obus Posted February 17, 2021 Author Share Posted February 17, 2021 (edited) 8 hours ago, amjsez said: My CPU core is not showing properly with bios 1202?) This is perfectly ok as long as all 20 threads are showing up and you have working PM. Spoiler This is a screenshot from my rig: Spoiler Edited February 17, 2021 by obus Link to comment Share on other sites More sharing options...
amjsez Posted February 18, 2021 Share Posted February 18, 2021 (edited) 9 hours ago, obus said: Hi @amjsez Can I have a look at your EFI. Hi @obus see attached. On the other more recent bios all the core shows up correctly and in order, but the USB-C does not shows up. I have not tried bios 3003 yet, have you tried that one? Current-EFI.zip Edited February 18, 2021 by amjsez additional info Link to comment Share on other sites More sharing options...
obus Posted February 18, 2021 Author Share Posted February 18, 2021 (edited) 5 hours ago, amjsez said: Hi @obus see attached. On the other more recent bios all the core shows up correctly and in order, but the USB-C does not shows up. I have not tried bios 3003 yet, have you tried that one I've tried all of the released bios:s from ASUS. The only one working with both ASM2142 and my two TitanRidge THB 3 card is 1202 bios: Spoiler Why are you using SMBIOS Mac Pro7,1? The most native SMBIOS for this mobo,CPU and GPU should be iMacPro1,1 I will take a look at your EFI. Edited February 18, 2021 by obus Link to comment Share on other sites More sharing options...
obus Posted February 18, 2021 Author Share Posted February 18, 2021 (edited) Hi @amjsez. Test this EFI as is. Don't change anything. This is with a new SMBIOS and OpenCore release 0.6.6. My 15 port:s USBPort.kext is for this configuration. If you don't want that later you can change it back to your old MacPro7,1 SMBIOS. You need to shut down clear NVRAM and restart your rig after applying the new EFI. Spoiler Back: 1 USB 2.0 port and 3 USB 3.0 port:s. (se attached picture) = total 7 port:s Front: 2 USB 3.0 ports = total 4 port:s Internal: 4 USB 2.0 = total 4 port:s TOTAL PORT:s = 15 EFI_C422_0_6_6_iMacPro1.1-BIOS_1202_RELEASE_FakeSMC.zip Edited February 18, 2021 by obus 2 Link to comment Share on other sites More sharing options...
etorix Posted February 18, 2021 Share Posted February 18, 2021 Working NVRAM! The DSDT defines an object with _HID "PNP0C02" and _UID "PCHRESV" which is similar (and actually cleaner) than the one defined by the DSDT of the C246-WU4. So a SSDT-PMC according to Dortania's template provides NVRAM. I could have tried that from the very beginning… I still have to do Thunderbolt testing, fan control through the BMC, a bit of cleaning and update to OpenCore 0.6.6 but it seems that the Gigabyte C621-SU8 will make a good Hackintosh Pro. 2 Link to comment Share on other sites More sharing options...
yapan4 Posted February 18, 2021 Share Posted February 18, 2021 (edited) 39 minutes ago, etorix said: fan control through the BMC This point can be interesting for ASUS WS C422Pro/SE users because this motherboard has cooling monitor and control via BMC chip Edited February 18, 2021 by yapan4 Link to comment Share on other sites More sharing options...
obus Posted February 21, 2021 Author Share Posted February 21, 2021 Updated to Big Sur 11.3 Beta 2 (20E5186d) without problems Spoiler 1 Link to comment Share on other sites More sharing options...
amjsez Posted February 22, 2021 Share Posted February 22, 2021 On 2/18/2021 at 2:22 PM, obus said: Hi @amjsez. Test this EFI as is. Don't change anything. This is with a new SMBIOS and OpenCore release 0.6.6. My 15 port:s USBPort.kext is for this configuration. If you don't want that later you can change it back to your old MacPro7,1 SMBIOS. You need to shut down clear NVRAM and restart your rig after applying the new EFI. Reveal hidden contents Back: 1 USB 2.0 port and 3 USB 3.0 port:s. (se attached picture) = total 7 port:s Front: 2 USB 3.0 ports = total 4 port:s Internal: 4 USB 2.0 = total 4 port:s TOTAL PORT:s = 15 EFI_C422_0_6_6_iMacPro1.1-BIOS_1202_RELEASE_FakeSMC.zip 8.07 MB · 3 downloads Hi @obus, I have used the EFI folder as is. System does not wake from sleep and the USB-C shows up as before but is still not functional. (I have USB-C drives and my iPhone connected with USB-C to Lightning cable) I also can't test the Thunderbolt port as Im still waiting for my Thunderbolt Dock. It should be here in a few days. Link to comment Share on other sites More sharing options...
