Sorry about the delay. I figured out that if I delete the pram.bin and nvram.bin before each boot it works fine. If they're present, it's random if it'll work or not, and if they're present from a different machine it almost certainly won't boot.

Here's the complete session log:

( 0.000s) loguru.cpp:665 INFO| arguments: bin/dingusppc.app/Contents/MacOS/dingusppc -d -b PMacG3_79d68d63.bin
( 0.000s) loguru.cpp:668 INFO| Current dir: /Users/arbee/emusrc/dingusppc/build
( 0.000s) loguru.cpp:670 INFO| stderr verbosity: 0
( 0.000s) loguru.cpp ...

So on the beige G3 it gets stuck in this loop:

FFF04698: 3C800000 lis r4, 0x0 ; out{ r4:0 }
FFF0469C: 60840004 ori r4, r4, 0x4 ; in{ r4:0 } out{ r4:4 }
FFF046A0: 7CA8242C lwbrx r5, r8, r4 ; in{ r8:F3008800 r4:4 } out{ r5:8400 }
FFF046A4: 7C0006AC eieio
FFF046A8: 70A40800 andi. r4, r5, 0x800 ; in ...

I've tried beige G3, 7500, and 6100, with and without appropriate guest OSes (7.5.5, 7.6.1, 8.1). It never even chimes so the guest OS is probably not in play though. I'll turn on full logging and get that for you later.

I had to install the Golden Gate beta on my MacBook Pro and everything works except DingusPPC. It starts up normally as far as the system info in the console and a window opens, but the emulated machine doesn't start up. No chime and black screen. This is with the latest git pull. Any ideas?

Check out one of the games booting up in this video.

I'll check in the pieces that make this possible this weekend, targeting the 0.290 release that will be at the end of September if all goes well.

Should be the next release, 0.289. Probably will be out some time next week.

Big update on this. Thanks to some major MMU and FPU bugfixing, MAME Github master boots and runs A/UX 3.1.1 on every machine Apple supported it on. That's the Mac II, IIx, IIcx, IIci, SE/30, IIfx, IIsi, Quadra 700, Quadra 900, Quadra 950, Quadra 800, Quadra/Centris 610, and Quadra/Centris 650 ...

Not as fast as original hardware still, but it's a noticeable bump at least.

To answer directly, press Delete (Fn-Delete on a MacBook) to enable MAME keys, then Esc will exit and Option-Enter will toggle full screen.

Yeah. Latest will run up to 9.1 now just like hardware.

Some great 68k improvements in now too - as I write this A/UX works on everything but the original Mac II, IIfx, and the Quadra 800/610/650 triplets. Locally I have that fixed so now IIfx is the last holdout. Also, the 68040 is now actually ...

I added branch instructions to my fork of risu from QEMU. risu captures to disk traces of instructions (random or not) from a machine (such as a real Power Mac). The same instructions can be ran on an emulated machine and compared with the trace. I think it would have been able to catch that bug ...

Will 9.1 eventually work?


I got 9.0 booting to Finder today. BEQL wasn't setting LR on fall through, only when the branch was taken. (The code in OS 9 that depends on that is *super* weird, too). Also 9 constantly flushes the instruction cache when you have a menu pulled down for who knows what ...

A/UX does not work in MAME. I haven't yet put a lot of effort into debugging why - we run a ton of 68K Unixes on other hardware so it's likely not super difficult to fix, just tedious because of the amount of loading that happens before it blows up.

If you pull current GitHub, a whole lot of things are far better with the 6100:

- Compatibility and stability are greatly improved by fixing a series of PowerPC core emulation issues
- FPU flags are implemented so the complete 13-minute Graphing Calculator demo completes with no errors
- RAM sizes ...

Thanks! That's a big info dump. I'm guessing I'll probably end up doing a built-in apprentice target because currently I can only run 7.1.2 and 7.5.0. 7.5.3/7.5.5/7.6.x all have issues. I haven't been able to find an 8.x CD that boots on a 6100, and I get identical results there in Dingus so I'm ...

What's a bus width transformer? It allows adding NuBus slots to a Mac that only had one PDS slot? There can be 14 slots (1 to E - Apple documentation) but Apple usually only used a max of 6 slots (9 to E)?

MAME as an optimization stores the emulated machine's memory at the host endianness but ...

In dev. I'm going to land the other PPC improvements soon (freeze for next release is Saturday) and likely the 030 and 040 PowerBook Duos.

I have the video modes and the RAM sizing fixed locally. Both of those features are borrowed from the LC III which I already had emulated elsewhere.

I've also wired up a bus width transformer that allows compatible NuBus cards that we emulate to work so you can have a dual-head PowerMac.

Most of ...

We've done some work with MAME's Lua engine to write a script that pokes the native mouse position into the Mac low memory globals, which gives the nice transparency. It works really well on some machines and pretty much not at all on others, depending on the type of ADB hardware involved. It should ...

On hardware, ScuzEmu lets the system directly see the files on the VFAT or exFAT-formatted SD card the SCSI emulator hardware is using. Normally you just put a hard disk or CD-ROM image on the card and the emulator presents it to the system as a hard disk or CD-ROM. ScuzEmu (the official BlueSCSI ...

2024 updates from my blog:
https://rbelmont.mameworld.info/?p=1626
https://rbelmont.mameworld.info/?p=1629
https://rbelmont.mameworld.info/?p=1638
https://rbelmont.mameworld.info/?p=1641

A summary of updates in 2024:
- Improved 680x0 FPU emulation, which fixed several games and the Jigsaw Puzzle ...

The Unitron ROM dumps I've seen are just copies of Apple's ROMs, including the 'Stolen from Apple Computer' icon. That's probably why the people involved don't want to go into coherent detail.

The 7.1.2 version of the Power Macintosh CD works and can be used to install a 6100 from zero. This disc is "pwrmaccd" in the MAME software list library if you have that, otherwise it's pretty easy to find. Note that the resulting 7.1.2 hard drive will also boot a Quadra 700 (and probably other ...

Just to clarify, there is some semi-working support checked in now. The big issue is that our PPC core wasn't reporting unrecognized instructions, it was just throwing a PPC illegal instruction exception silently. Installing MacsBug helped to figure out what was actually happening, and adding ...