Z490 & Z590 - Will Z590 ever have macOS Support ?

[dehjomz]

I have an MSI MEG Z590 ACE with an i9-11900K and OpenCore version 0.6.9. Most things seem to work except for sleep and Thunderbolt 4. Audio appears to be internal USB so no problems there. Intel Bluetooth seems to work with no kexts at all. No Airdrop, though. Wifi is Intel ant not working, but I don't need that. I still have to thinker with it, but seems very stable as it is right now. You do have to use SMBIOS 7,1 or it will not boot,. If anyone can help with Thunderbolt 4 and sleep, I will certainly appreciate it

If I remember correctly, your Ace's Thunderbolt controller is at root port #1, so for Thunderbolt hot plugging you can just use the Asus HERO hot plug SSDT, and change from RP05 to RP01. To find this SSDT you can go to https://hackindrom.zapto.org and go to the drop down menu for the HERO's SSDT and adapt it to your Ace. Use IORegistryExplorer and look for which root port your Thunderbolt controller's NHIO is located.

 

[iaksakkah]

If I remember correctly, your Ace's Thunderbolt controller is at root port #1, so for Thunderbolt hot plugging you can just use the Asus HERO hot plug SSDT, and change from RP05 to RP01. To find this SSDT you can go to https://hackindrom.zapto.org and go to the drop down menu for the HERO's SSDT and adapt it to your Ace. Use IORegistryExplorer and look for which root port your Thunderbolt controller's NHIO is located.

That's great info. I will work on it today, and I will post results later. Thanks for the reply.

 

[c-o-pr]

Had a disquieting experience with my build...

Asus released a production BIOS update. The installation said it was done, then the system went into a POST loop. It was late and I wasn't at my sharpest, so I just tried resetting and it came up then it said "please wait updating the BIOS". And I thought, didn't it just already update, but there was nothing I could do.

BTW—This new Asus BIOS version 0707 doesn't just cement the previous beta 0704, it has very significant changes to part of the UI layout, which I find surprising and inappropriate. To me a release following a beta should not fundamentally change the function nor presentation except to account for errors discovered in the beta. But Asus has their own way. My point is I have no idea what to expect as far as potentially detrimental changes, so I'm feeling a bit unnerved. Also, due to pitfalls in OC, I have been acclimated to the system sometimes crashing during macOS boot, for reasons of stuff like CFG lock, which even though patches shouldn't be needed. If one of the patches is disabled, macOS will sometimes crash once at boot, then start properly the next time around. (Huge aside, I'm sure I do not understand the nuances of BIOS POST and how config is managed underneath the OS. I'm pretty sure more is going on down there then I would prefer. Such as what exactly does it mean when Asus reports "The system POSTed in safe mode" then I hit continue... What mode is it in when it continues? Try to fiind out from Asus web site. Also NVRAM... What's going on in there? But this is another story.)

The system seemed to be working again, and I went looking for answers by double-checking all my BIOS config, which uses nothing fancy. For example macOS will crash at boot if Thunderbolt > Native OS Security is enabled.

It turns out the BIOS update threw away my saved profile, and when I had thought I saved a copy of the config to a flash drive, it appeared to complete but did something else rather than save, and I realized later I had used the save control wrong. So I had to review and restore all settings by hand.

I got the BIOS config worked out and booted into Big Sur and ran some benchmarks to see if things were set and verify cooling performance. Results were a little shy of my previous config, and I decided to play with RAM and CPU power tuning to see if I could get back to where I was. This tuning went well, and I left off within a hair's width of my best runs, and top-end performance according to web.

Today I thought I'll push RAM frequency a little more and see what happens, try to beat my best. It immediately got unstable so I upped voltage 0.15V which is still well within norms. From here I went into POST loop again. But, when I reset CMOS, it hung on black screen (the display receiving a signal but black) in POST. I'm not in OpenCore, yet. So I think I've maybe not reset properly and start to wonder about this new BIOS. Maybe Asus has messed up because they sure changed things. From here everything becomes completely unreliable. I pull my GPU and my add-on cards, and, when I do, I can get back into the BIOS. The post codes as various forms of "undefined" and "started IDE config". I reset CMOS, and it just loops and hiccups. Then, while I'm looking up how to reflash the old BIOS, it comes back up into the BIOS.

