held wrote: ↑Tue Feb 15, 2022 10:43 am
archive.org isn't a reputable source
A strange notion. Archive is a repository of individual sources. Some might be less or more "reputable" but it mostly means more or less known. In any case we shouldn't dismiss them out of hand.
Not really, there are loads of incomplete, broken, badly patched, bad uploads there. Developing with those is an accident waiting to happen. Better to use originals or a verified source. However for an end user archive.org is like candy-land
Sometimes that web is the only source for original versions. I found a copy of prehistorik which works on 8088 86 because it does not have the crack intro (which for some reason requires 286).
akeley wrote: ↑Tue Feb 15, 2022 1:41 pm
A strange notion. Archive is a repository of individual sources. Some might be less or more "reputable" but it mostly means more or less known. In any case we shouldn't dismiss them out of hand.
Not really, there are loads of incomplete, broken, badly patched, bad uploads there. Developing with those is an accident waiting to happen. Better to use originals or a verified source. However for an end user archive.org is like candy-land
Sometimes that web is the only source for original versions. I found a copy of prehistorik which works on 8088 86 because it does not have the crack intro (which for some reason requires 286).
Its not, the original version has disk based copy protection Original versions either come from disk or from a verified source. But I leave those things up to preservationists themselves.
But if you are interested, you should read this thread on vogons.org. It will give you a good idea on how complex everything is.
Like I said, archive.org is a good source for end-users. But marking those as being original is not a valid claim unless you have something to verify against
Re: A Real PC XT Core Perhaps?
Posted: Sun Feb 20, 2022 1:41 pm
by akeley
held wrote: ↑Sun Feb 20, 2022 12:54 pm
Like I said, archive.org is a good source for end-users. But marking those as being original is not a valid claim unless you have something to verify against
I don't really want to derail the thread arguing with this, but it's rather important to repeat that archive.org is just a repository and the stuff on it is as good or as bad as the sources it comes from. It's simply illogical to use sweeping generalizations and dismiss it as a whole.
There are numerous quality uploads on this site, "verified" by "preservationists", if we must, and which you can safely use for development (eg the likes of 4am for Apple II or the collection I linked to in the previous post.)
held wrote: ↑Tue Feb 15, 2022 10:43 am
archive.org isn't a reputable source
A strange notion. Archive is a repository of individual sources. Some might be less or more "reputable" but it mostly means more or less known. In any case we shouldn't dismiss them out of hand.
Not really, there are loads of incomplete, broken, badly patched, bad uploads there. Developing with those is an accident waiting to happen. Better to use originals or a verified source. However for an end user archive.org is like candy-land
Pretty sure that was his point, that archive org itself is just a general source, but what is shared there, might be a verified source. Like the ReDump stuff that is on there, archive is often what people are told to use to get those.
But archive org overall is not trying to do any verification and such, that is up to the individual uploaders.
When it comes to cracks and such, I would also add that sometimes the modified versions of games might actually be better, because crackers back in those days sometimes ended up fixing bugs that were in the original game. At a time when game did not otherwise get patches and updates, that is interesting I think that such groups might end up fixing games that probably would just have such bugs in the game forever for that system.
held wrote: ↑Tue Feb 15, 2022 5:06 pm
Not really, there are loads of incomplete, broken, badly patched, bad uploads there. Developing with those is an accident waiting to happen. Better to use originals or a verified source. However for an end user archive.org is like candy-land
Sometimes that web is the only source for original versions. I found a copy of prehistorik which works on 8088 86 because it does not have the crack intro (which for some reason requires 286).
Its not, the original version has disk based copy protection Original versions either come from disk or from a verified source. But I leave those things up to preservationists themselves.
Why does that change anything? Its not that big a problem these days to grab the copy protection and such when making disk images. Several ways of doing that, and even quite cheap ways now. Used to easily cost 100 dollars to get the hardware needed, but now it can be done for... 10 dollars I think? But then you might have to do some soldering yourself.
Re: A Real PC XT Core Perhaps?
Posted: Fri Apr 01, 2022 11:49 am
by held
Caldor wrote: ↑Fri Apr 01, 2022 7:04 am
Pretty sure that was his point, that archive org itself is just a general source, but what is shared there, might be a verified source. Like the ReDump stuff that is on there, archive is often what people are told to use to get those.
But archive org overall is not trying to do any verification and such, that is up to the individual uploaders.
It was
Caldor wrote: ↑Fri Apr 01, 2022 7:04 am
When it comes to cracks and such, I would also add that sometimes the modified versions of games might actually be better, because crackers back in those days sometimes ended up fixing bugs that were in the original game. At a time when game did not otherwise get patches and updates, that is interesting I think that such groups might end up fixing games that probably would just have such bugs in the game forever for that system.
Correct, copy protection can introduce problems that are not present in a virgin copy. But so did bad cracks. Companies had official distributions BBS's, so you could if they were available, you had a modem, and knew how to use it.
Caldor wrote: ↑Fri Apr 01, 2022 7:07 am
Why does that change anything? Its not that big a problem these days to grab the copy protection and such when making disk images. Several ways of doing that, and even quite cheap ways now. Used to easily cost 100 dollars to get the hardware needed, but now it can be done for... 10 dollars I think? But then you might have to do some soldering yourself.
Well considering the state of the AO486 core adding more complexity would be a real winner But I would like to see transcopy support when it has matured.
Re: A Real PC XT Core Perhaps?
