iM 4,1-14,4 EnableGop: Pre-OpenCore GOP support for EFI-era iMacs and Mac Pros

[internetzel]

I have the same problem:

serge@iMac-de-Serge-2 EnableGop % ./vBiosInsert.sh -a amdimac2.rom EnableGop.efi amdimac2new.rom
Auto-detecting GOP offset...
Compressing EFI using EfiRom...
Combining...
 - Not enough space within 128k limit - aborting!
serge@iMac-de-Serge-2 EnableGop %

I have the same iMac (iMac 12,2) and same graphic card (WX4130 Dell) and remove VGA section is also complex for me

If I understand , we must wait for a WX4130_EFI .rom

Now there is a directory EG:

ls -la IMAC-EFI-BOOT-SCREEN/EG
total 2136
drwxr-xr-x  13 serge  wheel     416 11 mar 14:18 .
drwxr-xr-x  16 serge  wheel     512 11 mar 14:18 ..
-rw-r--r--   1 serge  wheel   86181 11 mar 14:18 M4000-EnableGop-LVDS.rom.zip
-rw-r--r--   1 serge  wheel   86424 11 mar 14:18 M4000-EnableGop-eDP.rom.zip
-rw-r--r--   1 serge  wheel   84913 11 mar 14:18 M5100-EnableGop-LVDS.rom.zip
-rw-r--r--   1 serge  wheel   85034 11 mar 14:18 M6000-EnableGop.rom.zip
-rw-r--r--   1 serge  wheel   86824 11 mar 14:18 M6100-Elpida-EnableGop.rom.zip
-rw-r--r--   1 serge  wheel   86467 11 mar 14:18 M6100-HynixBFR-EnableGop.rom.zip
-rw-r--r--   1 serge  wheel  175050 11 mar 14:18 S7100X-6939-LVTMA-EnableGop.rom.zip
-rw-r--r--   1 serge  wheel   85873 11 mar 14:18 W5170M-EnableGop.rom.zip
-rw-r--r--   1 serge  wheel   86458 11 mar 14:18 W6150M-4GB-EnableGop.rom.zip
-rw-r--r--   1 serge  wheel  100058 11 mar 14:18 W7170M-EnableGop.rom.zip
-rw-r--r--   1 serge  wheel  102915 11 mar 14:18 WX7100_1243_EnableGop.rom.zip

a directory EG2:

