MBP 1,1-10,2 Force 2011 MacBook Pro 8,2 with failed AMD GPU to ALWAYS use Intel integrated GPU (EFI variable fix)

[JMVB]

I can't get external displays to work with my MBP late 2011. I know this thread is about forcing MBP to use the Integrated GPU because the AMD GPU failed. But both of my cards appear to be working, just won't switch (default) cards when I plug in display (via HDMI to Thunderbolt). MBP doesn't see display at all, but same display and connection/adapter works on a newer MacAir.

Is there anything I can do to get an external display to work while still using the internal GPU?

Thanks much!

There is some devices who can deliver video output by thunderbolt.
The problem is that it only work with specific apps

For my needs, it works very well

https://www.bhphotovideo.com/c/prod...bdlkulsdzminmon_Ultrastudio_Mini_Monitor.html

 

[nsgr]

So, its not much different than just running my own "bash roxy" each time that I reboot"

It's basically the same. The difference is that there are two files where one is for reboot and another for power off.

If you want, then you can do something more elaborate like in this post of 2011 and 2012. You must make the adaptations to use the nvram command.

https://bogner.sh/2012/11/os-x-run-commands-before-shutdown/

https://superuser.com/questions/252892/how-to-run-a-script-at-shutdown-with-snow-leopard

or

https://apple.stackexchange.com/questions/283763/run-a-script-before-shutdown

https://stackoverflow.com/questions...t-shutdown-not-log-off-or-restart-on-mac-os-x

http://www.launchd.info

 

[Bob B]

There is some devices who can deliver video output by thunderbolt.
The problem is that it only work with specific apps

For my needs, it works very well

https://www.bhphotovideo.com/c/prod...bdlkulsdzminmon_Ultrastudio_Mini_Monitor.html

Thanks for your suggestion. Unfortunately, the AMD GPU that feeds the TB port on my MBP doesn't work and thus no video output via the TB port.

Unless I find a way to replace the AMD GPU or run an external GPU somehow, I guess I am screwed.

 

[JMVB]

Thanks for your suggestion. Unfortunately, the AMD GPU that feeds the TB port on my MBP doesn't work and thus no video output via the TB port.

Unless I find a way to replace the AMD GPU or run an external GPU somehow, I guess I am screwed.

Dude, the thunderbolt port remain working

This device do not depend of the AMD GPU, it can work with the Intel one.
It is not a simple TB-HDMI adaptor

In fact, I'm using it with my dead AMD GPU 2011 macbook pro

 

[dosdude1]

Is there any way to 1) enable or replace the AMD Discrete GPU or 2) use an external GPU?
Would a TB dock like this one allow me to use an external monitor?

Desperately trying to get this to work somehow. Thanks for your help.

I'm working on getting my hardware solution to work with the external display port as well.

 

[Bob B]

Dude, the thunderbolt port remain working

This device do not depend of the AMD GPU, it can work with the Intel one

In fact, I'm using it with my dead AMD GPU 2011 macbook pro

Wow! I didn't think it would work via TB port with AMD GPU dead.
I am simply trying to connect a large Samsung monitor via HDMI to my TB port. I will be using the monitor to do audio production work using Logic X. The external monitor just needs to work, nothing fancy.

I am running OS X 13.6 High Sierra, so not sure it will work though.

Why do you think the device you recommend would only work with certain applications. I reviewed the specs on it so I guess I am missing something.

On Amazon Product Q & A found out that:

"Unfortunately this unit won't work as secondary monitor. "

Core Microsystems
Seller · April 10, 2016

"This is a software application driven device, it will not extend to a second monitor."

Snapxx
· April 10, 2016

Thanks.

 

[JMVB]

Wow! I didn't think it would work via TB port with AMD GPU dead.
I am simply trying to connect a large Samsung monitor via HDMI to my TB port. I will be using the monitor to do audio production work using Logic X. The external monitor just needs to work, nothing fancy.

I am running OS X 13.6 High Sierra, so not sure it will work though.

Why do you think the device you recommend would only work with certain applications. I reviewed the specs on it so I guess I am missing something.

Thanks.

Dude, again, it is not a secondary desktop device

It is not an external GPU card
It doesn't have the very same AMD GPU output characteristics

It is an applications output playout device

It will show your Final Cut final output
The same with Adobe premiere and the softwares listed down

I use my GPU-dead 2011 macbook pro for VTR to my ATEM switcher, via Propresenter 6 software, through Black Syphon app, with this device.

Mac OS X
Use your Favorite Software
UltraStudio is supported by a huge range of software applications for Mac OS X. Get drivers that work with fantastic editing software such as Apple Final Cut Pro X, Apple Final Cut Pro 7, Adobe® Premiere Pro® CC, Avid Media Composer, Avid Symphony and Avid News Cutter. For visual effects get full compatibility with After Effects® CC, The Foundry Nuke and Adobe Photoshop® CC. For audio get compatibility with ProTools, Steinberg Nuendo and Cubase. For streaming use Flash Media Encoder, Wirecast, Livestream Procaster."

In this video, is not the very same device. It is a Blackmagic Decklink Duo 2 PCIe into a TB Sonnet box.
BTW the results are the same with the ULtrastudio Mini Monitor

 

[Zom]

However, I'm afraid that upon restart I'll have trouble again.

How can I make this thing work even on restart??

I restarted my sisters computer 3x with it booting fully each time. I followed the MacOS version, not the Linux version.

The article says that only time this would have to be run again is after a smc or nvram reset.

It does take a really long time to boot now, but to her, it is preferable to buying a new MacBook at this time.

 

[JMVB]

I restarted my sisters computer 3x with it booting fully each time. I followed the MacOS version, not the Linux version.