I decide I've had it with new BIOS 0707, so I prep a flash drive and try Asus "Flashback". I think maybe Asus really wants its customers to appreciate this feature—I mean why do they put it in there if they aren't gonna break something?! So the flashing appears to go fine, then I reset, and it bounces once then says "Updating the BIOS" like it did before. So I figure OK this is just normal behavior: Flashing copies the BIOS into a fresh region, then reset checks it can flips the config over. (Srsly BIOS is as complicated as Mac OS System 7 was back in day.)

The system comes right out of POST, then tries to head off to boot Ubuntu, which is also attached to this box by SATA. I think WOW! OK, it's forgotten its boot config. So I go to try to boot, select the NVMe Rocket, but it is nowhere to be found. I look at the BIOS NVMe config and the device is listed, but as size 0.0. I reset again, and this time BIOS stops at POST and says Sabrent Rocket "NVMe failure immanent backup your data".

I pull the Rocket out and put it in a Sabrent type-C enclosure and plug it in and zilch. The case gets really hot really fast with no status LED showing. So I unplug, review, and it's dead, Jim.

Yeblarrrg

So I guess the POST problem was the drive going out. I have never seen an SSD / system collapse this way.

The other day, I also had the presence of mind to make a full system backup onto a EVO 970 NVMe, so I dropped this in and fixed up the EFI part and the system is up and working fine on it.

I submitted a ticket to Sabrent.

So question:
Can a BIOS upgrade kill a drive? Should I engage Asus on this too?

TIA

 

[dehjomz]

Had a disquieting experience with my build...

Asus released a production BIOS update. The installation said it was done, then the system went into a POST loop. It was late and I wasn't at my sharpest, so I just tried resetting and it came up then it said "please wait updating the BIOS". And I thought, didn't it just already update, but there was nothing I could do.

BTW—This new Asus BIOS version 0707 doesn't just cement the previous beta 0704, it has very significant changes to part of the UI layout, which I find surprising and inappropriate. To me a release following a beta should not fundamentally change the function nor presentation except to account for errors discovered in the beta. But Asus has their own way. My point is I have no idea what to expect as far as potentially detrimental changes, so I'm feeling a bit unnerved. Also, due to pitfalls in OC, I have been acclimated to the system sometimes crashing during macOS boot, for reasons of stuff like CFG lock, which even though patches shouldn't be needed. If one of the patches is disabled, macOS will sometimes crash once at boot, then start properly the next time around. (Huge aside, I'm sure I do not understand the nuances of BIOS POST and how config is managed underneath the OS. I'm pretty sure more is going on down there then I would prefer. Such as what exactly does it mean when Asus reports "The system POSTed in safe mode" then I hit continue... What mode is it in when it continues? Try to fiind out from Asus web site. Also NVRAM... What's going on in there? But this is another story.)

The system seemed to be working again, and I went looking for answers by double-checking all my BIOS config, which uses nothing fancy. For example macOS will crash at boot if Thunderbolt > Native OS Security is enabled.

It turns out the BIOS update threw away my saved profile, and when I had thought I saved a copy of the config to a flash drive, it appeared to complete but did something else rather than save, and I realized later I had used the save control wrong. So I had to review and restore all settings by hand.

I got the BIOS config worked out and booted into Big Sur and ran some benchmarks to see if things were set and verify cooling performance. Results were a little shy of my previous config, and I decided to play with RAM and CPU power tuning to see if I could get back to where I was. This tuning went well, and I left off within a hair's width of my best runs, and top-end performance according to web.

