Menu
▾
▴
arcem-user — Discussion and requests for help from users.
You can subscribe to this list here.
| 2003 |
Jan
|
Feb
|
Mar
|
Apr
|
May
|
Jun
|
Jul
|
Aug
|
Sep
|
Oct
(4) |
Nov
(18) |
Dec
|
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 2006 |
Jan
|
Feb
|
Mar
|
Apr
|
May
|
Jun
(2) |
Jul
(22) |
Aug
(6) |
Sep
|
Oct
|
Nov
|
Dec
|
| 2007 |
Jan
|
Feb
|
Mar
(6) |
Apr
|
May
|
Jun
|
Jul
|
Aug
|
Sep
|
Oct
|
Nov
|
Dec
|
| 2008 |
Jan
|
Feb
|
Mar
|
Apr
|
May
|
Jun
|
Jul
|
Aug
|
Sep
|
Oct
(5) |
Nov
(1) |
Dec
|
| 2012 |
Jan
|
Feb
|
Mar
|
Apr
|
May
|
Jun
|
Jul
|
Aug
(2) |
Sep
|
Oct
|
Nov
|
Dec
|
| 2013 |
Jan
|
Feb
|
Mar
|
Apr
|
May
|
Jun
|
Jul
(1) |
Aug
|
Sep
|
Oct
|
Nov
|
Dec
|
| 2018 |
Jan
(1) |
Feb
|
Mar
|
Apr
|
May
|
Jun
|
Jul
|
Aug
|
Sep
|
Oct
|
Nov
|
Dec
|
|
From: Derek S. <de...@q4...> - 2018-01-22 09:25:38
|
Hi, I have compiled the Linux version on Mint Linux X64. All works great, except the mouse is not working in Arcem. Is there something else I have to do you get the laptop touchpad working. -- Regards, Derek |
|
From: folkert <fo...@va...> - 2013-07-21 12:33:07
|
Hi, I would like to run Cyber Chess on ArcEm. I have the original disk laying here beside me but no floppy drive to read it. Does anyone have an image file of it? (I collect chess games: http://www.vanheusden.com/cchess/collection/ ) regards, Folkert van Heusden -- MultiTail is a versatile tool for watching logfiles and output of commands. Filtering, coloring, merging, diff-view, etc. http://www.vanheusden.com/multitail/ ---------------------------------------------------------------------- Phone: +31-6-41278122, PGP-key: 1F28D8AE, www.vanheusden.com |
|
From: Ralph C. <ra...@in...> - 2012-08-02 12:35:03
|
Hi Zack, > There is now a version of Slackware Linux for ARM that includes useful > programs like Firefox. Has anyone tried running it on ArcEm? Arcem emulates such an old set of ARM CPUs that I doubt any modern Linux would be happy. Linux/ARM development by Russell King did start out on Acorn's ARM machines but that was long ago. You've either got to go back to a very old Linux or look at an emulator for later Acorn ARM machines, e.g. http://www.marutan.net/rpcemu/, but even then you'll find it's too old for current Linuxes, you just won't have to go back quite so far when looking for an old Linux. You may stand more of a chance with current NetBSD. http://netbsd.org/ports/acorn26/ Cheers, Ralph. |
|
From: Zack S <zc...@ya...> - 2012-08-01 14:08:21
|
Hello, There is now a version of Slackware Linux for ARM that includes useful programs like Firefox. Has anyone tried running it on ArcEm? Thanks. |
|
From: Ralph C. <ra...@in...> - 2008-11-01 10:57:04
|
Hi Patrick,
Sorry for the delay; had paperwork deadlines. I suspect we're getting
more into developer territory below, so I'm CC-ing
arc...@li... as there's more people on the list and
they'll know the code changes that have gone on. I have set the
reply-to for there rather than arcem-users, subscribing you at the same
time.
> Yeah, sorry I was being lazy. I checked the 'arch/ReadConfig.c' where
> this file is read and it does appear to be $HOME for linux.
> Furthermore the supplied 'arcemrc' from the downloaded CVS sandbox
> needs to be ".arcemrc" in $HOME too.
>
> As a suggestion I think this should be handled by the makefile
> 'install' target. It doesn't work at all for (my) linux right now - it
> attempts to copy the wrong binary to /usr/local/bin.
Never noticed that before. I tend to not install and just do ./arcem.
> It should copy the correct binary and then copy the configuration file
> to $HOME and prefix a '.' to it.
OK, I've altered Makefile.
$ sed -n '/^install:/,/^$/p' Makefile
install: all
$(INSTALL) $(TARGET) $(INSTALL_DIR)
f=arcemrc; test -f $$HOME/.$$f || $(INSTALL) $$f $(HOME)/.$$f
$
$ make -n install
cp arcem /usr/local/bin
f=arcemrc; test -f $HOME/.$f || cp $f /home/ralph/.$f
$
> A 'man' page wouldn't go amiss either. It could be a simple subset of
> the online manual in nroff and copied into /usr/local/man as part of
> the makefile install target.
Yeah, it would be nice. I'd prefer to see the man page be the master;
a `groff -Thtml -man' version could always be put online.
> As a suggestion the location of the configuration file and (say) ROM
> should be overridden by commandline args to the 'arcem' binary and/or
> the ROM location can be entered in the configuration file (if not
> supplied via cmdline). Currently it's a PITA on my system as I must be
> in the actual directory of the arcem binary - otherwise it won't
> detect the ROM file.
Yes, this has always been a pain. I think it was discussed some years
ago, but nothing got done. If you've several ROM images, you end up
symlinking ROM to the one you want too.