obus Posted February 22, 2021 Author Share Posted February 22, 2021 (edited) 3 hours ago, amjsez said: Hi @obus, I have used the EFI folder as is. System does not wake from sleep and the USB-C shows up as before but is still not functional. (I have USB-C drives and my iPhone connected with USB-C to Lightning cable) I also can't test the Thunderbolt port as Im still waiting for my Thunderbolt Dock. It should be here in a few days. Load this bios with the attached settings in the CMO - file. After that send me your ioreg. 1202 patched_black.zip Edited February 22, 2021 by obus Link to comment Share on other sites More sharing options...
amjsez Posted February 23, 2021 Share Posted February 23, 2021 21 hours ago, obus said: Load this bios with the attached settings in the CMO - file. After that send me your ioreg. iMac Pro.ioreg Hi @obus, Pls see attached ioreg as requested. I had to enable Above 4G in the bios, other wise it would not boot. Thanks. iMac Pro.zip Link to comment Share on other sites More sharing options...
obus Posted February 23, 2021 Author Share Posted February 23, 2021 7 minutes ago, amjsez said: Hi @obus, Pls see attached ioreg as requested. I had to enable Above 4G in the bios, other wise it would not boot. Thanks. iMac Pro.zip 3.21 MB · 1 download Yes I know. It's not possible to save that settings in the *.cmo file. I will have a look at your ioreg. Link to comment Share on other sites More sharing options...
obus Posted February 23, 2021 Author Share Posted February 23, 2021 (edited) Everything is there as it should!!!!! You have 15 USB 3.0 port:s (XHCI) 2 thunderbolt 3/USB-C port:s (XHC2) and your rear and front ASM2142 USB-C respectively USB 3.1 port:s (XHC4 and XHC5). What is the Problem? Spoiler Edited February 23, 2021 by obus Link to comment Share on other sites More sharing options...
amjsez Posted February 24, 2021 Share Posted February 24, 2021 13 hours ago, obus said: Everything is there as it should!!!!! You have 15 USB 3.0 port:s (XHCI) 2 thunderbolt 3/USB-C port:s (XHC2) and your rear and front ASM2142 USB-C respectively USB 3.1 port:s (XHC4 and XHC5). What is the Problem? Hide contents Yes, the USB tree shows up and is completely as it should. But when USB-C drives, iPhone with USB-C to Lightning cable and the iPad with USB-C to lighting cable is plugged into the USB-C/Thunderbolt port it charges but nothing shows up on the USB tree and it is not detected by the system... Link to comment Share on other sites More sharing options...
obus Posted February 24, 2021 Author Share Posted February 24, 2021 (edited) 10 hours ago, amjsez said: Yes, the USB tree shows up and is completely as it should. But when USB-C drives, iPhone with USB-C to Lightning cable and the iPad with USB-C to lighting cable is plugged into the USB-C/Thunderbolt port it charges but nothing shows up on the USB tree and it is not detected by the system... I have no USB-C to Lightning cable so I can't test that but if I put one of my external USB-C disk in one of my thunderbolt port:s and the other one in my ASM2142 port:s I have this result in About this Mac --> System Report --> USB. Booth disks shows up as USB 3.1. The same in Hackintool,both disk:s shows up with correct speed too. Is your Titan Ridge card flashed? If so I think this could be the culprit. I've read somewhere that all devices is not showing up with flashed cards depending on which firmware you flashed. You need to investigate that a little bit further. Is your iPhone showing up if you connect it to your ASM2142 port with a lightning cable? If so it must have to be something with your flashed Titan Ridge firmware. (IPhone show up as a USB 3.0 device on the USB 3.0 bus even if it's connected to a thunderbolt 3 port on a original Mac.) Spoiler Edited February 24, 2021 by obus Link to comment Share on other sites More sharing options...
