Arculator v2.1 released!

[SarahWalker]

Only two years since the last release! I must be getting soft in my old age.

Changes since v2.0:

• New podules : Acorn AKA05 ROM Podule, Acorn AKA10 IO Podule (w/AKA15 MIDI Upgrade), Acorn AKA16 MIDI Podule, Aleph One 386PC/486PC Podule, Computer Concepts ColourCard, State Machine G16 Graphics Accelerator, Oak 16 Bit SCSI Interface
• Minipodule support for A30x0/A4000. Currently included : Acorn AKA12 MIDI/IO Podule, ICS A3INv5 IDE Interface, ZIDEFS IDE Interface
• Added A4 laptop emulation
• Added A500 prototype emulation
• Added HFE disc image support (read/write)
• Fixed MEMC1 timing
• Bugfixes

As ever, main site is at http://b-em.bbcmicro.com/arculator/ (and is in desperate need of some more modern styling...), github is at https://github.com/sarah-walker-pcem/arculator/.

[IanJeffray]

WOW! Impressive and boggling stuff. So much to play with! Totally boggled to see PC podule emulation.
Arculator support as an extension ROM is a smart idea too.
Great work. Thank you!

[NickLuvsRetro]

Thank you so much for continuing to work on this. It's been essential for going backwards and forwards with my own machine. Really appreciate it.

[Pernod]

Nice release, and good to see more podules supported, it saves me from having to reverse engineer them

What's the significance of the fixed MEMC1 timing? Does any particular software fail without this?

[SarahWalker]

No, but if you're aiming for cycle accuracy having broken timing is a bit embarrassing!

[steve3000]

Absolutely brilliant Sarah, as always!

Just tested RasterMan and very pleased it (still) runs nicely.

For the A500 prototype emulation - does this emulate the VIDC1 with sound byte format having D[7] as sign? (vs. VIDC2 used in production models, with D[0] sign). And is the FDC1793 emulated? Any other differences?

Is there any way to enter a different ARM3 speed, above 35MHz, without needing to modify source? BITD I had an IFEL 36MHz ARM3 upgrade and having recently overclocked a real ARM3 to 48MHz, I'd love to be able to do this (or more...) in emulation whenever I have something large to compress, compile, etc.

[SarahWalker]

For the A500 prototype emulation - does this emulate the VIDC1 with sound byte format having D[7] as sign? (vs. VIDC2 used in production models, with D[0] sign). And is the FDC1793 emulated? Any other differences?

Yes. Yes. ST-506 controller mapping is also different.

Is there any way to enter a different ARM3 speed, above 35MHz, without needing to modify source? BITD I had an IFEL 36MHz ARM3 upgrade and having recently overclocked a real ARM3 to 48MHz, I'd love to be able to do this (or more...) in emulation whenever I have something large to compress, compile, etc.

Not at present, no.

[timw]

This is brilliant - thanks, Sarah! I still remember reading in one of the magazines BITD that "emulation of an Archimedes on another machine will never be possible", or words to that effect...

Here are some notes for compiling on linux. Some peculiarities may be down to my particular build environment.

1. Add podules/morley_uap/src/Makefile to configure.ac
2. In podules/aka10/src/Makefile.am: change .o to .c and the same for morley_uap, otherwise the libraries don't get built and we get missing symbols at runtime:

   Code: Select all

   libaka10_la_SOURCES = ../../common/uart/6850.c aka10.c ../../common/joystick/joystick_sdl2.c ../../common/adc/d7002c.c ../../common/misc/6522.c

3. Add CFLAGS -std=c99 in podules/aka05/src/Makefile.am
4. #include <stdint.h> in wx-hd_new.cc and wx-joystick-config.cc
5. In wx-main.cc, exclude this windows call:

   Code: Select all

   #ifdef _WIN32
           SDL_SetHint(SDL_HINT_WINDOWS_DISABLE_THREAD_NAMING, "1");
   #endif

6. I get a compile error "wx-podule-config.cc:619:49: error: cast from ‘void*’ to ‘int’ loses precision". I don't know what's going on with these casts, but if I change int to long, it seems to work:

   Code: Select all

   cbox->SetSelection((long)val);

7. In wx-config.cc ConfigDialog::OnComboPodule(wxCommandEvent &event), the following doesn't return a valid string pointer:

   Code: Select all

   const char *sel_s = cbox->GetStringSelection().c_str();

   This breaks the function of the config menu, so I can't select any podules. I have to get the string as follows, instead:

   Code: Select all

   char sel_s[256];strncpy(sel_s, cbox->GetValue(), sizeof(sel_s));

8. Regenerate the makefiles:

   Code: Select all

   autoreconf --force --install
   ./configure --enable-release-build
   make

It works. Great stuff!

