Thread View: comp.sys.acorn.tech
4 messages
4 total messages
Started by Tom Kirby-Green
Wed, 16 Dec 1992 18:53
Reading !Run vars
Author: Tom Kirby-Green
Date: Wed, 16 Dec 1992 18:53
Date: Wed, 16 Dec 1992 18:53
15 lines
969 bytes
969 bytes
Hi quick question... What the best method of getting hold of a Application's Obey dir'. I'm using Desktop C - so basically I need the path that resspr etc uses to find its resources. At the mo. I'm launching the file with the usual !Run, but with the ugly line... Run <CADshell$Dir>.!RunImage "<CADshell$Dir>" Then using argv [1] to pick it up - theres gotta be a better way ;-) Thanks in advance - J.T.T.B. Tom. -- /---| The University of Westminster : BSc Computer Science. Hons, Year 3 |---\ | Tom Kirby-Green Email : gpcec@sun.pcl.ac.uk or Silence@Arcade | \----------| "Merry XMas to all, and a Windows(tm) free new year". |---------/
Re: Reading !Run vars
Author: Peter.Roozemaal.IS-assistent
Date: Thu, 17 Dec 1992 16:25
Date: Thu, 17 Dec 1992 16:25
21 lines
1380 bytes
1380 bytes
In article <1992Dec16.185334.11229@westminster.ac.uk> gpcec@westminster.ac.uk (Tom Kirby-Green) writes: >Hi quick question... > What the best method of getting hold of a Application's Obey dir'. >I'm using Desktop C - so basically I need the path that resspr etc uses >to find its resources. At the mo. I'm launching the file with the usual >!Run, but with the ugly line... > > Run <CADshell$Dir>.!RunImage "<CADshell$Dir>" > > Then using argv [1] to pick it up - theres gotta be a better way ;-) > > Thanks in advance - J.T.T.B. Tom. I would use the first part of argv[0]; that contains the full pathname with which the program was invoked. -- --- Peter W. Roozemaal --- Email: witaroo@dutinfh.twi.tudelft.nl -------------- -------------------------- Smail: Geitenkamp 45 ------------------------------- --- tel. +31-15-611389 ---------- 2623 BB Delft ---------- If it's urgent, --- --------------------------------- the Netherlands --------- don't phone me! ---
ST506 floppies/harddisks
Author: Peter.Roozemaal.IS-assistent
Date: Thu, 17 Dec 1992 16:32
Date: Thu, 17 Dec 1992 16:32
110 lines
6256 bytes
6256 bytes
I post this article on behave of Olaf Zevenboom who, due to serious
undergraduation doesn't have net access. Please send replies to the
above address and NOT to the address supplied in the signature
----------------------------------------------------------------------
Open letter to Acorn Computers Technical Department
===================================================
Problem: connecting a harddisk and two floppydrives to my Acorn Archimedes
A410/1
The harddisk is a ST506 type drive. It is a ST1100 or in full:
Seagate Imprimis Swift ST1100.
The floppydrives are both 3.5 Inch and should be connected to my AEDA
drivebuffer made and sold by Technomatic. The drives are:
(HD) Teac : FD-135HFN-419-U
(DD) YeData : YD 645C
Specifications:
A410/1 : upgraded to 4mb, temperature dependent fan, AEDA drivebuffer
The drivebuffer can controle four drives. My BASF 5.25 inch drive
works correctly with it, as did the Cannon (5.25 inch) drive I once
owned. As a reply to my letter with questions they send me a
catalogue with pricelist. You can imagine I won't spend another 30
pounds at Technomatic's! Unable to give your own product technical
support? Fabulous service guys!
ST1100 : 3.5 inch ST506 drive (83.75 mb formatted). 1072 cylinders,
10 heads, autoparking
The harddisk problem:
---------------------
When one tries to format a ST506 harddisk using the RISC-OS 3.10 format
program one will be warned by the program if the disk has more than 8 heads.
The program gives the following message:
WARNING - since the drive has more than 8 heads, the computer's 'low write
current/head select 3' links may need changing. -
This warning is omitted in the RISC-OS 2.00 formatter, but the effect is
the same!
This message is nowhere mentioned in manuals, etc. I do not have
PCB-schematics of my A410/1 so I cannot check which links are meant. I
suppose links LK12 & 13, but I am not sure.
When I continue with formatting without changing any links I get
discerror 16 (Sector ID has not been found?). Formatting with 8 heads is ok,
but 67mb out of 83mb is not entirely satisfactory. Besides when one tries to
verify the drive using the verify-menu option of the desktop-environment or
directly from the CLI (command line interpreter) ADFS starts complaining:
'Defect found at 000x0000' where x starts at 1 and increases by 1 at a time.
