
Ernst Dinkla
> I heard a figure of less than 150 (146ish) had payed their deposits to
> Acorn for an RPC2.
Reading the messages of potential and disappointed buyers the
rate of the last week before thursday 12.00 hrs must have been
quite high. Pricedrop and included software information is
recent so one should judge the figure with some extrapolation
in mind.
However a PRPC2 resurrection seems to me a dangerous adventure.
One should thoroughly investigate what is actually finished in
hardware and software and what can be dropped to keep the price
low without losing any aspect of the specifications as they are
low already.
The path towards a unix OS system with a dual boot platform must
the best and than it is better to merge RiscOs with one of the
Linux/Arm systems. I do remember a discussion with Neil Carson of
Chaltech in this newsgroup about RO running on the Cats board.
Main issue was the replacement of the VIDC by graphic cards and
the change of the OS that was needed. He and others thought this
was possible. RiscOs running on CATS may be easier than marketing
the PRPC2 in quantities and it also assures a better future.
> The company exists now, but isn't doing anything yet. This 'box' would
> pretty much be it's first product. It has nothing to do with Eidos
> other than the fact Stephen is connecting them.
Whether the names of Stephen and Eidos are on the box doesn't matter,
enough people will know there's some link and that is a positive
influence. A company 'Forbidden Technologies' seems as attractive
and that company must be able to give their first product a name
that is at the same level.
There is enough ARM based hardware available that could be adapted
one way or another to cover the near future demands.
Chaltech's Cats
Corels initiatives
Toshiba's Java portable
The Acorn klones.
Remains the question whether Acorn is willing to license the latest
version of RiscOs with sources etc. and allows any changes to it.
If they restrict the use to desktop systems and are no longer losing
money its development I can't see a commercial disadvantage for
them. If they aren't willing for one reason or another I wonder
whether a legal action against them might change their mind. Their
promises to the last moment and changing that by a complete turnaround
have an impact on the commercial activities of dealers, developers and
business users. Acorn isn't bankrupt.
The software that would be included with PRPC2 was to some extent
from third parties so if they participate the deal will not loose its
appeal.
> For me personally, I agree. If as much of the RISC OS 'Look and Feel'
> can be retained under Linux, I'd be delighted. And as a bonus, we all
> get an OS that's stable as hell and all the Pre-Emptive Multi-tasking
> you could wish for. Also, theoretically a distriution could be made to
> work on other platforms other than ARM to tempt some of the other 4
> million Linux users to the RISC OS GUI.
As someone else wrote in this thread do not restrict the RO GUI to
Linux make it available to unix in general. And a universal vector
format like Draw included and even more PostScript compatible than
Draw is now. Or has unix already something like that? Could be EPS
or PDF as well of course.
Ernst
--
Ernst Dinkla Serigrafie,Zeefdruk edi...@inter.nl.net
All views expressed are my own and may have no relation whatsoever
to the views of Acorn, Intel, Oracle, IBM, ARM, Sun, Compaq, Micro-
Richard J. Hesketh
> As it happens, I have this new company with some cash in it, but
> not enough on its own to launch RPC2 which would need UKP 2m-3m
> in working capital (to complete development and build stock).
Eeek! This is beginning to sound very interesting... Is this for /real/?!
> I wondered what people thought of these alternatives:
>
> 1) raise a few million quickly from Acorn users eg on this NG to
> complete and launch RPC2 next year (1999);
Hmmm... Phoebe looks good now, but by (say) Q2 1999, she'll look really
/very/ indequate! Too little even later, I think!
> 2) move the Acorn community on to an ARM-based Unix (eg Linux) box
> and spend the money on tools to assist this process;
Not bad - this would be a satisfactory quick fix...
> 3) build a new machine running eg Linux with say eight ARM10s
> ( see http://www.techweb.com/se/directlink.cgi?EET19980803S0011 )
> - such a new machine would have hardware FP, PMT, PCI for graphics
> cards and SCSI, standard Unix tools eg C++, and would be launched
> perhaps in a year from now.