Having just looked at what's involved, it seems there's code been added
by flibble since I was last involved that adds a --rom option, amongst
others. See arch/ArcemConfig.c's ArcemConfig_ParseCommandLine and
`arcem --help'. Perhaps it wasn't documented at the same time. Ah no,
I've now found http://arcem.sourceforge.net/manual.html#commandline
> If it's still an 'active' project I might update the code/makefile to
> implement the above. Let me know etc.
It's active in that once a year or so someone comes along and scratches
their itch, e.g. adds a port, etc. :-) This often results in more
cut-and-pasted lines, similar but slightly different, because none of us
have the time to do it right. :-) If you want CVS write access let me
know a SourceForge username and I'll add you.
If there's anyone out there with uncommitted stuff to CVS kicking about,
can you let the list know, else changes may give you the grief of
merging on the next update. It'll also give Patrick or me more free
reign to change stuff if we know we're not trampling others.
> Ta(2). It took me a while to work out why my 'double clicks' on my PC
> didn't execute !Configure. Thus I used F12 key to get cmdline prompt
> and did:-
>
> *configure Floppies 4
> *configure Harddiscs 2
>
> Then I worked out that the double-click delay and keyboard repeat
> rates were badly setup for my PC. Only a RAPIDO double-click launched
> the !Configure submenus. Then I managed to alter the mouse and
> keyboard delay rates for my PC ;-)
Ah, OK. *Configure Repeat and *Configure MouseStep are the RISC OS
options.
> Maybe online manual (+ man page?) should have a section on the mouse
> and keyboard delay rates and their configs.
Yes. They're under CVS too; hint, hint. It's the `webpages' module.
http://arcem.cvs.sourceforge.net/viewvc/arcem/
> OK... now I have managed to get that ELITE floppy image running - Yay!
> But now I need to work out the following:-
>
> 1. The whole emulation is running too fast. I need to slow it down.
> 2. I need to get a fullscreen of the ELITE game and/or ArcEm desktop.
I don't know if anyone's implemented having a scaled window, so you're
getting one X pixel to one RISC OS one, and at low-res RISC OS modes,
that's going to be a postage stamp.
The X code calls CopyScreenRAM() to get one row of pixels of video
memory into a temporary buffer, copies from there into an XImage, does
that for all the needed rows, and then calls XPutImage(3) to copy the
framebuffer to the X window. I guess the enlargement would happen
somewhere in that sequence, but it looks like the code has been copied
and changed several times as ports have come along.
Or perhaps the XVideo X extension is the thing to switch to as it can
put images with scaling using the graphics card IIRC. But that would be
more work and work on less platforms, and I doubt we need the speed.
It used to be that the X server's screen could zoom, something like
Ctrl-Alt-Keypad+ and Keypad- but that might not work by default in
Ubuntu set-ups. It was a solution for other emulators.
If the first method was implemented that might slow things down
sufficiently as a by-product. I'm a bit surprised that the game isn't
syncing itself with the 50Hz video vertical flyback signal. But perhaps
the emulator generates that as fast as it can! Other than that, I can
only suggest adding a delay at some core point, e.g. screen update or
every N emulated ARM instructions, just to see if it helps.
Cheers,
Ralph.
|
|
From: Patrick B. <pb...@ya...> - 2008-10-30 13:36:32
|
Hi Ralph, Thanks for the swift reply! I've got some further questions,comments etc - see inline.. [snipped] `strace -e trace=open ./arcem' suggests it's reading /home/ralph/.arcemrc for me, not ./arcemrc, so your disk params aren't being seen at the moment. Yeah, I know, $PWD for some files, $HOME for others. Yeah, sorry I was being lazy. I checked the 'arch/ReadConfig.c' where this file is read and it does appear to be $HOME for linux. Furthermore the supplied 'arcemrc' from the downloaded CVS sandbox needs to be ".arcemrc" in $HOME too. As a suggestion I think this should be handled by the makefile 'install' target. It doesn't work at all for (my) linux right now - it attempts to copy the wrong binary to /usr/local/bin. It should copy the correct binary and then copy the configuration file to $HOME and prefix a '.' to it. A 'man' page wouldn't go amiss either. It could be a simple subset of the online manual in nroff and copied into /usr/local/man as part of the makefile install target. As a suggestion the location of the configuration file and (say) ROM should be overridden by commandline args to the 'arcem' binary and/or the ROM location can be entered in the configuration file (if not supplied via cmdline). Currently it's a PITA on my system as I must be in the