Jump to content
fantomas1

Nvidia Web Driver updates for macOS Sierra (UPDATE 07/11/2018)

1,048 posts in this topic

Recommended Posts

Advertisement

To get it working I enabled clover nvram emulator (emu variable) and installed startup rc scripts to enable the driver.Then select Nvidia Web-driver from menu bar applet and reboot

 

Don't know if it's a clover issue a difference in the way Nvidia detects which driver is active ,be it OS X default or web.

Share this post


Link to post
Share on other sites

So I tried every method described below my post and have got none of them to work. I seem to have tried every combination of those and still anytime I boot without nv_disable=1 I get stuck in a reboot loop.

 

I am assuming that enabling nvram emulator in cover is the "EmuVariableUefi" driver under "Drivers UEFI 64 BIT" - On the install drivers screen, I have the following installed;

EmuVariableUefi

OsxAptioFix2Drv

VBoxHfs

And under extra drivers, OsxFatBinaryDrv UEFI

 

Boot Flags are just dart=0 and nada_drv=1, of course that isn't working causing the reboot loop

 

I have tried both MacPro 6,1 and iMac14,1 (which was working great before this)

 

Please let me know if anyone has any ideas....

Share this post


Link to post
Share on other sites

When trying to 

 

 

set manually nvda_drv=1 in the nvram

I used 'sudo nvram boot-args="nvda_drv=1"' and the response I get is "nvram: Error setting variable - 'boot-args': (iokit/common) general error"

 

I have tested the nvram and confirmed values, even boot arguments I write like "-v", are retained after reboot. Though, when I rebooted it did not boot in verbose mode. So I am not sure even finding a way to add nada_drv=1 to my nvram is going to help if it is ignoring it. 

 

Why all of a sudden is the nvda flag in clover not enough for macOS? Is there a different flag I can try to force it to use the web drivers? 

Share this post


Link to post
Share on other sites

Ok I was able to get it to work. First thing I had to do was uninstall "EmuVariableUefi" in clover. After reboot, I was able to see the hidden file "NVRAM.plist" in the root OS directory, duplicate, edit, delete original, and replace with edited plist to include boot-arg "nvda_drv=1". Once I switched back my SMBIOS to iMac 17,1 booted up fine with the web drivers working!! Had to double check that my board id was still set to none in the graphic policy kext as well, but it was.

 

Also, I uninstalled the web drivers in the system preference pane and reinstalled during this process so that also could have helped. 

 

So now I am booting without any bootflags in clover, everything seems fine even without "dart=0" and the web driver utility is actually telling me it is using web drivers (check box next to it) which it never had done that before, so that made me happy.....haha

 

Thanks everyone for your posts as it all helped me tremdously in finally finding this solution.

Share this post


Link to post
Share on other sites

Guys, I'm getting reboots when trying to boot with the Nvidia driver on a GTX 950.

 

It used to work with DP1. Then I updated to DP2, I thought it's the driver. So now I also updated the driver, and I got the same issue: reboot on second stage boot, when booting with nvda_drv=1. The only way I can boot right now, is with nv_disable=1. But, of course, that denies the purpose of having the Nvidia driver installed in the first place.

 

Any idea what's wrong here? Am I missing something?

Share this post


Link to post
Share on other sites

Guys, I'm getting reboots when trying to boot with the Nvidia driver on a GTX 950.

 

It used to work with DP1. Then I updated to DP2, I thought it's the driver. So now I also updated the driver, and I got the same issue: reboot on second stage boot, when booting with nvda_drv=1. The only way I can boot right now, is with nv_disable=1. But, of course, that denies the purpose of having the Nvidia driver installed in the first place.

 

Any idea what's wrong here? Am I missing something?

I had the same problem, but now with latest Nvidia Web Driver - 367.05.10.05b07 it is Ok.

and Activate manually the Webdriver usage in Configuration panel

 

Hope this helps

Share this post


Link to post
Share on other sites

I had the same problem, but now with latest Nvidia Web Driver - 367.05.10.05b07 it is Ok.

and Activate manually the Webdriver usage in Configuration panel

 

Hope this helps

Well, I do have that driver version. And that's exactly what I can't seem to be able to do: activate it.

 

I mean, I do activate it, it requires restart, and after restart, it goes into restart again (the issue described in my previous post). I think the driver doesn't get activated, for some reason. Just can't pin point the reason.

Share this post


Link to post
Share on other sites

Well, I do have that driver version. And that's exactly what I can't seem to be able to do: activate it.

