Contribute
Register

[SUCCESS] Screen Freezes using Sapphire Radeon Vega 64 Nitro+ in MacOS Mojave 10.14

Status
Not open for further replies.
I see, back to 59120003, then.
Now, needs more time to boot, but I will check my config again.

Thanks a lot for your time!


Now, jpg preview is not working any more.
Iam adding clover screenshots for help/
Screenshot 2018-10-23 at 18.49.03.pngScreenshot 2018-10-23 at 18.49.03.pngScreenshot 2018-10-23 at 18.49.16.png
 

Attachments

  • Screenshot 2018-10-23 at 18.48.49.png
    Screenshot 2018-10-23 at 18.48.49.png
    457.2 KB · Views: 137
Last edited:
@nextsak,

I cant see anything wrong with the settings in the screen shots you posted ....

Are you sure the IGPU is enabled in the BIOS ? and the DGPU is set as the primary adaptor ?

One quick way to check if the headless ig-platform-id is working is to look at the "Graphics/Displays" section in System Information (Apple Icon -> About this Mac -> System Report -> Graphics/Displays)

You should only see your DGPU listed, if you see the IGPU as well as your DGPU then something is wrong.
The usual cause is an issue with Lilu and/or WhatEverGreen kext not loading ...try deleting them and re install using the method in the guide.

Cheers
Jay
 
I am sure that I have enabled IGPU and the DGPU is set as primary, also it is headless as per your guide.
Jpeg preview problem seems fixed after adding device-id from FBPatcher at Devices-> Properties->device-id.
I also rebuild kext cache.

The only problem now is slow booting despite my Nvme 960 boot disk.
After Aorus logo, apple logo with very slow loading bar, then at middle loading point the screen flashes black and after few seconds loading bar reappears and continue to successful boot.

Iam adding my config iif you can kindly spot any problem.
Thanks in advance
 

Attachments

  • Screenshot 2018-10-25 at 08.08.11.png
    Screenshot 2018-10-25 at 08.08.11.png
    141.5 KB · Views: 105
  • config IGPU.plist
    6.2 KB · Views: 127
@patrickfogle

I would remove the Vega 64 , and re-install Mojave with just the IGPU as the default display adapter, install Lilu + WhatEverGreen using the correct ig-platform-id device property for your IGPU (not headless). If everything seems to be working edit the ig-platform-id and change to the headless value. Shutdown and install the Vega 64 and change the BIOS settings to IGPU enabled, DGPU as primary. Shutdown and swap the monitor cables over. Power up and see if the Vega and headless IGPU are now working.

Good Luck
Cheeers
Jay

@jaymonkey & @pastrychef (many thanks to both of you for your assistance so far!!),

OK. I am sorry for not replying with a report sooner, as I got very ill after we last communicated and have not been able to get on my computer again until today.

I did what both of you suggested, but I have some questions. First, let me explain what I have done so far this time around:

1. I remade my MacOS Mojave Install USB and totally started from scratch.
2. I removed my Vega64 (and every other unnecessary PCI card) and reinstalled MacOS with just the iGPU enabled in the BIOS.
3. When trying to boot from my SSD, I initially used @pastrychef 's EFI Unified 4.3.2 (High Sierra) as he had suggested, but it would not boot with that as it was (it went all the way to the point that it looked like MacOS was about to open, and then the monitor turned off and the PC rebooted spontaneously).
4. So, I kept @pastrychef 's same EFI and modified the config.plist in the EFI - Clover Configurator/Devices Tab according the suggested method that @jaymonkey had suggested according to his guide. This is what the EFI Devices Tab-Properties looked like before I edited it:

Screen Shot 2018-10-31 at 7.11.28 AM.png


The following is the same page after I edited according to @jaymonkey 's instructions for a NON-Headless Solo iGPU (Intel HD630 Skylake->Platformid: 0x59120000 or in the Properties Tab, 00002195)
Screen Shot 2018-10-31 at 6.37.52 AM.png


The following was also on the original Devices/Properties Tab (the first listed PciRoot Path with layout-id) and I left it unedited (as shown) when I edited the second PciRoot Path Properties Key:
Screen Shot 2018-10-31 at 6.37.37 AM.png


No other changes were made. Here is the new EFI if you want to take a closer look: EFI.zip

My results are ok so far (I have only used the installation with the iGPU). Here are my questions:

1. So far, with the above EFI, I see to have a stable system, but I do not have graphics acceleration. I know this mainly because my dock is not transparent (as shown below):
Screen Shot 2018-10-31 at 7.11.46 AM.png

Do I need to edit something in the EFI or otherwise to get accelerated graphics with just the iGPU installed?

2. Are the Boot Tab Custom Flag necessary (-alcbeta shikigva=60 -lilubetaall -wegbeta)? Or are these obsolete now with the new Lilu/WhateverGreen?

3.
@jaymonkey, you suggested that I get everything working right before I change over to the Vega64 (that is, change the current Deviceid to the Headless Deviceid of 30002195, install the Vega64, change over video cables, and change Bios to enable the iGPU with the Vega as Primary). Since I do not have acceleration on the solo iGPU, I assume my system is not ready for this switchover, right?

Thanks for you generous help!
 

Attachments

  • EFI.zip
    27.4 MB · Views: 117
@jaymonkey & @pastrychef (many thanks to both of you for your assistance so far!!),

OK. I am sorry for not replying with a report sooner, as I got very ill after we last communicated and have not been able to get on my computer again until today.

