Public Discussions > General Discussion

Retro Machine - DOS VDM - OpenOffice

<< < (2/4) > >>

Doug Clark:
Andrew,

In my experience eCS, and earlier versions of AOS, run windows and dos applications better than 5.1 on some machines.  Depends on the machine.  I am running AOS v 5.1 on three different machines (Ryzen R5/MSI motherboard, Thinkpad T530 and HP EliteDesk 800 G4).  The Thinkpad runs Win-OS/2 and Dos programs perfectly on AOS v 5.1 but would not run Win-OS/2 with earlier version of AOS.  The EliteDesk runs Win-OS/2 but in full screen mode only.  The Ryzen appears to run all Win-OS/2 programs in full screen, but some programs will freeze the machine in seamless mode - QuattroPro for Windows being the worst.  I also have problems with ReportSmith - and sometimes FrameMaker v 5.1 will freeze after awhile in seamless.  But again they all appear to run fine in full screen - many programs run fine in seamless mode.

I think it is both hardware and software.  I had a couple of earlier AMD  (FX4350 socket AM3+/ASUS motherboard) machines, one with eCS installed and one with AOS 5.02 installed.  Windows and Dos programs ran fine in on both machines.  When I upgraded the AMD AOS from v 5.02 to v 5.06 I started to have problems with Win-OS/2.

To test the theory I did a fresh install of AOS v 5.02 on a new hard drive on the AMD machine and transferred it to the Ryzen machine. (AOS v 5.02 will not install in the Ryzen machine). To get it to work I had to switch to a serial mouse and PS/2 keyboard because USB would not work and the network card stopped working - but the problematic windows programs (like Quattro Pro for Windows) ran fine.  So I think it is both the version of AOS and the hardware it is running on.

To run Win-OS/2 programs on the Thinkpad (before AOS v 5.1 came out) I ran my windows programs in a WIndows XP virtual machine running in VBOX on AOS.  That worked pretty good for me - although there is a trick to getting vbox to start without a hitch.

Andrew Stephenson:
Doug:... Many thanks for the discussion on the variable results you have had, running versions of OS/2 on sundry machines. Reading your account, I found myself thinking of that AOS booting-up oddity, whereby it pays to press Alt+F2 to ensure a small delay at a critical moment. Is it possible (Wild Guess Dept.) that we have here (sorry, Half-A**ed Attempt At Being Technical follows) timing clashes in a multi-processor m/c? If sequences of code got handled at different rates and the software was not written to cope, one might get problems?

BTW, I have been looking at Wikipedia's article on LibreOffice and wondered if anyone has investigated an OS/2 port. It's a HUGE article but the general notes point to LO being a lot more mature than its ancestor, OpenOffice (whose OS/2 port I bought and still have not installed -- mainly because of distractions).

Dave Yeo:
You might want to test with only one cpu enabled. Add /MAXCPU=1 to the PSD=ACPI.PSD line in config.sys.
As for LibreOffice, one of the first things they did when they forked from OpenOffice was remove the OS/2 stuff and since then so much development has happened that re-adding it would be very hard. Even building OpenOffice is next to impossible without the right environment which I understand only one person has.

Pete:
Hi andrew

I found this little trick to help with boot timing problems in the past:-

Open Editor (e.exe) and Save As [BootDrive]H:\OS2\BOOT\ALTF2ON.$$$- Yes, an empty "dummy" file.

This reports every driver that it has loaded during boot which seems to help with timing issues. Also useful if boot gets stuck at some point as you can see how far booting got before failing.


Regards

Pete

Roderick Klein:

--- Quote from: Andrew Stephenson on September 02, 2024, 04:32:16 pm ---
--- Quote from: JTA on September 01, 2024, 05:58:07 pm ---If one is considering multiboot solutions, then this could be a case where an AToF machine really shines for WordStar and other old non-OS/2 apps ... [...]

--- End quote ---
Thanks for the thoughts, though AToF is something I have yet to investigate.

The spate of bugs afflicting my DOS and Win-3.1, summarised earlier, started with AOS. The eCS-2 sessions worked beautifully, so maybe our SMOCs (Secret Masters Of Computing) tweaked something that needs to be un-tweaked. Or I could dig out a copy of eCS-2 (hmm, do I still have it?) and install that in its own disk partition, in the hope it can talk to modern-format files in the AOS-5.1 (&c) partitions -- or it can be HPFS-formatted and the AOS partitions could talk to it.

--- End quote ---

Sorry the conclusion that 3.1 and DOS is more stable under eCS we need waaay more data. Just be extremely careful with such conclusions!

First of all most all of the WIN/OS2 code in drivers in the config.sys and in the kernel is the same as in ArcaOS as in eCS.
This means the files in \MDOS\WINOS2. But also the VDD drivers. All of that code from MCP 2 fixpak 5 both OS'es are based on.

But it gets more complicated then that. When it comes to the base OS a LOT of code has been modified since eCS. This includes ACPI, Panorama, file system drivers such as JFS.
This can change timing in a way where something might have worked on an eCS system by accident.  Config.sys load orders can sometimes change things.
While this description is not 100% accurate I am just trying to point out there are waaay more moving parts to an OS then just some WIN/OS2 settings.

If you have an issue with a program you can reproduce on an ArcaOS 5.1.x system open a ticket at Arca Noae and see if they can look into this.
Always attach a testlog generic.

Roderick Klein


Navigation

[0] Message Index

[#] Next page

[*] Previous page

Go to full version