I do have a 9250 PCI with 128mb (64bit). But I don't know if it's the same partner model. I can easily provide pictures since I have that system in maintenance mode currently.

I found various photos of the rear of a Sapphire Radeon 9250 PCI that seems to be the same as your card, such as here, here, and here.

But unfortunately none of them are detailed enough to make out what value the resistor on the right is. 🙁 All I could see is that it appears to have 3 digits, so not a zero-ohm resistor. But we can see that the missing part on the left is definitely a capacitor. I'd guess it's probably the same value as the one next to it.

Whatever that missing resistor is, I suspect it's part of the feedback network of the RT9202 switching regulator nearby. If it's the 'bottom' resistor of the feedback divider (the one that connects to ground on one side), that means the RT9202 will always be thinking the regulated output voltage is too high, and so try to bring it down, probably ultimately bringing it to the RT9202's feedback reference voltage of 0.8V, which won't be enough to operate the GPU (if that's what it's supplying).

You can check if that missing resistor is indeed the lower part of the feedback network by checking with a meter if one pad is connected to ground, and the other pad is connected to pin 6 ('FB') of the RT9202.

HwAoRrDk wrote on 2025-11-28, 23:08:

I found various photos of the rear of a Sapphire Radeon 9250 PCI that seems to be the same as your card, such as here, here, and […]
Show full quote

I found various photos of the rear of a Sapphire Radeon 9250 PCI that seems to be the same as your card, such as here, here, and here.

But unfortunately none of them are detailed enough to make out what value the resistor on the right is. 🙁 All I could see is that it appears to have 3 digits, so not a zero-ohm resistor. But we can see that the missing part on the left is definitely a capacitor. I'd guess it's probably the same value as the one next to it.

Whatever that missing resistor is, I suspect it's part of the feedback network of the RT9202 switching regulator nearby. If it's the 'bottom' resistor of the feedback divider (the one that connects to ground on one side), that means the RT9202 will always be thinking the regulated output voltage is too high, and so try to bring it down, probably ultimately bringing it to the RT9202's feedback reference voltage of 0.8V, which won't be enough to operate the GPU (if that's what it's supplying).

You can check if that missing resistor is indeed the lower part of the feedback network by checking with a meter if one pad is connected to ground, and the other pad is connected to pin 6 ('FB') of the RT9202.

The regular Sapphire Radeon 9200 Atlantis AGP has the same regulator section, but only the variant with the switching regulator. There is an alternative BOM with linear regulation.

https://www.ixbt.com/video2/images/ref/r9200-scan-back.jpg

Not easier to find photos though.
Though if you measure like HwAoRrDk suggested, you should be able to determine which resistor it is. Nominal VDDC voltage for the RV280 is 1.62V.

I don’t know if these are still required but here are some photos.