Hi, do these cards have 512 KB or more video memory?
That would allow for 640x480 256c and 800x600 256c.
If the cards have 256KB of RAM, the limits are:
640x400 256c, 800x600 16c, 1024x768 mono
Edit: I really recommend to check how much RAM the cards have, just so be sure.
UniVBE and SciTech Display Doctor require at least 512KB, otherwise they simply refuse to work.
Second, the ET-3000 is old. Really old! The ET-3000AX is from September 1987.
https://dosdays.co.uk/topics/retro_review_et3000_pt1.php
At this time, Windows 2.03 was still brand new.
There were ET-3000/4000 drivers being made for Windows 2, even. Toastytech site has them in a zip file.
So I doubt that Windows 95 is unaware of this VGA chip and that external drivers are being required.
It's the predecessor of the famous ET-4000AX, after all.
If the card has 512KB of RAM, it should do 800x600 256c just fine, I think.
But that's not guaranteed. Maybe your card has unpopulated sockets for a memory upgrade, not sure.
The versions for 8-Bit slots are limited to 256KB, often.
My father had an Trident 8900 series card in his 386DX-40 PC and Windows 95 did detect it via hardware wizard, the drivers were part of Windows.
If you still have doubts, please check their RAM expansion.
Also, Windows for Workgroups 3.11 has dedicated Super VGA drivers for 256 colours in 640x480, 800x600 and 1024x768.
They contain detection code for a dozen popular graphics chips of their time.
They are slow and unaccelerated, maybe, but they usually work.
They also try to use VBE BIOS if available (not the case by default on ISA cards).
So they might be good for trouble-shooting, to check if the VGA cards work.
Alternatively, it's possible to experiment with PCem/86Box and configure for them for ET-3000AX and 512KB or more and see if Windows 95/98 can use it.
That way, a hardware fault can be ruled out.
Edit: This source of information is interesting: https://pdos.csail.mit.edu/6.828/2016/reading … gadoc/GENOA.TXT
It has evidences that the Tseng ET-3000AX and UMC 85C408 (Genoa 6400) are related.
Memory locations: […]
Show full quote
Memory locations:
$C000:$37 2 bytes start of info table from start of ROM
Usually $C000:$B4
$C000:x 1 byte $77
$C000:x+1 1 byte Version.
$00 Genoa 6200 or 6300
$11 Genoa 6400 or 6600
$22 Genoa 6100
$33 Genoa 5100 or 5200 (Tseng 3000)
$55 Genoa 5300 or 5400 (Tseng 3000)
$C000:x+2 2 bytes $6699
ID Genoa Chip Set:
ad:=memw[$C000:$37];
if (mem[$C000:ad]=$77) and (memw[$C000:ad+2]=$6699) then
case mem[$C000:ad+1] of
$33,$55:TSENG 3000 chips with special modes;
$22:GENOA_6100
0:GENOA_6200 or GENOA_6300
$11:GENOA_6400 or GENOA_6600
end;
Note that this would also ID the UMC 85c408 as a GENOA 6400, so test for UMC
first
Edit: Or let me put it this way, the ET-3000 is so historically old that it had been known to developers in former East Germany (GDR) behind the iron curtain.
It had been known via an parts ID, "U80621".
https://de.wikipedia.org/wiki/MME_U80601
I'm not saying that it's a bad chip, rather contrary. It was more advanced than ET-4000AX.
It had zoom and panning that made it great for CAD and wordprocessing under DOS/DCP.
I'm merely trying to say it's not an unknown chip.
For the ET-3000/4000, there were AutoCAD drivers (ADI) and VESA TSRs shipped on driver diskette.
VGA BIOS modes were probably similar between both chips, so basic SVGA drivers for 800x600 16c work with either card.
"Time, it seems, doesn't flow. For some it's fast, for some it's slow.
In what to one race is no time at all, another race can rise and fall..." - The Minstrel
//My video channel//