Jak zwykle pierwsza beta nowej serii WinUAE bogata jest w nowe poprawki - fani emulacji Amigi powinni się zrzucić na pomnik dla Toniego Willena, kiedyś gdy wypłyną ostatnie soki z ostatnich kondesatorów to dzięki jego pracy i Cloanto Amiga będzie wiecznie żywa;).
Ostatnia pełna wersja emulatora to WinUAE 4.0.1
Winuae 4.1.0 Beta 6 14/07/2018
- Full sub pixel accurate display emulation. AGA hires/shres BPLCON1 resolution is now fully supported, in all modes. Option in Chipset panel, not enabled by default. It is rarely needed and requires much more CPU power.
- Previously unemulated DIWHIGH H0/H1 AGA-only bits emulated. Does not require subpixel emulation mode.
- Emulated AGA borderblank undocumented feature: borderblank activates and deactivates horizontally 1 hires pixel later than on OCS/ECS. Only emulated if subpixel mode is enabled.
- Simple subpixel emulation mode (which only works if both odd and even playfield have same shift value) hires/shres resolution left border may have had wider than normal pixels in first visible pixel column.
- Added second fire button keys to built-in keyboard layout info strings.
- 68020 + memory cycle exact hung the emulator in some very tiny loops.
- Some programs caused continuous flood of "blitter is active, forcing immediate finish" log messages.
- Softfloat mode 68040/68060 nonmaskable FPU exceptions should be also disabled if Unimplemented FPU emu is ticked. It confuses 68040/68060.library FPU emulation.
- Blitter statesave with blitter active: log window opened and listed few lines of blitter debug information.
- CTRL+F11 quit ignored new "Warn when attempting to close window" option.
- Added new misc option which captures mouse immediately when windowed/full-window winuae window is activated.
- Added A4091 v40.4 and v40.9 to ROM scanner. v40.4 is most likely pre-release/beta version (but was dumped from working card), has serial debug logging and device name is "ncr.device", not "scsi.device".
- Fixed typo in NCR53C710 emulation (wrong bit), fixes A4091 v40.4 ROM boot hang.
- uaehf.device HD_SCSICMD didn't set scsi_SenseActual. Also set io_Actual=30 (=sizeof(struct SCSICmd))
- Allow also read commands (previously only metadata commands like INQUIRY were allowed) when accessing partition HDF via uaehf.device HD_SCSICMD.
- 53F94 emulation zeroed DMA transfer register at the end of transfer but it should be decreased by number of actually DMA transferred bytes (transfer won't necessarily complete fully). Fixes Blizzard 53F94 based SCSI controllers returning too large HD_SCSICMD scsi_Actual value after executing SCSI command that can return less than requested number of bytes.
- Fixed multimonitor mode mouse losing mouse capture at startup or mouse becoming unusable when first multimonitor window opened. Mainly x86 bridgeboard VGA related.
- Display port adapters that needed genlock data didn't work correctly since 4.0.0.
- Display port adapters didn't work in multimonitor mode. (I think they did work in some early 4.0.0 beta but got broken before final)
- "Warn when attempting to close window" option now works on the fly.
- Added CSA Twelwe Gauge emulation (partial, jumpers not yet supported)
- Added history list to statefile path string box.
PC Bridgeboard emulation rewrite :
- Replaced DOSBox and Fake86 x86 emulators with PCem emulator core. CPU, interrupt controllers (PIC), timers (PIT), DMA controller and IO and memory mapping. PCem emulates specific PC models in hardware level, DOSbox emulates sort of "generic" PC with generic BIOS which required too many hacks and had compatibility problems.
- DOS extenders, Windows 3.x enhanced mode and Windows 95 confirmed working.
- Floppy controller is still using my own implementation.
- (S)VGA, XTIDE, NE2000 expansions supported like previously.
- Added CL-GD542x linear frame buffer support.
- Optional CL-GD5429 SVGA chipset. 2M VRAM support and blitter which may be useful under Windows. Previous 5426 is still the default.
