OpenCore General Discussion

[Guest 5T33Z0]

Finally, the database inside of OCAT has become really meaningful:

 

https://github.com/5T33Z0/OC-Little-Translated/tree/main/F_Desktop_EFIs

 

Enjoy!

[Andrey1970]

On 12/3/2021 at 11:44 PM, tonyx86 said:

The Dortania guide discusses boot-arg igfxfw=2 here: https://dortania.github.io/GPU-Buyers-Guide/misc/bootflag.html and indicates that the boot-arg is for "9th Gen chipset or newer (ie Z390)."  Googling 9th Gen Intel Chipset (and variations of that string) produces ambiguous results for me.  Z390 chipset supports 8th and 9th Gen Intel CPUs (as does the Intel Q370 Chipset).  Did the Dortania guide mean to say that igfxfw=2 is for 9th Gen and later Intel CPUs (not Chipsets)?  If the guide did mean to reference 9th Gen Chipsets, what are the 9th Gen Intel Chipsets?

For chipsets with ME v12 (Z390, B360, H370, H310, Q370, C246 and newer).

Old chipsets with ME v11 not supported. Such as Z370 and other.

Use boot-arg igfxrpsc=1 for KBL+ with old chipsets.

 

[deeveedee]

Thank you.  I ran a few GeekBench5 Metal tests and the slight performance improvement is measurable and repeatable.  Note that I'm testing with Chipset Q370 and CPU i9-9900 (UHD630).

 

GeekBench5 Metal Benchmark without boot-arg igfxfw=2

Spoiler

 

 

GeekBench5 Metal Benchmark with boot-arg igfxfw=2

Spoiler

 

[gorans]

2 hours ago, Andrey1970 said:

For chipsets with ME v12 (Z390, B360, H370, H310, Q370, C246 and newer).

Old chipsets with ME v11 not supported. Such as Z370 and other.

Use boot-arg igfxrpsc=1 for KBL+ with old chipsets.

 

I have Intel Z490 chipset, i5-10600 and external GPU. iGPU is active for acceleration only.

VDADecoderChecker
Hardware acceleration is fully supported

I don't have igfxfw=2. Checked DRM and it's working. Is there something that I'm missing without this flag on my conf?

[Andrey1970]

It not for DRM. It for high performance IQSV.

[deeveedee]

@Andrey1970  He's confused, because the Dortania Guide implies that it is for DRM:

 

Spoiler

 

 

Looks like the Dortania Guide could be better written to describe the applicable chipsets and intent of the boot-arg.  Thanks again for your help.

 

EDIT: If you click on the "Fixing DRM" link, there is no mention of boot-arg igfxfw=2.

 

EDIT2: @Andrey1970 This issue seems to suggest that Apple Firmware does have something to do with DRM.  It's not clear that boot-arg igfxfw=2 triggers the Apple iGPU firmware update assocated with this issue, but that may be another source of confusion regarding the use of igfxfw=2.  I may be alone on this (and if I am, please ignore), but I think a better explanation of boot-arg igfxfw=2 is necessary, including the need for the boot-arg and expected results with the boot-arg.  Thank you.

Edited December 7, 2021 by tonyx86
Added EDIT2

[deeveedee]

The upgrade to OC 0.7.6 from 0.7.5 is very simple.  Thank you, developers!  My OC upgrade steps for my rig are here.  I'm still making the OC changes manually without the assistance of a configurator, so I'm sure there are "easier" ways to do this.

[telepati]

I am getting a problem with UIScale. After changing UIScale, the first stage is OK but after blink apple logo is getting bigger. I tested all scale modes but it didn't make any difference. If I set the UIScale as 0, first and second stage the same size. Yes, I reset NVRAM after each change. I made a video that the problem is clearly seen on this video. I want to see the logo as it was in the first stage. 

 

Hardware is Dell Monitor and Connection Type DP. 

 

I will add log after restart.