Yowzer!!! Given sufficient development of ARM Linux, this would be by far my
most favoured option! Not sure how many units it would shift - many of the
remaining die-hard Acorn enthusiasts are not UNIX-literate and don't want to
be - they just want another RISC OS box... However, I (and, I'm sure, many
hundreds like me) would love this option!
> In cases 2 and 3, we would create a RISC OS user-interface.
Sounds good...
> Moving to Linux would remove the cost of keeping RISC OS up to date
> and would allow support for the more modern 32 modes in ARM chips.
...all of which can only be a good thing!
> I'm off to Italy for a few days, but will look through this thread
> when I get back to see if anyone is interested in these ideas.
Funnily enough I'm off to Milan myself next week, providing that I can shake
off the bug that's keeping me in/near my bed this week!
Please, please, please keep me up to speed on this, Stephen - ANS will be
happy to act as a conduit for information on this project if it proves to be
viable, whichever path you take!
Regards,
Richard
--
o o Richard J. Hesketh ric...@glory-box.demon.co.uk o o
o o WebMaster, Acorn News Service http://www.acornusers.org/ANS/ o o
o o o o
o o DISCLAIMER: The Acorn News Service is an independent service. It is o o
o o in no way connected with, or supported by the Acorn Group PLC. o o
Sam Smith
This group will probably contact you themselves, but I believe a
lot of the groundwork has been done by a group trying to get Risc
OS on Linux. I *know* they exist. I don't know who they are. I
believe they are a couple of weeks from Beta.
Cheers
Sam
In article
<ant180523f7fxn%n...@ross.skarpsey.demon.co.uk>, Ross Tierney
<URL:mailto:kra...@argonet.co.uk> wrote:
>
> In article <na.5131774887....@argonet.co.uk>, Rob Hemmings
> <URL:mailto:rhem...@argonet.co.uk> wrote:
> > The first option sounds immediately attractive although perhaps the
> > others are more viable in the long term. What I would want to know
> > for options 2 or 3 is what would happen to existing Risc OS software?
>
> There are ways and means. Initial estimates say it would take about
> six months for us to get some system that would run a large proportion
> of code designed for RISC OS under Linux. Most of it would actually be
> fairly straight-forward, but time-consuming.
>
> We need to start something fairly soon, because the future of Optima
> is centered around this hypothetical product and if we're going to do
> a lot of work on it, we'll need to start fairly soon.
>
> The more interest we have, the more likely we are to do it.
>
> Cheers,
>
> Ross.
>
> (Not speaking in any official capacity what-so-ever!)
Sam Smith
mailto:Sa...@techie.com http://sams.base.org
Friends Spoiler List - mailto:Friend...@mindless.com
--
29 ways to decline a date :
25: I'd love to go out with you, but the last time I went out, I
never came back.
James Hammerton
Technologies' plan offers the best and most realistic hope of keeping
RiscOS, or rather a RiscOS-like platform using ARM chips alive. I
don't think buying up Phoebe would be a good idea, the machine IMHO
was simply not competitive with PCs. However the idea of creating a
RiscOS-like GUI for Unix is growing on me (despite my comments in
another thread), especially if it is accompanied by porting the Acorn
applications across to it such as Ovation, TechWriter, Artworks,
Avante, etc. This would potentially open up the market for the Acorn
applications to a wider market, especially if once the ARM Linux
version were created work started on making them available for other
Unix platforms. Thus the companies from the Acorn market could develop
ARM based Linux machines with this software being developed to run on
them and other Unix platforms. Existing users could run Linux/RiscBSD
on their RiscPCs. It may be worth making a ARM Linux port of BBC Basic
to run the RiscOS apps that have been written. It would be a painful
move since ultimately one cannot recreate the Acorn platform with 100%
accuracy on another OS but it would allow a relatively smooth
transition to the new platform. Consulting with the developers of ARM
Linux and machines based around it should probably be the first step.
I think this *could* work. After all Unix systems are much better
established than RiscOS, Linux is very widespread on peoples PCs let
alone anything else, and the RiscOS/GUI and software for Acorns does
have something to offer.