actual directory of the arcem binary - otherwise it won't detect the ROM file. If it's still an 'active' project I might update the code/makefile to implement the above. Let me know etc. > Now one of my goals here is to get ArcEm running the 'ELITE' game. So > I also downloaded an ".arc" image of it here:- > > http://www.iancgbell.clara.net/elite/arc/index.htm > > and renamed it to 'FloppyImage0' in my arcem directory. OK, b5052410.arc is a !SparkFS image, in other words, the equivalent of a ZIP or tar.gz in the RISC OS world. It's not a floppy image. There's also b5052420.zip for download off that page and that contains ELITE.ADF which is an 800KiB floppy image, suitable for calling FloppyImage0. Yay! That did the trick ;-) The ./hexcmos file from CVS seems to define no floppy drives and one hard disc exist. IOW, auto-hardware detection didn't exist on these machines, you used to change the setting in battery-backed RAM. To change this, you can run !Configure under Apps IIRC and then increase the number of Floppies to 1 and ST506 drives to 2. You should then see a floppy drive on the icon bar, click that to give a Filer window showing !elite, double-click that to put it on the icon bar, and then click it there to start the game. Ta(2). It took me a while to work out why my 'double clicks' on my PC didn't execute !Configure. Thus I used F12 key to get cmdline prompt and did:- *configure Floppies 4 *configure Harddiscs 2 Then I worked out that the double-click delay and keyboard repeat rates were badly setup for my PC. Only a RAPIDO double-click launched the !Configure submenus. Then I managed to alter the mouse and keyboard delay rates for my PC ;-) Maybe online manual (+ man page?) should have a section on the mouse and keyboard delay rates and their configs. If you still get it after setting up ~/.arcemrc, let us know. See above. This is working now AFTER I copied it manually to $HOME and renamed it to ".arcemrc" Run sha1sum(1) on it and see if it matches one of these digests. arthur-0.30.rom 6aebd686d97dfdf6726fa5f3246ef35b840b286d 524288 arthur-1.20.rom 1181ff9c2c2f3d6d414054ec33b2260404bafc81 524288 riscos-2.00.rom b82a78830dac386f9b649b6d32a34f9c6910546d 524288 riscos-3.00.rom b4fa494d51dd704c9da9653c2acc6acd08cc8cf4 2097152 riscos-3.10.rom 68533e0a93657a879c1697c04f32d517345b1a61 2097152 riscos-3.11.rom 3487729e87bebc9cb51665838c27beff22f7b3bd 2097152 riscos-3.19.rom ba8c7abe0e6c6c5d7138481c6c052425cbc8b8f4 2097152 riscos-3.50.rom c31318e712ee7de92e46f9df6c791b614d4c9304 2097152 riscos-3.60.rom 3cf75aa4d4dc8fe57110a124fbd1560d46c549b1 4194304 riscos-3.61.rom 71e67684e1d22f422ab9c586c6b0f900384e8f14 4194304 riscos-3.70.rom 42c3dd0ae43149849b674ea025cb2206512a5b4a 4194304 riscos-3.71.rom c5fe0645e48894fb4b245abeefdc9a65d659af59 4194304 riscos-3.80.rom 7e461a6aeb9b1cdc03ea734d0c3a9e1a65d3f98a 4194304 ncos-1.06.rom 4400bd142eebdc71b7ac1794f2cf7dc39de983e2 4194304 ncos-1.11.rom 7c9e2039bc096038ab236719211a236e9a28192f 4194304 arthur-0.30.rom MOS Utilities 0.30 (17 Jun 1987) arthur-1.20.rom MOS Utilities 1.20 (25 Sep 1987) riscos-2.00.rom MOS Utilities 2.00 (05 Oct 1988) riscos-3.00.rom MOS Utilities 3.00 (25 Sep 1991) riscos-3.10.rom MOS Utilities 3.10 (30 Apr 1992) riscos-3.11.rom MOS Utilities 3.11 (29 Sep 1992) riscos-3.50.rom MOS Utilities 3.50 (18 Feb 1994) riscos-3.60.rom MOS Utilities 3.60 (13 Apr 1995) riscos-3.70.rom MOS Utilities 3.70 (30 Jul 1996) riscos-3.71.rom MOS Utilities 3.71 (19 Feb 1997) riscos-3.80.rom MOS Utilities 3.80 (28 Apr 1998) ncos-1.06.rom MOS Utilities 4.50 (26 Oct 1996) ncos-1.11.rom MOS Utilities 4.50 (26 Oct 1996) Ok - did the cmd and it's matched your 3.10 rom image exactly - cheers! Well, at least that means it's not a tricky 32/64-bit problem. IIRC, Dr. David Gilbert, the original author, developed it on a 64-bit DEC Alpha, but the hands that followed, like mine, might not have been so careful. Yeah, it appears to be fine on 64bit linux right now - but I'll let you know if that changes. :-) OK... now I have managed to get that ELITE floppy image running - Yay! But now I need to work out the following:- 1. The whole emulation is running too fast. I need to slow it down. 2. I need to get a fullscreen of the ELITE game and/or ArcEm desktop. Cheers, Pat/ |
|
From: Ralph C. <ra...@in...> - 2008-10-30 00:25:30
|
Hi,
> > I then updated my './arcemrc' with (hopefully) the correct disk
> > params. Thus:-
>
> `strace -e trace=open ./arcem' suggests it's reading
> /home/ralph/.arcemrc for me, not ./arcemrc, so your disk params aren't
> being seen at the moment. Yeah, I know, $PWD for some files, $HOME
> for others.
I've checked into CVS an altered error that at least gives the path that
was being attempted to stop others getting caught out.
couldn't open config file: /home/ralph/.arcemrc
Cheers,
Ralph.