obus Posted February 24, 2021 Author Share Posted February 24, 2021 (edited) For working Hot Swap you need to connect pin 3 and 5 with a jumper like attached picture. For working USB you need to connect your Titan Ridge card to the internal USB 2.0 or 3.0. Spoiler Spoiler This is how my iPhone 12 (connected with a borrowed lighting cable) is showing up with my Titan Ridge card connected to the internal USB 2.0 port. If I connect my phone to ASM2142 USB-C port (front) it will show up exactly the same way on the USB 2.0 hub. Edited February 24, 2021 by obus Link to comment Share on other sites More sharing options...
etorix Posted February 24, 2021 Share Posted February 24, 2021 On 2/18/2021 at 11:25 PM, yapan4 said: This point can be interesting for ASUS WS C422Pro/SE users because this motherboard has cooling monitor and control via BMC chip Sorry for the late reply. The last few days have been frustrating: Although I can set the boot drive or create persisting entries in NVRAM, I still cannot run an OS X installer or updater. In swiching to OpenCore 0.6.6 I used another USB stick which had a Mojave installer and found that, contrary to Catalina, I could boot this one and get to the installer GUI—but then it would fail after the reboot. The issue seems to be in the way OS X installers operate rather than with my settings. As for fan control it was both very simple, and also frustrating. Very simple: It's all in the GUI. Frustrating: It's all in the GUI—and there is not much! Making matter worse Gigabyte's excuse for a documentation is a mere string of screenshots without any explanation. You actually have to plug the BMC to a LAN and power it up to learn how you should have wired the fans in your case… What fun! I have already set up IMPI fan control for my NAS (TrueNAS Core), based on Supermicro (server) motherboard. A fan profile can be selected from the BMC web GUI and customisation is done through the command line using ipmitool on the NAS console or ipmiutil for remote administration. It is important to adjust the BMC settings because defaults thresholds are meant for server fans, so the lower thresholds are above the usual operating range of (quieter) consumer fans: If a dynamic profile is used but suitable thresholds are not set for the fans, the BMC will slow down the fans, sense they are spinning at a few hundred RPMs, consider that the fans are stalling (below about 1000 RPM is a critical failure for a typical 4-5000 RPM server fan!) and immediately try to wake them up by pushing to the maximum, detect a low temperature and… slow down the fans… The result is pulsating fan noise: WHOOOSH! Whooshhhh… WHOOOSH! Whooshhh… Here is what a custom setting looks like on a Supermicro A2SDi-H-TF: root@NASflash[~]# ipmitool sensor list all | egrep FAN FAN1 | 600.000 | RPM | ok | 100.000 | 200.000 | 300.000 | 25300.000 | 25400.000 | 25500.000 FAN2 | 500.000 | RPM | ok | 100.000 | 200.000 | 300.000 | 25300.000 | 25400.000 | 25500.000 FAN3 | 500.000 | RPM | ok | 100.000 | 200.000 | 300.000 | 25300.000 | 25400.000 | 25500.000 FANA | 1500.000 | RPM | ok | 300.000 | 500.000 | 700.000 | 25300.000 | 25400.000 | 25500.000 NAME | current value | unit | condition | lnr (lower non-recoverable) | lc (lower critical) | lnc (lower non-critical) | unc (upper non-critical) | uc (upper critical) | unr (upper non-recoverable) Upper values are actually not set because these are of little use in my case; lower values are custom. Intake fans FAN2, FAN3 are Noctua NF-A9 PWM with a lower spinning rate of 400 RPM +/-20% so lower non-critical is set below the operating range to prevent any false alarm, the other lower thresholds are set accordingly. This is done and set with just one command (per fan): # ipmitool sensor thresh FAN2 lower 100 200 300 ... # ipmitool bmc reset warm In the same manner as "lower lnr lc lnc", upper thresholds would be set with "ipmitool sensor thresh FAN# unc uc unr". The changes are permanent but take effect only after a BMC restart—which can be done on the command line without resetting the whole computer. And on the BMC web interface, fan mode is set under Configuration, from a fixed list of four uneditable profiles: Here it is set to Full Speed rather than Standard or Optimal because speed control