Today I thought I'll push RAM frequency a little more and see what happens, try to beat my best. It immediately got unstable so I upped voltage 0.15V which is still well within norms. From here I went into POST loop again. But, when I reset CMOS, it hung on black screen (the display receiving a signal but black) in POST. I'm not in OpenCore, yet. So I think I've maybe not reset properly and start to wonder about this new BIOS. Maybe Asus has messed up because they sure changed things. From here everything becomes completely unreliable. I pull my GPU and my add-on cards, and, when I do, I can get back into the BIOS. The post codes as various forms of "undefined" and "started IDE config". I reset CMOS, and it just loops and hiccups. Then, while I'm looking up how to reflash the old BIOS, it comes back up into the BIOS.

I decide I've had it with new BIOS 0707, so I prep a flash drive and try Asus "Flashback". I think maybe Asus really wants its customers to appreciate this feature—I mean why do they put it in there if they aren't gonna break something?! So the flashing appears to go fine, then I reset, and it bounces once then says "Updating the BIOS" like it did before. So I figure OK this is just normal behavior: Flashing copies the BIOS into a fresh region, then reset checks it can flips the config over. (Srsly BIOS is as complicated as Mac OS System 7 was back in day.)

The system comes right out of POST, then tries to head off to boot Ubuntu, which is also attached to this box by SATA. I think WOW! OK, it's forgotten its boot config. So I go to try to boot, select the NVMe Rocket, but it is nowhere to be found. I look at the BIOS NVMe config and the device is listed, but as size 0.0. I reset again, and this time BIOS stops at POST and says Sabrent Rocket "NVMe failure immanent backup your data".

I pull the Rocket out and put it in a Sabrent type-C enclosure and plug it in and zilch. The case gets really hot really fast with no status LED showing. So I unplug, review, and it's dead, Jim.

Yeblarrrg

So I guess the POST problem was the drive going out. I have never seen an SSD / system collapse this way.

The other day, I also had the presence of mind to make a full system backup onto a EVO 970 NVMe, so I dropped this in and fixed up the EFI part and the system is up and working fine on it.

I submitted a ticket to Sabrent.

So question:
Can a BIOS upgrade kill a drive? Should I engage Asus on this too?

TIA

Sorry to hear that man. That BIOS update killed my one of friend's Thunderbolt ports. It only works with USB-c hosts now, but not Thunderbolt. Flashing back doesn't help.

Flashing a BIOS shouldn't kill peripherals or ports. It's never happened to me on Gigabyte or MSI...

 

[Middleman]

c-o-pr,

I'm sorry to hear that but it sounds like a corrupted BIOS upload there to me. What you should do is try using Memtest+ on a USB stick and run it and see how your system performs. More than likely it may experience corruption. In which case it might be advised that you simply reflash the BIOS. Also make sure your USB media (used for the firmware flash) is also not corrupted.

Sorry to hear that man. That BIOS update killed my one of friend's Thunderbolt ports. It only works with USB-c hosts now, but not Thunderbolt. Flashing back doesn't help.

Flashing a BIOS shouldn't kill peripherals or ports. It's never happened to me on Gigabyte or MSI...

Dehjomz, has your friend tried to flash their Thunderbolt via the Windows Intel TB drivers?

 

[dehjomz]

c-o-pr,

I'm sorry to hear that but it sounds like a corrupted BIOS upload there to me. What you should do is try using Memtest+ on a USB stick and run it and see how your system performs. More than likely it may experience corruption. In which case it might be advised that you simply reflash the BIOS. Also make sure your USB media (used for the firmware flash) is also not corrupted.



Dehjomz, has your friend tried to flash their Thunderbolt via the Windows Intel TB drivers?

As far as I know, the Windows Thunderbolt drivers don’t have the NVM firmware necessary for flashing Maple Ridge. Might you have more information on this?

 

[Middleman]

As far as I know the Windows Thunderbolt drivers don’t have the NVM firmware necessary for flashing Maple Ridge. Might you have more information on this?

