Jump to content

Hackintool v3.x.x

headkaze

By headkaze,
in Hackintosh Tools

 Share
1,057 posts in this topic

Recommended Posts

mnfesq
mnfesq
Posted
Posted (edited)

First off, headkaze, cudos to you for developing a very handy and versatile hacking tool for the OSx86 community.  

 

My question is only marginally related to your app. I used it to create a patch which included the VRAM increase to 2048 MB and applied it to my config.plist with success.  However, having 32 GB of RAM, I thought, why stop at 2048 MB?  According to Intel, my UHD 620 IGPU will use all available memory up to 32 GB.  While I would never want to go anywhere near that high, I thought I would double the patch and go from 2048 MB to 4096 MB.  Now the patch you use,  which I had seen elsewhere as a VRAM Framebuffer patch script that I downloaded about a year ago, has a code of 00000080.  The 8 refers to 8 times 256 MB.  To get 4096 MB, I would need the value to be 16, i.e., 2 times 8.  However, this has to be in hex code and the highest value for a single digit is F (i.e. 15), which gives me the 3840 MB depicted in the photo. 

Untitled.thumb.png.64bda97a6b75f7b754a26a8b94bde110.png

 

Do you know if there is a way to go above 15 times 256 MB or is that the maximum that this patch will allow?  Thanks for sharing any insight you may have.

Edited by mnfesq
Link to comment
Share on other sites
MacKonsti
MacKonsti
Posted
Posted (edited)
15 hours ago, headkaze said:

Okay that was a bug. Please check out 2.5.3.

 

Thank you, I confirm Everymac link now works as expected, many thanks.

 

Another missing ID to report, please, for my audio devices I get (null) on my NUC8i7BEH2, please see image.

 

The device IDs are (per lscpi):

 

Quote

00:1f.3 Audio device [0403]: Intel Corporation Coffee Lake High Definition Audio Controller [8086:9dc8] (Rev 30)

 

Also, what is the asterisk for, please?

 

Can you add in your internal database this Intel audio device for 8086:9dc8 ? Many thanks!

 

Audio Null.png

Edited by MacKonsti
  • Like 1
Link to comment
Share on other sites
headkaze
headkaze
Posted
  • Author
Posted
17 hours ago, mnfesq said:

However, having 32 GB of RAM, I thought, why stop at 2048 MB?

The reason why there's a 2048 MB patch is it fixes display glitching for some people using 4k. There's no useful reason to go higher than that so I wouldn't bother trying to.

Link to comment
Share on other sites
mnfesq
mnfesq
Posted
Posted (edited)
1 hour ago, headkaze said:

The reason why there's a 2048 MB patch is it fixes display glitching for some people using 4k. There's no useful reason to go higher than that so I wouldn't bother trying to.

 

That is the common understanding but I have a jigsaw puzzle app that, with a 3840x2160 image, uses a ton of graphics memory and would really be better served by a DGPU which, unfortunately, I have to disable since I cannot switch between graphics cards given my HP laptop BIOS.  With 3840 MB of VRAM, the puzzle pieces and the puzzle itself move much more smoothly, especially after working on it for more than 30 minutes.

 

I'd still like to know more about the patch code and the fact that it appears that only one digit changes VRAM and, in hex, you can only go up to 15 as a single digit. Thanks.

Edited by mnfesq
Link to comment
Share on other sites
headkaze
headkaze
Posted
  • Author
Posted (edited)
53 minutes ago, mnfesq said:

 

That is the common understanding but I have a jigsaw puzzle app that, with a 3840x2160 image, uses a ton of graphics memory and would really be better served by a DGPU which, unfortunately, I have to disable since I cannot switch between graphics cards given my HP laptop BIOS.  With 3840 MB of VRAM, the puzzle pieces and the puzzle itself move much more smoothly, especially after working on it for more than 30 minutes.

 

I'd still like to know more about the patch code and the fact that it appears that only one digit changes VRAM and, in hex, you can only go up to 15 as a single digit. Thanks.

VRAM is in bytes and stored in the framebuffer as a 32-bit value. 4096 MB is 4,294,967,296 bytes. When you try and store this value in a 32-bit integer it gets truncated (bit 32) so it will be zero. So the maximum value you can store is 4,294,967,295 bytes or ~4095 MB.

 

Technically the maximum value is 0xFFFFFFFF (all 32 bits set) which you would need to patch manually like so: 

                <key>framebuffer-unifiedmem</key>
                <data>
                /////w==
                </data>

PS Hackintool's Calculator is handy for things like this.

 

MaxVRAM.png

VRAMCalc.png

Edited by headkaze
  • Thanks 1
Link to comment
Share on other sites
NGF86
NGF86
Posted
Posted

Does this seem correct? After following the USB port config guide and making the kext, rebooting etc. I've got this showing in hackintool. Should only the ports with a device name shown, only be highlighted green (showing active port)? Or have I done something wrong....

USB grab.jpg

Link to comment
Share on other sites
headkaze
headkaze
Posted
  • Author
Posted (edited)
5 hours ago, NGF86 said:

Should only the ports with a device name shown, only be highlighted green (showing active port)?