The article says that only time this would have to be run again is after a smc or nvram reset.

It does take a really long time to boot now, but to her, it is preferable to buying a new MacBook at this time.

You might have an OS problem..
Which OS versión do you have?

I have 2 resurrected 2011 15" mbp.
One with El capitan (left, 8gb RAM) and other with Yosemite (right, 4 gb ram)
No one have ssd (both have the dark-boot mode)

Take a look:

 

[duckduckgoose]

I’m getting ‘This tool needs to be executed from the Recovery OS’ when I try to disable csrutil.

Also if I try to move or make a new file directory it just says ‘Operation not permitted’

Any ideas?

Thanks

There is not *one* guide to write up. Many roads lead to Rome. The best option would be if AppleMacFinder would update the first post of this thread pointing to the best alternatives.

Anyway. Even if this post now will quickly drown in the sheer length of this thread, I think this is currently one of the better guides:

#####__ The Guide __#####

This guide assumes that you run a stock system. Problem just occured. That means:
This guide assumes that all kexts are still in their default location /System/Library/Extensions.
Having all AMD-kexts there except one is beneficial for 'proper' operation.

To get some display acceleration back it will be necessary to force the machine to not boot into discrete graphics (dGPU) but directly into integrated graphics (iGPU). This will give you back your laptop – but you will lose some features: e.g. the ability to drive an external display. Thunderbolt data connections should work.

The initial procedure:

– To start from a clean slate: reset SMC and PRAM/NVRAM:

shutdown, unplug everything except power, now hold

<leftShift>+<Ctrl>+<Opt>+<Power>

release at the same time;

– Now power on again and hold

<Cmd>+<Opt>+<p>+<r>

at the same time until you hear the startup chime two times.

– Boot into Recovery by holding

<Cmd>+<r>+<s>

– Disable SIP:

csrutil disable

– disable dGPU on boot

nvram fa4ce28d-b62f-4c99-9cc3-6815686e30f9:gpu-power-prefs=%01%00%00%00

– enable verbose boot mode:

nvram boot-args="-v"

– reboot into single user-mode by holding

<Cmd>+<s>

on boot

– mount root partition writeable

/sbin/mount -uw /

– make a kext-backup directory

mkdir -p /System/Library/Extensions-off

– only move ONE offending kext out of the way:

mv /System/Library/Extensions/AMDRadeonX3000.kext /System/Library/Extensions-off/

– let the system update its kextcache:

touch /System/Library/Extensions/

– wait for the kextcache process to finish
then

reboot

Reboot normally:
you will have an accelerated iGPU display.


But the system doesn't know how to power-management the failed AMD-chip.
For that you have to either manaully load the kext after boot by:

sudo kextload /System/Library/Extensions-off/AMDRadeonX3000.kext

Automate this with the following LoginHook:

sudo mkdir -p /Library/LoginHook
sudo nano /Library/LoginHook/LoadX3000.sh

with the following content:

#!/bin/bash
kextload /System/Library/Extensions-off/AMDRadeonX3000.kext
exit 0

then make it executable and active:

sudo chmod a+x /Library/LoginHook/LoadX3000.sh
sudo defaults write com.apple.loginwindow LoginHook /Library/LoginHook/LoadX3000.sh

Preventive measures for future use

There are two further caveats to know: This is reversible when the SMC/PRAM/NVRAM is reset. If that happens the GPU-power-pref nvram can/has to be set again to force the use of the iGPU from boot-time.

Since this can happen quite easily (and is often erroneously recommended way too many times than it is actually useful), you should probably prepare for such a scenario and create a simple script to greatly speed up the process and also make entering the necessary variable much less error prone:

sudo nano /force-iGPU-boot.sh

– Enter the following content to this file:

#/bin/sh
sudo nvram boot-args="-v"
sudo nvram fa4ce28d-b62f-4c99-9cc3-6815686e30f9:gpu-power-prefs=%01%00%00%00
exit 0

– Now make that executable:

sudo chmod a+x /force-iGPU-boot.sh

In the future, when the SMC/PRAM/NVRAM gets reset to default values it is now possible to boot into SingleUser with:

<Cmd>+<s>

– And after mounting your boot-volume read-write to execute just:

sh /force-iGPU-boot.sh

This setup has now one kext in a place Apple's installers do not expect. That is why in this guide SIP has not been reenabled. If an update that contains changes to the AMD drivers is about to take place it is advisable to move back the AMDRadeonX3000.kext to its default location before the update process. Otherwise the updater writes at least another kext of a different version to its default location or at worst you end up with an undefined state of partially non-matching drivers.

After any system update the folder /System/Library/Extensions has to be checked for the offending kext. Its presence there will lead to e.g. a boot hang on Yosemite and Sierra, an overheating boot-loop in High Sierra.

Further: this laptop is overheating, no matter what you do. The cooling system is inadequate and the huge number of failing AMD chips are just proof of that.

To prolong the life of this now hacked machine it is advisable to abstain from really heavy lifting over prolonged stretches of time. Strictly follow the usual recommendations for laptops: use on hard surfaces, keep the fans and fins inside it clean. Using any fancontrol software with relatively aggressive settings should also help: like smcFanControl, MacsFanControl, or TGPro (the latter both commercial).


This is fairly complete and what I do recommend to everyone asking me.
Nevertheless. We're not done here, yet. Improvements are welcome. Share them!

 

[pckrullenbol]

I’m getting ‘This tool needs to be executed from the Recovery OS’ when I try to disable csrutil. Also if I try to move or make a new file directory it just says ‘Operation not permitted’
Any ideas?

