MP 1,1-5,1 Manually Configured OpenCore on the Mac Pro

[solaris8x86]

I think intel Mac is about to end in 2026. Each Mac cycle is about 7 years. As the Mac Pro 2019 is the last intel Mac introduced in the end of 2019. Therefore, its EOL is about in 2026 which is only 3-4 year away from now. After that, even the Mac OS to be introduced in 2026 will no longer support to Intel binary. By that time, the Opencore itself will come to an end because Mac OS will be solely support to M-processor only.

Of course, we can still run Linux or Windows with our intel Mac after that. Our Mac will still be useful.

 

[tsialex]

I think intel Mac is about to end in 2026. Each Mac cycle is about 7 years. As the Mac Pro 2019 is the last intel Mac introduced in the end of 2019. Therefore, its EOL is about in 2026 which is only 3-4 year away from now.

2020 5K iMac (iMac20,1 and iMac20,2), released back in August 2020, is the last Intel Mac ever released. Other Macs with Intel were released after 2019 Mac Pro, like 2020 MacBook Pro and Air.

 

[roobarb!]

Evening all, having some fun with my MP5,1 after swapping the NVME drive for a WD model (to avoid the TRIM delay on boot). That may be completely irrelevant, but here's what's happening.

Everything is fine, until I boot into Windows 11 (installed in UEFI mode using instructions from Post 1). As it was installing updates I used the CTRL-ENTER combo in OpenCore to select Windows as the system to boot by default, so I'd not end up back in macOS with every update reboot. However, updates complete I reboot the machine manually from Windows, expecting to see the OpenCore menu... but I was immediately landed back into Windows. Seems as though Windows was now booting directly in UEFI mode, not good.

After the usual roundabout of CTRL-ALT-P-R NVRAM reset to boot Mojave, then boot into Mojave recovery, mount and bless the correct /Volumes/EFI I'm back to OpenCore again. So:

- What have I missed which has apparently allowed Windows to rule the roost?
- How can I tell from Windows whether OpenCore booted the system or not?

I'm sure this wasn't a problem previously. Running OC 0.8.1, config.plist attached. TIA for any pointers.

 

[prefuse07]

I think intel Mac is about to end in 2026. Each Mac cycle is about 7 years. As the Mac Pro 2019 is the last intel Mac introduced in the end of 2019. Therefore, its EOL is about in 2026 which is only 3-4 year away from now. After that, even the Mac OS to be introduced in 2026 will no longer support to Intel binary. By that time, the Opencore itself will come to an end because Mac OS will be solely support to M-processor only.

Of course, we can still run Linux or Windows with our intel Mac after that. Our Mac will still be useful.

It looks like you might try to derail this thread to your topic, so before that happens, please take a look at THIS THREAD

 

[joevt]

- How can I tell from Windows whether OpenCore booted the system or not?

Is there somewhere in Windows to view the SMBIOS changes that OpenCore makes?
https://docs.microsoft.com/en-us/wi...ple-powershell-script-to-query-smbios-locally

 

[cdf]

- What have I missed which has apparently allowed Windows to rule the roost?
- How can I tell from Windows whether OpenCore booted the system or not?

You should be able to look up the spoofed board ID in Windows to verify that Windows indeed booted through OpenCore. I suspect that it did. Perhaps the boot picker just timed out before an image could show up on screen...

 

[molx]

ForceAquantiaEthernet not working on Mac Pro 5,1?

I have a 10Gbit Aquantia ethernet pci card in my mac pro.
It works in Monterey 12.2 with opencore 0.7.7 and the patch below.
But I cannot get it to work in Monterey 12.3 or 12.4 with now opencore 8.1 no matter what I try.
With 'ForceAquantiaEthernet true' the drivers are loaded, but no working network connection can be established.

Anyone running such a card?

Aquantia patch for Monterey 12.2

dict>
<key>Arch</key>
<string>Any</string>
<key>Base</key>
<string></string>
<key>Comment</key>
<string>AQC 107 10 GbE v2</string>
<key>Count</key>
<integer>1</integer>
<key>Enabled</key>
<true/>
<key>Find</key>
<data>
D4TAAgAA
</data>
<key>Identifier</key>
<string>com.apple.driver.AppleEthernetAquantiaAqtion</string>
<key>Limit</key>
<integer>0</integer>
<key>Mask</key>
<data>
</data>
<key>MaxKernel</key>
<string></string>
<key>MinKernel</key>
<string></string>
<key>Replace</key>
<data>
Zg8fRAAA
</data>
<key>ReplaceMask</key>
<data>
</data>
<key>Skip</key>
<integer>0</integer>
</dict>

 

[respect1]

hello to all, I have a small question, since the update of the plist file in 0.8.1, I have the mds_stores which work without stopping, in 0..8.0 I did not have that, it consumes more CPU and after this descent, anyone had this problem? Thank you for your answers.