- A2386SX VLSI chipset memory mapping features implemented (ROM shadowing, EMS pages).
- CPU panel x86 speed setting supported. "Max" option is gone because PCem emulates specific CPU frequency rate.
- FPU is not available, PCem does not seem to have FPU emulation, at least for pre-486 CPUs.
- Bridgeboard shared IO regions are now dynamically changed, same with mono/color vs VGA IO regions. Affects A2088+ only. Sidecar uses jumpers.
- PC Speaker (enable in bridgeboard settings) and Sound Blaster (various models) emulation added from PCem. Note that SB expansion GUI configuration does not do any SB model specific filtering.
- PC compatible mouse emulation, under Expansion/Custom expansions. Simple config: select which Amiga port gets routed to emulated ISA serial card. Serial (COM1), PS/2 and PS/2 Intellimouse supported. Last two aren't really possible in real world but it was easy to support in emulation (and may not work in emulation either). Implemented because bridgeboard AMOUSE is not Windows 95 compatible.
- Removed AT IDE Secondary, A2286 and A2386SX BIOS only supports primary IDE.
CSA Twelwe Gauge:
- A1200 68030 accelerator.
- v1.0 ROM (does newer ones exist?) is not KS 3.1 compatible. (KS 3.1 adds $08000000 RAM automatically, then CSA boot ROM does the same..)
- 5380 based PIO SCSI controller.
- "CSA Twelve Gauge(TM)/CSA1250(TM);P/N 22162A1000;Date: 93.JUN.25;Computer System Associates, Inc. (C)1993. All Rights Reserved"
Winuae 4.0.1 Beta 6 14/07/2018
- Statefiles now include full relative and absolute path (disks, harddrives etc), supports restoring statefile paths correctly even if relative path mode changed between save and restore. Old versions ignore new fields, old statefiles work like previously.
- If statefile was restored with disk image inserted in drive(s) but disk image was not found, saving statefile again made disk drive empty. (old bug)
- When tracing ('t' debugger command) STOP instruction in debugger, break first when STOP is executed (like previously), next trace waits until STOP instruction exits. Previously STOP activated debugger immediately again.
- Picasso96 statefile restore didn't restore screen state fully. (4.x bug)
Winuae 4.0.1 Beta 5 12/07/2018
- input.keyboard_type was always read as Amiga keyboard. If PC layout was set as default, keyboard layout was read incorrectly from config file.
- Amiga reset during RTG rendering in RTG multithread mode could have caused a crash.
- Modified Filesys path in hardfile dialog was not saved if dialog was exited with return key.
- Don't attempt to resolve symbolic links or shortcuts when saving config file paths. (b4)
- Only attempt to resolve symbolic links when directory filesystem accesses it (to simulate Amiga side link), not when opening disk images, hardfiles etc..
- Fixed clipboard sharing (b1 update broke it)
- Delay clipboard data transfer when screen mode changes to prevent situation where mode changed because of program taking over the system. (Emulator can't easily detect between all kinds of Amiga-side triggered native<>RTG mode changes or if it is user switching modes/alt-tabbing which triggers clipboard data transfer check)
- Added Harddrives and Disk Swapper drag'n'drop workaround that fixes corrupted graphics during drag operation. (ListView_CreateDragImage seems to been broken since Vista or so..)
- Probably still one more to go..
Winuae 4.0.1 Beta 4 10/07/2018
- Floppy, CD, Tape, HDF geometry, HDF filesystem, disk swapper, RTC and flash rom file paths also support b1 resolve environmental variables update. (b1 update had issues requiring partial rewrite so it was good idea to update all other paths too. Excluding special paths like ROM paths.)
- Previously only some SCSI controllers flashed CD led when emulating CD drive.
- Unfortunately b1 path handling change was more complex than I assumed so there will be some more delay.. (If I would have known this, I would not have done it yet..)
- Check that existing configs still work, including disk swapper.
