That TE really looks suspicious.

That is definitely weird. If I were you I'd just carefully peal that QC label off with a thin\sharp plastic thing (guitar pick, etc.) and move it to get a better look at the chip. I don't think any collectors would care about the QC sticker remaining in place as much as knowing what is printed under it in this case.

This card on DOS Days has a very similar layout, with some differences in components:
https://www.dosdays.co.uk/topics/retro_review_et4000_pt1.php

The MSI sticker on the back looks suspicious to me. It is extremely clean and free of scratches, and yet looks like it was cut\torn from a sheet haphazardly. I have never seen that logo before either. MSI was certainly around at this time, but I've never seen a video card with MSI markings on it from before the late 90s.

Is there an FCC ID printed on the card anywhere?

Also, what does the V-BIOS message say when you start up the PC? Dumping the BIOS could reveal something too.

Everything else about the card itself seems totally normal. The 80ns RAM looks right, even the plastic mold for the "TE4000" chip looks identical to any other ET4000. The other numbers printed on it are also similar to other chips.

Would be such a funny thing to have a typo on, but I guess it's possible. I actually remember something about a typo on a piece of hardware that I've seen in recent years... may have been a Voodoo 3. So, it can happen, but usually not on the printing that goes directly onto a chip package.

EDIT: Unless this is incorrect, MSI's logo was the same from 1986 to 2009 and it clearly had that "micro star" star shape on it. It was not red.
https://1000logos.net/msi-logo/

Considering the VGA BIOS is licensed by AcuMOS (which made chips like AVGA1 and AVGA2 and was later acquired by Cirrus Logic), could it be a relabeled AcuMOS/CL chip?

The VGA-Sync ROM is licensed by TsengLabs though... really weird.

TheMobRules wrote on 2025-12-02, 22:04:

Considering the VGA BIOS is licensed by AcuMOS (which made chips like AVGA1 and AVGA2 and was later acquired by Cirrus Logic), could it be a relabeled AcuMOS/CL chip?

No.
Acumos/Cirrus chips already had integrated RAMDAC...

The VGA-Sync ROM is licensed by TsengLabs though... really weird.

...and Sync ROM was a Tseng-only thing.

Anyway, it would be a good idea to look at the software side - dump the BIOS, and see how various sysinfo programs detect that card.

i do think its et4000 even though the te4000 printing doesn't look right, it has 144 pins.

The PCB seems to be identical to card 843 at https://www.vgamuseum.info/index.php/componen … -tseng-et4000ax , which has a chip labelled ET4000 as usual. Your card has an extra "MADE IN TAIWAN" print on it. Maybe this is a clone of that card using a factory reject ET4000 chip which someone printed TE4000 in it? Using partially damaged chips was quite common for RAM chips on cheap taiwanese stuff at that time, so I some manufacturer they might also have uesed other "not 100% OK" chips.

Very interesting to hear all of your insights! I took the sticker off and the print does indeed read TE4000 and also TC6058AF, which seems more correct. Unfortunately I don't have a suitable system to put the card in at the moment, not ready to go at least, but I might be able to put something together over the weekend. As for other clues in the meantime, there don't seem to be any. I can't find any other relevant stickers or printing on the backside except for 91 36 which I'm guessing might refer to year and week of manufacturing.

najls wrote on 2025-12-03, 21:06:

Very interesting to hear all of your insights! I took the sticker off and the print does indeed read TE4000 and also TC6058AF, which seems more correct. Unfortunately I don't have a suitable system to put the card in at the moment, not ready to go at least, but I might be able to put something together over the weekend. As for other clues in the meantime, there don't seem to be any. I can't find any other relevant stickers or printing on the backside except for 91 36 which I'm guessing might refer to year and week of manufacturing.

The attachment IMG_1266.jpg is no longer available

The attachment IMG_1267.jpg is no longer available

What a weird thing. If you compare the printing to other ET4000 chips, it is written exactly the same way. The "T" in ET\TE is a bit smaller with a shorter top than the T in the line below it... this is exactly how all other ET4000s are printed.

I'm not going to pretend to know how printing like this was done at the factory, but it seems strange that it's even possible to have any roll out of the factory like that and then make it onto a card and THEN make it to consumers.

najls wrote on 2025-12-03, 21:06:

I can't find any other relevant stickers or printing on the backside except for 91 36 which I'm guessing might refer to year and week of manufacturing.

That's likely. The date code on the RAMDAC is 9133, which is close enough to 9136, so that one was a quite fresh delivery. Most other chips are around 9116, which is "interesting". A factory with high output wouldn't have chips 20 weeks in storage, I suspect. What's really surprising is the RAM datecode, though. The RAMs are labelled 9032, so they were a year in old when the card was produced. RAM is expensive (just today, but also back in the day), and you need 8 chips per card. You just do not put RAM into stock for a whole year! This means the RAM chips are either mislabeled (if that's the case, likely intentionally), or they bought overstock RAM second-hand. This type of RAM was not obsolete by 1991, so it's not like you could get these chips new old stock for next to nothing. That would have been more like 1995 when these chips were no longer used in cosumer PC hardware.

Once again, very interesting to hear your thoughts. Only update I have is that Windows 98 recognizes the card as a plain old ET4000.

If I was to attempt dumping the VBIOS, or try to extract any other information for that matter, what would be good tools to use?

You can try running the SNOOP tool on it: SNOOP.EXE: (S)VGA adapter info tool and CRT Terminator config utility

and then compare against ET4000AX cards here: https://oummg.com/manual/adapters.html#tseng . In that table there is SNOOP column that has screenshots of running SNOOP on the card. Also running "SNOOP strings" dumps BIOS strings, that you could compare. There is a BIOS hash checksum too for comparison purposes.

najls wrote on 2025-12-07, 20:13:

If I was to attempt dumping the VBIOS, or try to extract any other information for that matter, what would be good tools to use?

NSSI can dump the VGA BIOS.

DEBUG also works - Re: Diamond Speedstar 64 ISA BIOS rom needed

Just wanted to say that in the back of my mind, some part of me wanted to change a driver INF file so that it lists the device name as "Tseng Labs TE4000AX" and then upload it here for you to try without saying anything about it... 🤣

(No, I wouldn't actually... but the mystery of this thing which is probably just a simple typo just hits me funny. 😅 )

Ozzuneoj wrote on 2025-12-08, 00:43:

Just wanted to say that in the back of my mind, some part of me wanted to change a driver INF file so that it lists the device name as "Tseng Labs TE4000AX" and then upload it here for you to try without saying anything about it... 🤣

😆

Very glad you didn't . Would probably have resulted in me trying and failing at making a Youtube documentary.

Anyway, this is what I got from SNOOP:

Grzyb wrote on 2025-12-08, 00:06:

DEBUG also works - Re: Diamond Speedstar 64 ISA BIOS rom needed

I dumped the BIOS as well if anyone's interested, but there really doesn't seem to be anything special about this card at all except for a misprint.