[arakeen]

Thanks for the update, however I am seeing an issue with the new version... On first launch of an A5000 machine I briefly get a Disc Error 23 or 22 for ADFS::5 (an HDF originally imaged from a physical A5000 ~10 years ago). Resetting the machine then boots without error.

Normally the machine is set to boot from this drive - this gives me the Error 23 at the initial boot screen. After a CMOS reset and before configuring boot back to the drive I get a short stall during the splash screen then Error 22 when trying to open the drive. Location is the same for each error (00000C00).

Screenshots attached. I reverted to 2.0 and confirmed that it still boots without error.

[SarahWalker]

Odd. Can you post your config please?

[arakeen]

I've included the .cfg and CMOS files, hopefully that covers it - thank you.

I noticed in the 2.0 release thread a mention of removing the 512 byte header from older images - is this worth a try, and if so what's the best/safest way to do that?

[arakeen]

I tried removing the top 512 bytes in WSL (dd if=hd5.hdf of=hd5new.hdf bs=512 skip=1), but I see the same behaviour when it's not configured as the boot drive (I haven't tried it as boot drive).

[SarahWalker]

Whoops, looks like secondary IDE drive support is broken on v2.1. Will be fixed for v2.2.

[garfield]

Whoops, looks like secondary IDE drive support is broken on v2.1. Will be fixed for v2.2.

Nice! Thanks Sarah!

[anightin]

Was just thinking how cool it would be to have floppy drive support on Arculator via a USB floppy or such.

Have been following the Drawbridge project and great to see that it's hooked into WinUAE for Amiga:

https://amiga.robsmithdev.co.uk/

I'm currently on the waiting list for the current batch of boards, and if successful will hopefully have a drawbridge-enabled drive to lend to anyone who has the skill-set to link in the drive support in a similar way.

Just a thought.

[sirbod]

When you edit a machine, there's a tagline below the machine name that says:

ARM2, 1MB RAM, MEMC1, Old IO, Arthur

Is that a leftover from a previous version or should it update to describe the machine config as list below in the drop-down menus? It doesn't appear to change.

[arakeen]

Whoops, looks like secondary IDE drive support is broken on v2.1. Will be fixed for v2.2.

Grand, thanks very much, I'll look forward to it

(Sorry for the delayed response, I swear all the posts following my last one only showed up for me today...)

[bakoulis]

Linux Mint 19.3 64bits
Arculator 2.1 file from homepage compiled with success.
Latest version with some fixes from github, stops with errors:

Code: Select all

mv -f .deps/arculator-wx-main.Tpo .deps/arculator-wx-main.Po
g++ -DPACKAGE_NAME=\"Arculator\" -DPACKAGE_TARNAME=\"arculator\" -DPACKAGE_VERSION=\"v2.0\" -DPACKAGE_STRING=\"Arculator\ v2.0\" -DPACKAGE_BUGREPORT=\"Sarah\ Walker\ \<b-em@bbcmicro.com\>\" -DPACKAGE_URL=\"\" -DPACKAGE=\"arculator\" -DVERSION=\"v2.0\" -DSTDC_HEADERS=1 -DHAVE_SYS_TYPES_H=1 -DHAVE_SYS_STAT_H=1 -DHAVE_STDLIB_H=1 -DHAVE_STRING_H=1 -DHAVE_MEMORY_H=1 -DHAVE_STRINGS_H=1 -DHAVE_INTTYPES_H=1 -DHAVE_STDINT_H=1 -DHAVE_UNISTD_H=1 -DHAVE_DLFCN_H=1 -DLT_OBJDIR=\".libs/\" -DHAVE_LIBZ=1 -DHAVE_LIBPTHREAD=1 -DHAVE_LIBOPENAL=1 -DHAVE_LIBASOUND=1 -I.    -I/usr/lib/x86_64-linux-gnu/wx/include/gtk2-unicode-3.0 -I/usr/include/wx-3.0 -D_FILE_OFFSET_BITS=64 -DWXUSINGDLL -D__WXGTK__ -pthread -I/usr/include/SDL2 -D_REENTRANT -O3 -MT arculator-wx-podule-config.o -MD -MP -MF .deps/arculator-wx-podule-config.Tpo -c -o arculator-wx-podule-config.o `test -f 'wx-podule-config.cc' || echo './'`wx-podule-config.cc
wx-podule-config.cc: In member function ‘void PoduleConfigDialog::OnCombo(wxCommandEvent&)’:
wx-podule-config.cc:351:140: warning: cast to pointer from integer of different size [-Wint-to-pointer-cast]
                                         update = item->function(this, item, (void *)(dynamic_cast<wxComboBox *>(obj)->GetCurrentSelection()));
                                                                                                                                            ^
wx-podule-config.cc: In function ‘void* podule_config_get_current(void*, int)’:
wx-podule-config.cc:588:59: warning: cast to pointer from integer of different size [-Wint-to-pointer-cast]
                         return (void *)cbox->GetSelection();
                                                           ^
wx-podule-config.cc: In function ‘void podule_config_set_current(void*, int, void*)’:
wx-podule-config.cc:619:49: error: cast from ‘void*’ to ‘int’ loses precision [-fpermissive]
                         cbox->SetSelection((int)val);
                                                 ^~~
Makefile:1510: recipe for target 'arculator-wx-podule-config.o' failed
make[1]: *** [arculator-wx-podule-config.o] Error 1
make[1]: Leaving directory '/home/user/arculator/src'
Makefile:398: recipe for target 'all-recursive' failed
make: *** [all-recursive] Error 1