 

I mean, I do activate it, it requires restart, and after restart, it goes into restart again (the issue described in my previous post). I think the driver doesn't get activated, for some reason. Just can't pin point the reason.

 

Do you have the variable nvda_drv 1 defined in your nvram? It seems like the bootarg option is no longer recognised by the web driver. Type sudo nvram nvda_drv=1 in the console and check that the line nvda_drv 1 appears in you nvram (sudo nvram -p) and that it is persistent between boots. boot-args nvda_drv=1 seems like it is being ignored by the web driver.

 

Hope this helps.

 

Cheers,

Share this post


Link to post
Share on other sites

Do you have the variable nvda_drv 1 defined in your nvram? It seems like the bootarg option is no longer recognised by the web driver. Type sudo nvram nvda_drv=1 in the console and check that the line nvda_drv 1 appears in you nvram (sudo nvram -p) and that it is persistent between boots. boot-args nvda_drv=1 seems like it is being ignored by the web driver.

 

Hope this helps.

 

Cheers,

 

Thanks, I'll check that tomorrow. And yes, I was using it as boot argument. In case it's not in nvram, how exactly should I go from there?

Share this post


Link to post
Share on other sites

Ok, so  nvda_drv=1 is not persistent in nvram. But neither is it on El Capitan. So I'm not sure that's the issue here.

 

What are the options for this case?

Share this post


Link to post
Share on other sites

Hi arsradu,

 

Assuming that you are using clover in UEFI only mode you need the EmuVariableUefi driver, the RC scripts and the clover control panel to be installed, if you have this options already install proceed as below, otherwise install these option and restart you machine.

 

Now, in the clover control panel go to the NVRAM variables tab and from the dropbox "Save NVRAM variables to disk" select "Always", I'm not sure if this is necessary but in my case only started to work after this.

 

Now set the nvda_drv variable to 1 in the NVRAM, type in the console sudo nvram nvda_drv=1 and press enter, check that the variable has been set to 1, type nvram -p and look for the line "nvda_drv      1", no "=" should appear. Restart you machine and hope for the best, this worked for me but as you can see in the posts above is much hit or miss with the Sierra beta 2.

 

Good luck!

Share this post


Link to post
Share on other sites

Had these troubles too with Sierra, not with El Cap before. Anyways on a hunch I also saved the nvram to the disk and now it seems to work right... but surprised by it but whatever works works...

So far have restarted 20'ish times just to make sure, both restart + cold start, and no black screens anymore =).

Share this post


Link to post
Share on other sites

Hi arsradu,

 

Assuming that you are using clover in UEFI only mode you need the EmuVariableUefi driver, the RC scripts and the clover control panel to be installed, if you have this options already install proceed as below, otherwise install these option and restart you machine.

 

Now, in the clover control panel go to the NVRAM variables tab and from the dropbox "Save NVRAM variables to disk" select "Always", I'm not sure if this is necessary but in my case only started to work after this.

 

Now set the nvda_drv variable to 1 in the NVRAM, type in the console sudo nvram nvda_drv=1 and press enter, check that the variable has been set to 1, type nvram -p and look for the line "nvda_drv      1", no "=" should appear. Restart you machine and hope for the best, this worked for me but as you can see in the posts above is much hit or miss with the Sierra beta 2.

 

Good luck!

EmuVariableUefi driver - checked

RC Scripts - checked

Save NVRAM variables to disk - set to Always, but it changes back to Auto after restart.

 

Also, setting the nvda_drv=1 in nvram is not persistent after reboot. I think I'm gonna reinstall, and start with the PB1 maybe. Hope that works.

Share this post


Link to post
Share on other sites
Save NVRAM variables to disk - set to Always, but it changes back to Auto after restart.

 

Do you have an existing nvram.plist at the root of your disk ? If so, try to delete it then recheck Clover prefpane on next startup.

Share this post


Link to post
Share on other sites

Do you have an existing nvram.plist at the root of your disk ? If so, try to delete it then recheck Clover prefpane on next startup/

Finally, fixed! :)

 

I think there was a bit of failure on my side, as well, since I was booting from the El Capitan SSD, instead of Sierra HDD. And even though they both have the exact same EFI content, same Clover version, I was calling for a nvram.plist that was not located on the HDD I was trying to boot from. :D So..yeah... My bad on that part.

 

Maybe my failures can help others learn so that they can avoid issues like these in the future. :)

 

By the way, now there is no nvda_drv=1 boot argument in my config anymore. And...so far so good.