RTFM..., it's really all in the excellent instructions from MikeyN, found at #875:
https://forums.macrumors.com/thread...fi-variable-fix.2037591/page-35#post-24956091

 

[ribof]

I ended up with a strange but not completely unwanted result: I tried following the MikeyN, found at #875:
https://forums.macrumors.com/thread...fi-variable-fix.2037591/page-35#post-24956091 instructions, and I also got the ‘This tool needs to be executed from the Recovery OS’ message (which a fix for IS NOT described in the instructions, I think it's a High Sierra thing). I figured out that to fix that I needed to type in 'csrutil disable' in the terminal that's available using the menus in the 'recovery OS' you to get to by pressing command-R on reboot.

Anyways, I think the first half of the instructions worked for me but somewhere around the second half either it didn't work or I didn't do it right (I'm not a super techie and I don't know anything about command line terminal stuff). I thought, oh well, let's see what happens once I restart. And what happened is my computer is now working except with 'safe mode' style limited graphics. But no crashes, and I can drive my external 24" monitor! Even after reboot or sleep it's still working this way with no crashes. I think my GPU isn't completely dead, maybe that's why it's workingish? As a backup machine I can live with 'safe mode' graphics, as photoshop works fine and browsing, email etc. aren't too painful.

But I wonder - can I somehow keep this setup but turn on Quartz Extreme on the internal gpu so my mac is almost normal?

 

[mikecwest]

I ended up with a strange but not completely unwanted result: I tried following the MikeyN, found at #875:
https://forums.macrumors.com/thread...fi-variable-fix.2037591/page-35#post-24956091 instructions, and I also got the ‘This tool needs to be executed from the Recovery OS’ message (which a fix for IS NOT described in the instructions, I think it's a High Sierra thing). I figured out that to fix that I needed to type in 'csrutil disable' in the terminal that's available using the menus in the 'recovery OS' you to get to by pressing command-R on reboot.

Anyways, I think the first half of the instructions worked for me but somewhere around the second half either it didn't work or I didn't do it right (I'm not a super techie and I don't know anything about command line terminal stuff). I thought, oh well, let's see what happens once I restart. And what happened is my computer is now working except with 'safe mode' style limited graphics. But no crashes, and I can drive my external 24" monitor! Even after reboot or sleep it's still working this way with no crashes. I think my GPU isn't completely dead, maybe that's why it's workingish? As a backup machine I can live with 'safe mode' graphics, as photoshop works fine and browsing, email etc. aren't too painful.

But I wonder - can I somehow keep this setup but turn on Quartz Extreme on the internal gpu so my mac is almost normal?

If your Nvidia GPU is failing, and you are running your Intel GPU, then you are pretty much going to lose your NVIDIA gnu, unless you go to an extreme measure, such as "reballing the GPU."

As far as "csrutil disable" is concerned, yes, for that you need to be in recovery mode. (You can hold CMD-R during boot to get there.) You can go to the Utilities menu to get to terminal prompt, and type csrutil disable, then press enter, type reboot, and press enter again.

I have recently learned, that you can press CMD-RS to load directly into "singe user recovery" at the terminal prompt. Here you can type csrutil disable, then press enter, type reboot, and press enter again.

for the command below, you just need normal single user mode, by holding CMD-S during boot.

nvram fa4ce28d-b62f-4c99-9cc3-6815686e30f9:gpu-power-prefs=%01%00%00%00
reboot

 

[ribof]

Doing any extreme measure to save my discreet gpu is beyond my capabilities, and it's not worth it to me spend money on my old 17" to have someone else do it. I just bought a used 2015 15" 2.2ghz (without a separate gpu - yeay!) as a replacement so my 17" is going to become a backup machine.

I did enter the "nvram fa4ce28d-b62f-4c99-9cc3-6815686e30f9:gpu-power-prefs=%01%00%00%00" command, which I assume did something since my computer isn't crashing now, but I'm still wondering how I got to this state where everything is normal except that the graphics are in 'safe mode' and I'm able to drive an external monitor. Has this happened to anyone else? And is there a way to turn on quartz extreme to get out of safe mode graphics but keep this state where my machine seems to be working without crashes?

 

[nsgr]

I have noticed that in the last updates of El Capitan and High Sierra the Recovery Mode Single User is no longer working - Command + R + S keys at boot - Recovery Mode Text.

Only Command + R keys at boot works - Recovery Mode Graphical.


Update:

In the case of MikeyN Guide, then you may have to use Recovery Mode Graphical to disable SIP - Command + R keys at boot.

 

[stormxeron]

My GPU failed 2 days earlier, and I am Lucky even to have the working guide. I now have successfully disabled AMD, by nvram setting>moving AMD*3000 to extensions-off(following MikeyN guide). I have one problem, the GPU Diode temperature hovers around 72-85 degree Celsius. My machine boots fine(now after disabling AMD), but I am little conscious about the GPU diode temperature.

I also have done the loginhook part and AMD*3000.kext is loaded after booting. Here are the screen shots:
1. My GPU diode temperature at near idle 72 degree Celsius.
2. Verification of loaded kexts

Is that temperature normal?, or Should I worry, and do something to Reduce that Temp??

What is weird to me is that GPU diode is in about 75 degrees but GPU proximity is in about 43 degrees. Isn't it weird? Also, The sensor freaked out at one time and was showing N/A for GPU diode and after a while it came back with reading of 75 again. What is happening, is the sensor thing bad or something?

One thing to note, is that, though GPU diode shows high temps, the machine is running a lot cooler than before. That is why I think there is the problem in system sensor itself. OR is it? I am confused.
Also Is Peak Load and Memory/Processor usage for Intel HD 3000 normal??(screenshot attached)

I am on latest High Sierra Version 13.6, And My Machine is Late 2011 15" MacBook Pro.