[timw]

Code: Select all

wx-podule-config.cc:619:49: error: cast from ‘void*’ to ‘int’ loses precision [-fpermissive]
                         cbox->SetSelection((int)val);
 

See my post item 6 further up in this thread for a quick fix for this!

[bakoulis]

Code: Select all

wx-podule-config.cc:619:49: error: cast from ‘void*’ to ‘int’ loses precision [-fpermissive]
                         cbox->SetSelection((int)val);
 

See my post item 6 further up in this thread for a quick fix for this!

Thank you mate, I saw this fix and now compiled but not all fixes is possible.
changes .o with .c produce errors when autoreconf.
Also 3.Add CFLAGS -std=c99 in podules/aka05/src/Makefile.am don't know what replace with it or where added.
Also 5.#ifdef _WIN32 have not string SDL_SetHint(SDL_HINT_WINDOWS_DISABLE_THREAD_NAMING, "1");
This command is many lines away from #ifdef _WIN32

[steve3000]

I think I've found a bug in Arculator v2.1 (windows version).

I've set it up as a 4mb A3010 with monitortype configured as 0 (TV/standard) and I've been trying to get a demo working which uses a MODE module to create a widescreen TV-resolution MODE, called MODE 99. But it wasn't displaying correctly, and although things were clearly being drawn, the output was scrambled.

Trying to get this working, I loaded the mode module and entered MODE 99 from the desktop... and very odd... it didn't enter the expected 416x272 mode, but rather 800x600 in 256 colours? Reset Arculator, and try without loading the mode module, and yes MODE 99 appears to be built in as 800x600 in 256 colours, and it even displays like this with the monitortype as 0. Maybe this is a test mode (and it certainly could be useful, perhaps as MODE 32), but there doesn't seem to be any way to override this built-in MODE 99?

[SarahWalker]

MODE 99 is in the Arculator support extension ROM. Disable that and MODE 99 ought to return to normal.

[narrowboat-paul]

Fantastic work! Appears to work okay Windows 11.

There is one game from the 90s I can now play. Always enjoyed playing it back in the day.
Using Virtual A5000, the game attempts to load, but the sound goes up the creek and picture is a bit limited to say the least.

Arculater however allows the game to work perfectly. Good picture and great sound. The game by the way is Fervour.

I recall back in the day, the school I was as had a number of quest trackerballs. These were fantastic for playing Fervour with. Can't say I was ever an ace at it or anything, but still does not detract from enjoying playing it. Great gameplay and backing track!

[narrowboat-paul]

Outstanding work with Arculator
Any chance of networking?

[thegiddler]

Decided to try Arculator 2.1 on Linux and all went well with installation, configuration etc.

Booted up OK and shows desktop with mouse pointer in the middle of the screen - all looking good.

When I click into the window, the mouse is captured but when I try to move the mouse pointer it just moves down to the bottom right of the screen and stays there. It is then not possible to move the mouse pointer at all.

Anyone else had this problem?

[pdjstone]

When I click into the window, the mouse is captured but when I try to move the mouse pointer it just moves down to the bottom right of the screen and stays there. It is then not possible to move the mouse pointer at all.

Are you running Linux in a VM or using any kind of non-standard mouse?

[davidb]

I find that, after the first click, the pointer behaves erratically. However, clicking again produces the pop-down notification in the browser about the application having control of the pointer.

[pdjstone]

I assume thegiddler was talking about the native Linux version of Arculator, not my browser port? (although I have fixed the erratic pointer in my version, I hope to release an update soon)

[davidb]

Sorry! I jumped in too quickly when I saw that it was you replying.

[thegiddler]

When I click into the window, the mouse is captured but when I try to move the mouse pointer it just moves down to the bottom right of the screen and stays there. It is then not possible to move the mouse pointer at all.

Are you running Linux in a VM or using any kind of non-standard mouse?

I was running Linux in a VM on my Mac and tried a few different mice just in case but all had the same problem.
I have since installed Arculator 2.1 on a real Linux box and it works perfectly (with the same mouse) so must have been something to do with the VM settings.
Now having a nostalgic trip though all my old 3.5" disks...

Thanks for the replies.