No, that was fine as far as I'm concerned. John did a good work, as far as I can tell. 😀
If I wasn't limited to a Raspbery Pi/ARM platform right now (PC broke), I would do some more testing here right now. 😢

Update. I tried to apply John's patch to my copy of Windows 3.0a's ega.drv, and it kinda works here.

Since I started out on the Raspberry Pi with that funny Okteta hex editor, -which I did not get along with-,
I had to move over to my old Presario running XVI32 in the middle of the process..
That's where some errors must have found their ways into the hex file.

Anyway, back to the EGA driver.
I got it to "work" first time in Windows Real-Mode (win /r), when my main group got deleted somehow.
After this, I also tried Standard-Mode (win /s) and it worked, too. Enhanced Mode (wina20.386 present in root),
partially worked, with the half of the screen being dark.

Update. John's patched EGA driver runs on Windows 3.1 in 286 Standard-Mode, 386 Standard-Mode and 386 Enchanted-Mode! 😀
I'll add some screenshots soon. Stay tooned.

Edit: Pictures added.
Edit: Just to be clear; I likely made a mistake during hex edit (I'm sorry !). 😢
The point of these screenshots is only to prove that it's possible to run Windows 3.x with such a patch in Enhanced and Standard Modes.

I have a Windows 3.1 EGA driver Remodeling to support 640x200 Succeeded.
The Windows 3.1 EGA driver works better than the Windows 3.0 driver.
This modified driver is designed to work from Windows 3.0 to Windows ME(4.9).

The screenshot looks like this.

My mod driver modified the resources to look like a CGA driver.
Therefore, the resource of the EGA driver is modified with the resource editor.

Charizard700 wrote on 2020-04-30, 17:39:

The screenshot looks like this.

Nice! I've recently restored a Commodore PC 10-II with the Plantronics colorplus card. How complex would it be to create a Windows 3.0 driver to supports its 16 color mode?

DigitalMan wrote on 2021-05-27, 21:03:

Charizard700 wrote on 2020-04-30, 17:39:

The screenshot looks like this.

Nice! I've recently restored a Commodore PC 10-II with the Plantronics colorplus card. How complex would it be to create a Windows 3.0 driver to supports its 16 color mode?

About as easy as a 16 color Tandy driver which hasn’t happened

rmay635703 wrote on 2021-05-28, 00:03:

DigitalMan wrote on 2021-05-27, 21:03:

Charizard700 wrote on 2020-04-30, 17:39:

The screenshot looks like this.

Nice! I've recently restored a Commodore PC 10-II with the Plantronics colorplus card. How complex would it be to create a Windows 3.0 driver to supports its 16 color mode?

About as easy as a 16 color Tandy driver which hasn’t happened

Not sure if that's possible, though.

Plantronics supports 4 colours only in 640x200 mode.
The 320x200 resolution was only used once by the MCGA driver in Windows 3.0 MME.
After it was never being heard of again.

"Problem" is, that Windows wants to have at least 20 colours for its own (as in - they are fixed/static).
http://www.fastgraph.com/help/windows_palette_manager.html

The normal 16 colours available in VGA mode 12h are somewhat borderline, already.

In short, 3.0 supports/knows three palette schemes only.
1-bit/monochrome, 4-bit/16c and 8-bit/256c

Not sure thus, how it would react to 2-bit/4c thus.
Maybe it goes monochrome, but lets the user do what ever he/she/they wants to do with these two extra colours (desktop colour etc).

It may behave a bit like GEM, perhaps.

Edit: A special case might be EGA perhaps (how fitting).
It had a 6-bit/64c in total, but only 16 colours could be picked and displayed at the same time.

Of course, changing the palette rapidly would make it look like as if there were more colours..

But Windows still wants its 20 static colours, so there's a conflict.

What I mean to say - not sure if Windows had another palette scheme burried deep inside.