James
--
James Hammerton, Research Student, School of Computer Science,
University of Birmingham | Home Page: http://www.cs.bham.ac.uk/~jah/
Connectionist NLP WWW Page: http://www.cs.bham.ac.uk/~jah/CNLP/cnlp.html
Replace "seemysigfile" with "james" in my email address
Paul Wilkinson
Forbidden Technologies <Forb...@surprise.demon.co.uk> wrote:
> 3) build a new machine running eg Linux with say eight ARM10s
> ( see http://www.techweb.com/se/directlink.cgi?EET19980803S0011 )
> - such a new machine would have hardware FP, PMT, PCI for graphics
> cards and SCSI, standard Unix tools eg C++, and would be launched
> perhaps in a year from now.
>
> In cases 2 and 3, we would create a RISC OS user-interface.
No question in my mind that this option would be the way to go. Phoebe was
primarily for the users who wanted more oomph anyway (although I was going
to buy one, I was dissappointed by the still absent hardware FP).
In addition to creating the WIMP for Linux, I believe a few more things
will need to be ported across as well:-
Drawfile support (by far the best of the commonly supported native Acorn
formats)
BBC Basic support (superb for writing even fairly large apps, and one of the
main reasons I use Acorn kit)
And perhaps most importantly, the anti-aliasing font display. I don't mind
if it uses TrueType fonts, but the Acorn font display is *without parallel*
on any other system. I'm far too used to it on my desktop to let it go now!
Oh, and of course, !Zap (no question about this I hope?!)
Also, I would like to add to the messages of sympathy for the ex-Acorn staff.
In my mind, you can't crap on talent like that (in addition to a v. loyal
user base) and surive. I can't see Acorn being around much longer ...
Cheers
Paul
--
Paul Wilkinson (paul.wi...@diamond.co.uk)
Andrew P. Harmsworth Esq.
> What *I'd* like to manage also is to get the focus of the developer
> community directed this way so they might be able to keep their jobs
> too. If Avante, Ovation Pro, Photodesk and Top Model to list but a
> very few of the great titles we currently have could be directed
> towards this solution it may be a saving grace for the whole
> community.
Don't forget Sibelius! I wonder what the brothers Finn think of all this?
Sibelius VMP (I have the PC demo) is a killer app., and then some. I hope
they finish the job with the Acorn version - I suspect they will, as they
have a large installed base of RiscPC users, who won't necessarily want to
'downgrade' to a PC or Mac.
--
Science Coursework http://start.at/scirep * Warwick School * * *
Warwick School http://welcome.to/warwick * WARWICK * Physics Dept
Solar System http://travel.to/theplanets * CV34 6PP, UK * 01926 776464
* * * * * * my own views * * * a...@warwick.warwks.sch.uk * * * *
Andrew P. Harmsworth Esq.
off is beginning to appeal! It is ironic that the company name is such -
seeing as Acorn have forbidden us the Phoebe.
Your attempt at a basic spec for the machine is much more elegant than what
Acorn had to cope with - although I don't pretend to understand all of it.
I think we'll all be watching this thread closely, while a modicum of faith
still remains in the continuation of the platform.
I also think that if you went to the DTI for additional investment capital,
backed by the chaps at the BBC who use Optima... then maybe you could pull
this off. There's an app called Elements (the talking one) which a chap at
the DTI said "why doesn't every school have this?" It runs under RiscOS...
APH
Simon John
John Rees <jw...@argonet.co.uk> wrote:
> In article <19980917....@rmtwenty.demon.co.uk>,
> dar...@rmtwenty.demon.co.uk (Darrell Taylor) wrote:
> > just because it has the name Eidos attached to it. I think it would
> > certainly get the attention of the computing industry.
> Two words spring to mind, TOMB RAIDER. The publicity a machine built
> by the people who brought the world Lara Croft would grab the attention of
> not just the computing industry, but a large chunk of the general public.
> One of Acorn's problems has been it is associated with schools and is seen
> as fuddy duddy. The kudos of being connected to one of the leading lights
> in the games world would do any new RISC OS machine a power of good.
Exactly - most of these sad PC owners do little but play Tomb Raider and Doom
- probably to let out their anger on Win95 crashing on them again! Our
computers made Tomb Raider and we have the best Doom on any platform - and
it's even compatible with the other platform's WADs etc!