I didn't realize it was for Maple Ridge. If it was for the earlier versions, perhaps it could've worked because Intel did release drivers for them before on the Z270 (for instance). This was discussed on a BIOS mod website before. It was the only way to enable them to work on macOS at the time.

Which macOS version is your friend using, BTW? Is it High Sierra? Did you know that Maple Ridge Thunderbolt is only supported by Catalina and Big Sur?

 

[dehjomz]

I didn't realise it was for Maple Ridge. If it was for the earlier versions perhaps it could've worked, because Intel did release drivers for them before on the Z270 (for instance). This was discussed on a BIOS mod website before. It was the only way to enable them to work on macOS at the time.

Which macOS version is your friend using btw? Is it High Sierra? Did you know that Maple Ridge Thunderbolt is only supported by Catalina and Big Sur?

Tested it in Windows... whenever a Thunderbolt device is plugged in, an error comes up and says that a Thunderbolt device was plugged into the wrong port, and to please use a Thunderbolt port. The same exact device was working on that same port prior to flashing to 0707... which by the way updates the Thunderbolt Maple Ridge firmware from the original NVM24.

 

[Middleman]

Tested it in Windows... whenever a Thunderbolt device is plugged in, an error comes up and says that a Thunderbolt device was plugged into the wrong port, and to please use a Thunderbolt port. The same exact device was working on that same port prior to flashing to 0707... which by the way updates the Thunderbolt Maple Ridge firmware from the original NVM24.

I see. In that case, has he looked into the other settings in the BIOS? It's quite possible it could be a setting somewhere that is causing this. For example the IOMMU may have something to do with it? Since that controls the Thunderbolt's IO system and PCIe device synchronicity.

 

[c-o-pr]

c-o-pr,

I'm sorry to hear that but it sounds like a corrupted BIOS upload there to me. What you should do is try using Memtest+ on a USB stick and run it and see how your system performs. More than likely it may experience corruption. In which case it might be advised that you simply reflash the BIOS. Also make sure your USB media (used for the firmware flash) is also not corrupted.

Dehjomz, has your friend tried to flash their Thunderbolt via the Windows Intel TB drivers?

Thanks for suggestion re memtest... But it's not RAM I'm struggling with.

It looks to me like this Rocket 4 NVMe was just dying, by cooincience, at the same time I was doing my BIOS update and tuning.

I will confirm the new BIOS 0707 is ok for me when I get a replacement drive. For now I am back on beta-0704. For now, if I think there is any change a BIOS update could whomp my drive I would be crazy to try to disprove this when I have no backup

As I am running a 10900, and 11th gen looks like less of a bump than I hoped for, plus some hackintosh downsides, I'm considering a WD Black replacement for Sabrent Rocket.

Sabrent responded with a request for more info on failure and this process will just take however long it takes. I expect Sabrent will make good on a replacement, but I would be unhappy to find the next one zonks the same way. It could be bad luck.. For now no way to know. There are no wide-spread reports of failure — well, that I can find...

I looked at Samsung and WD Black and the word on street is 980 soars for PCIe4 when it's cache is not over-run, but WD high-load performance on PCIe3 is more stable, at expense of lower peaks and bragging rights.

BTW—Re reports of 0707 disturbing USB: There is a report here that 0707 updates Maple Ridge, but I didn't see that happen. Maybe it did, but it wasn't shown during update. However I did see a specific TB update happen when originaly updating to beta-0704 last month. beta-0704 updated several sub units, including Intel ME. Well, it so happens that ME is also a separate download / update, so there's whatever way Asus thinks that customers want to move forward versus how Asus wants to move forward. Of course, it's understandable to me that enterprise users might want security updates for ME but nothing else. So what I am saying is there is pattern which is unfamilair to me with Asus update schedule and how/why they break out or include other parts of the firmware.

I'm trying to avoid my own tendency to religiosity in the face of lots of moving parts and unwanted events... Note to self: correlation != cause, etc...