serge@iMac-de-Serge-2 BootNatif % ls -la IMAC-EFI-BOOT-SCREEN/EG2
total 16920
drwxr-xr-x  33 serge  wheel     1056 11 mar 14:18 .
drwxr-xr-x  16 serge  wheel      512 11 mar 14:18 ..
-rw-r--r--   1 serge  wheel    40316 11 mar 14:18 CoreEG2.ffs
-rw-r--r--   1 serge  wheel    40260 11 mar 14:18 CoreEG2_x64.efi
-rw-r--r--   1 serge  wheel     7260 11 mar 14:18 EDIDParser.ffs
-rw-r--r--   1 serge  wheel     7204 11 mar 14:18 EDIDParser_x64.efi
-rw-r--r--   1 serge  wheel    63488 11 mar 14:18 M370-2820-EG2_.rom
-rw-r--r--   1 serge  wheel   185344 11 mar 14:18 M370.efi
-rw-r--r--   1 serge  wheel   131072 11 mar 14:18 M4000-EG2_adj.rom
-rw-r--r--   1 serge  wheel   131072 11 mar 14:18 M5100-ElpidaSamsung-EG2_adj.rom
-rw-r--r--   1 serge  wheel   131072 11 mar 14:18 M5100-HynixAFR-EG2_adj.rom
-rw-r--r--   1 serge  wheel   131072 11 mar 14:18 M5100-HynixBFR-EG2_adj.rom
-rw-r--r--   1 serge  wheel   131072 11 mar 14:18 M6000-EG2_adj.rom
-rw-r--r--   1 serge  wheel   189952 11 mar 14:18 NV14Python_EG2.efi
-rw-r--r--   1 serge  wheel   262144 11 mar 14:18 RX470-EG2.rom
-rw-r--r--   1 serge  wheel   524288 11 mar 14:18 RX480_4GB_EG_adj.rom
-rw-r--r--   1 serge  wheel  1048576 11 mar 14:18 RX5500XT-NV14-EG2.rom
-rw-r--r--   1 serge  wheel   524288 11 mar 14:18 S7100X-6930-LVTMA-EG2_adj.rom
-rw-r--r--   1 serge  wheel   185344 11 mar 14:18 VenusEG2.efi
-rw-r--r--   1 serge  wheel    61952 11 mar 14:18 Venus_XTA--6821.rom
-rw-r--r--   1 serge  wheel   131072 11 mar 14:18 W5170M-EG2.rom
-rw-r--r--   1 serge  wheel   131072 11 mar 14:18 W5170M-EG2_adj.rom
-rw-r--r--   1 serge  wheel   131072 11 mar 14:18 W6170M-4GB-EFI_adj.rom
-rw-r--r--   1 serge  wheel   131072 11 mar 14:18 W6170M-EG2_adj.rom
-rw-r--r--   1 serge  wheel   131072 11 mar 14:18 W7170M-EG2_adj.rom
-rw-r--r--   1 serge  wheel   524288 11 mar 14:18 WX4130-EG2_adj_ALT_VRAM.rom
-rw-r--r--   1 serge  wheel   524288 11 mar 14:18 WX4130_EG2_adj.rom
-rw-r--r--   1 serge  wheel   524288 11 mar 14:18 WX4150-EG2_adj_ALT_VRAM.rom
-rw-r--r--   1 serge  wheel   524288 11 mar 14:18 WX4150_EG2_adj.rom
-rw-r--r--   1 serge  wheel   524288 11 mar 14:18 WX4170-EG2_adj.rom
-rw-r--r--   1 serge  wheel   524288 11 mar 14:18 WX4170-EG2_adj_ALT_VRAM.rom
-rw-r--r--   1 serge  wheel   524288 11 mar 14:18 WX7100-EG2_adj.rom
-rw-r--r--   1 serge  wheel   524288 11 mar 14:18 WX7100-EG2_adj_ALT_VRAM.rom

and EFI directory is empty

I think you just have to be patient

thanks to all developers 👍

You could take the latest WX4130 VBIOS linked to on the first page of the GPU upgrade thread - that one has the VGA part already stripped (that's why it is described as lacking Bootcamp support).

 

[sergiodadi]

You could take the latest WX4130 VBIOS linked to on the first page of the GPU upgrade thread - that one has the VGA part already stripped (that's why it is described as lacking Bootcamp support).

I download the WX4130_GOP.rom from your post 15868 and I try but I have the same problem:

serge@iMac-de-Serge-2 EnableGop % ./vBiosInsert.sh -a WX4130_GOP.rom EnableGop.efi WX4130_GOP_EFI.rom
Auto-detecting GOP offset...
Compressing EFI using EfiRom...
Combining...
 - Not enough space within 128k limit - aborting!
serge@iMac-de-Serge-2 EnableGop %

I have flash your WX4130_GOP.rom to my graphic card to be sure that the vbios file download is good:

Here is the info of my graphic card (Dell WX4130 ) before the flash:

flash of the card:

My iMac work as before , no problem

Maybe I don't understand some things ?

 

[Borowski]

AMD-ROMs have a size-boundary, the complete bios-file including all images must not exceed 0x20000h bytes. In the linked rom is not eneugh empty space to add the GOP-enabler, perhaps only a few hundred bytes missing.

Three possible solutions:
- find a smaller GOP-image for this card
- remove the vga-driver from vbios-section, much work, and will lead to non-booting legacy os anymore
- add GOP-enabler to iMac-firmware.

Perhaps there are some other methods to reduce the rom, i don't have any experience w/ modding AMD-roms, sorry.

 

[sergiodadi]

AMD-ROMs have a size-boundary, the complete bios-file including all images must not exceed 0x20000h bytes. In the linked rom is not eneugh empty space to add the GOP-enabler, perhaps only a few hundred bytes missing.

Three possible solutions:
- find a smaller GOP-image for this card
- remove the vga-driver from vbios-section, much work, and will lead to non-booting legacy os anymore
- add GOP-enabler to iMac-firmware.