 

[nsgr]

My GPU failed 2 days earlier, and I am Lucky even to have the working guide. I now have successfully disabled AMD, by nvram setting>moving AMD*3000 to extensions-off(following MikeyN guide). I have one problem, the GPU Diode temperature hovers around 72-85 degree Celsius. My machine boots fine(now after disabling AMD), but I am little conscious about the GPU diode temperature.

I also have done the loginhook part and AMD*3000.kext is loaded after booting. Here are the screen shots:
1. My GPU diode temperature at near idle 72 degree Celsius.
2. Verification of loaded kexts

Is that temperature normal?, or Should I worry, and do something to Reduce that Temp??

What is weird to me is that GPU diode is in about 75 degrees but GPU proximity is in about 43 degrees. Isn't it weird? Also, The sensor freaked out at one time and was showing N/A for GPU diode and after a while it came back with reading of 75 again. What is happening, is the sensor thing bad or something?

One thing to note, is that, though GPU diode shows high temps, the machine is running a lot cooler than before. That is why I think there is the problem in system sensor itself. OR is it? I am confused.
Also Is Peak Load and Memory/Processor usage for Intel HD 3000 normal??(screenshot attached)

I am on latest High Sierra Version 13.6, And My Machine is Late 2011 15" MacBook Pro.
View attachment 779176 View attachment 779178 View attachment 779184

My tests:
Macbook Pro 15" Early 2011
High Sierra 10.13.6
LoginHook AMDRadeonX3000.kext
IstatMenu 5.32 - 14 day trial

https://bjango.com/mac/istatmenus/


Boot normal -> after login -> Macbook Pro 2011 in idle mode - only Active Monitor and Terminal open -> wait 4 minutes:

CPU Die Analog: 30 - 40 Celsius degrees

GPU Die Analog: zero / _ Celsius degrees

Verify if AMD GPU power off -> Finder -> Applications -> Utilities -> Terminal:

/System/Library/Extensions/AppleGraphicsControl.kext/Contents/MacOS/AGDCDiagnose -a | grep -i Power
  IG: FB0:on  FB1:N/A FB2:N/A 3D:idle HDA:N/A  Power:on
  EG: FB0:off FB1:off FB2:off 3D:idle HDA:idle Power:off
Policy:on GPUPowerDown:on Backlight Control:on Recovery:on
Power State Machine IG: 0 EG: 6

IG - Intel GPU
EG - AMD GPU


EG : FB0: off FB1: off FB2: off 3D: idle HDA: idle Power: off


Make sure your AMDRadeonX3000.kext of your LoginHook has a complete AMDRadeonX3000.kext.

Apple has put some updates with incomplete AMDRadeonX3000.kext without the Info.plist file.
This file is essential for the proper functioning of AMDRadeonX3000.kext.

It is good to have a backup of the complete AMDRadeonX3000.kext.
In this example the AMDRadeonX3000.kext is inside the /DisableExtensions directory.

ls -laR /DisableExtensions/AMDRadeonX3000.kext/

total 0
drwxr-xr-x@  3 root  wheel    96 Jun 29 22:38 .
drwxr-xr-x  33 root  wheel  1056 Aug 24 22:07 ..
drwxr-xr-x   7 root  wheel   224 Jun 29 22:39 Contents

/DisableExtensions/AMDRadeonX3000.kext//Contents:
total 40
drwxr-xr-x  7 root  wheel   224 Jun 29 22:39 .
drwxr-xr-x@ 3 root  wheel    96 Jun 29 22:38 ..
-rw-r--r--  1 root  wheel  8697 Jun 29 22:38 Info.plist
drwxr-xr-x  3 root  wheel    96 Jul  4 06:14 MacOS
-rw-r--r--  1 root  wheel     9 Jun 29 22:38 PkgInfo
drwxr-xr-x  3 root  wheel    96 Jun 29 22:39 _CodeSignature
-rw-r--r--  1 root  wheel   521 Jun 29 22:38 version.plist

/DisableExtensions/AMDRadeonX3000.kext//Contents/MacOS:
total 13088
drwxr-xr-x  3 root  wheel       96 Jul  4 06:14 .
drwxr-xr-x  7 root  wheel      224 Jun 29 22:39 ..
-rwxr-xr-x  1 root  wheel  6698064 Jul  4 06:14 AMDRadeonX3000

/DisableExtensions/AMDRadeonX3000.kext//Contents/_CodeSignature:
total 8
drwxr-xr-x  3 root  wheel    96 Jun 29 22:39 .
drwxr-xr-x  7 root  wheel   224 Jun 29 22:39 ..
-rw-r--r--  1 root  wheel  2428 Jul  4 06:14 CodeResources

Some websites with videos greatly increase system and GPU temperature. Test with Safari.

Firefox and Chrome are "toasters" when used on some websites with videos.

 

[stormxeron]

Hello, thanks for your reply,
When I checked using the command, the only difference between your result and mine was in the third line:
Yours is Policy: on and Mine is Policy: off, Is this the Issue?

Besides that, my version of loaded AMD*3000.kext has Info.plist, in fact I verified all the contents. If you find any difference, here is the result of second command:

Also, when I keep the Machine Idle, the GPU diode temperature hovers in 68-79 degree Celsius with Kext Loaded.

One Important Fact: When I keep The AMDRadeonX3000.kext Unloaded, the GPU Diode Automatically drops to 30-50 degree Celsius, of course the AMD Vcore is taking up power/Voltage. But When I load AMDRadeonX3000.kext, the AMD GPU Vcore is not taking power/Voltage but temperature goes high around 70-90 degree Celsius. Here is my screen shot: (see Vcore to confirm)
AMD*3000.kext not loaded: AMD*3000.kext Loaded:

Why is this so? When kext is loaded, temps go up, but when kext is unloaded, temps go down? Why?

