Contribute
Register

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

Joined
Jul 22, 2016
Messages
301
Motherboard
Gigabyte Z690 Aero G
CPU
i9-12900K
Graphics
RX 580
Mac
  1. iMac
  2. MacBook Pro
  3. Mac mini
  4. Mac Pro
Mobile Phone
  1. iOS
I just upgraded my test OC drive to OC 7.0 as I want to test this vector acceleration quirk BigAl is raving about.
On first boot, I got stuck with the error: hfsplus.efi at 4 cannot be found
So I rebooted and had a look at these. There's an openhfs whatever .efi, but no hfsplus.efi. So I copied hfsplus.efi over from my OC 0.6.8 EFI, and now it boots OK.
Is this a plain bad idea? I guess this openhfs thing is the successor of hfsplus? How can I make this work?
BTW, I am still on Mojave, but I guess that doesn't make a difference for a bootloader.

@zipb

Several versions of OC back Openhfs.efi was introduced in place of HfsPlus.efi by the OC developers in their base EFI folder. It caused problems for some. Since then @CaseySJ's guide EFI folders have been using the former HfsPlus.efi file which has been working fine for most. I've been using HfsPlus.efi without problems for months. There should be no problems using it.
 

CaseySJ

Moderator
Joined
Nov 11, 2018
Messages
18,934
Motherboard
Asus ProArt Z690-Creator
CPU
i7-12700K
Graphics
RX 6800 XT
Mac
  1. MacBook Air
  2. MacBook Pro
  3. Mac Pro
Classic Mac
  1. Quadra
Mobile Phone
  1. iOS
@zipb,

As @NCMacGuy pointed out, Acidanthera have been developing and releasing their OpenHfsPlus.efi, but it has caused a number of problems lately. I’m sure this driver will improve over time, but for now we recommend only Apple’s own HfsPlus.efi. The OpenCore folder that is provided in this thread does not use OpenHfsPlus.efi.
 
Joined
Jan 5, 2019
Messages
119
Motherboard
Gigabyte Z390 Designare
CPU
i7-8700K
Graphics
RX 580
For a number of months now, when I leave my Mac sleeping overnight or longer, the next time I login there is a resport the system restarted and the error is "Sleep Wake failure in EFI". It’s running latest Big Sur and OC config. In fact I wiped and replaced the EFI directory with a new copy that is using older serial info. Any things on the source/fix for this?

Sleep Wake failure in EFI.
 
Joined
Aug 25, 2012
Messages
1,167
Motherboard
Gigabyte Z390 Designare
CPU
i9 9900k
Graphics
RX480
Mac
  1. MacBook Pro
Classic Mac
  1. Performa
  2. Power Mac
  3. PowerBook
  4. Quadra
Mobile Phone
  1. iOS
@zipb,

As @NCMacGuy pointed out, Acidanthera have been developing and releasing their OpenHfsPlus.efi, but it has caused a number of problems lately. I’m sure this driver will improve over time, but for now we recommend only Apple’s own HfsPlus.efi. The OpenCore folder that is provided in this thread does not use OpenHfsPlus.efi.
I discovered your OC7.0 folder only after I did the upgrade, so I had to deal with Openhfsplus...
This gargantuan Z390 thread is amazing, but not a model of easy navigation.
Allright. I'll get rid of it. Back to OC6.8, I don't see anything in OC7.0 worth having. Except that my boot screen is the right size until about 2/3 through, then I get a blown up login screen, then after typing my pwd, the right resolution comes back. Didn't have that in Clover.
 
Joined
Jul 22, 2016
Messages
301
Motherboard
Gigabyte Z690 Aero G
CPU
i9-12900K
Graphics
RX 580
Mac
  1. iMac
  2. MacBook Pro
  3. Mac mini
  4. Mac Pro
Mobile Phone
  1. iOS
Allright. I'll get rid of it. Back to OC6.8, I don't see anything in OC7.0 worth having. Except that my boot screen is the right size until about 2/3 through, then I get a blown up login screen, then after typing my pwd, the right resolution comes back. Didn't have that in Clover.
@zipb

No problem using OC 0.7.0 and simply installing the HfsPlus.efi file and activating it in config.plist.