Attached logs.

IMG_0137.mov.zip config.plist

opencore-2021-12-07-143654.txt Lilu_1.5.8_21.1.txt

Edited December 7, 2021 by telepati

[Henties]

@eSaF I totally agree, I hate that some magic process/program configures my OC config.plist because when something with that background magic should go wrong I would be totally lost because I would not even know where to start looking for any anomalies enabling me to return to a working and stable system.

 

Of cause on the odd occasion, and time permitting, I am playing with OCAT but that is where my use of that "tool" ends. 

 

Greetings Henties 

[Guest 5T33Z0]

There's no "background magic" included in OCAT. All is does is merging new features and their default values into an existing config. If there's any "magic" included it comes from OC validate which points at conflicts arising from contradicting settings.

There a couple of cases where certain combinations of settings are prohibited to protect users.

 

For example: You enable CustomSmbiosGUID, well then you have to change PlatforimInfo > Generic from "Create" to scustom as well.

You want to enable a theme but can't do it? Well, better change PickerMode to external

etc, etc, etc.

 

Since OCAT userbase is growing the amount of threads about config errors is decreasing gradually. You won't see many posts about errors in the OpenCore. No more fiddling around with 5 different tool to keep the config, kexts, drivers and opencore up to date.

 

And now, since the database includes about 60 base configs and dropdown meus for quirks you won't have to have to lookt at the Guide for the 1000th time to check a config.

 

[MorenoAv]

I updated mine OpenCore to 0.7.7, and I'll do it by hand all the time... one time I used OCAT and it was a nightmare...

[miliuco]

@telepati

What is the reason of igfxrpsc=1 igfxfcms=1 boot args? Your config.plist seems okay to me.

You have " Failed to load image (1/1) Resources\Image\Acidanthera\GoldenGate\Background.icns prefix:Acidanthera\GoldenGate icon:0 - Not Found" in OC log, recheck you have all required icons in the theme folder.

[mnfesq]

Is there a way to configure the CPU bus speed in OC?  Mine is not accurately detected.  In About this Mac, it says I have a 2 GHz CPU when, in fact, it is a 1.8 GHz CPU.  I can manually configure bus speed in Clover but don't see where to do it in OC.  Thanks.

[Henties]

@5T33Z0 Ever since I started hacking many moons ago I have always enjoyed getting my hands dirty at the coal face, OCAT will not change that.

 

Greetings Henties

[telepati]

1 hour ago, miliuco said:

@telepati

What is the reason of igfxrpsc=1 igfxfcms=1 boot args? Your config.plist seems okay to me.

You have " Failed to load image (1/1) Resources\Image\Acidanthera\GoldenGate\Background.icns prefix:Acidanthera\GoldenGate icon:0 - Not Found" in OC log, recheck you have all required icons in the theme folder.

It for high performance IQSV to IGPU.

Its normal because official resourse does not have Background.icns.

 

Actually, I think this problem is related to DisplayPort. I recently replaced my monitor. My previous monitor was using mDP connection and there was no such problem. It has such a problem with the DisplayPort connection on the new monitor. 

 

I must check my IGPU port settings I guess.

[Guest 5T33Z0]

43 minutes ago, Henties said:

@5T33Z0 Ever since I started hacking many moons ago I have always enjoyed getting my hands dirty at the coal face, OCAT will not change that.

 

Greetings Henties

 

You can do what you do how you want to do it - old school style. But don't discredit an app just because it can do what you do manually in 5 seconds.

[Henties]

@5T33Z0 I will certainly not and have not discredited OCAT. OCAT in my opinion serves the need for those "hackers" that do not bother too much about what is going on behind the scenes or is it curtains ? I am just not one of those. As I wrote earlier I like to get my hands dirty and fully accept that others prefer not to.

 

Regards Henties

[MorenoAv]

For finishing this, it was my one and only experience with OCAT, I didn't say that the tool was the culprit but my ignorance in using it... 