|
|
From: Ralph C. <ra...@in...> - 2008-10-29 20:01:27
|
Hi Pat, > ..yep I'm running AMD64 cpu and 64bit version of Ubuntu Hardy (8.04). > For some reason I cannot get hard disk images to work. I've downloaded > from the ArcEm site the 20Mb and 64Mb formatted images and called them > 'HardImage0' and 'HardImage1' respectively. Yes, they're the correct filenames for the first two drives. > I then updated my './arcemrc' with (hopefully) the correct disk > params. Thus:- > > == > MFM disc > 0 612 4 32 256 > MFM disc > 1 1024 8 32 256 > == `strace -e trace=open ./arcem' suggests it's reading /home/ralph/.arcemrc for me, not ./arcemrc, so your disk params aren't being seen at the moment. Yeah, I know, $PWD for some files, $HOME for others. > Now one of my goals here is to get ArcEm running the 'ELITE' game. So > I also downloaded an ".arc" image of it here:- > > http://www.iancgbell.clara.net/elite/arc/index.htm > > and renamed it to 'FloppyImage0' in my arcem directory. OK, b5052410.arc is a !SparkFS image, in other words, the equivalent of a ZIP or tar.gz in the RISC OS world. It's not a floppy image. There's also b5052420.zip for download off that page and that contains ELITE.ADF which is an 800KiB floppy image, suitable for calling FloppyImage0. > floppy format ADFS 800KB used for drive 0's r/w, 428909 length, image. > inserting floppy image FloppyImage0 into drive 0: ok > > From the above, it appears to have detected the drive 0 and drive 1 - > only giving errors for drive2/3. But I cannot access them from ArcEm > GUI. The ./hexcmos file from CVS seems to define no floppy drives and one hard disc exist. IOW, auto-hardware detection didn't exist on these machines, you used to change the setting in battery-backed RAM. To change this, you can run !Configure under Apps IIRC and then increase the number of Floppies to 1 and ST506 drives to 2. You should then see a floppy drive on the icon bar, click that to give a Filer window showing !elite, double-click that to put it on the icon bar, and then click it there to start the game. > I also see (as pre previous posters in this list) the following:- > > == > Read data command exiting due to failed file seek > seek: cylinder address greater than or equal to configured, 0 >= 0 > == > > ..but I don't know if this is referring to the HardImage0/1 (which is > present) or HardImage2/3 - which is not configured. Can anyone help > with this error message? If you still get it after setting up ~/.arcemrc, let us know. > BTW: I have ROM image 3.10 (I think) which I managed to track down > after extensive googling/newsgroups (sigh) here's the byte count:- > > == > -rw-r--r-- 1 patb patb 2097152 2008-10-29 12:33 ROM > == Run sha1sum(1) on it and see if it matches one of these digests. arthur-0.30.rom 6aebd686d97dfdf6726fa5f3246ef35b840b286d 524288 arthur-1.20.rom 1181ff9c2c2f3d6d414054ec33b2260404bafc81 524288 riscos-2.00.rom b82a78830dac386f9b649b6d32a34f9c6910546d 524288 riscos-3.00.rom b4fa494d51dd704c9da9653c2acc6acd08cc8cf4 2097152 riscos-3.10.rom 68533e0a93657a879c1697c04f32d517345b1a61 2097152 riscos-3.11.rom 3487729e87bebc9cb51665838c27beff22f7b3bd 2097152 riscos-3.19.rom ba8c7abe0e6c6c5d7138481c6c052425cbc8b8f4 2097152 riscos-3.50.rom c31318e712ee7de92e46f9df6c791b614d4c9304 2097152 riscos-3.60.rom 3cf75aa4d4dc8fe57110a124fbd1560d46c549b1 4194304 riscos-3.61.rom 71e67684e1d22f422ab9c586c6b0f900384e8f14 4194304 riscos-3.70.rom 42c3dd0ae43149849b674ea025cb2206512a5b4a 4194304 riscos-3.71.rom c5fe0645e48894fb4b245abeefdc9a65d659af59 4194304 riscos-3.80.rom 7e461a6aeb9b1cdc03ea734d0c3a9e1a65d3f98a 4194304 ncos-1.06.rom 4400bd142eebdc71b7ac1794f2cf7dc39de983e2 4194304 ncos-1.11.rom 7c9e2039bc096038ab236719211a236e9a28192f 4194304 arthur-0.30.rom MOS Utilities 0.30 (17 Jun 1987) arthur-1.20.rom MOS Utilities 1.20 (25 Sep 1987) riscos-2.00.rom MOS Utilities 2.00 (05 Oct 1988) riscos-3.00.rom MOS Utilities 3.00 (25 Sep 1991) riscos-3.10.rom MOS Utilities 3.10 (30 Apr 1992) riscos-3.11.rom MOS Utilities 3.11 (29 Sep 1992) riscos-3.50.rom MOS Utilities 3.50 (18 Feb 1994) riscos-3.60.rom MOS Utilities 3.60 (13 Apr 1995) riscos-3.70.rom MOS Utilities 3.70 (30 Jul 1996) riscos-3.71.rom MOS Utilities 3.71 (19 Feb 1997) riscos-3.80.rom MOS Utilities 3.80 (28 Apr 1998) ncos-1.06.rom MOS Utilities 4.50 (26 Oct 1996) ncos-1.11.rom MOS Utilities 4.50 (26 Oct 1996) When you get to a `*' prompt for the CLI, e.g. F12 from the Desktop, the command `fx0' should print out the OS version too. > BTW(2): I have also tried 'linux32 ./arcem' to force 32bit linux on my > AMD64 - did not see any difference :-/ Well, at least that means it's not a tricky 32/64-bit problem. IIRC, Dr. David Gilbert, the original author, developed it on a 64-bit DEC Alpha, but the hands that followed, like mine, might not have been so careful. Cheers, Ralph. |
|
From: Patrick B. <pb...@ya...> - 2008-10-29 16:13:45
|
