Blocking the onboard with pci-probe-list fffbffff does not help?
This entry in the device list, vendor and device-ID, is just what i got for the 6200 before anything worked. 😕
I tried that a long time ago without any success, tis worth keeping in mind this is a Rev B G3 Beige Rage Pro onboard graphics, its not the same as your Rev C
but ill try it again soon just to be on the safe side!
indeed the Vendor and Device ID is what i expected with the SATA card its just a generic PC Card, I mentioned the numbers so people knew what to look for in my screen picture
Be prepared to setup the tty connection( Two Machine Mode ) to OF on the Beige, then I'm sure @joevt will have some insight into how to proceed.
sadly I dont think ill be able to do that at this point in time as I lack the required cables/adapters,
and also to a larger degree its very painful for me to cycle through computers as I have to drag them out of storage set them up and then put them away when the next machine needs testing
in my current setup I have only got space to have about 3 or 4 towers next to the test monitor, and even I have to shuffle them around the room depending on which im directly messing with
which again causes me a lot of pain due to my disabilities sadly
it really is very frustating PITA that I have to deal with sadly! and is why for example I was not able to contribute much with my G3 beige and 9600 during the 6200 saga as those where buried (under a 5500) and I need to shuffle a fair amount of machines to dig em out! all of which took me a good few weeks to do between moving a machine, recuperating and then moving the next one and so forth
and in general is why I am quite slow at doing things, I really have to pick and chose my battles/what to prioritise
so yeah I think someone else will have to pickup the torch on that front sadly
I missed this, that's not really a good sign.
I found with a PCI-E to PCI bridge on X86 host that some cards would not even allow the host to power on or post at all. Namely a PCI Radeon 9200 that was keyed PCI universal.
I suspect the card was not truly universal. As I recall the Beige G3 used either 5v or 3.3v PCI, but not both.
Tho if that were the issue here I don't think the system would power on at all, so there maybe some hope.
yeah I chucked the GT 120 (which I had to Hand from playing with my MacPro3,1) in there as sanity check, just to rule out any FCode shenanigans with the X1900 and 6600
I dont think its a voltage issue, the PCIe card is powered almost entire powered separately from what comes through the PCI slot, and the PCI to PCIe card does its own voltage/level shifting anyhow
but yeah as you say its quite strange and not good sign that even a completely non fCode GPU brings the system down, thats falling over right at the first hurdle there!
but at least we know the PCI to PCIe Bridge does work with at least 1 PCIe device, so hopefully the issues encountered are purely software/OF issues that can be overcome by the experts