 

[Gustav Holdoff]

hello to all, I have a small question, since the update of the plist file in 0.8.1, I have the mds_stores which work without stopping, in 0..8.0 I did not have that, it consumes more CPU and after this descent, anyone had this problem? Thank you for your answers.

are you sure? Maybe you just didn't pay attention? I noticed this with BigSur. I have MDS indexing disks for about ten minutes, since there are a lot of them (4 hdd x2tb, 2 nvme ssd x1tb and nvme ssd 2tb, external raid 4tb and time machine 2x8tb are also connected)

 

[respect1]

are you sure? Maybe you just didn't pay attention? I noticed this with BigSur. I have MDS indexing disks for about ten minutes, since there are a lot of them (4 hdd x2tb, 2 nvme ssd x1tb and nvme ssd 2tb, external raid 4tb and time machine 2x8tb are also connected)

yes me too I have a lot of records (8) but before this update I didn't have that, it's very weird, it's not constant but no longer present

 

[respect1]

are you sure? Maybe you just didn't pay attention? I noticed this with BigSur. I have MDS indexing disks for about ten minutes, since there are a lot of them (4 hdd x2tb, 2 nvme ssd x1tb and nvme ssd 2tb, external raid 4tb and time machine 2x8tb are also connected)

thank you for your reply

 

[Gustav Holdoff]

yes me too I have a lot of records (8) but before this update I didn't have that, it's very weird, it's not constant but no longer present

I checked and confirm,
You are right

You can follow the guide to create your custom config for 0.8.1 from scratch, or you can use the "WikiPost History" button to view the update notes for 0.7.7 to 0.7.8, ..., 0.8.0 to 0.8.1. Have an emergency plan in place (back up of ESP, rescue CD, etc.) and make sure to use ocvalidate.

with OC 0.8.1 mds has been indexing continuously for an hour
CPU performance dropped by 20%,
benchmarks: geekbench5 and corona renderer show a drop in cpu performance
also real loading time of control scenes in twinmotion increased by 10-20 percent.
I returned to OC 0.8.0 - performance returned to normal
later I will check, maybe I made a syntax error in the config
but I do the usual procedure, which I know and do almost automatically, so I'm more inclined to believe that the new OC version may have flaws
EDITED
I found what's wrong:
I removed syncretic's patch and inserted NoAVX kext, everything started working as it should
probably the patch does not work correctly with my system disk - Crucial P2 2Tb NVME ssd
(Earlier I resized and shrinked all my Apple_APFS ⁨Containers by 10% on each disk to get a raw unallocated space)

My CMP 4.1/5.1 dual cpu xeon 5680, 96gb RAM, radeonpro w5700, nvme crucial 2tb, evo plus 1tb, TB3 alpine ridge, oc 0.8.0, mac os 12.4

 

[respect1]

I checked and confirm,
You are right


with OC 0.8.1 mds has been indexing continuously for an hour
CPU performance dropped by 20%,
benchmarks: geekbench5 and corona renderer show a drop in cpu performance
also real loading time of control scenes in twinmotion increased by 10-20 percent.
I returned to OC 0.8.0 - performance returned to normal
later I will check, maybe I made a syntax error in the config
but I do the usual procedure, which I know and do almost automatically, so I'm more inclined to believe that the new OC version may have flaws
View attachment 2020873
My CMP 4.1/5.1 dual cpu xeon 5680, 96gb RAM, radeonpro w5700, nvme crucial 2tb, evo plus 1tb, TB3 alpine ridge, oc 0.8.0, mac os 12.4

I just return to 0.8.0, same thing

 

[Gustav Holdoff]

I just return to 0.8.0, same thing

Yes, mds starts, but mds works for me no more than 10 minutes.
check config 0.8.0 if the cdf instructions for the ASPP kext and for syncretic's patch are correctly followed.
second, may be you need to set another TRIM timeouts, as recommended earlier by tsialex, in case you have a problem with nvme ssd indexing?
For me, I came back from vacation and when I turned on MacBookAir 2017 (but with Cricial P2 nvme ssd) in Monterey,
I suddenly got a Hash Mismatch, with random restart- but MBAir 2017 is natively supports Monterey!
Yesterday I erased the macbookair disk, totally reformatted
and all day, downgraded to BigSur, and restored the working environment using Migration Assistant.
so I can assume that something similar can happen to you.
as to cMP, now I work on cdf 0.8.0 config, just replaced the whatevergreen with a new one (from 0.8.1)
the difference in 0.8.0 with new WhatEverGreen in the GB5 benchmark for open cl and in Blackmagic RAW appeared
old- 58747- new 61852 (but metal scores are the same)

