Article View: comp.sys.transputer
Article #1931Re: Atari/Perihelion Transputer Machine Spec
From: Hans Hansen
Date: Sat, 07 Nov 1987 05:24
Date: Sat, 07 Nov 1987 05:24
80 lines
5271 bytes
5271 bytes
In article <32936@sun.uucp> cmcmanis@sun.UUCP (Chuck McManis) writes:
In article <1034@titan.camcon.uucp> anc@camcon.uucp (Adrian Cockcroft) writes:
>Cost
>----
> Priced well below Mac II, base level entry price (no winchester or monitor)
> aimed at 1000 pounds according to Jack Lang.
> For now they will provide a set of 3 manuals, User Manual, Developers Manual
> and Technical Manual for 50 pounds, you then become a registered
> developer and get a priority place in the queue for developers hardware
> in December.
I was at Comdex and talked to Jack Lang, Tim King, and Jessica Lang at
length about this box. What I found out was quite enlightening to say the
least, let me say this first, "This is a really neat machine! ", then let me
say this second, "It ain't cheap, when you configure it like it Workstation,
it costs like a Workstation." So before I go on lets add some numbers to
these statements :
At the show the Perihelion folks were saying $2K for the ABAQ (Ay-bak)
and you had to have a Mega ST as well ($1.5K - $3K depending on which
one you have) and a monitor that can display the images your interested
in ($.5 - $3K, where 500 gets you a Multisync, and 3K gets you the
multisync II plus which can display 1K X 800 pixels), plus you really
need a hard drive so add another 1K to 1.5K depending on size (20 - 40Meg)
So minimum we're talking $5K to get a low-res color system with a 20Meg
hard drive, upto $8K for a 40meg High res systems (which is cheaper than
either a Mac II or a Color Sun). When the 'real' machines become
available the ST will be optional (and hopefully they will have redesigned
the clunky case) but they were murmering $5K sans monitor or hard drive.
So for those of you who want a Workstation for the cost of an Amiga or ST
will have to wait a few more years for the technology steamroller to get
a little be farther down the road.
As for the machine it was running some neat demos that primarily showed
off it's color capabilities. Since they were 'canned' images being replayed
aka ShowAnim or some other framebuffer animator it was not clear how fast
the actual machines were. One was running GEM on a 19" 1K X 1K monochrome
desktop, unfortunately most GEM applications couldn't really hack the
huge screen. I would be interested to know if Atari will be supporting
both GEM and Helios on the machines (all the really good questions hit
you after you've left the booth. :-)) from the looks of it, it did not
appear likely. The Perihelion people were quick to point out that the
slaved Mega was only providing I/O services and non of its memory was
being used by the ABAQ. So I assume if you wanted to run GEM you would
either emulate the 68000 or use a version of TOS in ROM that passed all
video requests on to the ABAQ.
Other interesting tidbits; the video memory (1 meg of video rams) was
separate from the ABAQ's program memory (shown with 4 meg). The internal
slots will accept boards with additional Transputers or memory on them
so you could continue to beef up this thing. And while some of the Demos
were written in Occam (INMOS's transputer language) Perihelion was
going to support C as the primary development language (what no BCPL?).
It also had a custom blitter that apparently was really fast however
only the demos were available so I couldn't really test drive them
all that well. Overall it looks like an interesting box and is probably
going to excite a lot of graphics folks. I don't see a whole lot of them
making it into the mythical home market, or the ever lucrative business
market. If Inmos has really conquered its manufacturing problems and
Helios lives up to its advance billing this should be worth a look at
least. And while there were boxes to play look at, at the show, I know
that more than half the work will be in the software.
So things to watch for :
o How Atari 'positions' this product. Depending on when it gets out the
door it will be competing with 386/486 boxes, Mac IIs, and Low end
Suns.
o What happens to the various ST owners. (In my opinion they should
split their product line like the Apple II/Macintosh computers are
truely separate.)
o Messages from people who have machines to play with of their own
(I am of course available for Beta Testing :-))
Sorry it dragged on so long, it was one of the more interesting aspects
of an otherwise PC/386/Laser Printer dominated show.
--Chuck McManis
uucp: {anywhere}!sun!cmcmanis BIX: cmcmanis ARPAnet: cmcmanis@sun.com
These opinions are my own and no one elses, but you knew that didn't you.
Technical Details
Message-ID:
<1204@omepd>Path:
arc1-ca.pugleaf.net!utzoo!mnetor!uunet!husc6!rutgers!ames!ucbcad!ucbvax!hplabs!hp-pcd!uoregon!omepd!hahReferences:
<1034@titan.camcon.uucp> <32936@sun.uucp>Show Full Headers
Complete Article Headers
Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP Path: arc1-ca.pugleaf.net!utzoo!mnetor!uunet!husc6!rutgers!ames!ucbcad!ucbvax!hplabs!hp-pcd!uoregon!omepd!hah From: hah@mipon3.intel.com (Hans Hansen) Newsgroups: comp.sys.transputer,comp.sys.atari.st Subject: Re: Atari/Perihelion Transputer Machine Spec Message-ID: <1204@omepd> Date: Sat, 7-Nov-87 00:24:58 EST Article-I.D.: omepd.1204 Posted: Sat Nov 7 00:24:58 1987 Date-Received: Tue, 10-Nov-87 02:21:44 EST References: <1034@titan.camcon.uucp> <32936@sun.uucp> Sender: news@omepd Organization: Intel Oregon Microcomputer Engineering Lines: 80