 

Thank you guys so much!

Share this post


Link to post
Share on other sites

@fantomas1,

 

Thank you for your Webdriver for Sierra public beta. It worked. 

I have GeForce GTX 960 4GB OC and I am running macOS Sierra Public Beta. 3 days in a row, when I played you tube video through Safari, after 3-4 songs, both montior screens suddenly turned black, then the disk icons reappeared while screen remaied black. Then the system crashed (unresponsive). Had to restart with restart button. Then, last night, I installed Chrome browser and played the same you tube videos, i did not experience black screens, freezes or system crashes even after 10 songs, It appeared this may be a Safari browser issue. Anybody has similar experience like me?  

Share this post


Link to post
Share on other sites

Hi Guys,

I have updated to DP3 with the latest web drivers but cannot boot up, only with nv_disable flag.

I used nvram patch and checked nvram -p looks OK.

Any other idea?

I don't know if my GFX supported under 10.2..

Share this post


Link to post
Share on other sites

Hi, seems I have the same a strange problem as Arsradu in post 69 here when using Sierra DP2  (Clover r3652) and Clover configurator 4.31

Even if I change the bootargument from nvda_drv=1 to nv_disable=1

On booting the both are remaining ! look on image below ( -> blackscreen)

 

So I have to modifiy it manually everytime at bootscreen and then boot is OK.

 

Never saw this before !

 

I have tried to nvram -c    but it comes back everytime ...

 

Seems it picks up nvram.plist parameters from my original bootdisk (Sierra DP2) from where I newely installed Sierra-SSD (PB2)

 

Is there a thing to change in Clover configurator preferencepane concerning NVRAM ? ( Auto save or never ?)

How usefull is NVRAM on hackintosh ? ( Tidbits releaved that resetting NVRAM desactivated localisation )

 

Thanks for your lights !

 

post-305243-0-75803000-1469373931_thumb.jpg

Share this post


Link to post
Share on other sites

Looks like you're right.

 

I got a GT 730 that absolutely needs webdrivers in El Capitan and below but runs just fine in Sierra without them.

for me no web drivers will boot with no gfx support, i have 750 TI

maybe my nvram related, how to clear nvram?

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

  • Recently Browsing   0 members

    No registered users viewing this page.

  • Similar Content

    • By WarDoc
      This Post is for helping users with issue that come with Whatevergreen

      Feel free to join in and help users if you know your stuff
      also get the dev's of whatever green to join as well


      Whatevergreen has a few issues

      Multi-monitor Support is lacking as 2 monitors seems to be the limit for some even if the devs say it's not users say it's true

      Some Id's are wrong or missing (hence the nano and 390X not working)

      some GPU names are wrong ( which has been reported many times)

      FOR RYZEN OWNERS with Polaris :
      Modified kext:/S/L/E/AMD9510Controller.kext//Contents/Info.plistfrom:<key>IOPCIMatch</key><string>0x67EF1002</string>to:<key>IOPCIMatch</key><string>0x67FF1002</string> Updating clover to Clover_v2.4k_r4152 and using the config.plist here https://pastebin.com/TYiTEVcMwithbusratio=36 and npci=0x3000 ALSO THIS THREAD IS NOT FOR WHINING in anyway please do not post off topic if you don't like given support in this forum that's all you but majority of use need it.
       
      Source code: repository. FAQ and documentation: link.  
      a real SSDT example https://www.dropbox.com/sh/el66izg1lmk1jiu/AADCZUEjyQyiWWELlmzW2ix-a?dl=0 Credits to Pavo for the hard work of decoding the mess
      Boot argumentsAdd -raddbg to enable debug printing (available in DEBUG binaries).Add -radvesa to disable ATI/AMD video acceleration completely.Add -radoff to disable WhateverGreen.Add -radbeta to enable WhateverGreen on unsupported os versions (10.13 and below are enabled by default).Add -rad24 to enforce 24-bit display mode.Add -radlogo to patch boot logo distortion.Add -raddvi to enable DVI transmitter correction (required for 290X, 370, etc.)Add radpg=15 to disable several power-gating modes (see FAQ, required for 7xxx GPUs). If you Don't Want To use WEG or LILU clover 4297 has a new patch that fixes the same issues without kexts 
      You have to set in config.plist <key>Graphics</key> <dict> <key>RadeonDeInit</key> <true/>
    • By Thereman
      Hello, I'm a noob at installing MacOS on my current PC (which is running Windows 10 x64 Pro).
      Can I do a dual boot with windows 10 and MacOS Mojave? I already have windows 10 installed and I would like to put and MacOS Mojave, can it? If not, should I reinstall from 0 MacOS Mojave and after Windows 10?
       