- Short description about path handling change: Previously any path was always converted to either full absolute or relative path with possible environmental variables resolved when config was loaded or path was typed/selected from dialog. When config was saved, old path (possibly containing environmental variables) was lost. Now paths are loaded unmodified from config file and only converted when needed. "Normal" path strings (no environmental variables) are still automatically converted to absolute/relative when saving config.
Winuae 4.0.1 Beta 3 08/07/2018
Possibly final beta, I wasn't supposed to do 4.0.1 vsync updates but below updates should have been in 4.0.0..
- If CD audio playing from real/virtual CD (not directly mounted image) was restarted because of config change, playback always started from beginning (or possibly failed to restart).
- 68060 FPU disable behavior change (68060 accelerator boards have "$f0" ROM to disable FPU because exec does not understand 68060 FPU and will crash if enabled), FPU is not anymore automatically disabled at hard reset if any emulated accelerator board is configured (matches now real world), if no accelerator board: if UAE boot ROM is at $f00000 ("old" mode), it disables FPU (just like real accelerator board do), if UAE Boot ROM is any other address: ("new" mode), m68k RESET instruction will disable FPU. Previously 68060 config got stuck in reboot loop because of not-disabled FPU if system was software reset.
- D3D9/11 vsync modes (lagless and standard) will now support 100/120Hz with optional black frame insertion.
- Always check interrupts immediately after MOVE to SR instruction (except if JIT enabled). Previously it was delayed until following instruction if fast 68020+ mode.
- Above FACOS JIT fix.
Winuae 4.0.1 Beta 2 03/07/2018
- QuikPak SCSI working. (Has same bug as tekscsi2.device 1.0: does not load RDB custom filesystems)
- 80-bit FACOS fix. Also removed old unneeded JIT FACOS workaround.
- If drive identity is read and it is CHS-only, don't enable CHS-only mode unless host OS also returns zero drive size. Some (old) Windows IDE drivers do support CHS-only drives. (3.6.0)
- Added checks to prevent crashes when running in lagless vsync under wine. (DISPLAYCONFIG_VIDEO_SIGNAL_INFO and IDXGIOutput1 not implemented)
- Very similar in high level compared to TekMagic, main difference is SCSI chip, 53C720 (was 53C710 in TekMagic).
- "tekscsi2.device (tekscsi2 2.1 (17.7.97) ©1997 Asimware Innovations Inc.)". Apparently v2.2 also exists.
Winuae 4.0.1 Beta 1 30/06/2018
4.0.0 bugs fixed:
- 80-bit native FPU mode FREM and FMOD had parameters swapped in x86 code.
- 64-bit FPU mode config mode was not loaded from config file.
- "Minimize when focus is lost" could cause crashes.
- "Minimize when focus is lost" incorrectly activated when switching modes in some situations. (same as above)
- If CPU panel select menu was active and then some other panel was selected: JIT got switched off.
- Don't abort with "tried to seek out of bounds" message when HDF/HD has virtual RDB and writing to too large offset.
- When inserting previously connected USB device, previous device type (Gamepad, CD32 pad etc) and autofire type (if any) was not restored.
- Do not copy data to Amiga side clipboard buffer if filesystem heartbeat signal is missing (=Some program took over the system), clipboard data can overwrite other program's data.
- Directory filesystem/UAE HDF uses indirect mode if new debugger is active. (Indirect is needed for debugger to detect memory reads and writes)
- "Minimize when focus is lost" minimized main emulation window when GUI was open and main window lost focus.
- RTG Multithread mode display refreshing was unreliable in palette modes when palette changed.
- Resolve hardfile/directory path environmental variables only when needed, loading and saving config now keeps original unresolved variables.
- Save also config file to crash dump file.
- QuikPak XP board emulation, not yet fully implemented.
- Added GVP v3.7, Blizzard 2060 v7.25 and QuikPak v2.1 to ROM scanner.
Tagi: Winuae, Amiga
Inne treści związane z tematem