Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
Status
The first post of this thread is a WikiPost and can be edited by anyone with the appropiate permissions. Your edits will be public.
I don’t rely only on change log, it’s very easy to see the differences in GitHub repo.
The only differences are version number and typo fixes in documentation.



1700594064163.png
1700594146531.png

If it’s working for you now, it’s not due to 1.3
 
Last edited:
I don’t rely only on change log, it’s very easy to see difference in GitHub repo.
The only differences are version number and typo fixes in documentation.

If it’s working for you now, it’s not due to 1.3
Agree that GitHub shows no significant differences. I'm not a KDK expert and don't fully understand the role of the KDK in the post-install patches, so this is a naive question: could it be that a KDK update might be the reason?
 
You are right 1.3 and 1.2.1 are not the same.
The version number is different and 2 typos in a documentation file have been fixed 🤓
They have replaced « those entires » by « those entries »
Huge difference
Not only I had a lot of WiFi Agent crashes with the last 1.2.1 and old 1.3.0n also - this is fixed now ;-) even when it is not in the changelog at the moment…
Agree that GitHub shows no significant differences. I'm not a KDK expert and don't fully understand the role of the KDK in the post-install patches, so this is a naive question: could it be that a KDK update might be the reason?
same KDK only change is the newest 1.3.0n don‘t worry I also have an eye on the repo - but I don‘t see any other reason so far… file size also too different btw… there's no need to get upset i thought the info might help others, that's it. (i'm incredibly grateful to be able to continue using my beloved hardware, just in terms of the latest xcode etc.) and your statement "If it’s working for you now, it’s not due to 1.3" is totally o.k. for me, but i know my system quite good, and for better comparability i even flashed the same nvram under linux - all meant in a relaxed way - take care.
 
Last edited:
Not only I had a lot of WiFi Agent crashes with the last 1.2.1 and old 1.3.0n also - this is fixed now ;-) even when it is not in the changelog at the moment…

same KDK only change is the newest 1.3.0n don‘t worry I also have an eye on the repo - but I don‘t see any other reason so far… file size also too different btw…
I don't know enough to dispute your claim, but it would carry much more credibility if you reverted to 1.2.1 (Release) and confirmed that the Wi-Fi Agent crashes returned with 1.2.1. You don't need to waste your time for my benefit (or anyone else's), but if you want to prove to yourself... Regardless, glad your Wi-Fi issue is resolved.
 
  • Like
Reactions: PropClear
Its hardware was not liking typos in accel.md doc file, or it prefers 1.3 version number, it’s the only logical explanation.
 
Has anyone encountered this bug yet?

I think, that only occurs on the google search page, on safari.
The font switches to to another font on hover, and even if not hovered it has some weird pixelization.
I have already tried deleting cookies and disabling all extension- but didn't helped.
It only seems to happen if it concerns a Google result that is displayed first with subtitles.
Likely a Safari bug, not OS bug reported in the article. 🧐
 
I don't know enough to dispute your claim, but it would carry much more credibility if you reverted to 1.2.1 (Release) and confirmed that the Wi-Fi Agent crashes returned with 1.2.1. You don't need to waste your time for my benefit (or anyone else's), but if you want to prove to yourself... Regardless, glad your Wi-Fi issue is resolved.
did that before the post, the Wifi Agent crashes where directly back with 1.2.1 release version... ;-) take care and thank you - im glad also without the annoying wifi agent crashes - so that was the reason for my post i thought that it may help others - take care deeveedee ;-)
 
  • Like
Reactions: deeveedee
did that before the post, the Wifi Agent crashes where directly back with 1.2.1 release version... ;-) take care and thank you - im glad also without the annoying wifi agent crashes - so that was the reason for my post i thought that it may help others - take care deeveedee ;-)
I admit that I'm stumped and can't explain it. Wouldn't be the first time ;)
 
14.2 Beta 3 works as well as any of the 14.2 Betas on MBP6,2 patched with OCLP 1.2.1. With OCLP, it continues to be a mistake to make blanket statements about macOS / OCLP performance when individual results will depend on Mac Model, use cases and apps. Still best to test and make one's own determination. Currently very happy with 14.2 Beta 3, but that's just my experience and does not guarantee another's.

EDIT: CPU utilization at idle and CPU fan activity remain low on my MBP6,2 running 14.2 Beta 3 (23C5047e) patched with OCLP 1.2.1 (Release).
Readng the past ten thread pages or so. It appears the OCLP v.1.2.1 non-metal patches seem less problematic than some of the metal iGPU patches in Ventura and Sonoma. 😏
 
  • Like
Reactions: hvds
Not only I had a lot of WiFi Agent crashes with the last 1.2.1 and old 1.3.0n also - this is fixed now ;-) even when it is not in the changelog at the moment…

It doesn't matter what's noted in the changelog! What matters it what's in the commit history. And in this regard, no substantial changes besides a version bump and some typo fixes have been made since OCLP 1.2.1. If you don't believe me, check it for yourself:


The only possible reason I can imagine why your wifi issues are gone NOW is that you haven't updated the EFI last time prior to patching so you were using older kext still.
 
