zyx030613
commented
3 years ago Did you solved this issue ? I meet the problem on M4800 with K1100M and 4k screen. The HDMI port is ok . But internal screen is always black or show strange...cty...
Open Xeon3D opened 4 years ago
zyx030613
commented
3 years ago Did you solved this issue ? I meet the problem on M4800 with K1100M and 4k screen. The HDMI port is ok . But internal screen is always black or show strange...cty...
Xeon3D
commented
3 years ago Sadly I have not. The OpenCore / Lilu guys are doing some progress with nvidia and DCB tables so I might revisit it in the future.
zyx030613
commented
2 years ago One year has passed, did you resloved M4800 with 4k internal display problem ? ^_^
Problem:
While the kexts load properly without any kind of injection, only external displays connected to HDMI or the DisplayPort work (with full resolution and acceleration).
IOReg shows 4 ports (00, 04, 05 and 07). If the external screen is connected to the HDMI port, it'll show up under
NVDA,Display-D@3 (port 07 00 00 00). If it's connected to the DisplayPort thru a DisplayPort to HDMI adapter, it'll show up underNVDA,Display-B@1 (port 04 00 00 00)Possible Causes / Solutions.
NVDA,Display-A@0 is connected to port 0 (
00 00 00 00) which under nvbios (envytools) doesn't have a DCB table, but there is a connector 0 which isn't connected to anything. macOS utility nvidiainfo does show an entry at DCB #0 which is not present in Linux). Maybe removing it from the table somehow, or force macOS not to "see" that port\connector, or even swap it to connector 8.A very strange NVCAP is shown on IOReg where all ports are in the primary and secondary display group (
05 00 00 00 00 00 **ff ff ff ff** 00 00 00 00 00 0e 00 00 00 00). Attempts to inject a more "proper" one by means of device injection have proven uneffective (01 00 OE 00).Proper display-cfg values have to be calculated accordingly to the DCB Table.