[SUCCESS] Gigabyte Designare Z390 (Thunderbolt 3) + i7-9700K + AMD RX 580

[snoby]

I've been trying for a while, I have a Clover Configuration running, but having difficulties getting it working with OpenCore 0.5.6. Does anyone have a working OpenCore configuration running? The mini guide at the beginning of the tread always boots to a circle with a slash through it.

 

[CaseySJ]

First of all a big thank you to you and your work!
My whole build is based almost entirely on your tutorials and guides. Thanks a lot for that.

Yesterday I made a fresh installation of Catalina 10.15.4. (I followed your tutorials and guides). Everything works perfectly, except for the problem with the mouse pointer:

• Cursor doesn't move smoothly, has frequent dropouts
• The scrolling behaves strangely, also here there are dropouts
• dropouts are quite sporadic but very frequent

I mainly use the Apple Magic Mouse 2 and Apple Magic Keyboard 2, but the problem also exists with the Logitech G602 mouse. The problems started already with 10.15.3, but not directly after installation. (directly after the installation of 10.15.3 everything was fine)

For Bluetooth I have switched off the original adapter (in the BIOS) and use the fenvi T919 WiFi card.

(I also screwed the AMD Radeon RX 5700 XT vertically with a riser cable, the fenvi T919 is under the riser cable. I don't know if this has anything to do with it, but subjectively this problem only occurred after the conversion to vertical GPU)

What could be the cause?

Edit:
Sometimes, when I leave the mouse alone, the cursor moves itself (in the range of about 1-2cm)

• Does this problem occur after a clean installation of 10.15.4 (no third-party apps or drivers)?
• Does this problem occur if all other USB devices are disconnected (except keyboard and mouse)?
• Does this problem occur if you temporarily disable the 2.4GHz WiFi band on your router?

 

[bugsyraz]

The IORegistryExplorer screenshot does not look correct. Under RP05 you should see the following (in the absence of any connected Thunderbolt device):

View attachment 460305

Suggestions:

1. Do not use the "search" field in IORegistryExplorer. Just scroll the device tree on left side until you see RP05. Do you still see a reduced list of devices compared with screenshot above?
2. Confirm that CLOVER/ACPI/patched no longer contains the file: SSDT-Z390-DESIGNARE-TB3HP-V4.aml
3. Re-read the Winbond chip with Raspberry Pi three times, compute checksums three times, and see if they match the checksum of DESIGNARE-Z390-NVM33-Elias64Fr.bin (shown below)

% shasum DESIGNARE-Z390-NVM33-Elias64Fr.bin
edbbe3cbf8e3fa4a9d991e0681f2a5702b248224  DESIGNARE-Z390-NVM33-Elias64Fr.bin

There was an issue with the GND pin that wasn't connected. Now I soldered it back and it should be working.

Only problem is - I can't start OSX anymore after the BIOS reset happened.

Gets stuck here
"End RandomSeed
++++++++++++++++"

Any ideas?

 

[CaseySJ]

@CaseySJ why was it decided to drop support for n-d-k OC builds? I quite liked their solution for dual booting Windows

It's just a matter of time management and priorities. OpenCore evolves quickly and parameter names get changed frequently. OpenCore is in Alpha stage (version number begins with 0.x.y) and I do not want to encourage widespread adoption at this time.

In Clover we can modify many aspects of config.plist right from the Clover Boot Menu, which is a significant boon for troubleshooting. This alone tips the balance significantly in favor of Clover. Try to see this from my point of view. If you had to support a large community of users, would you prescribe Clover or OpenCore?

It would be kind of crazy to prescribe OpenCore (for large numbers of people), much less two flavors of OpenCore. This is why the first sentence in the OpenCore Mini-Guide states, "Do not do this if you're new to Hackintoshing." However, experienced users can support n-d-k OpenCore by themselves.

 

[CaseySJ]

I've been troubleshooting this problem and I've been checking the GND pin on the chip, maybe there was a soldering issue there. Tested the ground loop with another GND on the case, looks like there's an interruption there.