is actually done through a very fine PERL script running in the background which uses ipmitool to pool CPU and HDD temperatures and adjust fan duty level per motherboard zone as required. This is very effective and can be adjusted to any choice of fans or use of the motherboard connectors (my build is actually the reverse of Supermicro default: CPU is on FANA and chassis fans for HDD on FAN1..3). Turning to the Hackintosh, Brew provides bottles for "impitool" and "ipmiutil" but there is no available driver for a local BMC; both programs are meant to be used over a network and are actually two different ports of UNIX impiutil. I picked ipmitool. Despite being the same Aspeed 2500 BMC as on the Supermicro board, the C621-SU8 is a wholly different experience. $ brew install ipmitool $ ipmitool -I lan -H 172.16.1.131 -U admin -P admin chassis power status Get Session Challenge command failed: Invalid data field in request Error: Unable to establish LAN session Error: Unable to establish IPMI v1.5 / RMCP session $ ipmitool -I lanplus -H 172.16.1.131 -U admin -P admin chassis power status Chassis Power is on $ ipmitool -I lanplus -H 172.16.1.131 -U admin -P admin sensor list | egrep FAN CPU_FAN1 | 600.000 | RPM | ok | na | 800.000 | 1200.000 | na | na | na SYS_FAN1 | na | | na | na | 800.000 | 1200.000 | na | na | na SYS_FAN2 | 0.000 | RPM | ok | na | 800.000 | 1200.000 | na | na | na SYS_FAN3 | 0.000 | RPM | ok | na | 800.000 | 1200.000 | na | na | na SYS_FAN4 | 900.000 | RPM | ok | na | 800.000 | 1200.000 | na | na | na SYS_FAN5 | 0.000 | RPM | ok | na | 800.000 | 1200.000 | na | na | na SYS_FAN6 | 0.000 | RPM | ok | na | 800.000 | 1200.000 | na | na | na So "lanplus" is the right interface— and something is to be done sooner than later about the default user/password! The Noctua NH-U12S DX3647 CPU cooler is reported OK although the defaults settings would qualify its normal operation as "critical failure" and many thresholds are simply not implemented. $ ipmitool -I lanplus -H 172.16.1.131 -U admin -P admin sensor thresh CPU_FAN 100 200 300 Valid threshold '100' for sensor 'CPU_FAN' not specified! $ ipmitool -I lanplus -H 172.16.1.131 -U admin -P admin sensor list | egrep CPU_FAN CPU_FAN1 | 600.000 | RPM | ok | na | 800.000 | 1200.000 | na | na | na $ ipmitool -I lanplus -H 172.16.1.131 -U admin -P admin bmc reset warm Sent warm reset command to MC ... $ ipmitool -I lanplus -H 172.16.1.131 -U admin -P admin sensor list | egrep CPU_FAN CPU_FAN1 | 600.000 | RPM | ok | na | 800.000 | 1200.000 | na | na | na Thresholds cannot be adjusted. Script control is not likely to succeed here, even with a crossed cable from the BMC LAN to one i210 LAN port to avoid going through a switch or router. This implementation is meant to be administered from the web GUI. "Sensor Scanning" to disable reporting. From the overview "Sensor", double clicking on a sensor brings to another screen, from where the thresholds can eventually be adjusted: This clears the red "Critical" flag. But these changes are lost if the computer is unplugged! So the reporting of sensor events is essentially useless because it can never be cleared of fallacious errors. The only setting which does persist over a loss of power is a customised 3-point fan curve under "Fan Control". The red temperature on the right indicates which sensor is used to adjust the fan. This is obvious for the CPU fan, not so much for SYS_FAN1 to SYS_FAN6… (1: Rear Right. 2: CPU MOS 3: North Side. 4: Near CPU Vccin (RAM) 5: BMC 6: PCH Too bad for the seventh temperature sensor in the bottom, which screams into the void.) There is apparently no way to assign another temperature sensor to a given fan port in the GUI, this has to be done through cabling. So setting fan control was quite short in my case, and could have been very, very short if Gigabyte had bothered to pay a technical writer to document what is actually implemented… I couldn't even find something about the BMC interface in the manual for the WS C422 Pro/SE so I have absolutely no idea what ASUS has devised and, well, it seems that Gigabyte's documentation is not even the worst out there. I hope the above can help as a starting point if you have issues. 1 Link to comment Share on other sites More sharing options...