Sibelius and Corel Xara came from Acorns too, let us not forget.
So is it going to be this Eidos/Forbidden Technologies lot or IMS who are
going to save our butts?!
--
Simon E. John
Email: sim...@argonet.co.uk
WWW: http://surf.to/simonsite
ICQ: 15267939
128-Bit encryption and you can *still* read this?!
Neil A. Carson
> The path towards a unix OS system with a dual boot platform must
> the best and than it is better to merge RiscOs with one of the
> Linux/Arm systems. I do remember a discussion with Neil Carson of
> Chaltech in this newsgroup about RO running on the Cats board.
> Main issue was the replacement of the VIDC by graphic cards and
> the change of the OS that was needed. He and others thought this
> was possible. RiscOs running on CATS may be easier than marketing
> the PRPC2 in quantities and it also assures a better future.
Indeed. It'd certainly be easier than rewriting chunks of RiscOS to run
on Linux. This would explode into a *massive* job and take too long for
it to be worthwhile.
Neil A. Carson
> Oh, and of course, !Zap (no question about this I hope?!)
Want to try direct screen access for different graphics modes inside X?
Be my guest. This thread gets more and more silly by the minute.
> Also, I would like to add to the messages of sympathy for the ex-Acorn staff.
> In my mind, you can't crap on talent like that (in addition to a v. loyal
> user base) and surive. I can't see Acorn being around much longer ...
Indeed (to the sympathy). I think Acorn might be around for a while yet,
however.
Neil
James Hammerton
> In article <ant180540345xn%n...@ross.skarpsey.demon.co.uk>, Ross Tierney
> <kra...@argonet.co.uk> wrote:
> > What *I'd* like to manage also is to get the focus of the developer
> > community directed this way so they might be able to keep their jobs
> > too. If Avante, Ovation Pro, Photodesk and Top Model to list but a
> > very few of the great titles we currently have could be directed
> > towards this solution it may be a saving grace for the whole
> > community.
>
> Don't forget Sibelius! I wonder what the brothers Finn think of all this?
> Sibelius VMP (I have the PC demo) is a killer app., and then some. I hope
> they finish the job with the Acorn version - I suspect they will, as they
> have a large installed base of RiscPC users, who won't necessarily want to
> 'downgrade' to a PC or Mac.
What Sibelius make of this is a good question. They probably will
produce "Sibelius for RiscOS" or whatever they'll call it, but will
they continue to develop it after its first release? I hope they do.
My feeling about this is that if the idea being float of porting
RiscOS to Linux and producing a ARM-based Linux machine running the
RiscOS for Linux goes ahead, the people developing it should contact
Sibelius and offer to help with porting their app to the new
platform. It could then be offered to existing users who obtain ARM
Linux for their machines, and potentially could be ported to other
versions of Linux.
rsim...@ewrcsdra.cut_this_out.demon.co.uk
: I don't understand why you all go on about Linux. Think more broadly,
: think Unix in general. The best OS I know is Unix. The best GUI I
: know is the Wimp. Write a standard lib + window manager that runs
: on Unix (or X should be OK too, but keep away from any high-level
: libs that force their dubious philosophy on you), and make that
: _any_ Unix. Wouldn't it be grand if you could just buy an Alpha
: box, get the Wimp environment for it and enjoy the best of both
: worlds?
desktop computer platform you can think of and a lot of servers as well.
Unfortunately, most current commercial Linux software is available for i386
only, despite the fact that in many cases only a recompile is required to
support the other platforms. The same problem would exist if some aspects
of RiscOS were ported to Linux. Open-Source software would be available for
all platforms, but commercial would mostly be available only for the most
popular. Going to Unix in general would not reduce this problem.
--
----------------------------------------------------------------------------
Richard Simpson
Farnborough, Hants, Uk Fax: 01252 392976
rsim...@ewrcsdra.demon.co.uk
James Hammerton
Does this hold for producing a RiscOS-style GUI from scratch, with an
eye to porting existing applications over?
Thomas Boroske
Dave Roberts <Da...@pharpech.demon.co.uk> wrote:
> These sound like good ideas, and some in .advocacy would be quick to point
> out they mentioned them first! I can see only one real downside, and that
> being, what would up and coming software like Avante do? It would surely be
> impossible without a *major* rewrite to make these work?