Checked the GND pin of the BIOS chip after that, ground loop works.

After that, BIOS has reset and after reverting the BIOS settings to the ones you specified in the guide, OSX won't start anymore and gets stuck at
"End RandomSeed
++++++++++++++++"

Any idea what might be causing this?

After a CMOS Reset it's likely that MSR 0xE2 has been re-locked. Let's check that first.

 

[bugsyraz]

After a CMOS Reset it's likely that MSR 0xE2 has been re-locked. Let's check that first.

Yes, MSR 0xE2 was re-locked indeed. Fixed that and it's still not booting.

Disabled Thunderbolt from BIOS, system boots. Thunderbolt Bus in system information shows the same thing, even with discrete Thunderbolt disabled in BIOS.

 

[CaseySJ]

Yes, MSR 0xE2 was re-locked indeed. Fixed that and it's still not booting.

Disabled thunderbolt from BIOS, system boots. Thunderbolt Bus in system information shows the same thing, even with discrete thunderbolt disabled in BIOS.

Ouch, it looks like a bad firmware flash. Let's read the Winbond chip and compare checksum with the Elias firmware file.

 

[bugsyraz]

Ouch, it looks like a bad firmware flash. Let's read the Winbond chip and compare checksum with the Elias firmware file.

Actually after reactivating it in BIOS, it ended up booting. TB Bus shows the same. Will re-read the chip and do the checksum.

 

[CaseySJ]

Actually after reactivating it in BIOS, it ended up booting. TB Bus shows the same. Will re-read the chip and do the checksum.

Does IORegistryExplorer still show the same short list of devices under RP05?

 

[c0c0p0ps]

Hey @CaseySJ I have been following all the Thunderbolt developments keenly for the last few weeks while recovering from Coronavirus and what you and the other guys on here have achieved is pretty amazing. My build is based off the @Stork Asus Maximus Hero X so I never thought Thunderbolt was going to be a possibility until I saw what first DSM then @Elias64Fr achieved with the firmware.

So... I have managed to flash the TitanRidge card with the patched NVM23 from your repository, jumped the header cable and am very pleased to say that I now boot up with a Thunderbolt tree using the SSDT files in the TitanRidge DROM thread. I had to replace RP05 for RP21 for my system, and modified the DROM and seem to have port 7 showing up. I also need to include the SSDT-DTPG in my patched folder otherwise link speed is only 20gb/s. With that DTPG I get a full 40gb/s. Is this a file that can simply be placed in or should it also be modified to be system specific?

Tests so far - If I cold boot my system with nothing connected then I get no hot plug on either port, and nothing seems to connect. Rebooting the system with a Thunderbolt 3 dock connected then gives me full hot plug that seems to be working on both Thunderbolt ports on the Titan Ridge. Cold booting the system with the dock connected shows nothing as connected when reaching desktop, however unplugging and replugging the cable then connects without a reboot. I have no other Thunderbolt devices to test, unfortunately, but connecting a series of USB devices to the dock they all correctly mount and populate in the USB device tree. I currently have no USB header cable connected internally and no power supply cables.

One slightly strange thing I have noticed is that the PCI device tree under system information is identifying the Thunderbolt controllers as Slot-4 which seems more like something that would be for a MacPro than a Hackintosh. Was this SSDT modified from the DSM2 / Nico files that were released maybe? Is this something that should potentially be edited or corrected in the SSDT or is it purely cosmetic? I did previously populate my PCI information from Hackintool but have now deleted everything until I sort out what is best for the Thunderbolt. Do you have any thoughts on this?

I have attached a save of my IOreg and modified SSDT if you possibly had the time to look it over and let me know if I need to correct anything further for my system. I am happy to share anything else that might be useful such as Clover or anything you recommend.

I still need to do more testing, especially for things like sleep which I have never had issues with on my build. Right now though I am pretty amazed at how well it seems to be working on a board that was never built to have Thunderbolt so again a huge thanks to you for the guides, and everyone who has made this possible!