yapan4 Posted February 25, 2021 Share Posted February 25, 2021 (edited) On 2/24/2021 at 6:55 PM, etorix said: Sorry for the late reply. The last few days have been frustrating: Although I can set the boot drive or create persisting entries in NVRAM, I still cannot run an OS X installer or updater. In swiching to OpenCore 0.6.6 I used another USB stick which had a Mojave installer and found that, contrary to Catalina, I could boot this one and get to the installer GUI—but then it would fail after the reboot. The issue seems to be in the way OS X installers operate rather than with my settings. You can try to clean NVRAM after each restart, but you've probably already tried it. I'm sure you'll find a solution. Thank you for sharing your experience about IPMI. ASUS has similar capabilities and similar GUI. I do not use the command line, but if necessary I will use your tips from this post. Thank you. I will note that the first versions of BMC firmware were very buggy and looked raw. Recent releases have significantly improved the situation, but there are still minor ones that have to be tolerated. Two little questions to @all 1) How to enable PMBus monitoring for the power supply on ASUS WS C422 Pro/SE or similar motherboards? 2) Could something motivate our developers to write a FakeSMC or VirtualSMC plugin for monitoring via BMC chip (something like BMCMonitor.kext)? Spoiler Spoiler Spoiler Spoiler Edited February 25, 2021 by yapan4 Link to comment Share on other sites More sharing options...
etorix Posted February 25, 2021 Share Posted February 25, 2021 1 hour ago, yapan4 said: Thank you for sharing your experience about IPMI. ASUS has similar capabilities and similar GUI. I do not use the command line, but if necessary I will use your tips from this post. Thank you. You're welcome! It is a very similar GUI indeed, only with the fan curves in a different place and defined by four points instead of three. The command line has two issues. 1/ Because real macs do not have BMCs there is no Darwin driver, so commands have to go through the network. And it may be even more difficult to motivate developers to port the Linux openipmi driver to Darwin than to motivate developers to write a FakeSMC/VirtualSMC plugin. 2/ The command line provides a way to read fan speed or temperature through standardised commands (there are much more commands than the few I have illustrated). But setting fan duty is done by sending raw hexadecimal commands to the BMC; there is no standard here, it all depends on the manufacturer's implementation. Supermicro's implementation is known (at least to gurus…) and the script I use on my NAS is written using this knowledge for Supermicro X10/X11 boards. Seeing how crippled Gigabyte's implementation of the BMC feels compared to Supermicro's, I fear that significant work would be required to find out the hex sequences to control the fans, and then to rewrite the script accordingly. It may be easier to hackintosh a X11SRA-F or X11SRL-F… 1 Link to comment Share on other sites More sharing options...
yapan4 Posted February 27, 2021 Share Posted February 27, 2021 (edited) @etorix I imagined it much easier because the BMC chip communicates through the LPC bus (just like the SMC chip). Maybe there are some thoughts what to do to make working PSU monitoring through PMbus. I bought two power supplies, both have a PMBus connector, both from Supermicro and PMBus monitoring do not work on both. Thank you. Edited February 27, 2021 by yapan4 Link to comment Share on other sites More sharing options...
yapan4 Posted March 5, 2021 Share Posted March 5, 2021 (edited) For those fans of C422 who do not want to have anything to do with IPMI, ASUS prepares the motherboard WS C422 DC https://www.asus.com/US/supportonly/WS C422 DC/HelpDesk/ Edited March 6, 2021 by yapan4 Link to comment Share on other sites More sharing options...
