Z790 Chipset & Raptor Lake

[etorix]

Maybe my configuration / conclusion is flawed, so let me know if I have missed something here.

Missed something? You went in great details on your frightening experiments.
I don't think that your configuration if flawed. But Intel is hitting hard into a thermal wall. Here's a shocking graph from a Threadripper 7000 review, which used Sapphire Rapids as well as consumer CPUs for comparison:

Threadripper 7000 sticks to its stated "350 W TDP" down to decimals.
Ryzen CPUs stay within 130% of their TDP in actual testing (per socket specification, maximal power is 135% of TDP).
Sapphire Rapids ventures above 140% of its 350 W TDP , which brings it to 500 W.
And your Core i9-14900K ramps up to an absurd 343% of is stated "125 W TDP", above the Threadripper 7980X which sports four time the thread count(!), and not far from the Sapphire Rapids Xeon W9-3495X. This is the 400+ watts monster you're trying to tame…

 

[schreiberstein]

Missed something? You went in great details on your frightening experiments.
I don't think that your configuration if flawed. But Intel is hitting hard into a thermal wall. Here's a shocking graph from a Threadripper 7000 review, which used Sapphire Rapids as well as consumer CPUs for comparison:

Threadripper 7000 sticks to its stated "350 W TDP" down to decimals.
Ryzen CPUs stay within 130% of their TDP in actual testing (per socket specification, maximal power is 135% of TDP).
Sapphire Rapids ventures above 140% of its 350 W TDP , which brings it to 500 W.
And your Core i9-14900K ramps up to an absurd 343% of is stated "125 W TDP", above the Threadripper 7980X which sports four time the thread count(!), and not far from the Sapphire Rapids Xeon W9-3495X. This is the 400+ watts monster you're trying to tame…

Yes, I agree in principle.

The AMD chips are superior in many regards - However, the AMD platform is not better in all aspects. Users are also plagued with stability issues and early adopter oddities. Also, we’re on a macOS-focussed forum here, and from what I gathered, Intel simply has the best compatibility (also in terms of specialized applications such as VSTs, DAWs, etc.)

Intel provided a good incremental upgrade and I am satisfied with that. I use macOS and I want to make extensive use of its hypervisor framework for virtualization. This is not possible on AMD, aside from all other potential incompatibilities I might face. In some ways, it is good that Intel has not fundamentally changed their platform during the last generations, as this might have introduced additional obstacles for running macOS. I perceive these platforms to be very reliable. My Z390 system has worked reliably for 5 years on macOS. I suppose Z790 will do the same. Either way, the days of macOS on Intel are numbered. And once this day comes, I guess I will migrate to Apple Silicon. For now, I enjoy running x86 software at native speed.

 

[dontblink]

For anyone else reading this, this is for a Gigabyte Z790 AERO running F9 BIOS, and a 14900K.
Has a Thermalright CPU frame and Noctua NH-d15 (using both fans).

Also keep in mind every single CPU is different even if sold as a 14900K, and some may not like these Core boost settings and vcore and fail to boot, and some can be pushed further than mine.

USE AT YOUR OWN RISK, there is nothing bad per se in this BIOS settings, like no increased in Vcore value, so nothing should melt, but under voltage can cause instability and fail to boot. This board has so far always allowed me to get back into the BIOS when I played with unstable settings.

Some BIOS settings, need a corresponding EFI settings to match in order to work with macOS, of the top of my head, ResizeGpuBars and VT-d setting are two important ones.
Also the iGPU is disabled.

---

Here is my current BIOS (F9) saved setting I'm using (there's a beta BIOS F10b, but would avoid that for now if loading my bios settings), tweaked for macOS but works great in Windows 11.

In the Advance menu, you'll see Vcore offset, set at -0.06. You can try as is, otherwise go down to -0.02 and then test in -0.005 increments. The first time I had the cooler installed and no CPUframe, the system wouldn't even boot at lower V-core offset.
In the advance CPU settings, you'll find the boost value 60 60 60 60 59 59 58 55, and the e-core there are 8 places set to 43 (as in 4.3GHz).