Perhaps there are some other methods to reduce the rom, i don't have any experience w/ modding AMD-roms, sorry.

I have download the vbios without the vga-driver :

"You could take the latest WX4130 VBIOS linked to on the first page of the GPU upgrade thread - that one has the VGA part already stripped (that's why it is described as lacking Bootcamp support)."

 

[Borowski]

Sorry, i didn't notice that.
I took a look into the WX4130_GOP.rom:
- vbios could be reduced a little bit, contains much padding at the end
- it has 2 compressed images after vbios, first one is the gop
- function of second image unknown to me

Perhaps the script doesn't like the second image after gop.

 

[luckasa]

I have download the vbios without the vga-driver :

"You could take the latest WX4130 VBIOS linked to on the first page of the GPU upgrade thread - that one has the VGA part already stripped (that's why it is described as lacking Bootcamp support)."

You can download the version that I have compiled here

 

[sergiodadi]

You can download the version that I have compiled here

Thank, I try tomorrow , I will tell you if it is OK

 

[internetzel]

Sorry, i didn't notice that.
I took a look into the WX4130_GOP.rom:
- vbios could be reduced a little bit, contains much padding at the end
- it has 2 compressed images after vbios, first one is the gop
- function of second image unknown to me

Perhaps the script doesn't like the second image after gop.

Yes, obviously I was too lazy to delete that unused EFI driver in there but just simply set the last image bit in the GOP driver image.
So that second EFI driver image (probably EdidParserDriver) can safely be overwritten with 0xFF bytes in order to make the script accept it.

 

[MacNB2]

Is it me or anyone else noticed a change in the boot loading progress bar ?

E.g. if I boot Catalina, or Big Sur or Monterey, I get the Apple Logo and the 1st Stage loading progress bar.
About a third of the way through the progress bar, the screen clears, there's a short pause (less than a second) and I immediately see the Login Screen.

That is, there's no longer 2nd stage progress bar after the screen clearing but straight to the Login Screen.
I only noticed this after adding EnableGop.ffs

 

[sergiodadi]

You can download the version that I have compiled here

thanks @luckasa it work on my iMac 27" mi 2011 (Dell WX4130) with WX4130_EFIBoot.rom

Apple Hardware Test work too:

Thanks also to @Bmju @internetzel @Ausdauersportler and others developers 👍

 

[Bmju]

@h9826790, @monoton - Could you test this version of OpenCore https://github.com/acidanthera/OpenCorePkg/actions/runs/4442242385 on Radeon VII, with and without DirectGopRendering, and send an OpenCore debug log for each setting? Just OpenCore directly, not EnableGop.efi. You could also try the EnableGop/EnableGopDirect from there too, if you like - though I believe it's unlikely to change anything. (I believe I'm right that this card also works (somewhat) in OpenCore even without DirectGopRendering, but always shows the cursor artefacts?)

 

[h9826790]

@h9826790, @monoton - Could you test this version of OpenCore https://github.com/acidanthera/OpenCorePkg/actions/runs/4442242385 on Radeon VII, with and without DirectGopRendering, and send an OpenCore debug log for each setting? Just OpenCore directly, not EnableGop.efi. You could also try the EnableGop/EnableGopDirect from there too, if you like - though I believe it's unlikely to change anything. (I believe I'm right that this card also works (somewhat) in OpenCore even without DirectGopRendering, but always shows the cursor artefacts?)

Sure, to confirm what you want me to test.

1) Remove EnableGop / EnableGopDirect from the Mac Pro's BootROM

2) Install this latest OpenCore beta debug mode

<key>Debug</key>
        <dict>
            <key>AppleDebug</key>
            <true/>
            <key>ApplePanic</key>
            <false/>
            <key>DisableWatchDog</key>
            <true/>
            <key>DisplayDelay</key>
            <integer>0</integer>
            <key>DisplayLevel</key>
            <integer>2151678018</integer>
            <key>LogModules</key>
            <string>*</string>
            <key>SysReport</key>
            <true/>
            <key>Target</key>
            <integer>65</integer>
        </dict>

3) Enable ProvideConsoleGop

4) Two boot test, one with DirectGopRendering, another one without DirectGopRendering. And save the logs.