Hello, I've downloaded the src from CVS and compiled on my Linux platform:- == Linux patb-desktop 2.6.24-21-generic #1 SMP Tue Oct 21 23:09:30 UTC 2008 x86_64 GNU/Linux == ..yep I'm running AMD64 cpu and 64bit version of Ubuntu Hardy (8.04). For some reason I cannot get hard disk images to work. I've downloaded from the ArcEm site the 20Mb and 64Mb formatted images and called them 'HardImage0' and 'HardImage1' respectively. I then updated my './arcemrc' with (hopefully) the correct disk params. Thus:- == MFM disc 0 612 4 32 256 MFM disc 1 1024 8 32 256 == Now one of my goals here is to get ArcEm running the 'ELITE' game. So I also downloaded an ".arc" image of it here:- http://www.iancgbell.clara.net/elite/arc/index.htm and renamed it to 'FloppyImage0' in my arcem directory. Now look at the diags from the cmdline:- == Version 1.10 (11/12/2005) + CVS, ARM2AS Couldn't open config file Could not open Extension Rom directory 'extnrom': No such file or directory extnrom_size = 0, extnrom_entry_count= 0 HDC: Couldn't open image for drive 2 HDC: Couldn't open image for drive 3 Shift/masks = r/g/b = 16/8,8/8,0/8 floppy format ADFS 800KB used for drive 0's r/w, 428909 length, image. inserting floppy image FloppyImage0 into drive 0: ok == Question: Why does it say 'Couldn't open config file' ? It is referring to './arcemrc' ? .. which is present and chmod'ed correctly (644 perms). From the above, it appears to have detected the drive 0 and drive 1 - only giving errors for drive2/3. But I cannot access them from ArcEm GUI. You can also see it has detected the floppy image. (ELITE .arc renamed). How can I run this image? I also see (as pre previous posters in this list) the following:- == Read data command exiting due to failed file seek seek: cylinder address greater than or equal to configured, 0 >= 0 == ..but I don't know if this is referring to the HardImage0/1 (which is present) or HardImage2/3 - which is not configured. Can anyone help with this error message? BTW: I have ROM image 3.10 (I think) which I managed to track down after extensive googling/newsgroups (sigh) here's the byte count:- == -rw-r--r-- 1 patb patb 2097152 2008-10-29 12:33 ROM == BTW(2): I have also tried 'linux32 ./arcem' to force 32bit linux on my AMD64 - did not see any difference :-/ Cheers, Patrick/ |
|
From: Patrick B. <pb...@ya...> - 2008-10-29 13:33:00
|
Hello,
Please could you enable me to post in the arcem-user list on Sourceforge? Thanks.
Regards,
Patrick Beilby
PS: If you are 'Ralph' reading this - Did you work at a Flight Simulator company in Sussex? ;-)
..and do you remember scrubbing the carpets there very late at night? ha ha.
|
|
From: Graeme J. <gr...@yu...> - 2007-03-22 17:13:35
|
On Thu, 22 Mar 2007, Ralph Corderoy wrote: > Hi grim, > >> I have RISC OS 2 ROMs which have always worked fine for me on Archie; >> but they don't seem to work too well with ArcEm. > > Just to make sure it's really 2.00, is the rom file's SHA1 digest > b82a78830dac386f9b649b6d32a34f9c6910546d? Yes, the checksum matches. Thanks - I hadn't thought to check that. >> The problem I have seems to be mainly with screen modes. When I boot >> with the RO2 ROMs, I see a "Supervisor" star-prompt. I issue *desk. >> and then get a mode-0-style, compressed desktop. Palette doesn't let >> me switch to any other mode. > > Looks like a CMOS RAM setting is constraining you, probably because your > hexcmos file contains RISC OS 3 settings, i.e. the same byte means > something different between the two OS versions. That makes sense - I think you must be right about this. > `*Status' may help, in particular the ScreenSize setting. `*Configure' > changes it. > > Alternatively, David Gilbert, author of arcem, provided a hexcmos file > suitable for RISC OS 2 back on the arcem-devel list around 2003-06-26; > the subject's "hexcmos for RiscOS 2". Sourceforge's archive appears to > be down at the moment. I'll definitely check that out. I have an Ibix The Viking disk image, and getting it to work would be the main aim of the exercise. I realise how sad that sounds! Thanks for your help, grim. |
|
From: Ralph C. <ra...@in...> - 2007-03-22 13:26:42
|
Hi grim,
> I have RISC OS 2 ROMs which have always worked fine for me on Archie;
> but they don't seem to work too well with ArcEm.
Just to make sure it's really 2.00, is the rom file's SHA1 digest
b82a78830dac386f9b649b6d32a34f9c6910546d?
http://inputplus.co.uk/ralph/#acornem
riscos-2.00.rom b82a78830dac386f9b649b6d32a34f9c6910546d 524288
> The problem I have seems to be mainly with screen modes. When I boot
> with the RO2 ROMs, I see a "Supervisor" star-prompt. I issue *desk.
> and then get a mode-0-style, compressed desktop. Palette doesn't let
> me switch to any other mode.
Looks like a CMOS RAM setting is constraining you, probably because your
hexcmos file contains RISC OS 3 settings, i.e. the same byte means
something different between the two OS versions. You can try a
"power-on reset" by holding down "R" during power-on IIRC. Don't know
how well that works in the emulator compared to the real hardware. :-)
`*Status' may help, in particular the ScreenSize setting. `*Configure'
changes it.
Alternatively, David Gilbert, author of arcem, provided a hexcmos file
suitable for RISC OS 2 back on the arcem-devel list around 2003-06-26;
the subject's "hexcmos for RiscOS 2". Sourceforge's archive appears to
be down at the moment.
We really need links to a selection of disc images and CMOS contents for
the different OSes that could be run.