Posted: Fri Apr 01, 2022 11:43 pm
by Caldor
When it comes to the AO486 core, some say that it is quite close to full capacity, and this is after the core having been optimized several times by Sorg and others. I have seen it said several times that x86 is very complex compared to other CPUs in consoles and other computers. But I still suspect much can be gained if a new PC core was started from scratch, maybe based on PCem or 86Box instead of the current codebase which was another emulator and it started out with some auto-conversion.
Either way, it will be a lot of work to do to make a better core. Simpler cores should be possible though, or just optimized AO486 to be more compatible.
Re: A Real PC XT Core Perhaps?
Posted: Thu Apr 07, 2022 4:38 am
by spark2k06
PCXT CGA core update for ZXUno:
ZXUno PCXT CGA (Release 2)
Fix in the instruction cache that fixes graphical glitches in several Dinamic games, like Army Moves, Freddy Hardest and Capitan Trueno, among others. Thanks gyurco for the contribution.
Glad to see that this core is moving forward ! Thank you @spark2k06
Re: A Real PC XT Core Perhaps?
Posted: Fri Apr 08, 2022 6:51 pm
by held
Caldor wrote: ↑Fri Apr 01, 2022 11:43 pm
When it comes to the AO486 core, some say that it is quite close to full capacity, and this is after the core having been optimized several times by Sorg and others. I have seen it said several times that x86 is very complex compared to other CPUs in consoles and other computers. But I still suspect much can be gained if a new PC core was started from scratch, maybe based on PCem or 86Box instead of the current codebase which was another emulator and it started out with some auto-conversion.
Either way, it will be a lot of work to do to make a better core. Simpler cores should be possible though, or just optimized AO486 to be more compatible.
A do over sounds nice, but like you said is a lot of work. And on top of that I think the problems will reoccur in the new project if not solved in the old one.
But personally I would split the project in 2 parts: One for gaming eg: like DOSBOX and one for running a computer eg: like PCE/PCEM/86Box. Maybe they can cut some of the FPGA load that way.
spark2k06 wrote: ↑Thu Apr 07, 2022 4:38 am
PCXT CGA core update for ZXUno:
ZXUno PCXT CGA (Release 2)
Fix in the instruction cache that fixes graphical glitches in several Dinamic games, like Army Moves, Freddy Hardest and Capitan Trueno, among others. Thanks gyurco for the contribution.
- Reworking of the CGA module to accept 16 bits accesses to the CGA registers, to be compatible with Next186. They are OUT DX, AX type accesses. Now the PAKU PAKU game works correctly. Thanks gyurco for the ideas about the special text mode used by the game.
* Atari protocol joystick support, mapped to numeric keypad
ZXUno PCXT CGA (Release 5)
* CTRL + ALT + DEL -> This is now a core-managed reset if the ZXUno cores menu offset is set via OUT 99h, COREnn. Otherwise, it will still be managed from the BIOS.
* JTOPL2 module update
Thanks @mcleod-ideafix for the ZXUno multiboot module and for giving me the idea of using an I/O port to set the core associated with the reboot.
With this new entry in the AUTOEXEC.BAT, the core will be instructed to reboot the core located at position 36 in the ZXUno cores menu when CTRL + ALT + DEL is pressed. If not prompted, CTRL + ALT + DEL will still be managed from the BIOS, and some games that have a hooked keyboard will not respond to this combo:
image_2022-04-17_07-09-52.jpg (31.07 KiB) Viewed 4724 times
The XTCGA36.DBG file is available in the HDD folder of the repository:
image_2022-04-17_07-10-51.jpg (33.14 KiB) Viewed 4724 times
If the PCXT core is located somewhere other than position 36, simply update it with the corresponding value in hexadecimal (here 0x24):
image_2022-04-17_07-12-18.jpg (5 KiB) Viewed 4724 times
Re: A Real PC XT Core Perhaps?
Posted: Fri Apr 22, 2022 1:44 am
by pgimeno
FWIW I've had that problem with trying to double scanlines via BIOS. Paradoxically, when PC emulators became a thing, scan doubling via writing directly to ports was more compatible than via BIOS.
Re: A Real PC XT Core Perhaps?
Posted: Fri Apr 22, 2022 2:56 am
by spark2k06
Thank you. The problem in this case is that before the use of CTRL + ALT + DEL was not intercepted in any way by the core itself, so its handling was exclusive to the BIOS, as it would be in any real PCXT. It is in this scenario that certain games can cause it not to work, forcing the user to physically RESET. This solution prevents this circumstance from occurring, forcing a RESET every time.
In a MiSTer port this would never be a problem because we would have access to the menu to force a RESET, but in a ZXUno this is not the case, and therefore it is an interesting feature, even more so when the user does not have a physical reset button at his disposal, which, on the other hand, even if he had one, would reset him to the Spectrum core by default, unless he had the PCXT core by default to boot.
Re: A Real PC XT Core Perhaps?
Posted: Sat Apr 23, 2022 10:21 am
by spark2k06
Tired of dealing with Next186's problems:
Re: A Real PC XT Core Perhaps?
Posted: Sat Apr 23, 2022 2:12 pm
by dmckean
spark2k06 wrote: ↑Sat Apr 23, 2022 10:21 am
Tired of dealing with Next186's problems:
Original versions either come from disk or from a verified source. But I leave those things up to preservationists themselves.
But I would like to see transcopy support when it has matured.