ASRock H310M-HDV - dual monitor problems

[Asural]

@brumas2025 Try setting device-id = <913E0000> and replacing WEG 1.7.0 with WEG 1.4.4.

Edited January 5 by Asural

[deeveedee]

device-id = <913E0000> (0x3e91) is the device-id for CPU i3-9100, so patching device-id with <913E0000> does not change anything for this hack.

 

As Baio77 states here, this reference indicates that CoffeeLake Refresh CPU (which includes i3-9100) does not need to be faked as 0x3e92 starting with macOS 10.14.4.  

 

However, later in the same document, it states that "For IGPU with dev.id 0x3E91 fake the id with 0x3E92 (device-id 923E0000)."

 

The document conflicts with itself, so our choices for this hack's device-id appear to be:

1. If we want to use device-id 0x3e91 (the device-id of i3-9100 which is in this hack), do not specify device-id in the PciRoot(0x0)/Pci(0x2,0x0) DeviceProperties
2. If we want to use device-id 0x3e92 (the device-id of Desktop CPU i5-9600), specify "device-id / data / <923e0000>" in the PciRoot(0x0)/Pci(0x2,0x0) DeviceProperties

Edited January 5 by deeveedee
fixed typo (changed e3 -> 3e)

[brumas2025]

11 hours ago, Asural said:

@brumas2025 Try setting device-id = <913E0000> and replacing WEG 1.7.0 with WEG 1.4.4.

I replaced it, tried to boot up - two black screens, and was able to connect via remote access.

Quote

 

iMac5-01-26.ioreg.zip

[deeveedee]

@brumas2025 Did you test with device-id = 0x3e92?

 

Also, please post the current config.plist that you are using as your testing baseline.  Note that DeviceProperty enable-hdmi20 is a WhateverGreen property that is not required for macOS 11+.

 

EDIT: I see that you did test device-id 0x3e92 here.  Sorry I missed that.

Edited January 5 by deeveedee

[brumas2025]

17 minutes ago, deeveedee said:

@brumas2025 Did you test with device-id = 0x3e92?

device-id = 0x3e92 end  with WEG 1.4.4. two black screens, and was able to connect via remote access.

config.plist

Edited January 5 by brumas2025

[deeveedee]

@brumas2025 Thank you.  sorry I missed that you had tested device-id 0x3e92 earlier in this thread.  Just to confirm, when you use device-id 0x3e92 with WEG 1.7.0, both displays "work" (still requiring you to turn one display off and back on).  Correct?

Edited January 5 by deeveedee

[brumas2025]

1 minute ago, deeveedee said:

@brumas2025

 

1 minute ago, deeveedee said:

@brumas2025 Thank you.  sorry I missed that you had tested device-id 0x3e92 earlier in this thread.  Just to confirm, when you use device-id 0x3e92 with WEG 1.7.0, both displays "work" (still requiring you to turn one display off and back on).  Correct?

Yes, it is

[deeveedee]

@brumas2025 When I first started hacking this HackMini8,1, audio over DP/HDMI was not supported by WhateverGreen/AppleALC when using graphics adapters, so audio over DP/HDMI needed to be disabled.  I'm not sure that this is affecting your hack, but we should test.  Please test the attached config.plist which adds DeviceProperty No-hda-gfx for PciRoot(0x0)/Pci(0x1f,0x3).  This is just a temporary test, so if the test does not resolve your graphics boot issues, the DeviceProperty No-hda-gfx should be removed from PciRoot(0x0)/Pci(0x1f,0x3).  Thank you.

 

config-H31-test-6-1.plist.zip

Edited January 5 by deeveedee

[brumas2025]

13 minutes ago, deeveedee said:

@brumas2025 This is just a temporary test, so if the test does not resolve your graphics boot issues, the DeviceProperty No-hda-gfx should be removed from PciRoot(0x0)/Pci(0x1f,0x3).  Thank you.

I tried to boot with your configuration, but the behavior has not changed - you need to turn off the power. Thank you for wanting to help!

[deeveedee]

@brumas2025 and @Asural What is the reason that ACPI patches now include PNLF and ALS0 (both are laptop patches for internal display brightness control and ambient light sensor)?

@brumas2025 Do you still have the testing patience to try the connector flags as I have stated here?