Cheers,
Ralph.
|
|
From: Graeme J. <gr...@yu...> - 2007-03-21 16:32:22
|
Hi all, I have RISC OS 2 ROMs which have always worked fine for me on Archie; but they don't seem to work too well with ArcEm. I run Ubuntu Edgy Linux, and built ArcEm from CVS a few months back. My RISC OS 3.1 ROMs work without problems. The problem I have seems to be mainly with screen modes. When I boot with the RO2 ROMs, I see a "Supervisor" star-prompt. I issue *desk. and then get a mode-0-style, compressed desktop. Palette doesn't let me switch to any other mode. Any thoughts? grim. |
|
From: robo_creeler <rob...@ya...> - 2007-03-21 15:27:27
|
Hi ArcEm crew, had my eye on this project for a while now. I vaguely remember trying to compile the project on 10.3.9 but got build errors. I'm on Intel now.. is anyone else running ArcEm on Intel macs? If so how is it running? Does sound output work on OS X? thanks for any help, robo |
|
From: Peter H. <fl...@us...> - 2007-03-20 11:31:16
|
On Mon, Mar 19, 2007 at 01:52:46PM -0000, giu...@in... wrote: > > Hello, > > i have wrote a sort of tutorial / howto about Archimedes and linux. > > I have found not so few problems using arcem ( and transferring > disk images ), so i have set up a page that _for now_ you can find at > http://www.webalice.it/giulio.canevari/Acorn/EmulatingAcornArchimedesWithLinux.html > > I would be happy to see it onto the arcem sourceforge page, in order to > make life easier for other people. > > Let me know what do you think. I'll copy this onto the website, once I get a chance. Thank you for writing this, I especially like the Disc transfer instructions, as I had no idea how to do that :) Is it alright for me to edit it, and change some of the formatting? Peter -- Peter Howkins pet...@ma... |
|
From: <giu...@in...> - 2007-03-19 13:52:52
|
Hello, i have wrote a sort of tutorial / howto about Archimedes and linux. I have found not so few problems using arcem ( and transferring disk images ), so i have set up a page that _for now_ you can find at http://www.webalice.it/giulio.canevari/Acorn/EmulatingAcornArchimedesWithLinux.html . I would be happy to see it onto the arcem sourceforge page, in order to make life easier for other people. Let me know what do you think. Thank you. -- Giulio Canevari ---------------------------------------------------------------------------- Vuoi rendere inaccessibili agli altri i tuoi documenti riservati? Prova Ksek! Il potente software di cifratura di Interfree. Scopri le sue funzionalita': http://ksek.interfree.it/ Lo Staff di Interfree ---------------------------------------------------------------------------- |
|
From: Ralph C. <ra...@in...> - 2006-08-27 22:05:24
|
Hi Dmitry, (Please keep these emails going to arc...@li... so the archives may help others.) > ah, after playing around a bit, I found out that you can "increase the > number of floppies" in the Configure program. I thought that I > accessed a floppy when I clicked on this ":4" icon! Yes. And `*configure floppies 4' would also work at the star-prompt (F12 from the Desktop). > btw, the arcem 1.0 runs so fast on my machine that I cannot even type > anything on the command line prompt, i get "ccccccaaaaaaatttttttttt" > instead of "cat" :) This is resolved in the current cvs version > though. Yes. I think the keyboard repeat rate in the emulated CMOS battery-backed RAM may have been too fast. `*configure repeat 50' would probably slow it down enough. They must have done that in CVS. > > This suggests the hard disc image file for drive 4 is a different > > size to what's configured in the arcemrc file. See > > http://arcem.sourceforge.net/manual.html#HDimages The error > > messages could do with a lot of improving. The image file for RISC > > OS drive 4 is `HardImage0', confusingly, because it's the first hard > > drive. > > In fact, I am using the blank hd image downloaded from arcem site (the > 20mb one) and I have the following lines in acemrc > > MFM disc 1 612 4 32 256 > > But I still get this error message ("Disc error 09 at :4/00000C00"). Looking at the source code, I think the first number should be in the range 0 to 3 inclusive to specify the size of the image files HardImage0 to HardImage3. The web page user guide is incorrect in giving an example using 1 and 2 and saying they're the only possible values AFAICS. I'll try and get it corrected. So the problem may be you have HardImage0 but your arcemrc file is saying `1' instead of `0'. The error handling should be improved to tell you this. Cheers, Ralph. |
|
From: Dmitry S. <dmi...@gm...> - 2006-08-27 19:29:35
|
On 8/27/06, Ralph Corderoy <ra...@in...> wrote: > > Hi Dmitry, > > > I am using the cvs versoin, compiled on 2.6.16 linux and whenever I > > try to access a floppy, I have a message box "Disc error 09 at > > :4/00000C00" and a lot of messges on the terminal: > > OK, the `:4' means drive 4 and RISC OS treats drives 0 to 3 as floppies, > and 4 to 7 as hard drives. How do you know it's a floppy you're > accessing? Are you in the Desktop? If so, the icon would have `:0' > under it for drive 0. If you have a `*' command line prompt, which you > get with F12 in the Desktop, then the command `cat :0' would catalogue > drive 0 IIRC. Type `help cat' for more. > ah, after playing around a bit, I found out that you can "increase the number of floppies" in the Configure program. I thought that I accessed a floppy when I clicked on this ":4" icon! btw, the arcem 1.0 runs so fast on my machine that I cannot even type anything on the command line prompt, i get "ccccccaaaaaaatttttttttt" instead of "cat" :) This is resolved in the current cvs version though. > > Seek - cylinder address greater than configured (=0) > > Read data command exiting due to failed file seek > > Seek - cylinder address greater than configured (=0) > > Read data command exiting due to failed file seek > > Seek - cylinder address greater than configured (=5) > > Seek - cylinder address greater than configured (=0) > > Read data command exiting due to failed file seek > > This suggests the hard disc image file for