5) Install EnableGop into the BootROM, and test boot with ProvideConsoleGop and DirectGopRendering disabled.

6) If negative / buggy result from 5, try the same thing with EnableGopDirect

Do I miss anything?

 

[h9826790]

(I believe I'm right that this card also works (somewhat) in OpenCore even without DirectGopRendering, but always shows the cursor artefacts?)

I can't remember if I test ProvideConsoleGop without DirectGopRendering in OpenCanopy. But from memory, I did that in the text mode boot picker. Which no boot picker show on the screen (most likely it's there, but just I can't see it), however, once timeout, I can see the loading screen straight away (which means the boot scree function is somehow working).

Base on that experience, and the test we've done here. I believe you are right. If no EnableGop / EnableGopDirect, and no DirectGopRendering, but just ProvideConsoleGop, most likely the OpenCanopy boot picker will work but with artefacts.

 

[Bmju]

Do I miss anything?

No, nothing! Thank you.

 

[Bmju]

No, nothing! Thank you.

@h9826790 - For the debug logs, if you could please make sure to set some value in OC screen Resolution setting. Max or any resolution is fine. Anything other than empty or 0x0. Thank you.

 

[h9826790]

Just finished the test with Max setting. Will PM you the logs.

 

[MacNB2]

Is it me or anyone else noticed a change in the boot loading progress bar ?

E.g. if I boot Catalina, or Big Sur or Monterey, I get the Apple Logo and the 1st Stage loading progress bar.
About a third of the way through the progress bar, the screen clears, there's a short pause (less than a second) and I immediately see the Login Screen.

That is, there's no longer 2nd stage progress bar after the screen clearing but straight to the Login Screen.
I only noticed this after adding EnableGop.ffs

No one seeing this issue on your modified modified BootROM with EnableGop ??

 

[Bmju]

No one seeing this issue on your modified modified BootROM with EnableGop ??

Yes, this happens. But are you saying this is different from booting the same OSes from OpenCore without EnableGop.efi installed? I believe it should not be. Perhaps you could re-test, and send your config if you test and confirm it is different when you do not have EnableGop installed.

 

[armdn]

Well, this is strange...

For the Quadro K600 ROM

Auto-detecting GOP offset...
Compressing EFI using EfiRom...
Adding Nvidia header...
Combining...
Verifying (starting at 0xF400)...
Found compressed EFI ROM start at 0x4c
Input size: 110004, Output size: 11967, Scratch size: 13376
UEFI decompression failed!

For the Quadro K2000 ROM the same thing...

Auto-detecting GOP offset...
Compressing EFI using EfiRom...
Adding Nvidia header...
Combining...
Verifying (starting at 0xF400)...
Found compressed EFI ROM start at 0x4c
Input size: 146868, Output size: 11967, Scratch size: 13376
UEFI decompression failed!

What can be cause the problem? ROMs downloaded from TechPowerUp.

 

[Bmju]

Well, this is strange...

For the Quadro K600 ROM

Auto-detecting GOP offset...
Compressing EFI using EfiRom...
Adding Nvidia header...
Combining...
Verifying (starting at 0xF400)...
Found compressed EFI ROM start at 0x4c
Input size: 110004, Output size: 11967, Scratch size: 13376
UEFI decompression failed!

For the Quadro K2000 ROM the same thing...

Auto-detecting GOP offset...
Compressing EFI using EfiRom...
Adding Nvidia header...
Combining...
Verifying (starting at 0xF400)...
Found compressed EFI ROM start at 0x4c
Input size: 146868, Output size: 11967, Scratch size: 13376
UEFI decompression failed!

What can be cause the problem? ROMs downloaded from TechPowerUp.

Please send the ROMs.

I did make an update to the script attempting to support these newer Nvidia cards. Not published yet and I don't know if it works - the previous user who hit this issue (see thread from here onwards: https://forums.macrumors.com/thread...-era-imacs-and-mac-pros.2378942/post-32018613 ) had problems with their machine and has been unable to test so far.