So, Isn't it better to leave AMDRadeonX3000.kext unloaded if I want my GPU Die temps to remain under 50s degree Celsius? What do you suggest?

And I always use Safari. Infact, I have clean installed the OS, so not much messing around.

 

[nsgr]

Hello, thanks for your reply,
When I checked using the command, the only difference between your result and mine was in the third line:
Yours is Policy: on and Mine is Policy: off, Is this the Issue?

View attachment 779356

Besides that, my version of loaded AMD*3000.kext has Info.plist, in fact I verified all the contents. If you find any difference, here is the result of second command:

View attachment 779357

Also, when I keep the Machine Idle, the GPU diode temperature hovers in 68-79 degree Celsius with Kext Loaded.

One Important Fact: When I keep The AMDRadeonX3000.kext Unloaded, the GPU Diode Automatically drops to 30-50 degree Celsius, of course the AMD Vcore is taking up power/Voltage. But When I load AMDRadeonX3000.kext, the AMD GPU Vcore is not taking power/Voltage but temperature goes high around 70-90 degree Celsius. Here is my screen shot: (see Vcore to confirm)
AMD*3000.kext not loaded: AMD*3000.kext Loaded:
View attachment 779358 View attachment 779359

Why is this so? When kext is loaded, temps go up, but when kext is unloaded, temps go down? Why?

So, Isn't it better to leave AMDRadeonX3000.kext unloaded if I want my GPU Die temps to remain under 50s degree Celsius? What do you suggest?

And I always use Safari. Infact, I have clean installed the OS, so not much messing around.

1 - What software are you using to measure the temperature?

2 - What is the result of the Terminal?

nvram -p

I'm testing the agdc=2 option on boot-args. Only in Recovery Mode - SIP is enable.
I think that's why Policy=on was enabled.

nvram boot-args="-v agdc=2"

In this post I only had the boot-args="-v" and the result was Policy=off. #1790
Temperature: CPU Die analog=30 Celsius and GPU Die analog=zero with kextload AMDRadeonX3000.kext.

 

[stormxeron]

1 - What software are you using to measure the temperature?

-> I have iStat Menus v. 6.20 and Temperature Gauge pro both are showing same GPU die analog temp at 72 degree celsius

2 - What is the result of the Terminal?

nvram -p

=>Result

Idyll:~ idyll$ nvram -p

SystemAudioVolume    %db

boot-gamma    %10%06%00%00%a3%9c%00%00%00%00%00%00%be%00%00%00%00%00%00%00%0e%00%01%04%8a%00%03%0d%be%03%84%12%14%07%06%19}%0c%08"%8a%17L2%18.%8e:%1e8W]%e6\a%86b%82%aa%a9|%ab%ac%b32%b9%ae%bbw%c2%b1%c6`%cd4%d2%d3%d6%0d%00%01%04w%00%03%0dZ%03%04%12%05%06%06%191%0b%08"%25%15%0d4%9c+%0f>L6%9ak%99`d%93%0a%82%ab%ad%f3%9e%b2%c9`%c1%f9%e5a%df%bd%f4@%f0%0e%00%c1%04r%00%03%0d%9c%02%06%19%ba%08H!%d9%0fL3%cd!%0f>d+%dez%16[*%a9%f4|%ac%b1S%84s%cd%b7%a1%fa%e9G%c8%fb%ef%c8%d2%fd%f6%ba%e1%fe%fa%1c%ee

boot-args    -v agc=0

backlight-level    %f4%00

gpu-policy    %01

bluetoothInternalControllerInfo    %1a%82%ac%05%000%11%fa%b8%8d%12)-%ff

prev-lang:kbd    en:0

SystemAudioVolumeDB    <

previous-system-uuid    A9B8826A-0097-3FBF-90D1-31A71529AF8B

efi-boot-device    <array><dict><key>IOMatch</key><dict><key>IOProviderClass</key><string>IOMedia</string><key>IOPropertyMatch</key><dict><key>UUID</key><string>6722FA11-18C2-455E-BB9E-FC9AC255B5F8</string></dict></dict><key>BLLastBSDName</key><string>disk1s2</string></dict><dict><key>IOEFIDevicePathType</key><string>MediaFilePath</string><key>Path</key><string>\CFE257D3-49A2-4577-9E2F-0024B9CA446C\System\Library\CoreServices\boot.efi</string></dict></array>%00

bluetoothActiveControllerInfo    %1a%82%ac%05%00%00%000%11%fa%b8%8d%12)-%ff

efi-boot-device-data    %02%01%0c%00%d0A%03%0a%00%00%00%00%01%01%06%00%02%1f%03%12%0a%00%00%00%00%00%00%00%04%01*%00%02%00%00%00(@%06%00%00%00%00%00`%d2%97;%00%00%00%00%1a%bd%d8kmT%80A%b8%85%fe%13p%dd%0a%9a%02%02%04%03$%00%f7%fct%be|%0b%f3I%91G%01%f4%04.hB%11%fa"g%c2%18^E%bb%9e%fc%9a%c2U%b5%f8%04%04%9a%00\%00C%00F%00E%002%005%007%00D%003%00-%004%009%00A%002%00-%004%005%007%007%00-%009%00E%002%00F%00-%000%000%002%004%00B%009%00C%00A%004%004%006%00C%00\%00S%00y%00s%00t%00e%00m%00\%00L%00i%00b%00r%00a%00r%00y%00\%00C%00o%00r%00e%00S%00e%00r%00v%00i%00c%00e%00s%00\%00b%00o%00o%00t%00.%00e%00f%00i%00%00%00%7f%ff%04%00

csr-active-config    w%00%00%00

fmm-computer-name    Idyll

LocationServicesEnabled    %01

I'm testing the agdc=2 option on boot-args. Only in Recovery Mode - SIP is enable.
I think that's why Policy=on was enabled.

nvram boot-args="-v agdc=2"

=>What does it do? I am going to test it.
->Well, I tested it agdc=2 cooled the gpu a bit further to around 66 degrees.

