Contribute
Register

Intel HD Graphics Framebuffer Edits - Desktop

Status
Not open for further replies.
If your profile is correct, no Intel graphics. Off topic.

H toleda, my profile was so outdated. Sorry

I´ve updated it and this is what I´m currently using:

Mobo: GA-H97N-WIFI
CPU: Intel Core i3-4360
Graphics: Intel® HD Graphics 4600
I would really appreciate your help, the short below video shows the issue. I´m having.
I included my config.plist from clover


It only happens when the TV I have connected through HDMI is turned off.
I´ve tried:
- Changing the HDMI cable
- Removing Inject Intel from Clover (causes show wrong info in about my mac). So I turned it back on.
- The issue only happens with iMac 14,1 or 14,2 since it should be the right model based on processor. When choosing MacPro 6,1 it does not give me that issue but when using the VDADecoderCheckerSierra with MacPro it does not give me the
Hardware acceleration is fully supported, so no 4K video properly playback.
- Tried Kexts for IntelGraphicsFixup.kext

It seems like the monitor is plugged in and out. And I think it might have to be something to Airplay as seen on the video happens randomly and when I select the Airplay icon above.
 

Attachments

  • config.plist
    6.4 KB · Views: 252
Post #1/Problem Reporting; attach all requested files

Hi Toleda, sorry for the delay on my response.

I was troubleshooting the issue first. And I found the kind of solution.

The issue only happens when the HDMI TV is turned off. Then system log starts sending errors and the monitor starts turning off and on (blinking) also cursos starts to be sluggish.

So the only solution I found was to turn off image on my TV but not turning it off completely.

If there is a fix for that issue that you know it would be great if not I think I can live with the TV image turned off while I'm not using it.

Thanks.
 
Hey guys,
So I installed Sierra 10.12.5 on my build (see signature).

Setup:
Video ram to 64 MB in Bios, no other PCIe Video card
QE/CI only works if I inject the proper ig-platform-id for Haswell Desktop (0x0d220003) and the fake ID 0x04128086
Bios is latest (had to add entry in UEFI Shell to get to the bootloader)
Clover is running the latest version
I don't use HDMI audio, but ALC892 optical
I get 2560x1600 using pixel clock patch through HDMI, otherwise, I only get 1920x1080
I use SwitchRes X to get 1920x1200 HiDPI@2560x1600 (actually I have to set up the scaled resolution to 3840x2476 to get a proper 1920x1200 HiDPI, otherwise I got two 10~12 pix horizontal bars above and under the screen... strange.)
The port in IOreg is 0x0 (while it should be 0x7) Framebuffer@2
no DSDT or SSDT

Issue after I've checked the above:
Display sleep wouldn't work properly and freeze the PC, so I was trying to fix that with a framebuffer patch.

I used the 100 FB patch to apply the DP2HDMI to the port 0x7
I got the port back in IOReg, the display goes to sleep and wakes up (no S3 sleep for now, but that's another topic), but, weird thing:
as soon as I rebooted with the port 0x7 DP2HDMI patch, the display changed to super sharp pixels.
I had to lower the sharpness of the display in its OSD to 0 and even though, it still looks different than before the patch (sharpness was set at 50 before, for as long as I remember), like display would receive a YPbPr signal now, while it received an RGB signal before, or similar.

So my question is, how can I make sure to get the same rendering as before (with sharpness at 50, better looking), but with the display sleep working properly.

Thanks for your insights

Attached files:
Config.plist
IOReg raw file
 

Attachments

  • config.plist
    5.9 KB · Views: 242
  • Haswell-Asrock.ioreg
    5 MB · Views: 162
Last edited:
Hello,
Display is connected through HDMI
Pixel clock patch allowed me to get a proper 2560x1600 resolution, without it, I would only get 1920x1080
(SwitchRes is only for HiDPI), the Pixel clock patch didn't changed the port 0x0 to 0x7 though.

I didn't test the FB patch without the pixel clock patch. I'll let you know. Thanks for the quick answer.

Edit: the issue was actually a wrongly identified display, with the EDID override, the display is fine, and works as intended in RGB with QE/CI enabled.
 
Last edited:
is there already a clover patch for broadwell integrated 5500 and high sierra?
 
Status
Not open for further replies.
Back
Top