[Solved] automate-eGPU EFI - eGPU boot manager for macOS and Windows
Successfully using this EFI to boot into Windows using Razer Core X with 1070 on 2016" MBPr (Pro 460) for a few weeks already and it's pretty stable. A couple of issue were encountered but none of them critical:
1. When switching from macOS to Windows I always have to boot into Windows twice in a row. First time Windows boots using the integrated graphics card even though the eGPU is connected. I have to boot into Windows once again to have the 1070 detected and working.
2. No way to change display brightness in Windows. I wish there was some terminal command to do that. Or maybe even forcing some fixed brightness level to always be used in Windows. Because right now looks like the only way to change it is to boot into macOS and then boot back into Windows twice because of the issue above.
3. In some rare cases Windows throws some kind of BSOD when loading but that's very rare and third reboot always helped so far.
But even with all of that the EFI is a very usable solution so once again thanks for all the hard work @goalque !
To avoid use of a USB stick, what is the possibility of adding these efi boot files to my internal SSD's EFI partition?
I was previously successful at this by adding a separate tiny partition to my internal drive and copying the automate-efi folder to it and would allow me to choose it on startup separate of my main EFI boot partition but now that I have redone my bootcamp installation, I forgot to leave a bit of empty drive space to make a separate efi boot partition and as fragile APFS is, I don't want to boot to windows to resize my NTFS volume smaller, lol. Any tips?
The main post has been updated to include the automate-eGPU EFI Installer and some insight into what the installer does.
Here's an attempt to a step-by-step for my 2012 Mini with AMD eGPU.
- Open a Terminal.
- If your shell is bash, the default Terminal shell, go to the next step - otherwise type bash. Apparently, mac_editors command line is not valid in (at least) tcsh, which happens to be my favourite shell.
- Plug in an USB stick that you don't mind getting overwritten. Note its name on the desktop. Make sure it's name does not contain EFI.
Note: I had to repartition the stick with exFAT (it's a 16GB stick and DOS/FAT is ancient enough to have trouble with that size) because information on it got damaged somehow in an earlier attempt.
- Optional: Eject and unplug all other writable USB media to prevent accidentally overwriting the wrong device (your backups, for example).
- Copy-paste mac_editors command in Terminal and follow the steps there (that's where the name of your USB stick comes in):bash <(curl -s https://egpu.io/wp-content/uploads/2018/11/automate-eGPU-EFI-install.txt)
- Remove the file apple_set_os.efi from EFI/CLOVER/drivers64UEFI/ on your USB stick. The 2012 Mini is one of those Mac's that don't need it. My eGPU hasn't activated once with this file in the EFI stick, so it matters.
- Eject the USB stick.
- Shut down the Mac.
- Unplug the thunderbolt cable; I also powered off the eGPU and let everything settle for a few minutes.
- Plug in the USB stick. The stick doesn't need to be plugged directly in an USB port on your Mac, BTW, an USB hub should work just fine.
Steps for every boot of your Mac:
- Turn on the eGPU enclosure (if you turned it off), but don't connect the Thunderbolt cable to the Mac yet.
- Turn on the Mac with the Option-key (Alt on a PC keyboard) pressed until you see a list of disks to choose from. If you see the Apple logo instead, something went wrong; perhaps the keyboard didn't get detected, or Option is the wrong key after all (I don't use Apple keyboards).
- Choose your USB stick to boot from.
- Press q
- Start Mac OS from the menu.
- Log in.
- Plug in the Thunderbolt cable.
- Log out (Cmd-Shft-Q).
- At this point, the eGPU should activate.
If you noticed that there are no pmset commands in those steps, that's correct. I don't seem to need them on my Mini.
At step 7, I usually see the eGPU symbol in the menu bar for a moment. Is it normal that it disappears?
At step 8, it seems that if I let MacOS churn through the entire session starting back up, that too much time passes before logging out and the eGPU doesn't come up. Does that make sense in any way?