HfsPlus.efi.jpg

 
Joined
Aug 22, 2017
Messages
408
Motherboard
Z390 Designare
CPU
i9-9900k
Graphics
RX 580
Mac
  1. iMac
Mobile Phone
  1. iOS
In the process of investigating whether I should update to Big Sur, I ran across a lot of banter regarding AppleVTD. Correct me if I'm wrong, but several folks here have mentioned that it's something that is 'required' to have a working Antelope audio interface with the Z390 board. I myself have had a fair share of Thunderbolt issues when trying to get my Slate VRS8 going a while back, a process I documented in this post with references to other posts detailing some of the work. While I can't comment on the TB3 to Ethernet working (I don't own the adapter), I can say that using the lessons learned on the Slate VRS8, I built a system identical to mine for a friend with a fully working Antelope Descrete 4. He has been running it ever since without any issues, including all the Mojave updates from 10.14.2 - 10.14.6. We are both still running Mojave. My VRS8 and UAD Apollo X4 still work perfectly. His Descrete 4 also works perfectly. He is still on F6 BIOS, running Clover r5117 with the following boot flags:
Screen Shot 2021-07-03 at 3.32.17 PM.png

I, on the other hand recently implemented OC 0.7.0 and both of my devices, Slate VRS8 and Apollo X4 work. So my question is: what again is the point of pursuing and implementing AppleVTD?
 

CaseySJ

Moderator
Joined
Nov 11, 2018
Messages
18,934
Motherboard
Asus ProArt Z690-Creator
CPU
i7-12700K
Graphics
RX 6800 XT
Mac
  1. MacBook Air
  2. MacBook Pro
  3. Mac Pro
Classic Mac
  1. Quadra
Mobile Phone
  1. iOS
@luckyal,

The answer to your question is simple. It’s a good question that many will ask, but it also has a very simple answer:

If all the devices you need to use are working properly, then there’s nothing else to do. The old adage, if it ain’t broke, don’t fix it totally applies here.

On the other hand, Experiment Update #3 provides reasons for making this change — but only for those affected.
 
Joined
Dec 13, 2013
Messages
117
Motherboard
Gigabyte Z390 Designare
CPU
i9-9900K
Graphics
RX 6900 XT
Mac
  1. MacBook Pro
  2. Mac mini
  3. Mac Pro
Mobile Phone
  1. iOS
OK, I've been having a really weird issue with OC0.7.0 when after my monitors go to sleep, my main monitor won't wake up. My hack is still awake/running, just the monitors have gone to sleep. Wiggle the mouse, hit the keyboard, everything wakes up, but not my main Dell. If I let my hack go completely to sleep and then wake it up, then both monitors wake just fine. It's really confusing. I'm running Big Sur 11.4, iMacPro1,1 with a RX 6900 XT attached to a Dell U3417W on DP and a 1024x600 stats display on HDMI. Again, this only happens with OC0.7, so I've reverted to 0.6.9 and now everything works normally again. Thoughts? My rig is a Z390 Designare with an i9-9900K and 128GB of 2600MHz RAM running the official F9i BIOS.

1625360415418.png
 
Joined
Aug 22, 2017
Messages
408
Motherboard
Z390 Designare
CPU
i9-9900k
Graphics
RX 580
Mac
  1. iMac
Mobile Phone
  1. iOS
If all the devices you need to use are working properly, then there’s nothing else to do. The old adage, if it ain’t broke, don’t fix it totally applies here.

On the other hand, Experiment Update #3 provides reasons for making this change — but only for those affected.
Fair enough. I don't plan to make any changes on my own system but rather trying to be helpful to those that have Antelope interfaces but have been unable to use them. The default instructions on page one suggest that VT-d should be "Disabled" in BIOS. Though I understand why, I don't necessarily agree because there's a good reason why people bought Designare in the first place. That is to have a fully functioning (or as close to a real mac as possible) Thunderbolt. Unlike UAudio's Apollos which use onboard DSP to achieve low latency, most of Thunderbolt audio interfaces out there, require direct access to memory (DMA) typically reserved for virtualization (VT-d). I have helped a good half dozen people setting this up for a range of audio interfaces, from MOTU, Slate, Presonus and others.

I believe (I could be wrong) that in order to have a fully working Thunderbolt implementation, the following need to be true:

  1. Enable VT-d in BIOS (assuming Thunderbolt devices are already enabled with 'no security' in BIOS)
  2. Remove dart=0 (Boot>Arguments in Clover / NVRAM>Add>boot-args in OC)
  3. Disabled SIP (if the installed driver doesn't produce a security prompt to allow driver to function)
Recently I have been looking at building my own multi 10G switch for my homelab and some of the things this guy talks about, with respect to DMA and specific BIOS configurations, made a lot of sense and reminded me of the reasons why I had previously struggled to get my VRS8 going. It has to do with latency. Give this article a look as well and focus on the "Bounced Buffer" section.
 
Top