 

[monofase]

Hi all,
Anyone had success in installing Windows 10 with failed AMD GPU?
macOS is working fine with software solution. But now I need to make it working under Windows. I installed Windows 10 (EFI Mode) and works only in safe mode. Is there a way in Windows to disable the AMD dead chip?

 

[JMVB]

Hi all,
Anyone had success in installing Windows 10 with failed AMD GPU?
macOS is working fine with software solution. But now I need to make it working under Windows. I installed Windows 10 (EFI Mode) and works only in safe mode. Is there a way in Windows to disable the AMD dead chip?

Hello, thanks for your reply,
When I checked using the command, the only difference between your result and mine was in the third line:
Yours is Policy: on and Mine is Policy: off, Is this the Issue?

View attachment 779356

Besides that, my version of loaded AMD*3000.kext has Info.plist, in fact I verified all the contents. If you find any difference, here is the result of second command:

View attachment 779357

Also, when I keep the Machine Idle, the GPU diode temperature hovers in 68-79 degree Celsius with Kext Loaded.

One Important Fact: When I keep The AMDRadeonX3000.kext Unloaded, the GPU Diode Automatically drops to 30-50 degree Celsius, of course the AMD Vcore is taking up power/Voltage. But When I load AMDRadeonX3000.kext, the AMD GPU Vcore is not taking power/Voltage but temperature goes high around 70-90 degree Celsius. Here is my screen shot: (see Vcore to confirm)
AMD*3000.kext not loaded: AMD*3000.kext Loaded:
View attachment 779358 View attachment 779359

Why is this so? When kext is loaded, temps go up, but when kext is unloaded, temps go down? Why?

So, Isn't it better to leave AMDRadeonX3000.kext unloaded if I want my GPU Die temps to remain under 50s degree Celsius? What do you suggest?

And I always use Safari. Infact, I have clean installed the OS, so not much messing around.

According con Realmacmods:

"The safe boot clears OS level GPU preferences, that may interfere with the following process.

This will now cause your MacBook Pro to stop switching to the Radeon automatically, but it will still draw power, create heat, and be visible to the OS.

We discovered that simply removing 1 resistor will resolve this.

The resistor in question is R8911 on the 17″ MBP and R8911 on the 15″ MBP
a 1 Ohm resistor that provides a current path to the ISL6263C DC to DC Converter.

Just remove it. The preferred method is a professional reflow station, but an iron and a steady hand will get you where you need to be. If you used flux to remove it (not needed), make sure you clean up with a little Alcohol or other suitable solvent.

That is basically it. Next time you boot up you will notice your GPU defect issue is gone, and you will no longer see the AMD GPU as installed hardware.

A small warning: It has been tested that this mod will survive a SMC reset as well as Battery Removal. It is not yet known if it will survive a NVRAM reset. For this (and other security related) reasons, it is recommended that you set an EFI Password. This will make it impossible to accidentally reset your NVRAM. Just don’t ever forget this password. Seriously, don’t. Ever.

In the event that a NVRAM reset removes the EFI mod, replacing R8911, or even bridging the pads for R8911 will restore functionality so you can reapply the EFI mod. Really though, lets just stay away from the NVRAM resets. They rarely fix anything that couldn’t be fixed otherwise anymore.

ADDITIONAL STEP FOR HIGH SIERRA

High Sierra requires an additional hardware mod. This jumper wire is different between 15″ and 17″, both are shown below. Without this mod, your screen WILL NOT turn on after going to sleep. On the 15″ we just need to put a small jumper wire (enamel wire) between PIN 2 of R9704 and PIN 1 of C9711. Logic board does not need to be removed for this step. Only bottom cover removed.

I'm in El capitan, and I have not intentions to perform an OS upgrade, so, I don't need the second mod, I only did the first one.

 

[nsgr]

-> I have iStat Menus v. 6.20 and Temperature Gauge pro both are showing same GPU die analog temp at 72 degree celsius


=>Result

Idyll:~ idyll$ nvram -p

SystemAudioVolume    %db

boot-gamma    %10%06%00%00%a3%9c%00%00%00%00%00%00%be%00%00%00%00%00%00%00%0e%00%01%04%8a%00%03%0d%be%03%84%12%14%07%06%19}%0c%08"%8a%17L2%18.%8e:%1e8W]%e6\a%86b%82%aa%a9|%ab%ac%b32%b9%ae%bbw%c2%b1%c6`%cd4%d2%d3%d6%0d%00%01%04w%00%03%0dZ%03%04%12%05%06%06%191%0b%08"%25%15%0d4%9c+%0f>L6%9ak%99`d%93%0a%82%ab%ad%f3%9e%b2%c9`%c1%f9%e5a%df%bd%f4@%f0%0e%00%c1%04r%00%03%0d%9c%02%06%19%ba%08H!%d9%0fL3%cd!%0f>d+%dez%16[*%a9%f4|%ac%b1S%84s%cd%b7%a1%fa%e9G%c8%fb%ef%c8%d2%fd%f6%ba%e1%fe%fa%1c%ee