drive 4 is a different size > to what's configured in the arcemrc file. See > http://arcem.sourceforge.net/manual.html#HDimages The error messages > could do with a lot of improving. The image file for RISC OS drive 4 is > `HardImage0', confusingly, because it's the first hard drive. > In fact, I am using the blank hd image downloaded from arcem site (the 20mb one) and I have the following lines in acemrc MFM disc 1 612 4 32 256 But I still get this error message ("Disc error 09 at :4/00000C00"). > > > PS. I am no Acorn user, I am playing with the emulator to get ArcElite > > running :) > > My Elite playing was on a BBC Master. :-) > Well, to be honest I only played Elite on a ZX Spectrum, but I was seduced by the rumor that ArcElite is the best Elite version ever :) -- Dmitry Sustretov ------------------------------------------------------------ -- PhD student at Langue et Dialogue, LORIA -- http://www.loria.fr/~sustreto/ ------------------------------------------------------------ |
|
From: Ralph C. <ra...@in...> - 2006-08-27 12:01:24
|
Hi Dmitry, > I am using the cvs versoin, compiled on 2.6.16 linux and whenever I > try to access a floppy, I have a message box "Disc error 09 at > :4/00000C00" and a lot of messges on the terminal: OK, the `:4' means drive 4 and RISC OS treats drives 0 to 3 as floppies, and 4 to 7 as hard drives. How do you know it's a floppy you're accessing? Are you in the Desktop? If so, the icon would have `:0' under it for drive 0. If you have a `*' command line prompt, which you get with F12 in the Desktop, then the command `cat :0' would catalogue drive 0 IIRC. Type `help cat' for more. > Seek - cylinder address greater than configured (=0) > Read data command exiting due to failed file seek > Seek - cylinder address greater than configured (=0) > Read data command exiting due to failed file seek > Seek - cylinder address greater than configured (=5) > Seek - cylinder address greater than configured (=0) > Read data command exiting due to failed file seek This suggests the hard disc image file for drive 4 is a different size to what's configured in the arcemrc file. See http://arcem.sourceforge.net/manual.html#HDimages The error messages could do with a lot of improving. The image file for RISC OS drive 4 is `HardImage0', confusingly, because it's the first hard drive. > I have a file HDFORMAT.ADF, downloaded from arcem site, in the working > directory of the emulator, symlinked to FloppyImage0. OK, although if you get one of the pre-formatted for RISC OS hard drive images from http://arcem.sourceforge.net/index.html#downloads then you shouldn't need to format it. > ROM Image is from support2.zip file. RISC OS 3.11 IIRC. > Anyone knows how to resolve this? Hope this is some help. The developers' list has people on it that know these answers. The users' list is even more quiet and I think I'm the only one here to answer. https://lists.sourceforge.net/lists/listinfo/arcem-devel > PS. I am no Acorn user, I am playing with the emulator to get ArcElite > running :) My Elite playing was on a BBC Master. :-) Cheers, Ralph. |
|
From: Dmitry S. <dmi...@gm...> - 2006-08-26 14:42:07
|
Hello, I am using the cvs versoin, compiled on 2.6.16 linux and whenever I try to access a floppy, I have a message box "Disc error 09 at :4/00000C00" and a lot of messges on the terminal: Seek - cylinder address greater than configured (=0) Read data command exiting due to failed file seek Seek - cylinder address greater than configured (=0) Read data command exiting due to failed file seek Seek - cylinder address greater than configured (=5) Seek - cylinder address greater than configured (=0) Read data command exiting due to failed file seek I have a file HDFORMAT.ADF, downloaded from arcem site, in the working directory of the emulator, symlinked to FloppyImage0. ROM Image is from support2.zip file. Anyone knows how to resolve this? -- Dmitry Sustretov ------------------------------------------------------------ -- PhD student at Langue et Dialogue, LORIA -- http://www.loria.fr/~sustreto/ ------------------------------------------------------------ PS. I am no Acorn user, I am playing with the emulator to get ArcElite running :) |
|
From: Ralph C. <ra...@in...> - 2006-08-14 14:37:43
|
Hi Tim, > If I use the Ctrl key at all, it seems that it does the software > equivalent of getting stuck - ie Ctrl-lock. You'll get more replies on the https://lists.sourceforge.net/lists/listinfo/arcem-devel mailing list as that's where most people that work on the source hang out. Can I suggest you subscribe to that and post again; it's very low volume. Cheers, Ralph. |
|
From: Tim V. <tim...@uk...> - 2006-08-14 02:56:03
|
Using CVS checked out version 1.10 And Release version 1.00 On Linux i686 (Fedora Core 5) Risc OS v 3.1 If I use the Ctrl key at all, it seems that it does the software equivalent of getting stuck - ie Ctrl-lock. No messages about this event appear at the console. I've tried both versions and the behavior is the same. Obviously frustrating to be unable to use the Ctrl key on what is otherwise an excellent emulation. Is there anything in terms of Debugging I can do to find this problem? |
|
From: Ralph C. <ra...@in...> - 2006-07-24 09:53:07
|
Hi Charles,
> adfs: module not supported by Novell, setting U taint flag.
> ADFS-fs error (device loop0): adfs_checkmap: crosscheck != 0xff
> Unable to handle kernel NULL pointer dereference at virtual address 00000000
> ...
> Call Trace:
> [<c01d2c5d>] vsprintf+0xd/0x10
> [<d0c3ba33>] __adfs_error+0x23/0x90 [adfs]
> [<c01089a5>] do_IRQ+0xe5/0x120
> [<c0106cf8>] common_interrupt+0x18/0x20
> [<c03d007b>] md_setup_drive+0x16b/0x5c0
> [<c015007b>] sys_fchdir+0x5b/0x80
> [<d0c3bb2d>] adfs_calczonecheck+0x2d/0xe0 [adfs]
> [<d0c3bc10>] adfs_checkmap+0x30/0xc0 [adfs]
> [<d0c3c12a>] adfs_read_map+0x1ca/0x200 [adfs]
Well, the good news is you've found a bug in the kernel.
adfs_error(sb, NULL, "map corrupted");
That `NULL' shouldn't be there. The bad news is the bug triggered
because the kernel's adfs thinks your ADFS disc map is corrupted.