I highly recommend using the Intel Extreme Tuning Utility in windows (I didn't install any gigabyte utilities, only drivers one by one from gigabyte) and live test v-core offset and max boost speed in case things aren't stable, and check temps, then manually set those in the bios as value tested in the Intel utility won't stick. that will get you closer, then like I did last was Geekbench 6 in macOS, since it appeared to be the most sensitive to max boost/v-core offset tweaks (it simply stops, so no score displayed, but doesn't close fully, and if you check the console.app under crash report, you'll get confirmation that it did crash during cpu benchmark)

I have my fan profile set to 100% fan speed once reaching 60C (you can likely push that to 70C or even 80C), and just progressively slower under that, so that it's silent at 40~45C (like right now basically idle, hovers around 33~40C), none of my fan are high rpm even at 100%, I have two Noctua 140mm 1200RPM for my front intake, and the included fan in the NZXT H5 flow also 1200RPM (120mm) for bottom intake and rear exhaust.

Do you have an updated EFI that goes with the BIOS profile?

 

[eliaden]

Do you have an updated EFI that goes with the BIOS profile?

only slightly different from my last post.

This is my working config with OC 0.9.5, but I would assume updating to OC 0.9.6 should be flawless, but just in case would rather post a tested working version.

ACPI were created with 14900K and F9 bios, so not sure how other cpu would be (I assume 14900KF is the same)
and has specific stuff for the 6950XT

and of course fill in the
SystemSerialNumber, SystemUUID, ROM, MLB

Sleep doesn't work:
Sleep ACPI and kext are still in there and active, but don't work as expected, so can be removed, but otherwise don't seem to hurt either, just don't let your computer sleep at all. (last time I tried 1st sleep/wake "works" but second round has black screen and fans spin at 100%, only way out is force reboot)

Sounds from MOBO audio front and back ports, partially works, but had artifacts when I tested, so wouldn't consider it fully functional, didn't investigate further as I use an out of the box macOS supported USB DAC for my headphones and speakers.
Sound through DisplayPort from the GPU does seem to work well (in case you want to plug headphones to your display), but haven't tested extensively.

Ethernet does work (at least at gigabit speeds, 2.5 Gb/s does show up in the system info, but I don't have a 2.5Gb device to test with)

I replaced the wifi card with a Broadcom one, and stayed in Ventura, so my usb map keeps that bluetooth internal usb active.

Other thing that didn't seem to work in macOS is DisplayPort to USB C passthrough (it gets artifact, like a weak HDMI cable, and sometime black screen, maybe a better cable would work, but this was the included one with my monitor, and seemed to work in linux last time I tried). But that USB C port is usable as a fully functional USB C port for external SSD and such.

 

[eliaden]

Maybe my configuration / conclusion is flawed, so let me know if I have missed something here.

You will get a "significant" single core performance running at 6Ghz compared to 5.3ghz, I started like you, with any core load (whether it was 1 or 8) set to 5.3~5.4Ghz, but one or two cores at 6Ghz pulls less than 100watt, so the cooler (at least the nh-d15) can easily handle that sustained load and did increase cinebench score by 25% or so, not an insane leap but not insignificant either, I see it as a worthy boost.
The drawback with higher single core boost is you will likely have to reduce you vcore offset to hit the 6Ghz and stay stable, so you would get a slightly hotter all core load (but if all core load is still under the cooler capability I don't see anything wrong with that; however if it does lower all core boost capability then lower single core and higher vcore offset is then possibly preferable depending on usage)

When more core are active it will automatically limit the max boost to your bios settings, like 5.3Ghz and keeping the wattage to a level your cooler can handle.

But I fully agree that avoiding any throttling altogether is better, performance is basically the same with a cooler CPU (1~2% at most, depending on the situation can even be worse)
My system now never throttles, even with 6Ghz single/dual core boost.

