I dont know much about that motherboard but I found this:

You can also occasionally get memory resource conflicts by installing a new AGP video card into an old AGP 1.0 motherboard. The video card will work properly until you install the display driver. Once you try to install the driver, a memory conflict shows up. The range of conflicting addresses varies from case to case. This problem is very unusual and when it happens it is rarely possible to fix it. I'm not sure exactly what causes the problem but apparently the motherboard and the video card are incompatible in some way which prevents Windows from properly assigning memory addresses to the video card.

Its odd it happens with so many cards. It was on this page: https://www.playtool.com/pages/agpcompat/agp. … versalagp30card

Mainboard is AGP 1.0 .... will only do 1x 2x @ 3.3v .....had the same issue here. only cards with AGP 1.0 /3.3v worked

Yep in other words: use a video card that is in spec with a motherboard. Never have tried an 8x AGP in 2 x motherboard, always worried it would fry something... Universal or not.
There not just the voltage but chipset, the signaling and bios support...added: AFAIK those universal are for 4x or 8x boards, not 2x

gf4mx and FX should definitely work without a hitch on 440BX, people even put them into socket7 systems. Something is wrong with the motherboard. 440BX 2/3 AGP divider means 133 FSB = 88 Mhz AGP, but GF4MX should handle that?
OP you can try setting FSB to 100MHz

Thanks for all the replies.

@Grem Five @giantclam @Horun: I get all of you that on paper they shouldn't probably work but before opening this post I searched in the forum, there are positive reports of users with same cards (or even Geforce 6xxx) and motherboard of mine. So maybe my motherboard is less tolerant or defektive like @rasz_pl said. By the way I tried the 100Mhz FSB with the slot-1 PIII (100x5).

Well at least I didn't fried anything since after I tested all the cards in the Athlon PC.

Disable Sideband Adressing via RivaTuner.

So I had to find the correct setting (it was under NVStrap page), set to "force off" but nothing changed. Still the blinking cursor at first boot. Tried also to disable fastwrite.

What I haven't said before it is that after the failed first boot, Win98 starts normally (aka not in safe mode), the card is correctly displayed in device manager but resolution and color are stuck to lowest settings. To try again I have to unistall the card from device manager.

Just tested also a TNT2 M64, still universal AGP but it is a 4x. Same issue.

Well, that's odd. Do you have any ATi Radeon AGP cards laying around to test?

I don't have any ATI AGP cards... I'm kinda a Nvidia guy 🤣

But I do have a Voodoo3 AGP. I tested and it works perfect as expected in this motherboard since it is an AGP 2x. Tried with OpenGL and D3D games.

Besides some other oem branded Matrox G100\200 and a Nvidia 6600GT (which I think there is really no point in trying it), these are all the AGP cards I have.

Voodoo3 is more of a fast PCI pretending to be AGP 😀
Reinstalling windows, even better if different sku like xp/2000 would also help diagnose if it isnt a software problem

mrfusion92 wrote on 2023-09-17, 21:43:

I have this rebranded Asus P3B-F rev 1.04 motherboard. It's currently running with a Coppermine PIII 933 Mhz mounted on a slotket.
<snip>
After I install the drivers what I get is just a blinking cursor after the Windows 98 boot screen.

You're running the AGP frequency at 89Mhz... What happens if you run your PIII at only 100Mhz FSB (700Mhz)?

I will try to test on my P3B-F, with a Slot1 P3 866Mhz.

UPDATE: Tested with my P3B-F 1.04 and MX440 8x, I could not replicate your issue. It works fine.

I agree with rasz, try Win2k/XP if you have it. I ran a Ti4200 on a P2B-F under Win2k without issue. The 980XGL is higher end but it's still the same family of cards/GPUs.
I don't think Win2k even needs a chipset driver for the 440BX, you can just install the nVidia driver and it should work.

Also keep the FSB at 100MHz and AGP at 2/3, so the AGP clock is at the correct spec 66MHz until you figure out what's going on.

I already tried a CPU with FSB @ 100 Mhz but I will try again and this time with 2000\XP.

mockingbird wrote on 2023-09-19, 03:50:

UPDATE: Tested with my P3B-F 1.04 and MX440 8x, I could not replicate your issue. It works fine.

Oh well... That doesn't look good for mine motherboard then, I'm starting really to think that there is a hardware issue somehow.

Just to add, my P3B-F also has no issues with Geforce 4 and Geforce FX series cards.

I'm kinda in a streak of bad luck this last couple of weeks. First my Abit VH6T broke on me without any visible damange, then I found out my Voodoo2 had bent pins in the FBI chip (but at least this I managed to fix it) then this partially faulty(?) P3B-F.

mrfusion92 wrote on 2023-09-19, 06:45:

I'm kinda in a streak of bad luck this last couple of weeks. First my Abit VH6T broke on me without any visible damange, then I found out my Voodoo2 had bent pins in the FBI chip (but at least this I managed to fix it) then this partially faulty(?) P3B-F.

Wait a moment -- don't be so quick to draw conclusions from posts by strangers on internet forums... In fact, I just tested with two different ti4200 8x cards, Windows starts up alright, the driver works with it, Quake3 launches with it, but when I max out the display settings (1024x768, 32 bit, trilinear, etc) in Quake3, it crashes upon re-entering the game after it applies the settings. This is Quake3 1.32c, and nVidia driver version 45.23.

I did not have this issue with my older, original (read: non-8x) ti4600 card when I tested it.

So maybe there is something to what you're saying.

I would have to test with a different BX motherboard to confirm my findings (in that it is unique to the P3B-F).

Reinstall Windows. FX 5600 works fine on my 440BX.

Check the position of the Vio jumper.
And hardware monitoring to.
AGP8X cards react very poorly to voltages greater than ~3.5V when running on AGP2X.
Most often they hang under load.