If you send the ROMs, I can make test versions using my updated script. But don't test them unless you have a hardware way to restore your vBIOS if it doesn't work (CH341A + SOIC clip which you can run from another machine, or from your machine with another GPU in it, basically).

 

[MacNB2]

Yes, this happens.

Under what circumstances does this happen ?

But are you saying this is different from booting the same OSes from OpenCore without EnableGop.efi installed? I believe it should not be. Perhaps you could re-test, and send your config if you test and confirm it is different when you do not have EnableGop installed.

I had added EnableGop.ffs to my cMP BootROM when I noticed this behaviour with OpenCore 0.9.0. Anyway instead to rolling back to previous OpenCore versions I thought I go back to the BootROM without EnableGop to test.

So I just spent the whole morning trying to recover from flashing back the original (reconstructed) BootROM (without EnableGop.ffs). Nightmare as I got NO OpenCore boot picker...no Apple logo...no progress bar...it just booted straight into Catalina and Login screen.
This what I did:

1. Flashed Reconstructed BootROM (without EnableGop.ffs)
2. Booted into Recovery and blessed OpenCore (bless --mount /Volumes/EFI --setBoot --file /Volumes/EFI/EFI/BOOT/BOOTx64.efi)
3. Restarted and NO OpenCore boot picker..black screen...booted straight into first macOS it found (Catalina) and got Login screen
4. Checked if OpenCore was the actual boot loader by using kextstat | grep -v com.apple and saw Lilu & Whatevergreen kexts listed
5. Restarted into recovery and tried bless command again. No joy....no boot picker.
6. Restarted and held Alt/Option ...no joy...no boot picker screen.
7. Did several NVRAM Resets....no joy
8. Downloaded DEBUG version of OpenCore and installed that to get a log.

The log showed some strange issues relating to GOP and screen resolution :

...
...
00:445 00:003 OCAE: Screen resolution has 0-dimension
00:448 00:003 CRSCR: AppleEvent->RegisterHandler[0] returned Success
00:451 00:003 OC: Driver CrScreenshotDxe.efi at 5 is successfully loaded!
00:455 00:003 OC: Driver XhciDxe.efi at 6 (Beta support for USB 3.0 boot, only works via a USB 3 hub, otherwise, NO boot) is skipped!
00:458 00:003 OC: Driver ToggleSipEntry.efi at 7 (Allow SIP control) is being loaded...
00:463 00:005 OCABC: EfiBootRt candidate - <nil>
00:467 00:003 OCABC: IsEfiBootRt 0 (BP 1, Apple 0)
00:470 00:003 OCB: Arch filtering 0(69632)->75485018(69632) caps 4 - Success
00:473 00:003 OC: Driver ToggleSipEntry.efi at 7 is successfully loaded!
00:476 00:003 OC: Driver ResetNvramEntry.efi at 8 (Allow NVRAM reset) is skipped!
00:480 00:003 OC: Connecting drivers...
03:787 03:307 OC: Connecting drivers done...
03:791 00:003 OC: Found 3 pointer devices - Success
03:794 00:003 OC: Requested resolution is 0x0@0 (max: 1, force: 0) from Max
03:797 00:003 OCC: Requesting 0x0@0 (max: 1) resolution, curr 0, total 0
03:800 00:003 OCC: Current FB at 0x0 (0x0), format 0, res 0x0 scan 0
03:804 00:003 OCC: No compatible mode for 0x0@0 (max: 1) resolution
03:807 00:003 OC: Changed resolution to 0x0@0 (max: 1, force: 0) from Max - Not Found
03:810 00:003 OCC: Switching to direct GOP renderer...
03:813 00:003 OCC: Found EG2 support 10002
03:816 00:003 OCC: Got rotation 0 degrees from EG2
03:820 00:003 OCBLT: 0: shl:0 shr:16 mask:FF
03:823 00:003 OCBLT: 1: shl:0 shr:0 mask:FF00
03:827 00:003 OCBLT: 2: shl:16 shr:0 mask:FF0000
03:830 00:003 OCBLT: Bytes per pixel: 4
03:833 00:003 OCBLT: 0: shl:0 shr:16 mask:FF
03:836 00:003 OCBLT: 1: shl:0 shr:0 mask:FF00
03:839 00:003 OCBLT: 2: shl:16 shr:0 mask:FF0000
03:843 00:003 OCBLT: Bytes per pixel: 4
03:846 00:003 OC: Setting UIScale to 1 - Success
03:850 00:003 OCC: Using builtin text renderer scale 1 mode 1
03:853 00:003 OCC: ASCII Text Reset [RenderResync] - Load Error
03:856 00:003 OCC: Setup ASCII Output - Load Error
03:859 00:003 OC: Requested console mode is 0x0 (max: 0) from
03:863 00:003 OC: Requested not to use audio
03:866 00:003 OC: OcMiscLoadSystemReport...
03:869 00:003 OC: OcLoadAcpiSupport...
...
...