 

[respect1]

Yes, mds starts, but mds works for me no more than 10 minutes.
check config 0.8.0 if the cdf instructions for the ASPP kext and for syncretic's patch are correctly followed.
second, may be you need to set another TRIM timeouts, as recommended earlier by tsialex, in case you have a problem with nvme ssd indexing?
For me, I came back from vacation and when I turned on MacBookAir 2017 (but with Cricial P2 nvme ssd) in Monterey,
I suddenly got a Hash Mismatch, with random restart- but MBAir 2017 is natively supports Monterey!
Yesterday I erased the macbookair disk, totally reformatted
and all day, downgraded to BigSur, and restored the working environment using Migration Assistant.
so I can assume that something similar can happen to you.
as to cMP, now I work on cdf 0.8.0 config, just replaced the whatevergreen with a new one (from 0.8.1)
the difference in 0.8.0 with new WhatEverGreen in the GB5 benchmark for open cl and in Blackmagic RAW appeared
old- 58747- new 61852 (but metal scores are the same)
View attachment 2021131

thank you so much for your help, I think I waiting the next paquet from Martin for the update, now with 0.8. it"s ok.

 

[Gustav Holdoff]

The guide has been updated to OpenCore version 0.8.1. Also included are instructions for AVXpel. Happy updating!

Since there are several strategies described in the introduction to Your post for OpenCore,
in particular the Martin's strategy differs from the one you proposed.
Maybe it will be useful for cdf and for h9826790
Depending on my mood, I use either the cdf config, or the Martin's config,
and I can share my own experience:
Starting from BigSur, I began to notice a visual difference in how my desktop looks, so after the system boots I can immediately see which opencore bootloader I have (of course, I don’t mean that I see the difference, because the opencore icons themselves are different for your configs)
Martin's strategy leads to a clearer and brighter screen picture and sharper fonts (in programs too)
But the cdf strategy shows the best score in the "geekbench5 compute" benchmark and in Heaven benchmark
below is a comparison of benchmarks - those that are better are cdf (61852 and 69071)- the others are Martin's

In real tasks (Twinmotion, UE5, Archicad) I don't see the difference
In the cdf config, I seem to understand more what is being done and why it is being done
Martin's config is more incomprehensible to me, except for the general sections: kexts and patches for which I read a lot in discussions are clear to me
But I always use Martin's config for updates because of the quick and clear strategy for editing the config for updates (with cdf I need to make more changes and it is less convenient to do it in a text editor).
The other day I tried OCLP for my macbooks, first I patched the system, and then I began to study the resulting config.
there is also a lot of things in the config that are unfamiliar to me
And since I don’t understand the OCLP strategy at all, I’ll refrain from using it for the time being because of the automatic and incomprehensible prescribing of patches in my system.
Especially if something goes wrong, then I will not be able to find the reason
and since I am not a developer, but a user, then probably in patches I will not understand anything for a very long time
So open core is more understandable and enjoyable for me
My CMP 4.1/5.1 dual cpu xeon 5680, 96gb RAM, radeonpro w5700, nvme crucial 2tb, evo plus 1tb, TB3 alpine ridge, oc 0.8.0, mac os 12.4

 

[cdf]

Martin's strategy leads to a clearer and brighter screen picture and sharper fonts (in programs too)
But the cdf strategy shows the best score in the "geekbench5 compute" benchmark and in Heaven benchmark

Thanks for sharing your experience. I'm not sure what could be causing these discrepancies. Perhaps it's the spoofed model.

 

[Gustav Holdoff]

Thanks for sharing your experience. I'm not sure what could be causing these discrepancies. Perhaps it's the spoofed model.