EGA was very popular in early 16-Bit Windows.
This went so far, that early VGA drivers reduced VGA resolution for maximum compatibility with EGA applications.
Edit : Shortened. Already mentioned that one. 😅

There is a 4 color Tandy driver
Primary + black if memory serves

Trouble is just the act of rewriting a driver to fit that memory structure

Jo22 wrote on 2019-11-16, 09:31:

Update. John's patched EGA driver runs on Windows 3.1 in 286 Standard-Mode, 386 Standard-Mode and 386 Enchanted-Mode! :) I'll ad […]
Show full quote

Update. John's patched EGA driver runs on Windows 3.1 in 286 Standard-Mode, 386 Standard-Mode and 386 Enchanted-Mode! 😀
I'll add some screenshots soon. Stay tooned.

Edit: Pictures added.
Edit: Just to be clear; I likely made a mistake during hex edit (I'm sorry !). 😢
The point of these screenshots is only to prove that it's possible to run Windows 3.x with such a patch in Enhanced and Standard Modes.

It is just Standard Mode ol bean plain and simple.

http://gunkies.org/wiki/Windows_3.1
http://gunkies.org/wiki/Windows_3.0

Hi! I'm speaking under correction, but I think on 3.1, Standard-Mode is available in
both krnl286.exe and krnl386.exe.
Windows 3.0 only had one in krnl286.exe.

Jo22 wrote on 2021-05-30, 01:10:

Hi! I'm speaking under correction, but I think on 3.1, Standard-Mode is available in
both krnl286.exe and krnl386.exe.
Windows 3.0 only had one in krnl286.exe.

Doesn't matter what kernal it is old chap, the startup routine detects what processor it is running on. It is still just Standard Mode running the same as it would on a 286. In other words paging 64k chunks of memory. Reading the second link explains quite clearly.

Nice try though.

^Hi again. I'm sorry, I don't fully understand what you mean to say.
The Windows kernals (in Win 3.0) reside in kernal.exe, krnl286.exe and krnl386.exe.

On Windows 3.0, kernal is using Real-Mode, while krnl286 (+DOSX) is used for Standard-Mode. Krnl286 is using 16-Bit Protected-Mode.
By contrast, krnl386 is used for Enhanced-Mode. Krnl386 is using 32-Bit Peotected-Mode and V86.

In Windows 3.1 it's not as simple.
WIN.COM will check the CPU type and if it finds a 386 or higher, will always use krnl386:
Someone can delete krnl286.exe and WIN /S still works.
To execute the "true" Standard-Mode, on 386+, DOSX located in the SYSTEM directory must be executed. It will try to load krnl286.exe then.
However, it must be executed from within the Windows directory, if memory serves.

Edit: Long story short - I merely tested the patched EGA driver with both kernels for the sake of completeness.
This seemed reasonable to me,
because it was a Real-Mode compatible driver, originally.

Edit : I guess, someone could also say that Windows 3.1 has one Standard-Mode,
but two implementations of it. 😉
After all, a kernal is the very heart of a system.

Holy cow, I did not expect this request to go so far. The 286 I was running this on actually had a tantalum capacitor fail a few months ago and I haven't gotten around to even disassembling the machine to assess the damages.

This is absolutely amazing though, I always thought it would be cool to write my own drivers and someone here actually did. Amazing!

FAMICOMASTER wrote on 2021-06-09, 03:48:

Holy cow, I did not expect this request to go so far. The 286 I was running this on actually had a tantalum capacitor fail a few months ago and I haven't gotten around to even disassembling the machine to assess the damages.

This is absolutely amazing though, I always thought it would be cool to write my own drivers and someone here actually did. Amazing!

Some clever folk here alright 😀

Caluser2000 wrote on 2021-06-09, 03:56:

FAMICOMASTER wrote on 2021-06-09, 03:48:

Holy cow, I did not expect this request to go so far. The 286 I was running this on actually had a tantalum capacitor fail a few months ago and I haven't gotten around to even disassembling the machine to assess the damages.

This is absolutely amazing though, I always thought it would be cool to write my own drivers and someone here actually did. Amazing!

Some clever folk here alright 😀

For sure, it's why I keep coming back after all this time.

Hello all!
How complicated would be making a win2/3 Plantronics ColorPlus driver for 640x200 in 4 colors?
Any ideas if such driver already exists? I've spent a lot of time searching, but no results..
Thank you!

Now what I'd really like to see if an EGA 640x35o driver that can use the 64 colour palette...sort of like how the modified VGA driver could use all 256 colours by taking back the 20 from the GUI.