I did what both of you suggested, but I have some questions. First, let me explain what I have done so far this time around:

1. I remade my MacOS Mojave Install USB and totally started from scratch.
2. I removed my Vega64 (and every other unnecessary PCI card) and reinstalled MacOS with just the iGPU enabled in the BIOS.
3. When trying to boot from my SSD, I initially used @pastrychef 's EFI Unified 4.3.2 (High Sierra) as he had suggested, but it would not boot with that as it was (it went all the way to the point that it looked like MacOS was about to open, and then the monitor turned off and the PC rebooted spontaneously).
4. So, I kept @pastrychef 's same EFI and modified the config.plist in the EFI - Clover Configurator/Devices Tab according the suggested method that @jaymonkey had suggested according to his guide. This is what the EFI Devices Tab-Properties looked like before I edited it:

View attachment 361888

The following is the same page after I edited according to @jaymonkey 's instructions for a NON-Headless Solo iGPU (Intel HD630 Skylake->Platformid: 0x59120000 or in the Properties Tab, 00002195)
View attachment 361891

The following was also on the original Devices/Properties Tab (the first listed PciRoot Path with layout-id) and I left it unedited (as shown) when I edited the second PciRoot Path Properties Key:
View attachment 361889

No other changes were made. Here is the new EFI if you want to take a closer look: EFI.zip

My results are ok so far (I have only used the installation with the iGPU). Here are my questions:

1. So far, with the above EFI, I see to have a stable system, but I do not have graphics acceleration. I know this mainly because my dock is not transparent (as shown below):
View attachment 361896

Do I need to edit something in the EFI or otherwise to get accelerated graphics with just the iGPU installed?

2. Are the Boot Tab Custom Flag necessary (-alcbeta shikigva=60 -lilubetaall -wegbeta)? Or are these obsolete now with the new Lilu/WhateverGreen?

3.
@jaymonkey, you suggested that I get everything working right before I change over to the Vega64 (that is, change the current Deviceid to the Headless Deviceid of 30002195, install the Vega64, change over video cables, and change Bios to enable the iGPU with the Vega as Primary). Since I do not have acceleration on the solo iGPU, I assume my system is not ready for this switchover, right?

Thanks for you generous help!

There's no acceleration because the ig-platform-id is wrong. If you are using ONLY the Intel HD 630, it should be 00001259. If you are using your Vega, it should be 03001259.

The Custom Flags don't matter. If you don't want them, you can delete them. They are not obsolete.
 
@jaymonkey & @pastrychef (many thanks to both of you for your assistance so far!!) .... My results are ok so far (I have only used the installation with the iGPU). Here are my questions:

Thanks for you generous help!

@patrickfogle,

You have entered the PlatformID of 59120000 incorrect, you have entered 00002195 when it should be 00001259
We only reverse the byte order, not the bytes themselves ...

Cheers
Jay

Note: I've recently added more info on Headless configurations in the guide, it covers things in a bit more detail with more examples.
 
@patrickfogle,

You have entered the PlatformID of 59120000 incorrect, you have entered 00002195 when it should be 00001259
We only reverse the byte order, not the bytes themselves ...

Cheers
Jay

Note: I've recently added more info on Headless configurations in the guide, it covers things in a bit more detail with more examples.

@jaymonkey & @pastrychef,

Thanks to both of you for your prompt responses. It seems that I misunderstood the meaning of "reverse the byte order" just as Jay mentioned. I made the single change to the PlatformID as mentioned, and now I am getting accelerated video, but with some really erratic, crazy glitchy stuff. Almost too bad to write this reply. Looks like a lightening storm in the twilight zone! You can see what I mean from this linked video I uploaded:

Any idea what that is about or how to fix that?
 
@jaymonkey & @pastrychef,

Thanks to both of you for your prompt responses. It seems that I misunderstood the meaning of "reverse the byte order" just as Jay mentioned. I made the single change to the PlatformID as mentioned, and now I am getting accelerated video, but with some really erratic, crazy glitchy stuff. Almost too bad to write this reply. Looks like a lightening storm in the twilight zone! You can see what I mean from this linked video I uploaded:

Any idea what that is about or how to fix that?

How much RAM did you allocate to the IGPU in BIOS?
 
How much RAM did you allocate to the IGPU in BIOS?
@pastrychef,

As far as I can tell, in the Asus BIOS, there is only an option for how much RAM you allocate to the iGPU when you have iGPU Multi-Monitor enabled. Here is a BIOS screen that shows a perameter like what you’re asking about, but this parameter goes away when you disable the “iGPU Multi-Monitor” setting. It is 128MB when it is enabled. I don’t know what the RC6 (Render Standby)” is. I’ve not seen it mentioned anywhere in my readings:

A3AF4114-705C-441F-9074-92179810BF0A.jpeg


Is there another place that you can change what you’re asking about?
 
@pastrychef,

As far as I can tell, in the Asus BIOS, there is only an option for how much RAM you allocate to the iGPU when you have iGPU Multi-Monitor enabled. Here is a BIOS screen that shows a perameter like what you’re asking about, but this parameter goes away when you disable the “iGPU Multi-Monitor” setting. It is 128MB when it is enabled. I don’t know what the RC6 (Render Standby)” is. I’ve not seen it mentioned anywhere in my readings:

View attachment 361936

Is there another place that you can change what you’re asking about?

That looks fine. I have the RC6 thing enabled, but I don't think that matters... I don't know why you get the graphics anomalies.
 
Status
Not open for further replies.
Back
Top