If you've access to the original machine, you could try a `*CheckMap' on
it. That would show if the ripping went OK. And if there's problems
then Druck's DiscKnight may be able to sort them out. Whether this is
also possible under arcem on a disc image, I don't know. Druck may also
be able to provide a private copy of DiscReader that would work on the
image under Linux.
http://www.armclub.org.uk/products/discknight/
http://groups.google.co.uk/group/comp.sys.acorn.programmer/browse_thread/thread/c14e10aed4b14498
Cheers,
Ralph.
|
|
From: charles c. <cha...@co...> - 2006-07-24 09:30:27
|
Hi Ralph,
here's the output from the commands you sent.
cheers,
Charles.
adfs: module not supported by Novell, setting U taint flag.
ADFS-fs error (device loop0): adfs_checkmap: crosscheck != 0xff
Unable to handle kernel NULL pointer dereference at virtual address 00000000
printing eip:
c01d2709
*pde = 00000000
Oops: 0000 [#1]
Modules linked in: adfs uhci_hcd ehci_hcd ath_pci nvram usbserial parport_pc
lp
parport cpufreq_userspace speedstep_ich speedstep_lib freq_table thermal
process
or fan button battery ac snd_pcm_oss snd_mixer_oss snd_intel8x0 snd_ac97_codec
s
nd_pcm snd_timer snd soundcore snd_page_alloc ipv6 af_packet wlan_wep ds edd
sub
fs joydev sg st sd_mod sr_mod intel_agp agpgart scsi_mod ide_cd cdrom
ath_rate_o
noe wlan ath_hal yenta_socket pcmcia_core evdev e100 mii dm_mod usbcore
CPU: 0
EIP: 0060:[<c01d2709>] Tainted: P U VLI
EFLAGS: 00010213 (2.6.8-24.24-default SL92_BRANCH-20060721040626)
EIP is at vsnprintf+0x39/0x510
eax: cca2dca7 ebx: cca2dca8 ecx: 00000000 edx: 335d2358
esi: cca2dca8 edi: 00000014 ebp: ffffffff esp: cca2dc74
ds: 007b es: 007b ss: 0068
Process mount (pid: 6544, threadinfo=cca2c000 task=c03dc550)
Stack: c6332934 d0c3c60f d0c3c59d 335d2358 cca2dca8 00000000 cca2dca8 d0c3c5ab
00000014 ce121600 c01d2c5d cca2dd40 d0c3ba33 c03c1800 00000000 00000046
00000000 c01089a5 cbbfdc00 00000000 00000000 00000000 c0106cf8 cbbfdc00
Call Trace:
[<c01d2c5d>] vsprintf+0xd/0x10
[<d0c3ba33>] __adfs_error+0x23/0x90 [adfs]
[<c01089a5>] do_IRQ+0xe5/0x120
[<c0106cf8>] common_interrupt+0x18/0x20
[<c03d007b>] md_setup_drive+0x16b/0x5c0
[<c015007b>] sys_fchdir+0x5b/0x80
[<d0c3bb2d>] adfs_calczonecheck+0x2d/0xe0 [adfs]
[<d0c3bc10>] adfs_checkmap+0x30/0xc0 [adfs]
[<d0c3c12a>] adfs_read_map+0x1ca/0x200 [adfs]
[<d0c3c383>] adfs_fill_super+0x223/0x3a0 [adfs]
[<c01d2c17>] snprintf+0x17/0x20
[<c017fc81>] disk_name+0x61/0x70
[<c01570d2>] get_sb_bdev+0xb2/0x110
[<c03dc550>] test_wp_bit+0x0/0x70
[<d0c3c50e>] adfs_get_sb+0xe/0x20 [adfs]
[<d0c3c160>] adfs_fill_super+0x0/0x3a0 [adfs]
[<c01572f7>] do_kern_mount+0x87/0x180
[<c0169cda>] do_new_mount+0x5a/0x90
[<c016a2d2>] do_mount+0x172/0x1c0
[<c013a53e>] __alloc_pages+0x13e/0x470
[<c016a08a>] exact_copy_from_user+0x2a/0x60
[<c016a64c>] sys_mount+0x6c/0xc0
[<c0105c89>] sysenter_past_esp+0x52/0x79
Code: 54 24 0c 0f 88 8e 00 00 00 8b 74 24 10 8b 44 24 0c 89 4c 24 14 8d 6c 30
ff
89 f0 48 39 c5 73 0b 89 f2 83 cd ff f7 da 89 54 24 0c <80> 39 00 75 1a eb 40
39
ee 77 06 88 06 8b 4c 24 14 46 89 c8 41
|
|
From: Ralph C. <ra...@in...> - 2006-07-24 08:54:49
|
Hi Charles,
> mount -t adfs ./IMAGE.000 /mnt
>
> gives me a segfault on the newly ftp'd copy, which I suppose is
> progress from 'incorrect fs type'. Any thoughts where I should go from
> here?
What does
dmesg -c >/dev/null
mount -o loop,ro -t adfs ./IMAGE.000 /mnt
dmesg
as root give?
Cheers,
Ralph.
|
14 messages has been excluded from this view by a project administrator.