etorix Posted March 12, 2021 Share Posted March 12, 2021 I could not find a commercial description of the WS C422 DC (nor a shop selling it…), but I found the manual. And the motherboard diagram shows a AST2510 BMC This board appears to have fan control by the regular BIOS but a BMC for the Data Center ("DC"?). Strange… If I were to go C422 I think I would still prefer the WS C422 Pro/SE: Thunderbolt and one more x8 PCIe slot for expansion or putting a NVMe drive directly on the CPU rather than on the chipset. WS C422 Pro-ACE offers better USB, M.2 and U.2 though. I also found the ipmitool commands to control fan duty on the WS C422 Pro/SE! It seems one only has to ask ASUS: One good point for their user support. Spoiler Please use 05h command to control PWM FAN. 1. All Fan 50% Duty (32h = 50) ipmitool raw 0x30 0x05 0x01 0x32 0x32 0x32 0x32 0x32 0x00 0x00 0x00 2. Auto Fan mode ipmitool raw 0x30 0x05 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 C422 Fan: PWM1 => CPU Fan1,2 PWM2 => PUMP Fan PWM3 => AIO Fan PWM4 => Front Fan PWM5 => Rear Fan PWM6 => None PWM7 => None PWM8 => None Table 9-13 ASUS OEM Commands Net Function = OEM(30/31) LUN=0 Code Command ============ 05h Set FAN Duty: Request Data ========== Byte 1 – 0x00 Auto Mode, 0x01 Manual Mode Byte 2 – PWM1 FAN Duty(%) Byte 3 – PWM2 FAN Duty(%) Byte 4 – PWM3 FAN Duty(%) Byte 5 – PWM4 FAN Duty(%) Byte 6 – PWM5 FAN Duty(%) Byte 7 – PWM6 FAN Duty(%) Byte 8 – PWM7 FAN Duty(%) Byte 9 – PWM8 FAN Duty(%) Response Data ========== Byte 1 – Completion Code During the last week I very much enjoyed the BMC of the C621-SU8 as I investigated the older firmwares. Flashing is through the BMC, and the BMC is also very convenient for the initial configuration: Flashing resets settings, and video output then defaults to the BMC, not to the dGPU. Setting the BIOS through IPMI saved me fiddling with cables and monitor inputs, or bringing a second monitor. Short story: All firmwares F5 to F8 accept the patch from @metacollin to unlock MSR E2h and 1AAh. Firmwares F5 and F6 have a "small" 418 kB DSDT… which defines 112 potential ACPI processors (_SB_.SCK0.C000 to C06F; did the engineers expected to put in a Xeon Platinum 9000 CPU??? ). Booting can proceed to PCI initialisation without CPU wrapping, but I never managed to complete PCI initialisation and actually boot OS X on firmwares F5 and F6. Firmware F7 has a big 969 kB DSDT, with 56 ACPI "devices" for the processors (_SB_.SCK0.CP00 to CP37) like firmware F8, requires CPU wrapping like firmware F8—and boots straight into OS X with my EFI folder from firmware F8. So I'm back to patched firmware F8. Next I tried various booting related quirks, including those under UEFI>Protocol Override. Still no installation or update for Mojave or Catalina. I tried this method for creating a "msu-product-url" key in NVRAM and install Big Sur on X299 systems without working NVRAM. Still no luck with Mojave/Catalina installers/updates. But it did work for installing Big Sur: No hiccups, my hack rebooted automatically as many times as necessary until Big Sur 11.0.1 was ready for user configuration. Update to 11.2.3 proceeded without any issue and without even requiring a manual intervention. Along the way, I noticed a few debug messages about NVRAM being successfully set or retrieved. So, I put my regular working EFI folder onto the BigSur installer stick, a new drive, cleared NVRAM and tried to install Big Sur without the "msu-product-url" trick… Guess what? My C621 hack, which cannot install or update Mojave or Catalina, can install and update Big Sur as if it were a real Mac! Fully working NVRAM up to Big Sur requirements!!! On C621! I read so many threads about new firmwares breaking working hacks and/or Big Sur breaking working hacks (and of course all the warnings about C621 having broken NVRAM, weird memory map and so on) that I certainly did not expect Big Sur to work BETTER than Mojave or Catalina. None of the booting quirks I tried are required for that. I can even disable the Booter>RebuildAppleMemoryMap quirk (along with the custom slide quirks). It is very frustrating that Mojave/Catalina installers/updaters stall around the point where IOHDIXController creates a disk image. I will probably end up writing a guide to install Big Sur on C621-SU8 while I will actually use Mojave (installed and updated by moving the drive to a C246 hack) until I no longer have active 32-bit applications. 1 Link to comment Share on other sites More sharing options...