For that reason anyone can use what one want and if the tool is useful continue to use it...

Thanks

[deeveedee]

 I tested boot-arg igfxrpsc=1 with my KBL-R (i5-8250U / UHD620) HP Envy x360 15m laptop (SMBIOS MBP15,2, Open Core 0.7.6, Big Sur 11.6.1, WEG 1.5.5, Lilu 1.5.8).  After several trials with each, I think GeekBench5 Metal performance for my laptop is better without with the boot-arg.  My laptop doesn't boot with igfxfw=2, but that's to be expected.  If there are other tests that I should perform to confirm the value of igfxrpsc=1, I'm open to suggestions.

 

GB5 Metal without igfxrpsc=1

Spoiler

 

 

GB5 Metal with igfxrpsc=1

Spoiler

 

Edited December 8, 2021 by tonyx86
Metal benchmark is better WITH boot-arg igfxrpsc=1

[Tecnicaso Rico]

It's true that in the past OCAT was problematic but that is normal with new tools. Now, as of today, it's mature and very stable. I have all popular configurators available and I notice OCAT is the one, that almost all times it's the first configurator updated when necessary after a new commit that requires it.  For does who screw up or mess their config in the past: it won't happen anymore if you use the included EFI backup tool.  And for those, like me, who most of the time updates right after a new commit: it's the most convenient one because of how fast it's being updated and well maintain. The developer it's open minded and always welcome new ideas that enhance further more his tool. It reminds me to hackintool which is a beast of a tool and an all star one by the way.

 

Edited December 7, 2021 by Tecnicaso Rico

[deeveedee]

IMHO - Regardless of which "tool" or manual OC update process you use, it is wise to configure a test version of your EFI on a bootable USB drive.  Only after confirming proper operation of the new EFI should the EFI be transferred to your production boot drive.  I don't trust any EFI until I've thoroughly tested it in a sandbox.

[theroadw]

1 hour ago, tonyx86 said:

 I tested boot-arg igfxrpsc=1 with my KBL-R (i5-8250U / UHD620) HP Envy x360 15m laptop (SMBIOS MBP15,2, Open Core 0.7.6, Big Sur 11.6.1, WEG 1.5.5, Lilu 1.5.8).  After several trials with each, I think GeekBench5 Metal performance for my laptop is better without the boot-arg.  My laptop doesn't boot with igfxfw=2, but that's to be expected.  If there are other tests that I should perform to confirm the value of igfxrpsc=1, I'm open to suggestions.

 

GB5 Metal without igfxrpsc=1

  Hide contents

 

 

GB5 Metal with igfxrpsc=1

  Hide contents

 

 

One of your benchmarks is for OpenCL and the other is for Metal.

 

I made the test in my rig and there was a slight improvement using igfxrpsc=1.

From 4275 to 4314

[deeveedee]

@theroadw Your powers of observation are incredible!  I will rerun to be consistent with my benchmarks.  Thank you!

 

EDIT: @theroadw  I updated my post to ensure consistent METAL benchmarks.  As you indicated, performance is better with boot-arg igfxrpsc=1.  Thank you.

Edited December 8, 2021 by tonyx86

[pkdesign]

On 12/7/2021 at 4:06 PM, Henties said:

@5T33Z0 I will certainly not and have not discredited OCAT. OCAT in my opinion serves the need for those "hackers" that do not bother too much about what is going on behind the scenes or is it curtains ? I am just not one of those. As I wrote earlier I like to get my hands dirty and fully accept that others prefer not to.

 

Regards Henties

This entire discussion is off topic. Move it to:

 

[SavageAUS]

Is there any reason I might get constant reboots after the first update stage? My AMD machine boots and runs fine but updating to 12.1 just doesn’t want to happen.
After the first reboot I choose macintosh hd then not long after it reboots.


Sent from my iPhone using Tapatalk