Yes - unless some genius comes up with a nice new idea to emulate
RiscOS, but I doubt that. We can only hope that existing developers
would support the switch, and port their apps.
> One of the main disadvantages of Linux (listening only to what other
> people have said) is the lack of decent commercial software. Maybe we have
> what they are looking for?
:-) Nice thought, but consider that people saying that are comparing
with the Windoze market, ie there ARE commercial apps for Linux.
Kind regards,
--
Thomas Boroske
Thomas Boroske
Steven Purchase <st...@technium.demon_DOT_co_DOT_uk> wrote:
> Forbidden Technologies <Forb...@surprise.demon.co.uk> wrote:
>
> > 3) build a new machine running eg Linux with say eight ARM10s
> > ( see http://www.techweb.com/se/directlink.cgi?EET19980803S0011 )
> > - such a new machine would have hardware FP, PMT, PCI for graphics
> > cards and SCSI, standard Unix tools eg C++, and would be launched
> > perhaps in a year from now.
> >
> > In cases 2 and 3, we would create a RISC OS user-interface.
>
> I think that might be a good way forward. Collaborating with the two
> existing ARM-based machines that already exist would be a way to reduce
> the time and money it takes to get a working machine available.
OTOH, work on the user-interface could be started NOW on any
Linux machine (PCs, Alphas, whatever) that have stable Linux
versions and toolchains available.
In the last 5 years, the ARM has never been the ideal desktop
processor - at the moment, a plain PC seems significantly more
attractive for such a system. So I really think such a
project should try to support multiple hardware platforms,
among them ARM (of course), but surely PC, and anything else
would be a nice extra. I mean, doing an ARM version first
means you have to do a stable Linux - how good is ARMLinux
actually ?
Option 3) mentions an interesting spec, some would say dream machine.
But I feel any hardware produced for the Acorn userbase should
ideally justify itself, ie not only by being the only hardware
the new RiscOS 5 runs on. That way, if it proves to be too
expensive / not viable to build such a machine, or the next
ARM-based machine even further in future, you can easily sack
it and choose another processor platform.
> As for software, I feel there would be a bright, if slow future for
> development of software on a new OS.
Agree.
> > Moving to Linux would remove the cost of keeping RISC OS up to date
> > and would allow support for the more modern 32 modes in ARM chips.
>
> ...And open the door to many more technologies that already exist for
> Linux/Unix systems and will be ported in the future... I see a far better
> future for a machine like this than I did for RISC-OS machines.
I've been advocating that approach for ages :-)
> Please, if anyone has the money/power/will to make a go of a project like
> this, go for it. I'm sure you'll be able to count on support for a very
> large percentage of the Acorn community.
Here's one !
Thomas Boroske
deh...@Informatik.TU-Muenchen.DE (Andreas Dehmel) wrote:
> I don't understand why you all go on about Linux. Think more broadly,
> think Unix in general. The best OS I know is Unix.
Yes. But people would probably want to buy a complete package.
The OS would be Linux in this case. Why not ?
> The best GUI I
> know is the Wimp. Write a standard lib + window manager that runs
> on Unix (or X should be OK too, but keep away from any high-level
> libs that force their dubious philosophy on you),
X is a *must*, otherwise you use the capability to view
remotely on any Unix box, XTerm or PC.
> and make that
> _any_ Unix. Wouldn't it be grand if you could just buy an Alpha
> box, get the Wimp environment for it and enjoy the best of both
> worlds?
Yes - or even a PC and have a games machine + the machine that
runs RiscOS x ?
I'm quite sure you get Linux for Alpha, btw.
> Although most people would probably use Linux PCs for it
> I think it'd be a very bad move to restrict yourself to just one
> Unix variant.
The problem with supporting ANY Unix variant is that some
components may not be available, or be implemented differently.
Hell, some of the SunOS boxes here don't even have a standard
X11 distribution, but some wierd openwindows system instead.
Then there's all those differences in command line tools -
it's easy to write a GUI frontend for ps or top, but
it means a lot of work if you want to support all those
slightly different options and output formats.
