So you are the author of the APU scanout fix driver on Github ? If you have a fix for the problem, I will certainly be interested in testing it. My system (4600G / B550M K) is currently disconnected but I shall find time to test it again once your purported fix is available.
That's certainly nice to hear. I have been working for the past few weeks attempting to get Windows 7 to play well with a Cezanne processor(5800U), so I would very much like to see, if you are the author of the scanout fix, that is
Not sure if it actually works. I have released a new version. It is only tested on 5500u and 5825u,and it seems that it may need patches for different ids. Not sure if it actually works. I have released a new version. It is only tested on 5500u and 5825u,and it seems that it may need patches for different ids. 4600g is another id Not sure if it actually works. I have released a new version. It is only tested on 5500u and 5825u,and it seems that it may need patches for different ids.
Different ids as in different CPUs and/or different integrated graphics hardware? If so this will be troublesome, the work to make this work with all possible CPUs will be difficult and overwhelming. I shall find time to test it later, and I would have a strong interest in seeing this work if I were to own a laptop with a suitable Ryzen CPU. But since I don't own such a laptop, if it doesn't work I don't really mind too much. I have made Windows 7 x64 and Windows 8.1 x64 work on my hardware with a supported discrete graphics card and I shall use that if it is really necessary.
I should clarify my previous wording. The current patch is not actually tied to a specific CPU or PCI device ID. The 5500U (164C) and 5825U (15E7) have different PCI IDs, but both are already using the same desktop/surface handling rule successfully. The patch identifies the relevant allocation/surface characteristics rather than simply checking the CPU model or device ID. So a different PCI ID does not necessarily mean that a new patch is required. The 4600G is simply untested at the moment. If it exhibits a different failure mode or uses a different display path, additional work may be needed, but it may also work with the existing rules.
I have tested your updated version on my B550M K / Ryzen 5 4600G running Windows 7 x64, and unfortunately the result was the same, either via the HDMI port or DisplayPort (using a DP-to-VGA adapter). After installing the driver and restarting, the screen just went black following the appearance of a white screen. But this time I could not even shutdown the system by pressing the power button and the keyboard was also unresponsive. I had to power off the system by holding down the power button.
If you can come up with a fix I am certainly looking forward to testing it. As I have said before, I have already gotten the system to run Windows 7 x64 with a GTX 750 Ti with the help of your ACPI fix (and Windows 8.1 x64 works directly). So I am in no hurry. Please take your time.
Based on what I could find(amateurishly) regarding Cezanne laptop CPUs...CSM does not seem to be an option that can be usable, at least, I can't figure it out. So does use CSM or otherwise on a UEFI-only system, such as with the 5850U or 5800U?
That should be useful for laptops which do not have CSM (at least not without flashing a modded BIOS), if it actually works, that is. For desktop systems like mine, I shall stick with CSM mode with Windows 7 / Windows 8.1 (and Windows 10). I don't see an updated version of the driver at the Github site. Is it still under testing right now?