      P.S: I have already make a topic about "Can I run MacOS" and the export boys on the forum said to me I can run it.
    • By fusion71au
      This is a brief guide on how to create a vanilla El Capitan (also same process for Sierra) OS X Installer USB with an updated prelinked kernel containing FakeSMC.

      It is tailored for those users who want to understand the “nuts and bolts” of how to create an installer and also to help brush up on their terminal skills (rather than have one made for them with the numerous automated “tools” available or even Apple’s createinstallmedia) :
       
       
      Specifically, it is also a “Proof of Concept” which shows that installation is possible without even having to rely on boot loader kext injection.

      Prerequisites
      Existing Yosemite installation (or Mavericks - see post#4 for steps 7,8) “Install OS X El Capitan.app" downloaded to the Applications folder Pacifist FakeSMC.kext - Slice or Kozlek branch Bootloader - Clover or Chameleon 8GB or larger USB drive (16GB recommended), formatted HFS+ (MBR or GUID) named “Installer”  
      Procedure
      1.  Boot into Yosemite with the kext-dev-mode=1 boot flag
      2.  Open OS X terminal and type the following lines, followed by <Enter> after each line.
          The image restore and file copying may take a while to complete, and at the end of the process, the Installer volume is renamed to “OS X Base System"....
       
      sudo -s hdiutil attach /Applications/Install\ OS\ X\ El\ Capitan.app/Contents/SharedSupport/InstallESD.dmg asr restore -source /Volumes/OS\ X\ Install\ ESD/BaseSystem.dmg  -target /Volumes/Installer -erase -format HFS+ -noprompt -noverify rm /Volumes/OS\ X\ Base\ System/System/Installation/Packages cp -av /Volumes/OS\ X\ Install\ ESD/Packages /Volumes/OS\ X\ Base\ System/System/Installation cp -av /Volumes/OS\ X\ Install\ ESD/BaseSystem.dmg /Volumes/OS\ X\ Install\ ESD/BaseSystem.chunklist /Volumes/OS\ X\ Base\ System diskutil unmount /Volumes/OS\ X\ Install\ ESD exit 3.  Right click on the “OS X Base System” Volume and click “Get Info”
       

      4.  Click on the lock icon and untick “Ignore ownership on this volume”
       

      5.  Extract/Copy the El Capitan Kernels folder into /System/Library/ of the USB with Pacifist.  NB It is found in the "Essentials.pkg" in /System/Installation/Packages
       

      6.  Delete or Rename the original /System/Library/PrelinkedKernels/prelinkedkernel —> OG.prelinkedkernel
       

      7.  Copy FakeSMC.kext and other necessary kexts (e.g. VoodooPS2Controller.kext for laptops) into the /Library/Extensions folder of the installer USB using Finder.
       

      8.  Back in terminal, type the following lines, followed by <Enter> after each line to rebuild the prelinkedkernel…..
       
      sudo -s chmod -R 755 /Volumes/OS\ X\ Base\ System/Library/Extensions chown -R 0:0 /Volumes/OS\ X\ Base\ System/Library/Extensions touch /Volumes/OS\ X\ Base\ System/System/Library/Extensions kextcache -u /Volumes/OS\ X\ Base\ System exit Any errors should be noted but the output below is normal e.g.


      9. Install your Bootloader targeting the OS X Base System volume
       
       
       
       
      10.  Boot your system with the USB without injected kexts into the OS X Installer GUI....
       
       
       
       
       
      Post Install
      The original prelinked kernel in a fresh install of El Capitan will also lack FakeSMC ie it will only be linked to Apple signed kexts.  In order to boot into El Capitan the first time around without boot loader kext injection, the PLK needs to be rebuilt for the El Capitan volume like we did for the installer:
       
      1.  Boot into Yosemite with the kext-dev-mode=1 boot flag
      2.  Delete or Rename the original /System/Library/PrelinkedKernels/prelinkedkernel for the El Capitan volume —> OG.prelinkedkernel
      3.  Copy FakeSMC.kext and other necessary kexts (e.g. VoodooPS2Controller.kext for laptops) into the /Library/Extensions folder of El Capitan using Finder
      4.  Back in terminal, type the following lines, followed by <Enter> after each line to rebuild the prelinkedkernel.  In this example, the El Capitan volume is named "El_Capitan" - change if you have named it something else...
      sudo -s chmod -R 755 /Volumes/El_Capitan/Library/Extensions chown -R 0:0 /Volumes/El_Capitan/Library/Extensions touch /Volumes/El_Capitan/System/Library/Extensions kextcache -u /Volumes/El_Capitan exit 5. Install your Bootloader targeting the El Capitan volume.  This step is only necessary if you are installing El Capitan on a new drive without existing boot loader (not required if installing on a disk with existing boot loader beside Yosemite).
      6. Reboot your system without injected kexts into El Capitan!
       
