Date   : Sun, 24 Jul 2011 00:26:40 +0200
From   : rick@... (Rick Murray)
Subject: Risc PC (Was 'Minitel in France')

[crossposted to BeebSoc; we probably ought to continue there]


On 23/07/2011 16:16, F. Haroon wrote:

> I personally find the Web TV (wwitv.net ? Forgot URL I think) is
> sufficient for me than bothering with boxes, subscriptions and being
> controlled by the gov't and media corps.  Anyway...

Rather harder to watch contemporary programming that way. Plus, it is 
somewhat harder to watch episodes of Bleach dubbed into Italian! ;-)

I used to like jpopsuki, but it looks like it recently facepalmed 
itself. <clicky><clicky> Ah, you can get it by throwing the following 
into WinAMP:
   http://jpopsuki.ddo.jp:8888/stream.nsv


> without ever having got an ARM machine although used enough of the
> Archies up to 3.1 (premature ceasure of production?).

Premature cease? The A300/A400 range? Well, it is logical given the 
A5000 and how much of a step forward that was over the older machines 
(ARM3, faster, better spec, IDE...). Sadly it was still based around the 
original chipset (so your 800x600 SVGA was somewhat limited).

Still, it was a good upgrade pre-RiscPC.


> and the early ones at least had no backwards compatibility.

!!??

AFAIK the early ones only missed the 16 bit sound. What, in particular, 
failed as compatibility?

Don't forget *all* (no exceptions) RISC OS 2 programs that used the
   SYS "OS_UpdateMEMC", 64, 64
trick to speed things up *failed* (no exceptions) on RISC OS 3 which 
behaved somewhat differently.
And by fail, I mean power-cycle required to unfreeze the machine (at 
least on my A3000).

Given the quantum leap [*] between the older machines and the RiscPC, 
compatibility was actually pretty good.


> Then they hyped Phoebe (RPC2) too early

To early for hardware that sounded good, but was fairly quickly 
overtaken. The period circa '95 - '05 was pure madness. Previously, 
there was a steady progression of 386s, 486s, Pentiums, and clock speeds 
rose into 100s of MHz. But then it all went a bit mental. Domestic kit 
seems to have flattened out at 1.5-2.5GHz, but to eke out more, we're 
looking at dual core, quad core, and all sorts of OMFG-style pipelining 
prediction tricks. Wasn't there a processor that actually follows 
branches, starts executing *both* paths of code, then tosses away the 
one not taken? Not to mention some versions of the Atom (IIRC) that 
don't execute code as it is written, but rearrange for better performance.
We shaln't even discuss what graphics cards can do. I don't think Phoebe 
would have made the grade. :-(


<aside>
Mmm... Mom's just called through that Amy Winehouse has apparently been 
found dead. Why does she think I care? There's plenty of people more 
deserving of death... like that guy in Oslo. Grrrr...
</aside>


Perhaps Acorn never really got over the BBC Micro and its educational 
successes? They never quite understood that increasing power meant that 
fewer people were bothering to learn how/why. Consider the Beeb, the MOS 
is something one person alone could understand. They could, if they 
wanted, roll their own. I wouldn't be surprised if people such as JGH 
didn't have a half-written replacement MOS they began as a rainy day 
project out of a mixture of boredom and curiosity.

Fast forward to RISC OS. It is possible that a single person could 
understand it, but it would be difficult. There are many nuances, and 
there's a lot to remember and correlate. Certainly a single person could 
conceivably write a replacement, but it would be painful. Can anybody 
hope to match years of development by the very people who designed the 
processor?

Fast forward to Windows and Linux. You know what? Give up. Something 
that takes a half hour to install is not something you will stand much 
chance of understanding. You might be a specialist in your own little 
part (i.e. I understand RISC OS's IIC quite well), but the whole thing? 
Every line of code? Unlikely.

To this end, while the Beeb was aimed at a generation of kids who wanted 
to get inside the machine (at my school, I wasn't the only person who 
knew 6502, and I wasn't the best at it either - though with an opcode 
chart blutack'd to the wall behind my bunk, I was probably the 
geekiest). However in the '90s the world split into two. The geek camp 
who didn't mind that the Slackware book started with instructions on 
compiling your own kernel (yes, chapter one page one!). The rest went 
Windows (pedestrian) or Mac (pompous) for a machine that they could use 
with minimal grief. I suppose the artefact of constantly rebooting for 
driver changes in Win16 is a hang-over of the fact that there isn't a 
lot of management needed other than periodic defrags.


Put it like this. It work we're tossing Windows and going Ubuntu. There 
are two types of people in the company. Those who were taught what a 
spreadsheet is, and those who were taught Excel.
The first lot adapted to Linux and, IIRC, OpenOffice, in no time. Those 
taught Excel have come completely unstuck as the Ubuntu version doesn't 
look and feel *exactly* like Excel.


How does this relate to Acorn? Simple. More and more, people wanted 
computers that "does stuff". It was led by the desktop revolution (does 
it do Word? does it do Photoshop?) but is more and more led by the 
on-line revolution (does it do Facebook? does it do eBay?) [note, also, 
the implicit suggestion of the dumbing-down of humanity]


> The company had Stan Boland which decided to dismantle the ship
> completely rather than try to rescue it.

Some laud his decision. Some want to baste him in lard and spit-roast 
his carcass do the dulcet sounds of [his] screaming.

I think I'm more annoyed about the demise of Acorn than anything else, 
but in the end it seemed more and more inevitable. Another annoyance is 
the Iyonix being floored by RoHS testing requirements. My God, I had a 
fluo-low-energy lamp that splattered RFI across the spectrum and it had 
an RoHS sticker on it...


> Better, faster and programmable machines that are the user's servants
> rather than their masters (almost)

Almost? It's been downhill ever since Clippy. While I respect Google and 
Amazon's search capabilities, there are times when it bugs the hell out 
of me.

Search: InPhone

Did you mean iPhone? Showing results for "iPhone". Click here
to see results for "InPhone".

Gah! If I *meant* i-****ing-Phone, I would have typed that!

I meant THIS: http://www.youtube.com/watch?v=hE3k1yHk5Fs

That's quite a bizarre and spectacular advert for... a new sort of phone 
socket. I guess anything rolled in the eighties. ;-)

Point is - as processing power increases, attempts to second-guess us 
increase likewise. Sadly, however, this isn't perfect, and it seems for 
some inexplicable reason people are quite willing to believe something a 
machine tells them. People that drive into lakes following their GPS? I 
can understand that, there's been some new roads built near here and 
Google's Navigation starts freaking out if you take a new road and 
appear to be flying at 60 through a field of wheat (papa! nicole!). ;-)
But rather than thinking "oh, it's not a road on the map" or "oh, if I 
turn left I'll get kinda soggy", people go ahead and, like, do it. 
Because a machine told them to. Ummm...? [dumbing down, I swear it!]

See also: http://xkcd.com/906/


> and don't take an eon to start and shut down leaving myself
> yearning for Linux and my hair grey.

I yearn for RISC OS. Both Linux *and* Windows are slow to boot. At least 
Linux gives you lots of nerdy text, followed by cute light-bugs to look at.

Actually, I don't mind much. My netbook suspends itself correctly when I 
close the lid. It's been about two weeks since the last reboot, and that 
was only for an Avast! update.


Best wishes,

Rick.

-- 
Rick Murray, eeePC901 & ADSL WiFI'd into it, all ETLAs!
BBC B: DNFS, 2 x 5.25" floppies, EPROM prog, Acorn TTX
E01S FileStore, A3000/A5000/RiscPC/various PCs/blahblah...