Last edited:
It doesn't matter what's noted in the changelog! What matters it what's in the commit history. And in this regard, no substancial changes besides a version bump and some typo fixes have been made since OCLP 1.2.1. If you don't believe me, check it for yourself:


The only possible reason I can imagine why your wifi issues are gone NOW is that you haven't updated the EFI last time prior to patching so you were using older kext still.
Causality can be a bitch.
 
It doesn't matter what's noted in the changelog! What matters it what's in the commit history. And in this regard, no substancial changes besides a version bump and some typo fixes have been made since OCLP 1.2.1. If you don't believe me, check it for yourself:


The only possible reason I can imagine why your wifi issues are gone NOW is that you haven't updated the EFI last time prior to patching so you were using older kext still.
nope done EFI and root patches as always... 👨‍🔧
 
  • Like
Reactions: K two
Yesterday I was working on a project & suddenly my desktop crashed giving black screen and brought me to the login screen. The error report popped up showing windowserver. I am on mac Os Sonoma with OCLP 1.3.0n. All was smooth earlier. Now this brings up a question of unreliability in back of my mind.
Should I go back to Ventura or stick around and keep ignoring these crashes.:rolleyes:
 
  • Like
Reactions: luckyduke23
Yesterday I was working on a project & suddenly my desktop crashed giving black screen and brought me to the login screen. The error report popped up showing windowserver. I am on mac Os Sonoma with OCLP 1.3.0n. All was smooth earlier. Now this brings up a question of unreliability in back of my mind.
Should I go back to Ventura or stick around and keep ignoring these crashes.:rolleyes:
Which Sonoma version ?
 
Yes, it should work. Some have had problems with Wi-Fi, so if possible connect with ethernet. You will have to run the root patches again after the OS is updated.
In addition: Hotspot/iPhone via USB is also an option, assuming there is a sufficient data plan in place.
 
Agree that GitHub shows no significant differences. I'm not a KDK expert and don't fully understand the role of the KDK in the post-install patches, so this is a naive question: could it be that a KDK update might be the reason?
Experimented a bit with 14.2b3 and 1.3.0n (from source, downloaded yesterday, behaviour should be same as 1.2.1).

MBP5,2: the easy case. Installed 14.2b3 over 14.1.1 from USB installer (InstallAssistant available via gibmacos). 1.3.0n in EFI and for root patches. The KDK made for b3 is loaded during root patching. No problems (as always for this machine, needed external keyboard/mouse before root patches are applied).

MBP11,1: as reported earlier in this thread, login loop due to windowserver crash, also here.
(1) installed 14.2b3 to external SSD from USB installer over 14.1.1. Using 1.3.0n in EFI. windowserver crash during login and even earlier during boot. Boot in safe mode ok. In /Library/Extensions there were two extra kexts, AppleIntelHD5000Graphics and AppleIntelFramebufferAzul, left over from 14.1.1 which don't go away when installing over. Removed them manually.
(2) normal boot and login afterwards. No wifi, no graphics accel of course.
(3) applied 1.3.0n post-install patches. No KDK is loaded ("KDK-less" patching for this machine). Root patching re-installs the two mentioned kexts to /L/E. Results in login loop. - NB a CCC data clone also puts them there.
(4) removed only AppleIntelFramebufferAzul and left AppleIntelHD5000Graphics in /L/E. Login ok, no graphics accel. Graphics resolution too high like when both kexts removed or running in safe mode.
(5) the other way around with only AppleIntelFramebufferAzul in /L/E. Login ok, no graphics accel. Graphics resolution low now so there is a visible effect.

So anyway it is easy to recover and avoid the windowserver crashes. Is one of these two kexts the culprit or their interplay with CoreDisplay, SkyLight? Don't know, needs more fine-grained investigation.

More importantly, 14.1.1 on internal SSD runs just fine on the MBP11,1. Big thank you to developers.
 

Attachments

  • Bildschirmfoto 2023-11-22 um 09.18.31.png
    Bildschirmfoto 2023-11-22 um 09.18.31.png
    155.9 KB · Views: 54
  • WindowServer crash 142b3 130n 2.txt
    43.7 KB · Views: 51
Last edited:
Reading the last comments on the last two pages it seems that the topic has become something just about OCLP updates and RC versions or betas of macOS Sonoma.

Guys, OCLP is just a facilitator, it is not necessary to launch macOS Sonoma on unsupported macs, the premise is pure OpenCore. OCLP's job is only to facilitate configuration, but you can do all of this from scratch if you know what you are doing.