boot-args    -v agc=0

backlight-level    %f4%00

gpu-policy    %01

bluetoothInternalControllerInfo    %1a%82%ac%05%000%11%fa%b8%8d%12)-%ff

prev-lang:kbd    en:0

SystemAudioVolumeDB    <

previous-system-uuid    A9B8826A-0097-3FBF-90D1-31A71529AF8B

efi-boot-device    <array><dict><key>IOMatch</key><dict><key>IOProviderClass</key><string>IOMedia</string><key>IOPropertyMatch</key><dict><key>UUID</key><string>6722FA11-18C2-455E-BB9E-FC9AC255B5F8</string></dict></dict><key>BLLastBSDName</key><string>disk1s2</string></dict><dict><key>IOEFIDevicePathType</key><string>MediaFilePath</string><key>Path</key><string>\CFE257D3-49A2-4577-9E2F-0024B9CA446C\System\Library\CoreServices\boot.efi</string></dict></array>%00

bluetoothActiveControllerInfo    %1a%82%ac%05%00%00%000%11%fa%b8%8d%12)-%ff

efi-boot-device-data    %02%01%0c%00%d0A%03%0a%00%00%00%00%01%01%06%00%02%1f%03%12%0a%00%00%00%00%00%00%00%04%01*%00%02%00%00%00(@%06%00%00%00%00%00`%d2%97;%00%00%00%00%1a%bd%d8kmT%80A%b8%85%fe%13p%dd%0a%9a%02%02%04%03$%00%f7%fct%be|%0b%f3I%91G%01%f4%04.hB%11%fa"g%c2%18^E%bb%9e%fc%9a%c2U%b5%f8%04%04%9a%00\%00C%00F%00E%002%005%007%00D%003%00-%004%009%00A%002%00-%004%005%007%007%00-%009%00E%002%00F%00-%000%000%002%004%00B%009%00C%00A%004%004%006%00C%00\%00S%00y%00s%00t%00e%00m%00\%00L%00i%00b%00r%00a%00r%00y%00\%00C%00o%00r%00e%00S%00e%00r%00v%00i%00c%00e%00s%00\%00b%00o%00o%00t%00.%00e%00f%00i%00%00%00%7f%ff%04%00

csr-active-config    w%00%00%00

fmm-computer-name    Idyll

LocationServicesEnabled    %01

=>What does it do? I am going to test it.
->Well, I tested it agdc=2 cooled the gpu a bit further to around 66 degrees.

agc = AppleGraphicsControl -> AppleMuxControl.kext

agdc = AppleGraphicsDeviceControl -> AppleGraphicsDeviceControl.kext

These options are not documented by Apple.
We go in blind flight. Trial and error.


Test:

1 - Turn off your Wi-Fi connection.

2 - Turn off your bluetooth -> System Preferences -> Bluetooth.

3 - Disconnect ethernet cable.

4 - Turn off the updates -> System Preferences -> App Store -> disable "Automatic check for updates" and disable "Automatic download apps".

5 - Shutdown Macbook Pro.

6 - Reset SMC -> press (left-side) Shift-Control-Option along with power button just once.

7 - Reset NVRAM/PRAM -> Press power button -> Immediately press and hold Command-Option-P-R keys.

8 - Boot Single User Mode (Command + S keys at boot) and set up gpu-power-prefs to Intel GPU and reboot:

nvram fa4ce28d-b62f-4c99-9cc3-6815686e30f9:gpu-power-prefs=%01%00%00%00
reboot

No nvram gpu-policy=%01
No nvram boot-args="-v agc=0"

If you want you can leave only the verbose boot. Boot verbose does not interfere with anything.

nvram boot-args="-v"

9 - Boot normal -> after login -> Macbook Pro 2011 idle with LoginHook or kextload AMDRadeonX3000.kext -> wait 4 minutes and check temperature.

When the GPU Die Analog heats up (50, 60, 70 Celsius degrees), then the zone next to the MagSafe / External microphone connector / above the ESC / F1 / F2 / F3 keys heats up a lot (aluminum).

When the temperature of the GPU Die Analog is at zero, then the aluminum in the above zones is cold or slightly warm.

You can compare the aluminum temperature between the ESC, F1, F2 (left side) and the F11, F12 (right side) key zones with your fingers.

------------
Resetting the System Management Controller (SMC)
The System Management Controller (SMC) is a chip on the logic board that controls all power functions.
If the computer is experiencing any power issue, such as not starting up, not displaying video, sleep issues, or fan noise issues, resetting SMC may resolve it.


To reset SMC:

1. If computer is on, turn it off by choosing Shut Down from the Apple () menu.

2. Connect power adapter to computer and to a working power source.

3. On built-in keyboard, press (left-side) Shift-Control-Option along with power button just once. Important: Use Shift-Control-Option keys on left side of keyboard.

Note: When the LED on the MagSafe connector is orange/amber, resetting the SMC will change it to green for a few seconds, indicating that SMC was correctly reset.

4. Wait 5 seconds and press power button to start computer.

Note: If bottom case is removed, you can alternately reset SMC by disconnecting both power adapter and main battery, and holding down power button for 5 seconds.


Resetting Parameter RAM (PRAM)
PRAM stores certain system and device settings in a location that Mac OS X can access quickly. Exactly which settings are stored in the computer’s PRAM varies depending on the type of computer as well as the types of devices and drives connected.

To reset PRAM:

1. If computer is on, turn it off by choosing Shut Down from Apple () menu.

2. Locate the following keys on the keyboard: Command, Option, P, and R. You will need to hold these keys down simultaneously in Step 4.
Note: If keyboard does not have an Option key, use Alt key instead.