 

EDIT: @brumas2025 If you want to test different connector flags, test the attached config.plists which are as follows:

• config-H31-test-7-1.plist: framebuffer platform-id 0x3e9b0007 with framebuffer-conX-flags 0x03c7 (unchanged)
• config-H31-test-7-2.plist: framebuffer-conX-flags 0x0498
• config-H31-test-7-3.plist: framebuffer-conX-flags 0x0187
• config-H31-test-7-4.plist: framebuffer-conX-flags 0x01C7
• config-H31-test-7-5.plist: framebuffer-conX-flags 0x0098

 

If you test the attached config.plists, report any changes from your baseline behavior (two displays working after turning one display off and back on).  Thank you.

 

NOTE: Start testing with config-H31-test-7-1.plist.  If config-H31-test-7-1.plist does not work for any reason, stop testing and report the failure.  Do not test the other config.plists.

config-H31-test-7-1.plist.zip config-H31-test-7-2.plist.zip config-H31-test-7-3.plist.zip config-H31-test-7-4.plist.zip config-H31-test-7-5.plist.zip

Edited January 5 by deeveedee

[brumas2025]

2 hours ago, deeveedee said:

@brumas2025  Do not test the other config.plists.

 

• Test results :
• config-H31-test-7-1.plist:   = download is normal turning on/off the monitor power.
• config-H31-test-7-2.plist: framebuffer-conX-flags 0x0498 = two black screens and a cyclic reboot.
• config-H31-test-7-3.plist: framebuffer-conX-flags 0x0187 = download is normal, turning on/off the monitor power.
• config-H31-test-7-4.plist: framebuffer-conX-flags 0x01C7 = download is normal, turning on/off the monitor power.
• config-H31-test-7-5.plist: framebuffer-conX-flags 0x0098 = two black screens and a cyclic reboot.

Edited January 5 by brumas2025
maybe I spelled it wrong,

[deeveedee]

@brumas2025 Thanks for testing.  That is strange.  config-H31-test-7-1.plist should be the same as your previously working config.plist.  Could you please test the attached config.plist and confirm that it is "working?"  Thank you.

 

Please make sure you are using WhateverGreen.kext 1.7.0.

config-H31-test-7-0.plist.zip

[brumas2025]

8 minutes ago, deeveedee said:

@brumas2025 Thanks for testing.  That is strange.  config-H31-test-7-1.plist should be the same as your previously working config.plist.  Could you please test the attached config.plist and confirm that it is "working?"  Thank you.

 

Please make sure you are using WhateverGreen.kext 1.7.0.

config-H31-test-7-0.plist.zip 6.79 kB · 0 downloads

maybe I spelled it wrong, but 7-1, 7-3,7-4 - the download is normal, followed by enabling disabling...

[deeveedee]

@brumas2025 I'm confused by your test results.  Could you please restate your test results here? It seems to me that you are saying that all configurations produced two black screens, but now I think that I am misunderstanding.  Thank you.

 

EDIT: The reason that I am confused is that your previously "working" baseline was that you booted with two displays that had black horizontal lines.  Then you powered one display off and back on and the displays were fixed.  Now it appears you are saying that the displays boot to black screens.   Thank you for helping me to understand.

Edited January 5 by deeveedee

[brumas2025]

10 minutes ago, deeveedee said:

@brumas2025 I'm confused by your test results.  Could you please restate your test results here? It seems to me that you are saying that all configurations produced two black screens, but now I think that I am misunderstanding.  Thank you.

 

EDIT: The reason that I am confused is that your previously "working" baseline was that you booted with two displays that had black horizontal lines.  Then you powered one display off and back on and the displays were fixed.  Now it appears you are saying that the displays boot to black screens.   Thank you for helping me to understand.

The "working" baseline really lies in the fact that it is loading with two displays with black horizontal lines. Then I turned off and turned on one display again, and the displays were fixed. Sorry for confusing

Edited January 5 by brumas2025
translation problems..

[deeveedee]

No worries.  Now that you have edited the post, I understand.  The connector flags do affect the display behavior, but not in the way that I hoped.  