For example, I am only waiting for the monthly release of the OpenCore version, which is currently in OpenCore 9.6, OCLP only serves me to install the updated root patch, which is nothing more than making the iGPU (Intel HD Graphics), bluetooth and WiFi.

This thing about reverting versions of OCLP doesn't make any sense, because to make macOS Sonoma work perfectly, you need to keep the kexts updated, if you revert to an old version of OCLP, you're demonstrating that you just don't know what you're doing and the meaning of this topic.
 
Reading the last comments on the last two pages it seems that the topic has become something just about OCLP updates and RC versions or betas of macOS Sonoma.

Guys, OCLP is just a facilitator, it is not necessary to launch macOS Sonoma on unsupported macs, the premise is pure OpenCore. OCLP's job is only to facilitate configuration, but you can do all of this from scratch if you know what you are doing.

For example, I am only waiting for the monthly release of the OpenCore version, which is currently in OpenCore 9.6, OCLP only serves me to install the updated root patch, which is nothing more than making the iGPU (Intel HD Graphics), bluetooth and WiFi.

This thing about reverting versions of OCLP doesn't make any sense, because to make macOS Sonoma work perfectly, you need to keep the kexts updated, if you revert to an old version of OCLP, you're demonstrating that you just don't know what you're doing and the meaning of this topic.
I would like to see your self created OpenCore configuration booting your 2012 Mac into Ventura or Sonoma. When you are done with failing experiments you will know why OCLP is much more than a simple OC configurator for real Macs.
 
Last edited:
Reading the last comments on the last two pages it seems that the topic has become something just about OCLP updates and RC versions or betas of macOS Sonoma.

Guys, OCLP is just a facilitator, it is not necessary to launch macOS Sonoma on unsupported macs, the premise is pure OpenCore. OCLP's job is only to facilitate configuration, but you can do all of this from scratch if you know what you are doing.

For example, I am only waiting for the monthly release of the OpenCore version, which is currently in OpenCore 9.6, OCLP only serves me to install the updated root patch, which is nothing more than making the iGPU (Intel HD Graphics), bluetooth and WiFi.

This thing about reverting versions of OCLP doesn't make any sense, because to make macOS Sonoma work perfectly, you need to keep the kexts updated, if you revert to an old version of OCLP, you're demonstrating that you just don't know what you're doing and the meaning of this topic.
Right. The importance of OCLP (on top of the rather hidden OC) comes from the history of patching macos for unsupported macs.
These patchers replaced/downgraded kexts etc and provided stubs for missing functions or binary patches e.g. for a function to always return "true" (for non-metal machines) and put them directly into the system on disk. OCLP is superior and more general by using OC, but OC itself remains behind the scenes for most of us.
 
Last edited:
Experimented a bit with 14.2b3 and 1.3.0n (from source, downloaded yesterday, behaviour should be same as 1.2.1).
I have a 2012 10,1 rMBP that got caught in the boot-loop that has been documented here.
Was meaning to either downgrade to 14.1 or wait for a possible OCLP that works.
Doing something similar (other kexts) to what you tried makes the machine reasonably usable for now, as the boot-loop can be avoided and a bonus in that Wifi started working again.
No acceleration of course, but very helpful stop-gap solution. Thanks!
 
  • Like
Reactions: hvds
Right. The importance of OCLP (on top of the rather hidden OC) comes from the history of patching macos for unsupported macs.
These patchers replaced/downgraded kexts etc and provided stubs for missing functions (for non-metal machines) and put them directly into the system on disk. OCLP is superior and more general by using OC, but OC itself remains behind the scenes for most of us.
In fact it does the same downgrade/replacement/stub fixing now for all metal macs. The non metal methods and tools are used to continue with metal patching now.

From this perspective changing an OCLP version can also change the patching and system behavior, updates are meant to change it in a positive way offering more or better hardware support, one cannot exclude the possibility a downgrade may do the same thing in certain cases and for a few machines.

Mostly macOS updates cause the new issues though. When Apple changes libs the OLCP patches may become incompatible and need to be re-engineered or replaced.
 
Last edited:
Reading the last comments on the last two pages it seems that the topic has become something just about OCLP updates and RC versions or betas of macOS Sonoma.

Guys, OCLP is just a facilitator, it is not necessary to launch macOS Sonoma on unsupported macs, the premise is pure OpenCore. OCLP's job is only to facilitate configuration, but you can do all of this from scratch if you know what you are doing.

For example, I am only waiting for the monthly release of the OpenCore version, which is currently in OpenCore 9.6, OCLP only serves me to install the updated root patch, which is nothing more than making the iGPU (Intel HD Graphics), bluetooth and WiFi.

This thing about reverting versions of OCLP doesn't make any sense, because to make macOS Sonoma work perfectly, you need to keep the kexts updated, if you revert to an old version of OCLP, you're demonstrating that you just don't know what you're doing and the meaning of this topic.

This must be the most uninformed post about OCLP ever.
 
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.