Avoiding throttling is a bit like an automatic car vs manual shifting, the automatic is reactive (like CPU throttling), where manual can be proactive (reducing Ghz based on core load, thus reducing vcore/wattage, before it even gets a chance to overheat)

 

[ramazarusx]

Missed something? You went in great details on your frightening experiments.
I don't think that your configuration if flawed. But Intel is hitting hard into a thermal wall. Here's a shocking graph from a Threadripper 7000 review, which used Sapphire Rapids as well as consumer CPUs for comparison:

Threadripper 7000 sticks to its stated "350 W TDP" down to decimals.
Ryzen CPUs stay within 130% of their TDP in actual testing (per socket specification, maximal power is 135% of TDP).
Sapphire Rapids ventures above 140% of its 350 W TDP , which brings it to 500 W.
And your Core i9-14900K ramps up to an absurd 343% of is stated "125 W TDP", above the Threadripper 7980X which sports four time the thread count(!), and not far from the Sapphire Rapids Xeon W9-3495X. This is the 400+ watts monster you're trying to tame…

My W9-3475x, if let unchecked (removed all the BIOS limitations), eats up 900w+.

 

[dontblink]

Curious what your Gigabyte biscuit scores were? Anyone care to post? I had mine at 83 and topped out at -70mv.

 

[ramazarusx]

1800W+ OCd to 6GHZ

 

[schreiberstein]

Fro

only slightly different from my last post.

This is my working config with OC 0.9.5, but I would assume updating to OC 0.9.6 should be flawless, but just in case would rather post a tested working version.

ACPI were created with 14900K and F9 bios, so not sure how other cpu would be (I assume 14900KF is the same)
and has specific stuff for the 6950XT

and of course fill in the
SystemSerialNumber, SystemUUID, ROM, MLB

Sleep doesn't work:
Sleep ACPI and kext are still in there and active, but don't work as expected, so can be removed, but otherwise don't seem to hurt either, just don't let your computer sleep at all. (last time I tried 1st sleep/wake "works" but second round has black screen and fans spin at 100%, only way out is force reboot)

Sounds from MOBO audio front and back ports, partially works, but had artifacts when I tested, so wouldn't consider it fully functional, didn't investigate further as I use an out of the box macOS supported USB DAC for my headphones and speakers.
Sound through DisplayPort from the GPU does seem to work well (in case you want to plug headphones to your display), but haven't tested extensively.

Ethernet does work (at least at gigabit speeds, 2.5 Gb/s does show up in the system info, but I don't have a 2.5Gb device to test with)

I replaced the wifi card with a Broadcom one, and stayed in Ventura, so my usb map keeps that bluetooth internal usb active.

Other thing that didn't seem to work in macOS is DisplayPort to USB C passthrough (it gets artifact, like a weak HDMI cable, and sometime black screen, maybe a better cable would work, but this was the included one with my monitor, and seemed to work in linux last time I tried). But that USB C port is usable as a fully functional USB C port for external SSD and such.

From what I gathered, this “second sleep - freeze - high fanspeed” has to do with “Above 4G Decoding” and “Resizable Bar”. The same has been introduced to Z690 after a certain BIOS revision. I have the same problem, so I will check if it works with these options disabled. I wish some ACPI/firmware person could look into this and provide a fix for this.

 

[BigHack]

I'm testing out the ASRock Z790 PG-ITX/TB4 and it boots successfully with a Z690-like opencore config and appears to be stable. However, I am having annoying ethernet issues (similar to @ramazarusx on the ProArt) where it just won't detect the ethernet cable. The chip is an Intel Killer E3100X, also known as i225-K rev 2, and appears to be supported by the native AppleEthernetE1000 driver. On Linux it uses the same driver as the other i225-family controllers. Any ideas?

Hello
Did you get the Intel Killer E3100X to work on the ASRock PG-iTX?