The cause found:
I have 2 monitors (Samsung 70" 4k tv) connected one through the display port to hdmi (plus a mirror through USB-C to hdmi), and the other mini display port to hdmi.
when loading windows, the opencore picker appears in the first input, the windows loading window is in first input,
then a black screen appears and in order to see the login window, I have to switch to the second input.
in Windows there is no difference in resolution,
but in macos, despite the fact that the resolution of both inputs is written 3840x2160,
in reality it seems that the second input gives out less than what is written
Some versions of opencore in the cdf edition behaved like this: the window with the opencore picker appeared on the first input, and the login window on the second, where I remained to work
that's why my erroneous conclusion arose

 

[_yesman_]

Having trouble installing OCLP on my cMP 5,1 with non-flashed rx580. In the OCLP troubleshooting guide, it says:

Booting with a non-flashed GPU​

For Mac Pro, Xserve and iMac users with non-flashed GPUs, you can still easily boot OpenCore and view the entire boot process. To do so, make sure SIP is disabled(to allow NVRAM write access) and run the following:

sudo bless --verbose --file /Volumes/VOLNAME/EFI/OC/OpenCore.efi --folder /Volumes/VOLNAME/EFI/OC --setBoot

• Note you will need to replace VOLNAME with the Volume name of your USB or hard drive with OpenCore
• Note 2: Once done, you can re-enable SIP
• Note 3: The EFI partition OpenCore was installed on must be mounted, if it's unmounted simply rerun "Install OpenCore to drive"

So, I'm trying to bless the OC EFI partition (on my disk 'Untitled') manually, right? But it's not working for me. I get:

EFI found at IODeviceTree:/efi
No mount point for /Volumes/Untitled/EFI/OC
Can't determine mount point of '/Volumes/Untitled/EFI/OC' and ''

I did follow the note 3 advice, and still no joy. What am I doing wrong?

 

[NC12]

Having trouble installing OCLP on my cMP 5,1 with non-flashed rx580. In the OCLP troubleshooting guide, it says:



So, I'm trying to bless the OC EFI partition (on my disk 'Untitled') manually, right? But it's not working for me. I get:

EFI found at IODeviceTree:/efi
No mount point for /Volumes/Untitled/EFI/OC
Can't determine mount point of '/Volumes/Untitled/EFI/OC' and ''

I did follow the note 3 advice, and still no joy. What am I doing wrong?

You could try running the bless opencore tool from Martins Package

 

[cdf]

Having trouble installing OCLP on my cMP 5,1 with non-flashed rx580. In the OCLP troubleshooting guide, it says:



So, I'm trying to bless the OC EFI partition (on my disk 'Untitled') manually, right? But it's not working for me. I get:

EFI found at IODeviceTree:/efi
No mount point for /Volumes/Untitled/EFI/OC
Can't determine mount point of '/Volumes/Untitled/EFI/OC' and ''

I did follow the note 3 advice, and still no joy. What am I doing wrong?

Try blessing from recovery mode. See Basic Setup > First Boot in post #1. These instructions have also recently been added to the OCLP docs with pictures:

OpenCore-Legacy-Patcher/docs/BOOT.md at main · dortania/OpenCore-Legacy-Patcher

Experience macOS just like before. Contribute to dortania/OpenCore-Legacy-Patcher development by creating an account on GitHub.

github.com

 

[_yesman_]

Try blessing from recovery mode. See Basic Setup > First Boot in post #1. These instructions have also recently been added to the OCLP docs with pictures:

OpenCore-Legacy-Patcher/docs/BOOT.md at main · dortania/OpenCore-Legacy-Patcher

Experience macOS just like before. Contribute to dortania/OpenCore-Legacy-Patcher development by creating an account on GitHub.

github.com

OK Thanks, I'll give that a try.

 

[EdMun]

On this MP3,1 with APFS firmware transient boot always defaults to the first SATA HFS volume starting from bay #1, not necessarily bay #1 as common wisdom would have me believe and peculiar if not impossible.

Edit: I have OC under RefindPlus on a HFS data SSD default boot, unblessed. Am I missing something or is my config queer?

 

[6DecadesLater]

I have upgraded to OC 0.8.1 and then did an update to Monterey 12.4 using option #2 from post #1. Monterey update went smoothly - all seems good.
When checking my nvram OC version it comes up unknown. Also lost the graphic boot picker. I did an nvram reset (3 chimes including first one) but still OC unknown. I have had that before so not concerned. Perhaps a 5 chime reset will work but unless I see any odd behaviour I am Ok with it as booting into Mojave takes a long time to rebless. I like the graphic boot picker but can live with the text version - hardly ever reboot anyway but may look at the config sometime when I have a moment.
Everything seems to work fine, acceleration, etc.

As always, the effort that has been put into this is greatly appreciated.

 

[cdf]

When checking my nvram OC version it comes up unknown.

Maybe you have ExposeSensitiveData=0. Note that unless the second bit of ExposeSensitiveData is set, it's normal to see UNK-000-0000-00-00 for the version.

Also lost the graphic boot picker.

Depending on which OC version you updated from, you may need to update the OpenCanopy files.