• Test results :
• config-H31-test-7-1.plist:   = download is normal turning on/off the monitor power.
• config-H31-test-7-2.plist: framebuffer-conX-flags 0x0498 = two black screens and a cyclic reboot.
• config-H31-test-7-3.plist: framebuffer-conX-flags 0x0187 = download is normal, turning on/off the monitor power.
• config-H31-test-7-4.plist: framebuffer-conX-flags 0x01C7 = download is normal, turning on/off the monitor power.
• config-H31-test-7-5.plist: framebuffer-conX-flags 0x0098 = two black screens and a cyclic reboot.

Based on your test results, framebuffer-conX-flags 0x3c7, 0x187 and 0x1c7 all behave the same, booting to two displays with horizontal black lines and requiring power-off/power-on to fix displays.

framebuffer-conX-flags 0x498 and 0x0098 fail.

 

Unfortunately, I don't think that there's anything helpful that we can learn from these test results.  Thank you for your continued patience with this tedious experimentation.

 

Note that in your config.plist, you can remove the ACPI > Add PNLF and ALS0 (they aren't applicable to your hack) and you can remove DeviceProperty enable-hdmi20 which is for an older version of macOS.

 

EDIT: You may want to review the framebuffer-con0-flags bit definitions here to see if you find any flag patterns that should be tested.

Edited January 5 by deeveedee

[brumas2025]

7 minutes ago, deeveedee said:

Note that in your config.plist, you can remove the ACPI > Add PNLF and ALS0 (they aren't applicable to your hack) and you can remove DeviceProperty enable-hdmi20 which is for an older version of macOS.    Yes, I will do that. Thank you.

 

EDIT: You may want to review the framebuffer-con0-flags bit definitions here to see if you find any flag patterns that should be tested. I think it's going to be difficult for me.

 

[deeveedee]

I'm out of ideas for now, so my suggestion would be for you to continue researching this topic if you have the patience to continue.

 

I searched and found a few suggestions that might help (and are included in the attached config.plists for you to test):

• Add property igfxfw=2 (which we had at one time as a boot-arg, but which was commented out in your last config.plist).  I have added this back as a DeviceProperty.
• Experiment with framebuffer-conX-pipes 0x11 and 0x12 on con1 and con2
• Experiment with framebuffer-conX-flags 0x4c7 (which is a mystery to me, since I don't know the extra bit flag)

 

Feel free to test the attached config.plists and report your finding if any of them seem to be better than what you currently have.  If you find any other suggestions, post them here for discussion.

 

The attached config.plists are as follows:

• config-H31-test-7-6.plist: Add DeviceProperty igfxfw=2, change con1 and con2 flags to 0x04c7, framebuffer-con1-pipe = 0x11, framebuffer-con2-pipe = 0x12
• config-H31-test-7-7.plist: Add DeviceProperty igfxfw=2, leave con1 and con2 flags unchanged at 0x03c7, framebuffer-con1-pipe = 0x11, framebuffer-con2-pipe = 0x12
• config-H31-test-7-8.plist: Add DeviceProperty igfxfw=2, leave con1 and con2 flags unchanged at 0x03c7, framebuffer-con1-pipe = 0x12, framebuffer-con2-pipe = 0x11

 

If it seems to you that I am guessing, you're right.  I had hoped that changing framebuffer platform-ids (which changes framebuffer-flags) or changing framebuffer-conX-flags would make a difference.  This hack continues to stump me.

config-H31-test-7-6.plist.zip config-H31-test-7-7.plist.zip config-H31-test-7-8.plist.zip

Edited January 5 by deeveedee

[Asural]

On 1/4/2026 at 10:01 PM, brumas2025 said:

during the removal of ioreg, I did not have usb 3 devices connected.

Currently configured as a normal Hackintosh,
1. Updated to the latest BIOS.
2. Reviewed BIOS settings.
3. Configured ACPI.
4. Currently reviewing config.plist, mainly DeviceProperties settings.

I would like to perform other operation tests, so please configure USBMap.kext.

There were some errors in the operating instructions, so I will correct them.
 

USB connector to use with USBMap.kext

 

Internal USB connector not to use with USBMap.kext

 

First, select D.

Available USB connectors will be displayed, so insert a USB into the USB connector to be used with USBMap.kext to recognize it.
If you insert a USB 3.0 device into a USB 3.0 connector, the USB 2.0 port will also be recognized, but if you insert a USB 2.0 device, the USB 3.0 port will not be recognized.

Press q to exit the recognition screen, and the following menu will appear. Select K to create USBMap.kext, then press q to quit.

