It looks like an SMD capacitor array, possibly 0612 size, probably in the 10nF-100nF range. You could remove a similar one from the board and measure it, but 100nF 16-25V will most likely be OK. The remains of this cap array could be shorting one of the GPU power rails, you could try removing it …

I have never seen a post office in a grocery store or a pharmacy over here. The lockdown is in effect for at least 1 more month, in the meantime I'll see if I can find something cheaper. If not, I'll send it via regular post when I'm allowed to.

Glad to hear you fixed it! Drag soldering works quite well with the right flux, right tip and the correct temperature. There are many types of flux out there, some are really good, some are crap. Solder wick is really useful if the drag soldering approach doesn't work, same as the flux, some brands …

No problem, I hope it works fine after soldering it. It's not really a good idea to use hot air on such old ICs , the epoxy package had ~20 years to absorb moisture. Just use a small tip on your soldering tool, with a little solder on it (preferably the type with lead in it), and apply flux on the …

I know, it's ridiculous. 60E is the cheapest option I found (it's basically 60E for a letter) others are even more expensive (DHL or TNT was ~80 if I recall correctly) . The postal service is still up, but we're currently under quarantine and not allowed to leave our homes except for emergencies or …

Try gently poking all the TMU pins and see if any of them move.

You're right, the 3 tranistor looking components can't cause this error. Those are just some diodes on the RGB lines, the switching is done by the PI5C3384Q or equivalent IC close to those diodes. FB* are ferrite beads, all of them should read as a short. I can send you a NOS FBI but shipping is …

The photo you uploaded is fine. The marked pin on the top-left side of the FBI is not connected, you can ignore it. The marked pin on the bottom-left side of the FBI goes to the 3.5V rail. You could route a piece of wire from the FBI pin through the hole near C7 and connect it to C68 on the bottom …

Sorry to hear that. I don't have a datasheet for the FBI chip. Can you post a photo of the affected pins/pads?

The card was made by Creative. I use small diameter needle probes, no scraping is required. I've never seen a QS5244 or equivalent IC go bad. But it is unlikely that IC is related to your issue. It buffers HSYNC/VSYNC for the SLI connector and a couple of other signals (FBI to TMU if I recall …

The via just looked a bit weird. I managed to insert a small piece of wire in the via and soldered it at both ends. You don't need to remove the soldermask for probing the VIAs, just push slightly harder with the probes and it should be OK. If the outer layers are connected, I wouldn't worry about …

I doubt changing those caps to 100nF will do anything. As for the cap between M66EN and GND, if the card doesn't connect M66EN to GND for 33MHz operation and leaves it floating/pulls it up for 66MHz operation, the cap is added there to make the pin act as a signal GND. IIRC this cap should be 10nF …

On V2 cards, AS2830 regulator (or equivalent) should output 3.5V, so that's fine on your card. The RAMDAC should not output 135MHz back to the FBI, CLK1 should output 90MHz (when the cards isn't idle) and CLK0 varies depending on the selected video mode. In idle CLK1 should be ~44MHz, and CLK0 ~ …