Linux (or NetBSD or whatever) has the advantage that it
is available for 'enough' platforms, and that all these
platform can use an identical system, in effect.
Sure, if it turns out that it's only a matter of a recompile,
then do it for any platform you like. But I guess that
to get a 100% flawless system you have to rely on there
being a standard distribution underneath.
> The additional effort in keeping it portable across
> Unix flavours would be relatively small,
Do you think this is so trivial ?
> the benefits for people
> willing to get some serious equipment (or install it on their
> workstation at work, even) would be massive.
It would be ideal, but I happen to think that Linux or NetBSD or
some freely available Unix would be a reasonable compromise.
Just don't restrict it to one hardware platform only !
> IMHO going Unix is the best solution. The best hardware is there and
> will probably always be there so we'll never have to worry about
> slow processors or missing hardware FP again.
<fx: dreams come true ...>
> The OS is unrivalled
> in stability and power. It's just the current GUI that's completely
> and utterly pathetic, and projects like KDE who're just trying to
> emulate the latest Windows version are going in the wrong direction
> entirely too.
Yes :-) Projects trying to emulate and *better* RiscOS are
the way forward !
> Also I don't like the suggestions about using ARM Unix machines
> only, it's even worse than restricting yourself to x86 machines.
/me hugs Andreas with affection
:-)
> Don't make the same mistake Acorn did when they developed their
> OS/GUI: KEEP IT PORTABLE!!! Since we're only talking the GUI here
> that's even a lot simpler than if it involved a whole new OS too.
Agree one hundred times.
> So by all means, port the Wimp to Unix (I've been toying with the
> idea myself and with yesterday's devastating news there's no way
> to go for me but in that direction). Now that Acorn programmers
> _have_ to get their act together something like this might even
> be done faster than you'd think.
> Also it shouldn't be too hard to write something like OSlib for Unix
> to make programs written in high-level languages more portable.
> Alas the days of ARM-hacking are probably gone forever, but that's
> a little price to pay if the Wimp can be safed and Acorn developers
> are willing to migrate to Unix/Wimp -- just imagine Ovation Pro on
> an Alpha box... :-)
OK. Now what is required for this project ? I suddenly realise that
I don't know an awfull lot of Unix programming...
Obviously, a windowmanager/desktop would be required. Plus some
standard toolset library for menus/dialogue boxes.
How can messages/events be passed around ? I guess the X api
contains functions for that ? Must be the natural place, since
applications can be running on different systems, with the
X server being their only common link.
How are filetype actions done ? You could just define file$type_xxx
and alias$run variables as usual, and the filer could search these,
but would that be efficient ? Maybe we want a *better* system too ?
And what about application directories ? And the RiscOS-way of actually
navigating and starting apps in the filer ? I think applications
should be linked to (a requirement for multiuser), but should we
stay with the application directory structure ? I think it would
be nice if apps didn't have to be "seen" to work.
What about a /usr/riscos/bin, /usr/riscos/boot, /usr/riscos/resource
directory (/usr/riscos/bin could of course contain !run files too,
the actual runimage would then reside in resource - or this could
always be the case). .../boot contains all the filetype-claims
and such, and they could all be run at once without the system
having to search - clashes would have to be dealt with, but
it's time for that anyway. .../resource (could be called .../lib too)
would contain the "rest" of the applications in individual
application directoriess - these would also contain physical links to
the files inside .../boot and .../bin
The nice thing here is that it's easy for a shell to search just
the one bin-directory.
How do we get the usual RiscOS filer-appearance now ? First,
users could have ~/riscos directories. Inside this, they
have dirs with their private user data as usual, but they
could also link to the appdirs inside /usr/riscos/lib - the
filer could recognize it, put a "!" in front of it and
display an icon and open the directory on shift-doubleclick.
As standard, users would get an "app" directory that just contains
all wimp applications, maybe grouped in a way.
OK so far, but would the RiscOS-way of "install this app, try it,
delete it if it isn't interesting" work ? Yes, probably -
after all, apps are completely contained in one directory.
