Did you commit every file? I get an error compiling starting with 386.c with no timer.h. Using MinGW+Msys.
No, I seemingly screwed up that commit. Now fixed (hopefully).
Thanks! It works.
SarahWalker, please add these features to PCem:
1. Access to host time and date disablement
2. Shuttle HOT-433 with Award BIOS machine
3. Pentium emulation
I'm not joking!
You could use MESS if you really want those latter two features in the meantime, if you are 'not joking'.
Interesting project!
I tried it today, and it works well for me on the main PC.
On a slower system (Intel Atom) it puts out sound (clicks and pops) most of the time, regardless of the settings.
--> ISA Soundcard Overview // Doom MBF 2.04 // SetMul
wrote:2. Shuttle HOT-433 with Award BIOS machine
Perhaps if you need this feature that badly, you could explain what possible use it might have?
wrote:wrote:2. Shuttle HOT-433 with Award BIOS machine
Perhaps if you need this feature that badly, you could explain what possible use it might have?
I made that request because the HOT-433 AMI BIOS doesn't fully work in PCem.
wrote:wrote:wrote:2. Shuttle HOT-433 with Award BIOS machine
Perhaps if you need this feature that badly, you could explain what possible use it might have?
I made that request because the HOT-433 AMI BIOS doesn't fully work in PCem.
Yes? And why should anyone care?
For that matter, what do you mean by "doesn't fully work"?
If I may reply to that question too:
The old HOT-433 BIOS v1.3 is working fine for me: 433AIP16.ROM. With the text mode windowed BIOS user interface.
The later BIOS'es have reboot issues, or do not boot at all. Though at the time of trying them, I was also messing around with the proper settings for the 40MB harddisk image.
--> ISA Soundcard Overview // Doom MBF 2.04 // SetMul
wrote:If I may reply to that question too:
The old HOT-433 BIOS v1.3 is working fine for me: 433AIP16.ROM. With the text mode windowed BIOS user interface.The later BIOS'es have reboot issues, or do not boot at all.
I have no documentation for the UMC chipset used by the HOT-433, so was forced to guess the shadow BIOS controls (which are required by BIOSes at this point in time). I didn't guess entirely right, which is why very few BIOSes for this board work. This isn't going to change unless someone comes up with a UM8881F datasheet, and to be honest it's not something I'm that concerned about - the AMI BIOS on my site largely works, and there are other boards that can be emulated.
I initially thought it would give the benefit of PCI cards, but all the graphics ROMS work with the 486 VLB systems too.
I was also messing with the 286, which is new for me.
Unfortunately EMM386 does not work and EMM286 takes too much base mem, but there is no other way to get EMS. Maybe an emulated EMS board would fit here.
Regarding the CPU: playing wolfenstein 3D with a SX CPU works fine. But selecting a FPU equipped DX system gives lower framerates, and slightly warped walls.
Not complaining, just having 'fun' with PCem. 😀
--> ISA Soundcard Overview // Doom MBF 2.04 // SetMul
Heh, you can load just about any random AMIBIOS and it will boot on this emulator.
To the developer, if you want to emulate the SiS 471 ISA/VESA and 496/497 PCI/ISA/VESA boards, the datasheets are here: SiS 486 Chipset Datasheets
I'm glad the Dogz bug is fixed, though now they tend to flip every few frames or so.
Also Microsoft Plus! installs properly now.
You know what would be awesome? A Linux port. Specifically because my fastest computer only runs Linux.
I said something too soon, Plus! tab in Display Properties cause a rundll32 crash.
Got the Windows Sound System working with another card's drivers 😀
wrote:I said something too soon, Plus! tab in Display Properties cause a rundll32 crash.
Fixed!
So just out of curiosity, what causes a problem like that?
Usually for issues in Windows it's bugs with page fault handling, when CPU state is not restored correctly following the faulting instruction. In this case chance stack alignment meant that a far CALL instruction faulted between pushing the return selector and address, and the stack pointer was not restored correctly.