It looks like it's saying that the requested screen resolution is 0x0 but I had not changed the config.plist at all.
After lots of iterations with DirectGopRendering to ON/OFF and UIScale from 0 to 2, I finally managed to get OpenCore boot picker back (I think UIScale did the trick as I have a 4K monitor attached to the RX 580).

But the OpenCore debug out still shows issues:

...
...
00:561 00:003 OCAE: Screen resolution has 0-dimension
00:564 00:003 CRSCR: AppleEvent->RegisterHandler[0] returned Success
00:568 00:003 OC: Driver CrScreenshotDxe.efi at 5 is successfully loaded!
00:571 00:003 OC: Driver XhciDxe.efi at 6 (Beta support for USB 3.0 boot, only works via a USB 3 hub, otherwise, NO boot) is skipped!
00:575 00:003 OC: Driver ToggleSipEntry.efi at 7 (Allow SIP control) is being loaded...
00:580 00:005 OCABC: EfiBootRt candidate - <nil>
00:584 00:004 OCABC: IsEfiBootRt 0 (BP 1, Apple 0)
00:588 00:003 OCB: Arch filtering 0(69632)->75492018(69632) caps 4 - Success
00:591 00:003 OC: Driver ToggleSipEntry.efi at 7 is successfully loaded!
00:595 00:003 OC: Driver ResetNvramEntry.efi at 8 (Allow NVRAM reset) is skipped!
00:599 00:003 OC: Connecting drivers...
03:906 03:307 OC: Connecting drivers done...
03:910 00:003 OC: Found 3 pointer devices - Success
03:913 00:003 OCC: GOP exists on ConsoleOutHandle and has 0 modes
03:917 00:004 OCC: Looking for GOP replacement due to invalid mode count
03:921 00:003 OCC: Alternative GOP status is - Success
03:925 00:003 OC: Requested resolution is 0x0@0 (max: 1, force: 0) from Max
03:929 00:003 OCC: Requesting 0x0@0 (max: 1) resolution, curr 4294967295, total 8
03:932 00:003 OCC: Current FB at 0x80000000 (0x0), format 0, res 0x0 scan 0
03:936 00:003 OCC: Mode 0 - 2560x1440:1
03:939 00:003 OCC: Mode 1 - 640x480:1
03:943 00:003 OCC: Mode 2 - 800x600:1
03:946 00:003 OCC: Mode 3 - 1024x768:1
03:950 00:003 OCC: Mode 4 - 1280x1024:1
03:954 00:003 OCC: Mode 5 - 1400x1050:1
03:957 00:003 OCC: Mode 6 - 1600x1200:1
03:961 00:003 OCC: Mode 7 - 1280x960:1
03:965 00:003 OCC: Setting mode 0 with 2560x1440 resolution
03:971 00:006 OCC: Changed resolution mode to 0
03:974 00:003 OC: Changed resolution to 0x0@0 (max: 1, force: 0) from Max - Success
03:978 00:004 OC: Selected UIScale 1 based on 2560x1440 resolution
03:982 00:003 OC: Setting UIScale to 1 - Success
03:986 00:003 OCC: Using builtin text renderer scale 1 mode 1
03:990 00:003 OCC: Install console control (752900B0/0/0), current - Success
03:993 00:003 OCC: Setup ASCII Output - Success
03:997 00:003 OC: Requested console mode is 0x0 (max: 0) from
04:001 00:003 OC: Requested not to use audio
04:004 00:003 OC: OcMiscLoadSystemReport...
04:008 00:003 OC: OcLoadAcpiSupport...
...
...