If a user installs/copies such a directory (=application)
using the filer, it could be recognized (have a file named
!!riscosapp inside the directory as a marker) and the boot and
run-files be copied into a private copy of the aforementioned
dir-structure (so that running from the shell works too).
Similar when deleting.
Not quite thought out, but what do you think ? The tricky
thing is to retain the good sides of RiscOS, while getting
the good sides of Unix on top...
Please get the discussion going.
Nick Boalch
Forbidden Technologies <Forb...@surprise.demon.co.uk> wrote:
[snip]
> 1) raise a few million quickly from Acorn users eg on this NG to
> complete and launch RPC2 next year (1999);
Well, its a nice idea but I don't think you'd be able to get the
capital, since it would be too risky an investment. Even the enthusiast
would be doubtful...
> 2) move the Acorn community on to an ARM-based Unix (eg Linux) box
> and spend the money on tools to assist this process;
Sounds OK. Linux /is/ excellent, and with a decent RiscOs-alike
window manager we'd be home and dry.
> 3) build a new machine running eg Linux with say eight ARM10s
> ( see http://www.techweb.com/se/directlink.cgi?EET19980803S0011 )
> - such a new machine would have hardware FP, PMT, PCI for graphics
> cards and SCSI, standard Unix tools eg C++, and would be launched
> perhaps in a year from now.
Brilliant. Again, Linux with a RiscOs window manager (for X
perhaps?) would be a great solution. And the spec sounds impressive.
With the ability to add all the expansion cards that RiscOS doesn't
support ATM, this would be rather good...
The only other things you'd need would be support for Drawfiles.
Hopefully RiscOS developers could be encouraged to port their apps.
I would hate to be without such gems as Zap and Messenger. And as
for Avante...
Options two and three would seem to me to be much safer investments
since releasing the Phoebe next year would just be plain stupid. It
would be so under-specced (?) by then that you'd never get your money
back.
Well, I'm definitely interested in 2/3. Keep in touch.
Cheers, Nick
--
!nick boalch
--> ni...@greenbeak.demon.co.uk
--> http://www.greenbeak.demon.co.uk/
--> ICQ# 17407400
Professor: Someone who talks in somebody else's sleep.
.
J.William Kay
<URL:mailto:seemys...@tardis.ed.ac.uk> wrote:
> What Sibelius make of this is a good question. They probably will
> produce "Sibelius for RiscOS" or whatever they'll call it, but will
> they continue to develop it after its first release?
Well, according to Ben Finn, posting to the Opus Mailing List:
> We have not made a decision on this (only having had a few hours to
> ponder it) but on an initial analysis this development at Acorn clearly
> would make a big difference to the viability of new-Sib for Acorn.
> Firstly because I doubt anyone would buy it other than as an upgrade,
> and secondly because a lot of users may decide to get a PC or Mac
> instead. So I can't say it looks like there will be much demand for it.
> However, we will consider this further and come to a decision in the
> next few days. You'll be the first to know.
> Whatever happens, you can rest assured that we'll continue supporting
> all of our customers whether they continue with Acorn or follow the
> PC/Mac route.
So, no firm decisions as yet from the Sibelius camp.
Regards,
William.
--
-----------------------------------------------------------------------------
J.William Kay Acorn StrongARM RiscPC 202MHz/26MB
-----------------------------------------------------------------------------
... Once you understand your computer it is obsolete
The Sherratt Clan
Tony Hopstaken <webr...@xs4all.nl> wrote:
> > What about a /usr/riscos/bin, /usr/riscos/boot, /usr/riscos/resource
> > directory (/usr/riscos/bin could of course contain !run files too,
> > the actual runimage would then reside in resource - or this could
> > always be the case). .../boot contains all the filetype-claims
> Arrgh. Just like windows. Splatering files all over the place....... :-(
Um, not as I read it. It is sort of like now but inside outŚ instead of
!app.!boot , !app.!run , !app.!etc
you have /boot.app , /run.app , /etc.app
If it is set up properly, it could work OK. But !App dirs are best for
neatness.
I like option 3Ś 1 is probably DoDo'd. By the time it is out it'll be
what, Q2 1999? It'll seem so lame by thenŚ
TTFN, Karl