 

 

I believe there is only one USB device, so selecting 1 will complete the conversion for the Tahoe, then press q to quit.
 

 

I believe the correct number of connectors is 8, so please attach the USBMap.kext you created for confirmation.
 

[Asural]

22 hours ago, deeveedee said:

@brumas2025 and @Asural What is the reason that ACPI patches now include PNLF and ALS0 (both are laptop patches for internal display brightness control and ambient light sensor)?

Sorry, I don't know which one you're talking about.

I think there was a similar guide on Form, but I was referring to this guide.
WEG 1.4.1 enable-metal: Added igfxmetal=1 boot argument (and enable-metal property) to enable Metal on offline IGPU.
When I checked the current FAQ.IntelHD.en.md, it seems to have been deleted. I didn't do enough research.
 

I'm currently considering a different test than the framebuffer settings.
I have a question for you: Have the HP800's PciRoot(0x0)/Pci(0x2,0x0) settings remained the same since Catalina?
 

Edited January 6 by Asural

[brumas2025]

10 hours ago, deeveedee said:

The attached config.plists are as follows:

• config-H31-test-7-6.plist: Add DeviceProperty igfxfw=2, change con1 and con2 flags to 0x04c7, framebuffer-con1-pipe = 0x11, framebuffer-con2-pipe = 0x12
• config-H31-test-7-7.plist: Add DeviceProperty igfxfw=2, leave con1 and con2 flags unchanged at 0x03c7, framebuffer-con1-pipe = 0x11, framebuffer-con2-pipe = 0x12
• config-H31-test-7-8.plist: Add DeviceProperty igfxfw=2, leave con1 and con2 flags unchanged at 0x03c7, framebuffer-con1-pipe = 0x12, framebuffer-con2-pipe = 0x11

 

If it seems to you that I am guessing, you're right.  I had hoped that changing framebuffer platform-ids (which changes framebuffer-flags) or changing framebuffer-conX-flags would make a difference.  This hack continues to stump me.

 

Test results (-config-H31-test-7-6) The monitors load without problems and turn on immediately!!! (I still don't believe in success)

I've tried overloading more than one tray, and everything is always fine.

Thank you for your help!!!!

Quote

https://transfiles.ru/fm53t

iMac 06-01-26.ioreg.zip

[deeveedee]

@brumas2025 That's great!  Let's wait a day or two to make sure that the displays continue to work normally.  If this is indeed the fix, would you be open to testing a bit more to isolate the change that fixed the problem?

 

Thanks again for your patience.

 

8 hours ago, Asural said:

I'm currently considering a different test than the framebuffer settings.
I have a question for you: Have the HP800's PciRoot(0x0)/Pci(0x2,0x0) settings remained the same since Catalina?

 

The same DeviceProperties do work in Catalina through Tahoe; however, in Catalina, port type DVI also worked for DP->DVI adapters.  If you look at my thread, you'll see that I needed to change from port type DVI to HDMI, since port type DVI is no longer supported by macOS.

 

EDIT: @brumas2025 I love your movie!  It is strange that the Apple initially appears on both displays.  Has that always been the case or did that behavior change when we changed DeviceProperties.

Edited January 6 by deeveedee
fixed type (changed DP->HDMI to DP->DVI)

[Asural]

10 minutes ago, deeveedee said:

tThe same DeviceProperties do work in Catalina through Tahoe; however, in Catalina, port type DVI also worked for DP->HDMI adapters.  If you look at my thread, you'll see that I needed to change from port type DVI to HDMI, since port type DVI is no longer supported by macOS.

Thanks, I was wondering why HDMI doesn't work on my PC with Tahoe WEG 1.4.1, only the extended VGA works.
I've used your thread as a reference, but it's quite different from the HP 400 series.
 

[deeveedee]

@Asural If you start a thread dedicated to your HP 400 series, I'd be happy to help you there.

[Asural]

20 minutes ago, deeveedee said:

@Asural If you start a thread dedicated to your HP 400 series, I'd be happy to help you there.

This is off-topic, but...
I usually use the HP6300SFF as my main computer, and I post guides here, but if I still have the energy to write a post of this size for the HP400, I'd like to create one.

 

I've added an explanation of flags to the file attached here, but I never thought that flags would be enough to solve the problem.
I never thought of it like you did, and I'm both surprised and impressed that I was able to succeed thanks to you.