For the record, my 2012 Mini detects as having Thunderbolt 2. Throughput in System Information shows the expected 10 Gb/s though.
I just followed these steps to test my Aorus gaming box 1080 enclosure with Msi rx560 aero itx and I'm in 10.14.2 Mojave currently, but nothing worked... LED didn't turn on and the fan of the card did not run, but the small fan of the power worked, is that normal? Cause the LED and the fan worked great with the original gtx1080 even I did't login with EFI boot.
I got also a weird problem now...
A member told me to take a look into this thread after I had some issues with installing clover. It was a great hint so I installed the automate-eGPU. For the first it look very well and I was able to play some games with the eGPU now.
The weird part is that if I play a game like Pupg everything is fine but if I want to start games like Black Ops 4 or Warframe the eGPU doesn't work.
If this question is already asked once somewhere else I'm sorry for this reply. If there is any solution already I would be grateful for a helpful answer.
@xabazios just for kicks let's give this a try:
1. Boot into windows WITHOUT the eGPU connected.
2. Run DDU and clean your system of all remnants of any drivers.
3. Once you reboot after DDU, try installing Adrenalin drivers from here (start with v18.8.2)
4. Reboot after installation ( DO NOT CONNECT eGPU)
5. Once windows loads shut down.
6. Boot into windows but this time plug in your eGPU.
Let me know what happens.
I followed your steps and got this result:
When I boot, it takes a few seconds to identify the external monitor, then I'm allowed to login I via the external monitor. In the meanwhile the internal display freezed at the boot-logo of windows and its starts to flicker. I can play games and use all applications via the external monitor - full usage use of windows 10, BUT my internal MacBook Display just stays at the boot logo (windows) and flickers.
Do you (someone) know(s) what's wrong? I don't really need the internal display, so if there's a way to disable it, I could game on this mac by using only the external monitor. But actually the flickering is very terrible, also I don't know if this isn't good for my MacBook.
thanks very much, I hope you can help me
a couple of weeks ago I managed to get the following system working in mac 10.14.2 and windows 10 using purge-wrangler and purge-nvda, used this guide https://egpu.io/forums/builds/mid-2012-15-macbook-pro-non-retina-rx58010gbps-tb1-gigabyte-gaming-box-via-tb3-to-tb1-adapter-macos10-13-4-win10-clarketus/
- Macbook pro 15' retina 2012 Intel HD graphics 4000 + NVIDIA GeForce GT 650M TB1
- Core X
- Gigabyte AMD Radeon XR 580
- Using eGPU displayport to connect directly to a external screen, this screen set up as main screen
Once I got back to my desk and tried to reenable purge-nvda macox booting process will stuck at the black screen with the open logo and the white progress bar, no matter how long I left it running. I managed to reboot after doing a nvram restore –which reverts back System Integrity Protection– after that I have tried to replicate the results of purge-wrangler and purge-nvda with no luck. Each time I run purge-nvda, press option 1 (AMD eGPUs ) and reboot the booting process will stuck at the black screen with the open logo and the white progress bar.
I found automate-eGPU EFI searching for a solution so I tried this instead. First uninstalled purge-wrangler and purge-nvda and then followed the instructions. I am able to boot on Macosx and the eGPU is recognized after login –icon showed at the bottom right corner and so on– but I am having the same behaviour as before, external display connected directly to eGPU remains black and integrated display appears as secondary screen. On system preferences my 3 graphics appears, radeon 580, HD graphics 4000 and NVIDIA GeForce GT 650M. From my previous successful seting I understood that I need to disable the GeForce GT 650M in order to make the external display (connected to the eGPU) work again. Any ideas?
I may be missing something very obvious, went through this super long post but I couldnt find any explicit reference to this issue.
Many thanks to everyone with enough patience to read this, I would really appreciate any suggestions or advice as I am stuck with no eGPU at all now…