macOS Native CPU/IGPU Power Management

[Blindedbythelights]

Background process will prevent the CPU from sitting at 800MHz too long, but you should at least see it bounce down there once in a while.

The actual log shows it is hitting 800Mhz between spiking up so maybe it is ok (far right column)

 

[pastrychef]

The actual log shows it is hitting 800Mhz between spiking up so maybe it is ok (far right column)

It appears power management is working. I don't think you have anything to "fix". Remember, if you have lots of background processes running, you may not see 800MHz much.

 

[Blindedbythelights]

It appears power management is working. I don't think you have anything to "fix". Remember, if you have lots of background processes running, you may not see 800MHz much.

Thanks once again man you are always so helpful!

 

[RYlNP]

System Definition/macpro6,1: socket LGA 2011 only
Broadwell/5xxx - frequency vectors required
GitHub - Piker-Alpha/freqVectorsEdit.sh

freqVectorsEdit.sh shows me a whole bunch of possible choices. how do i know which one i need?

 

[Wolfster]

Regarding this instruction

GPU: All Systems/CLOVER/config.plist/ACPI/DSDT/Patches/Item 0/Intel GPU PM- Rename GFX0 to IGPU

• copy: config-audio_cloverHDMI+.plist.zip/ACPI/DSDT/Patches/Item 0/
• paste: to config.plist/ACPI/DSDT/Patches/

Your instructions imply to add the 'rename GFX0 to IGPU' only, yet your attached plist contains two different patches. Is it necessary to copy both the 'GFX0 to IGPU' and 'PEGP to GFX0' patches, or either one? I am using a Radeon VII as primary gpu but have the igpu enabled in bios.

 

[Ennio1985]

Regarding this instruction

GPU: All Systems/CLOVER/config.plist/ACPI/DSDT/Patches/Item 0/Intel GPU PM- Rename GFX0 to IGPU

• copy: config-audio_cloverHDMI+.plist.zip/ACPI/DSDT/Patches/Item 0/
• paste: to config.plist/ACPI/DSDT/Patches/

Your instructions imply to add the 'rename GFX0 to IGPU' only, yet your attached plist contains two different patches. Is it necessary to copy both the 'GFX0 to IGPU' and 'PEGP to GFX0' patches, or either one? I am using a Radeon VII as primary gpu but have the igpu enabled in bios.

To add to this question, I using WEG and Lilu for my rx580, and it supposed that WEG is in charge of that renames I use the HD630 as headless.

and the audio of my system is from the MB, not the display.

so in that scenario still has to made the renames and install the config-audio_cloverHDMI+.plist ?

 

[Jmd89]

When I used iMac19,1, the CPU never throttled down to 800MHz. When I use iMacPro1,1, it does.

Sorry to Necropost but yes, even my 9600k with iMac19,1 doesn't throttle to 0,8. With iMacPro1,1 all the states are up and running well... Even my rx590 behaves sensibly better, with less power drain and idle clocks. Strange right?

EDIT: It seems that the iMac19,1 SMBIOS has no problem now, maybe the cp states are injected in the Os thanks to the previous SMBIOS change.

 

[Ennio1985]

GPU: All Systems/CLOVER/config.plist/ACPI/DSDT/Patches/Item 0/Intel GPU PM- Rename GFX0 to IGPU

a question about this step. I have an i7-7700k, and a RTX 580. the system definition that im using is iMac 18.3. and I using lilu+ WEG. so using them it suppose that any change or rename to the GPU it will be done automatically. is this necessary?

I ask because checking the hacking tool now (Native CPU PM work) the platform-id appear to be: 0x59120003, instead of 0x59120000. the one that finish with 3 defines my system as "mobile", like a macbook so I read I could cause problems.

I facing some strange things, like reading a book pdf in preview, and the texts after appear in screen it takes 1 second or less to clarify in the screen. so maybe it suggest to be a problem with headless configuration.

so that's why I ask if this part is necessary. or can I change the platform id making the patch by using the patch mode in Hackintool ?

 

[Archangeliques]

GPU: All Systems/CLOVER/config.plist/ACPI/DSDT/Patches/Item 0/Intel GPU PM- Rename GFX0 to IGPU

I have a question too. My CPU PM is working, I'm using dGPU with iGPU in headless mode. dGPU PM is working as well. When I check from Intel Power Gadget iGPU's freq is always 1.2Ghz. For the systems that use iGPU in headless mode, iGPU PM is required? or Having it done with the rename have any benefits?

@Ennio1985 I see you have iGPU in headless mode too. Have you done the rename? Did it make any difference? Worth trying?

Thanks!

 

[Ennio1985]

I have a question too. My CPU PM is working, I'm using dGPU with iGPU in headless mode. dGPU PM is working as well. When I check from Intel Power Gadget iGPU's freq is always 1.2Ghz. For the systems that use iGPU in headless mode, iGPU PM is required? or Having it done with the rename have any benefits?

@Ennio1985 I see you have iGPU in headless mode too. Have you done the rename? Did it make any difference? Worth trying?

Thanks!

well the difference is that when I applied the CPU Native PW change my platform id to mobile, so I made the patch using hacking tool and then recognize the correct platform-id and I see a little improvement in the score in Geekbench