yapan4 Posted March 14, 2021 Share Posted March 14, 2021 (edited) On 3/13/2021 at 1:02 AM, etorix said: I could not find a commercial description of the WS C422 DC (nor a shop selling it…), but I found the manual. And the motherboard diagram shows a AST2510 BMC This board appears to have fan control by the regular BIOS but a BMC for the Data Center ("DC"?). Strange… If I were to go C422 I think I would still prefer the WS C422 Pro/SE: Thunderbolt and one more x8 PCIe slot for expansion or putting a NVMe drive directly on the CPU rather than on the chipset. WS C422 Pro-ACE offers better USB, M.2 and U.2 though. A small note - according to the manufacturer's website, the AST2510 is the only 2D VGA chip https://www.aspeedtech.com/server_ast2510/ I would also prefer a motherboard with IPMI but would like to try something like WS C422 ACE or WS C422 DC(if released). The reason is regular HW monitoring ...and possibly faster POST. On 3/13/2021 at 1:02 AM, etorix said: I also found the ipmitool commands to control fan duty on the WS C422 Pro/SE! It seems one only has to ask ASUS: One good point for their user support. I want to find the full version of user guide for ipmitool from ASUS. Probably not for public use at all. Thanks for link, very useful. On 3/13/2021 at 1:02 AM, etorix said: Next I tried various booting related quirks, including those under UEFI>Protocol Override. Still no installation or update for Mojave or Catalina. I tried this method for creating a "msu-product-url" key in NVRAM and install Big Sur on X299 systems without working NVRAM. Still no luck with Mojave/Catalina installers/updates. But it did work for installing Big Sur: No hiccups, my hack rebooted automatically as many times as necessary until Big Sur 11.0.1 was ready for user configuration. Update to 11.2.3 proceeded without any issue and without even requiring a manual intervention. Along the way, I noticed a few debug messages about NVRAM being successfully set or retrieved. So, I put my regular working EFI folder onto the BigSur installer stick, a new drive, cleared NVRAM and tried to install Big Sur without the "msu-product-url" trick… Guess what? My C621 hack, which cannot install or update Mojave or Catalina, can install and update Big Sur as if it were a real Mac! Fully working NVRAM up to Big Sur requirements!!! On C621! I read so many threads about new firmwares breaking working hacks and/or Big Sur breaking working hacks (and of course all the warnings about C621 having broken NVRAM, weird memory map and so on) that I certainly did not expect Big Sur to work BETTER than Mojave or Catalina. None of the booting quirks I tried are required for that. I can even disable the Booter>RebuildAppleMemoryMap quirk (along with the custom slide quirks). It is very frustrating that Mojave/Catalina installers/updaters stall around the point where IOHDIXController creates a disk image. I will probably end up writing a guide to install Big Sur on C621-SU8 while I will actually use Mojave (installed and updated by moving the drive to a C246 hack) until I no longer have active 32-bit applications. Apple is probably encouraging Gigabyte C621-WU8 users to install Big Sur in this way One more like from me for such an interesting sequel. Edited March 14, 2021 by yapan4 1 Link to comment Share on other sites More sharing options...
etorix Posted March 16, 2021 Share Posted March 16, 2021 On 3/14/2021 at 1:29 PM, yapan4 said: A small note - according to the manufacturer's website, the AST2510 is the only 2D VGA chip https://www.aspeedtech.com/server_ast2510/ Thanks for your support! I learned something about Aspeed chips… but then I wonder how useful this 2D-only chip is. For OS X, it may have to be disabled for graphic acceleration to work properly. On 3/14/2021 at 1:29 PM, yapan4 said: I want to find the full version of user guide for ipmitool from ASUS. Probably not for public use at all. Thanks for link, very useful. Have you tried asking ASUS user support? With sensor reporting by ipmitool and the raw commands to set fans, it would be possible to make fan control scripts through the network. What is missing is an OS X driver for the BMC chip: This would require a developer with spare time and a motherboard with an AST2500 (not necessarily a C422 or C621 board, but few people take a Supermicro or AsRockRack server board to make a desktop hackintosh…). 1 Link to comment Share on other sites More sharing options...
yapan4 Posted March 17, 2021 Share Posted March 17, 2021 22 hours ago, etorix said: I wonder how useful this 2D-only chip is. For OS X, it may have to be disabled for graphic acceleration to work properly. I tried BMC video a long time ago (on HS or Mojave). Worked in VESA mode, ofcourse without acceleration. On Catalina and Big Sur it will probably not work because of the unsupported VGA port, I don't know, need to try but not enough time. 23 hours ago, etorix said: Have you tried asking ASUS user support? No, I haven't applied yet. But I will apply if I do not find workarounds. 23 hours ago, etorix said: but few people take a Supermicro or AsRockRack server board to make a desktop hackintosh… I agree, the quota of hackintoreshers on server or workstations is insignificant ... but I hope someone, after all, will take up this case Link to comment Share on other sites More sharing options...
Balamut Posted March 23, 2021 Share Posted March 23, 2021 Hey guys I know its W-2xxx section but I was wondering if any of you have Xeon W-32xxx and managed to get it working properly? Link to comment Share on other sites More sharing options...
Recommended Posts