Specifically..GOP exists on ConsoleOutHandle and has 0 modes followed by Looking for GOP replacement due to invalid mode count

In the config.plist file, UIScale is set to 0.

EnableGop.ffs is NOT the issue now as the without it, I still see no 2nd Stage Apple Logo and progress bar. But the above log shows maybe the problem ?

Shall I raise an Issue on Acidanthera ?

 

[Bmju]

Under what circumstances does this happen ?

Mostly, normally, typically - in my experience. I am not totally ruling out that it can be avoided with the right settings, though I suspect not, on many non-natively-supported cards. Maybe anyone else can chime in?

It looks like it's saying that the requested screen resolution is 0x0 but I had not changed the config.plist at all.

This just means you are using Max resolution.

It's not clear to me that anything else in the rest of your post is non-normal. (Yes, it can be annoying and fiddly getting back to OpenCore after an NVRAM reset - which is all that reflashing a clean ROM really amounts to, here - with no native boot picker, but that's not an issue for the Acidanthera bugtracker.)

 

[MacNB2]

Mostly, normally, typically - in my experience. I am not totally ruling out that it can be avoided with the right settings, though I suspect not, on many non-natively-supported cards. Maybe anyone else can chime in?

OK thanks. I'll ask others on the OpenCore thread if anyone else is seeing this behaviour.

This just means you are using Max resolution.

It's not clear to me that anything else in the rest of your post is non-normal. (Yes, it can be annoying and fiddly getting back to OpenCore after an NVRAM reset - which is all that reflashing a clean ROM really amounts to, here - with no native boot picker, but that's not an issue for the Acidanthera bugtracker.)

I'm surprised GOP exists on ConsoleOutHandle and has 0 modes is normal or OK as it is an invalid mode count.

Anyway, I rolled back to OpenCore 0.7.3 as a test and it too exhibits the same behaviour of no 2nd Stage logo. I could swear I have seen it before on the cMP.....but probably I'm mistaking it for one my Hack systems.

It's almost as if the macOS drivers initialise but don't know where to send the 2nd stage screen output until after the initialisation has completed and WindowServer up and running.

I guess it's a cMP specific "attribute" as I don't see this behaviour on my Hacks.
That is, Hackintosh behaves more like a real Mac than a real cMP with OpenCore 🤷‍♂️

 

[Bmju]

OK thanks. I'll ask others on the OpenCore thread if anyone else is seeing this behaviour.



I'm surprised GOP exists on ConsoleOutHandle and has 0 modes is normal or OK as it is an invalid mode count.

Anyway, I rolled back to OpenCore 0.7.3 as a test and it too exhibits the same behaviour of no 2nd Stage logo. I could swear I have seen it before on the cMP.....but probably I'm mistaking it for one my Hack systems.

It's almost as if the macOS drivers initialise but don't know where to send the 2nd stage screen output until after the initialisation has completed and WindowServer up and running.

I guess it's a cMP specific "attribute" as I don't see this behaviour on my Hacks.
That is, Hackintosh behaves more like a real Mac than a real cMP with OpenCore 🤷‍♂️

I see the same behaviour on my hack, tbh. But let me ask around and confirm whether this is or isn't considered normal.

I would 100% agree with you that this is not the perfect behaviour of a native Mac with all native compatible components, of course.

About 'GOP exists on ConsoleOutHandle and has 0 modes', just reading you own log as posted, this is immediately followed by 'Looking for GOP replacement due to invalid mode count', and then 'Alternative GOP status is - Success' - so this is just ProvideConsoleGop doing its job.

 

[MacNB2]

I see the same behaviour on my hack, tbh. But let me ask around and confirm whether this is or isn't considered normal.

I would 100% agree with you that this is not the perfect behaviour of a native Mac with all native compatible components, of course.

About 'GOP exists on ConsoleOutHandle and has 0 modes', just reading you own log as posted, this is immediately followed by 'Looking for GOP replacement due to invalid mode count', and then 'Alternative GOP status is - Success' - so this is just ProvideConsoleGop doing its job.

Thank you for indulging my curiosity.