The drive is ok (writing directories/files, loading,saving, *map & * compact
worked fine). So why the errors. Neither the formatter is complaining.
The guy I bought the drive from said I should set 'Precompensation
Cylinder', 'Parking Cylinder' and 'Low current cylinder' to 1072, because
these options should not be used with my drive. Besides this I should remark
that the cables he gave me are cut: wire 8 of the thin-cable (20 wires) and
wire 3 of the thick-cable (34 wires). None of these cables are 'twisted'.
QUESTIONS:
- how can I get the best out of my drive (ie highest capacity)
- why does *verify behave strangely
- why are the wires cut and is this ok. What are the meanings of those
wires?
The floppydrive problems:
-------------------------
No matter which link combinations on the drives I try I can not get them
to work. I bought the Teac-drive at the dump, without a frontplate and
instructions on cabling, etc. I know I can (unfortunately) not use it in
HD-mode, but it should be usable as an ordinary DD drive. I bought the
YeData drive new at a sell-out. I have been able to get the Teac drive
working correctly with the Master128/512 of a friend of mine. Changing the
drive-links has an effect on the drive spin-ups and the drive-led, but a
proper catalogue or format is impossible.
An A4-sheet was included with the sealed YeData drive stating the
drivecable should be 'twisted' (wires 10-16 inverted). The links of both
drives got different names (except the obvious DS (drive select)). Why does
every manufacturer thinks he knows things better?
The standard documentation of the AEDA-interface consists of two
A4-sheets. It shows you how to configure AEDA for several combinations of
internal/external drive(s). Using 5 links (as standard not made) the
following options can be set (I quote):
- Data Cable Power Supply: if your drive can be powered through the data
cable, make links 2 and 4 using two short pieces of wires.
- External Disc Eject: If your external disc drive has the facility for disc
eject signal (as the Archimedes internal drive), make link 1.
- In-Use signal: make link 3
- Disc Present: link 5 when made will enable this signal.
This AEDA (Archimedes External Disc Drive Adaptor) sells currently at 10
pounds. When one buys the most expensive drivebuffer (30 quid) (at the time)
one expects one gets jumper-facilities to (un)make links and proper service!
The names of the links on the drives are:
TEAC : MS, ds0-3, IR, RY, DC , FG.
YEDATA : ds0-3, TM or TW, TD, MR, OP (0P?), P1, P2, P3.
The yedata drive links MR, OP and P1, P2 can be mode by jumpers (and of
course the ds-links).
QUESTION: I have a PSU, an old Master Compact-case and a couple of drives.
How can I make them work?
All reactions varying from comments, tips to complete solutions are mostly
welcome. I can be contacted at:
Olaf Zevenboom
roozemaa@dutiws.twi.tudelft.nl
--
--- Peter W. Roozemaal --- Email: witaroo@dutinfh.twi.tudelft.nl --------------
-------------------------- Smail: Geitenkamp 45 -------------------------------
--- tel. +31-15-611389 ---------- 2623 BB Delft ---------- If it's urgent, ---
--------------------------------- the Netherlands --------- don't phone me! ---
Re: Reading !Run vars
Author: Denis Howe
Date: Fri, 18 Dec 1992 22:34
Date: Fri, 18 Dec 1992 22:34
19 lines
1302 bytes
1302 bytes
In article <1992Dec16.185334.11229@westminster.ac.uk> gpcec@westminster.ac.uk (Tom Kirby-Green) wrote: >I'm launching the file with the usual !Run, but with the ugly line > > Run <CADshell$Dir>.!RunImage "<CADshell$Dir>" In article <BzEwAo.Cqw@dutiws.twi.tudelft.nl> roozemaa@dutiws.twi.tudelft.nl (Peter.Roozemaal.IS-assistent) wrote: >I would use the first part of argv[0]; that contains the full pathname >with which the program was invoked. What wrong with SWI OS_ReadVarVal? In fact, you can use Obey$Dir directly until you start multitasking so often you don't even need to define an extra private directory variable. -- Denis Howe <dbh@doc.ic.ac.uk> So Biggs, you're the idiot who bought all these IBM PCs. You're fired!
Thread Navigation
This is a paginated view of messages in the thread with full content displayed inline.
Messages are displayed in chronological order, with the original post highlighted in green.
Use pagination controls to navigate through all messages in large threads.
Back to All Threads