No, the ports showing a device name indicate a device plugged into the port. The green means at some point it had a device plugged into the port. The green highlighting is to help you keep track of which ports are active and which ports are not active. Once you have identified all the active ports, delete the rest.

Edited by headkaze
  • Like 2
Link to comment
Share on other sites
NGF86
NGF86
Posted
Posted
4 hours ago, headkaze said:

No, the ports showing a device name indicate a device plugged into the port. The green means at some point it had a device plugged into the port. The green highlighting is to help you keep track of which ports are active and which ports are not active. Once you have identified all the active ports, delete the rest.

Ok cool done that, many thanks!

Link to comment
Share on other sites
mnfesq
mnfesq
Posted
Posted (edited)

I continue to explore the capabilities of this app and am so impressed by them.  However, I would like to make a suggestion for future builds.  I maintain a very large collection of downloaded files.  In fact, I keep them on a separate storage drive along with videos and music (in different folders, of course.). However, I deleted my users/mnfesq/Downloads folder and replaced it with a shortcut that directs downloads to the Downloads folder on my storage drive.  As a result, I get an error when using the kext updater.  See the image below.

1832306910_KextsInstalled.thumb.png.d737b5182f1cb0e5779a4e341a4b55c5.png 

Might it be possible to allow the destination folder to be configurable so that I don't get this error? Thanks.

Edited by mnfesq
Link to comment
Share on other sites
headkaze
headkaze
Posted
  • Author
Posted
2 minutes ago, mnfesq said:

However, I deleted my users/mnfesq/Downloads folder and replaced it with a shortcut that directs downloads to the Downloads folder on my storage drive.

Why not use a Symbolic Link?

Link to comment
Share on other sites
mnfesq
mnfesq
Posted
Posted (edited)
1 hour ago, headkaze said:

Why not use a Symbolic Link?

 

It turns out that I had created symbolic links and not actual aliases.  I tried deleting the link I had and creating a new link.  After that, the kext updater worked fine.  Not sure what that was about but thanks for the reply.

Edited by mnfesq
  • Like 1
Link to comment
Share on other sites
headkaze
headkaze
Posted
  • Author
Posted (edited)
47 minutes ago, mnfesq said:

EDIT:  As you can see, it turns out that I did create symbolic links rather than actual aliases since I used the code "ln -s".  If that meant it should have worked with the kext updater, it didn't.

Symlinks work on the OS level so there shouldn't be any need for me to do anything. Maybe you need to make a symlink for /Users/mnfesq/Downloads also?

EDIT: I read your update. Glad you got it working :)

Edited by headkaze
  • Thanks 1
Link to comment
Share on other sites
DoZe
DoZe
Posted
Posted

@headkaze :

 

Hi. Just want to signal On HP ZbooK 15uG4, my codec is recognized at CX8200, but it's a CX7700 (I verify directly on the chip)

(Perhaps a problem with AppleALC.Kext ?)

 

Keep up the VERY good work :)

 

Regards, Ray

  • Like 1
Link to comment
Share on other sites
headkaze
headkaze
Posted
  • Author
Posted
3 minutes ago, DoZe said:

@headkaze :

 

Hi. Just want to signal On HP ZbooK 15uG4, my codec is recognized at CX8200, but it's a CX7700 (I verify directly on the chip)

(Perhaps a problem with AppleALC.Kext ?)

 

Keep up the VERY good work :)

 

Regards, Ray

 

CX8200 is 0x14f12008. What's yours?

Link to comment
Share on other sites
DoZe
DoZe
Posted
Posted (edited)
48 minutes ago, headkaze said:

CX8200 is 0x14f12008. What's yours?

 

@headkaze

 

Under Windows : VEN_14F1&DEV_2008&SUBSYS_103C8290&REV_1001

 

See here (page 37) : http://www8.hp.com/h20195/v2/GetPDF.aspx/c05322740.pdf    

 

I put this with subsys because i had the problem with AppleALC.kext, MacPeet made a custom AppleALC.kext 1.3.6 (bootarg: Alcid=15) version for me, because it did not work before.

 

Like Probook 4x0 G4 series, HP Zbook 15uG4 had CX7700 onboard, not CX8200
 

Edited by DoZe
Link to comment
Share on other sites
headkaze
headkaze
Posted
  • Author
Posted (edited)
1 hour ago, Hiko said:

I am using i9-9900k, which one should I choose, Ice Lace (LP) or (HP)?

Thanks!

There's no Ice Lake support in macOS. I'll be removing it from Hackintool in the next release to avoid confusion. You're going to have to use a Coffee Lake fb. I would do some searching on the forums to find out what others have been using to configure this CPU.

Edited by headkaze
Link to comment
Share on other sites
kaoskinkae
kaoskinkae
Posted
Posted 
a good idea would be in the part corresponding to the PCI to be able to sort them by type as DPCIManager does
Thank you for following my advice and modifying the PCI and being able to sort them by type

 

 

 
 
 
  
 
 
  • Like 1
Link to comment
Share on other sites
 Share
Go to topic listing
×
×
  • Create New...