3. Press power button.

4. Immediately press and hold Command-Option-P-R keys.

Important: You must press this key combination before the gray screen appears.

5. Hold down keys until computer restarts, and you hear startup chime a second time.

6. Release keys.


-----------------

Reset SMC / Reset NVRAM PRAM #1916

MikeyN Guide deacrese AMD GPU - System temperature - Page 35 - #875

Rename old AMDRadeonX3000.kext - move new AMDRadeonX3000.kext after update - Page 57 - #1425

Shutdown-Restart Macbook-black screen-Page 57-#1425

Close Lid-Sleep-Wake Pages 43-50 #1066 #1243

 

[stormxeron]

agc = AppleGraphicsControl -> AppleMuxControl.kext

agdc = AppleGraphicsDeviceControl -> AppleGraphicsDeviceControl.kext

These options are not documented by Apple.
We go in blind flight. Trial and error.


Test:

1 - Turn off your Wi-Fi connection.

2 - ..........

Thank you for your reply.
I did what you said:

>reset smc (orange changed to green) >reset nvram(pressed keys until booted second time) > set nvram .......... gpu-power-prefs=%01%00%00%00 > reboot in verbose > load AMDRadeonX3000.kext

I did not set 'agc' or 'agdc' values this time.

Before loading the kext: GPU Die Temperature= 45 degree celsius.
After Loading the kext: GPU Die Temperature=68 degree celsius.
After Waiting 20 mins with Full Fan Speed: GPU Die Temperature= 52 degree celsius.
->I did not find significant temperature difference between the left side(above esc, f1, f2, f3) and right side(above f10,f11,f12), slightly warm but I would call it cold still. It certainly did not feel like it was above 50s.

As of writing this now with normal fan speed: GPU Die temperature= 59 degree celsius.

After nvram -p in the terminal, the result was very less than before:

bluetoothInternalControllerInfo    %1a%82%ac%05%000%11%fa%b8%8d%12)-%ff

fmm-computer-name    Idyll

SystemAudioVolumeDB    >

backlight-level    %f4%00

SystemAudioVolume    %db

LocationServicesEnabled    %01

boot-args    -v

Any other suggestions? because this temp I think is way above since you mentioned it should be '0'. Isn't it?
I am confused still: the temperature is very low when I did not load the kext, but after loading the kext, it goes up suddenly. Isn't that opposite of what should happen?

And lastly, should I set agdc=2 or agc=0 or 9? Will it help/work in High Sierra?

 

[nsgr]

Thank you for your reply.
I did what you said:

>reset smc (orange changed to green) >reset nvram(pressed keys until booted second time) > set nvram .......... gpu-power-prefs=%01%00%00%00 > reboot in verbose > load AMDRadeonX3000.kext

I did not set 'agc' or 'agdc' values this time.

Before loading the kext: GPU Die Temperature= 45 degree celsius.
After Loading the kext: GPU Die Temperature=68 degree celsius.
After Waiting 20 mins with Full Fan Speed: GPU Die Temperature= 52 degree celsius.
->I did not find significant temperature difference between the left side(above esc, f1, f2, f3) and right side(above f10,f11,f12), slightly warm but I would call it cold still. It certainly did not feel like it was above 50s.

As of writing this now with normal fan speed: GPU Die temperature= 59 degree celsius.

After nvram -p in the terminal, the result was very less than before:

bluetoothInternalControllerInfo    %1a%82%ac%05%000%11%fa%b8%8d%12)-%ff

fmm-computer-name    Idyll

SystemAudioVolumeDB    >

backlight-level    %f4%00

SystemAudioVolume    %db

LocationServicesEnabled    %01

boot-args    -v

Any other suggestions? because this temp I think is way above since you mentioned it should be '0'. Isn't it?
I am confused still: the temperature is very low when I did not load the kext but after loading the kext, it goes up suddenly. Isn't that opposite of what should happen?

What remained was to test with another AMDRadeonX3000.kext.

0 - Move or rename this AMDRadeonX3000.kext with problems to put the new AMDRadeonX3000.kext into the /DisableExtensions or /System/Library/Extensions-off directory.


1 - Test another AMDRadeonX3000.kext that is complete with Info.plist.

I'm not sure if these two updates have the kexts inside.

https://support.apple.com/kb/DL1974?viewlocale=en_US&locale=en_US

https://support.apple.com/kb/DL1973?viewlocale=en_US&locale=en_US


2 - You have to mount the .DMG file in the Finder and extract the pkg file with the Pacifist software with Adminstrator permissions - #1674

https://www.charlessoft.com

To ensure, change the root and wheel permissions on the AMDRadeonX3000.kext that was extracted.

sudo chmod -R 755 /DisableExtensions/AMDRadeonX3000.kext
sudo chown -R root:wheel /DisableExtensions/AMDRadeonX3000.kext

3 - Disable LoginHook -> reboot -> boot normal -> and test by manually loading AMDRadeonX3000.kext with kextutil:

sudo defaults delete com.apple.loginwindow LoginHook
reboot

4 - Load new extract AMDRadeonX3000.kext with kextutil - example /DisableExtensions directory

kextutil /DisableExtensions/AMDRadeonX3000.kext

5 - Wait 4 minutes and check temperature.


Update:

agc=0 or agc=9, then the AMDRadeonX3000.kext power off AMD GPU.
agc=1, 2, 3, 4, 5, 6, 7, 8, so AMDRadeonX3000.kext does not power off AMD GPU.

agdc is still in the testing phase.


Update 2:

Or test with the AMDRadeonX3000.kext that comes in the installer (usb stick / pendrive) or an older update like 10.13.3 or 10.13.4.