It looks mechanically cracked off. Measure what both pads connect to, if its 3.3/5 supply and ground then any 100nF will do.
Does the card not work without it? Missing cap rarely results in dead board, and since it looks mechanical I would take a closer look at all the solder joints on big chips.

The card does not output any video/sync signal - not sure about the sound. It was working a few months ago but when I tried to test it recently it did not work.

Testing it on a P3 based machine that has onboard video and a set of LEDS for POST status shows it is not working but also not causing issues when the onboard video is enabled, but when switching to PCI video it fails POST with the lights indicating PCI bus error.

This is the same machine it was working on before and another PCI video card works fine.

Hmmm if it did blow would be a bit odd being a SMT type but if did might have taken out something with it OR the opposite: something died and blew that cap. Would expect the Tants to short or blow first if a supply/age issue.
I would check those Tants for a short before moving forward. Sorry have nothing else to add.

Slight update - tried replacing the missing/blown capacitor with a 100nF and still nothing. When plugged into PCI it has power all over the board but nothing coming out. The yellow tantalum is testing OK.

It is very odd - next step is taking the bios chip out and reading it and hopefully finding the BIOS image online to try in a new chip.

The board is behaving like it has no power (chips don't warm up, nothing comes from the outputs) but you can read the voltages etc all over it.

Further update - after a bit of fiddling around with the BIOS chip to no avail it was noticed that these two chips LF353 feel different - one was warming up and the other was cold. Without a schematic it’s difficult to know what these do etc.

nothing that would prevent the card from at least initializing as a VGA, judging by location they are either used for D/A or as headphone opamp
Did you measure U18 output on leg 5 (bottom of the card)? do you still have a bodge element soldered to R320? Did you try reseating video bios chip? can you dump it to check if its all there? What about what looks like a PTC in series with L10 in upper left corner?

Hi mate - the bodge hardware soldered to R320 is factory issue! We have removed the bios chip, dumped it and checked it with no issue. Tried installing a different BIOS (for a 2000 series card, as couldn't find a 3000 on line) but that did not change anything. PTC is testing OK and the fact the board has power makes me think the PTC is OK as well.

Have not measured U18 / leg 5 output, will give it a go today.

thanks for your help/advice 😀

U18 is whats powering the low voltage parts of the card, Im guessing 3.3V
you said
>can read the voltages
>board has power
is the 3.3V clean (need scope to look at it while booting)? all L5 L6 L7 not blown?

>fails POST with the lights indicating PCI bus error

so not even VGA initialization but straight up PCI fails to enumerate. I still think that cap was damaged mechanically, I would poke at all NV1 chip legs.

OK some good and some bad with the card today. U18 l5 was 5V and leg 4 3.6v from memory.

I put the card into a Pentium 233MMx running win98 alongside another video card - Windows detected the Edge 3D (well, it detected the presence of a PCI video card) and I was eventually able to load a driver, however the Diamond Edge self extracting exe driver program claimed there wasnt a diamond edge card on board (perhaps detecting the other card first). With the driver loaded Win98 then would hang whilst booting (unless in safe mode).

When I tried loading a bare NV1 driver it really didnt like it and dumped the boot during PCI config.

pictures attached.

why 3 graphic cards in same system? maybe cant be initialized because you are running out of resources?

There was only two - for some reason it loaded the S3 card driver when it loaded the Edge 3D driver as well. I ended up deleting that S3 card/driver but it made no difference.

I am wondering if Win98 is picking up the bios from the Edge 3D but there is something wrong with the NV1 hence no video output. I’ve found a pin out for the SGS Thompson version of the NV1 which labels the pins so when I get a chance I’ll try to see if I can pick up any signs of life with a multimeter.

Might also see if I can get an ISA card working on the motherboard and perhaps it won’t conflict with a card on the PCI bus.

Well with an ancient VGA card in as ISA slot it booted windows in safe mode but still would hang during normal Win98 startup.

I found some information on the SGS Thompson version of the the NV1 (The STG2000) which has pinouts of the chip as below

I tested with a multimeter the VDD leg and it was 3.3V, GND pins were 5V. I don't have an oscilloscope so I am quite limited in what is testable.

The full spec for the STG2000 is here;

The Diamond Edge cards are supposed to be very similar except that they use vram instead of dram.