       
      Other links
      The All-In-One Guide to Vanilla OS X for beginners
       
      Updates for Sierra and High Sierra   Custom Prelinkedkernel Generator Tool I have made a custom prelinkedkernel generator "PLK.tool" for Sierra + El Capitan.  Instructions:   1.  Download and extract the attached BaseSystem_PLK.tool.zip into your ~/Downloads folder.   2.  Copy any extra kexts necessary for booting your hack to ~/Downloads/BaseSystem/ExtraKexts (e.g. FakeSMC, VoodooPS2Controller), making sure SIP is disabled. 3.  Copy BaseSystem.dmg to ~/Downloads/BaseSystem  4.  Open terminal and run the following commands... cd ~/Downloads/BaseSystem chmod +x PLK.tool ./PLK.tool ---> supply your admin password ---> will place your new custom prelinkedkernel on the desktop.       macOS High Sierra bypass Firmware and MBR checks in post#13.
      BaseSystem_PLK.tool_ElCap.zip
      BaseSystem_PLK.tool_Sierra.zip
      BaseSystem_PLK.tool_Mojave.zip
    • By fantomas1
      Hi InsanelyMacaliens    
       
      Use this thread to link / talk about of the future Nvidia Web Driver updates for macOS High Sierra.
       
      10.13.6
      Nvidia Web Driver - 387.10.10.15.15.108 --> build 17G2307 (thanks to marcozardi)  New!
      Nvidia Web Driver - 387.10.10.10.40.105 --> build 17G65 (thanks to sl0wtarget)
       
      10.13.5
      Nvidia Web Driver - 387.10.10.10.35.106 --> build 17F77 (check here)
       
      10.13.4
      Nvidia Web Driver - 387.10.10.10.30.107 --> build 17E202 (check here)
      Nvidia Web Driver - 387.10.10.10.30.106 --> build 17E199 (thanks to BreBo)
      Nvidia Web Driver - 387.10.10.10.30.103 --> build 17E199 (thanks to ricoc90)
       
       
      10.13.3
      Nvidia Web Driver - 387.10.10.10.25.161 --> build 17D102 (thanks to Cyberdevs)
      Nvidia Web Driver - 387.10.10.10.25.160 --> build 17D2102 (thanks to Cyberdevs)
      Nvidia Web Driver - 387.10.10.10.25.159 --> build 17D2012 (thanks to cyberdevs)
      Nvidia Web Driver - 387.10.10.10.25.158 --> build 17D102 (thanks to cyberdevs)
      Nvidia Web Driver - 387.10.10.10.25.157 --> build 17D2047 (thanks to cyberdevs)
      Nvidia Web Driver - 387.10.10.10.25.156 --> build 17D47 (thanks to darthsian) 
       
       
      10.13.2
      Nvidia Web Driver - 378.10.10.10.25.106 --> build 17C2205 (thanks to KGP-X99)
      Nvidia Web Driver - 378.10.10.10.25.105 --> build 17C2120 (see this post)
      Nvidia Web Driver - 378.10.10.10.25.104 --> build 17C205 (thanks to ReddestDream)
      Nvidia Web Driver - 378.10.10.10.25.103 --> build 17C89 (thanks to Il Francy MacPc) 
      Nvidia Web Driver - 378.10.10.10.25.102 --> build 17C88 (thanks to piiggggg)
       
       
      10.13.1
      Nvidia Web Driver - 378.10.10.10.20.109 --> build 17B1003 (see this post)
      Nvidia Web Driver - 378.10.10.10.20.108 --> build 17B1002 (thanks to BreBo)
      Nvidia Web Driver -378.10.10.10.20.107 --> build 17B48 (thanks to piiggggg) 
       
       
      10.13.0
      Nvidia Web Driver - 378.10.10.10.15.121 --> build 17A405 (thanks to Asgorath)
      Nvidia Web Driver - 378.10.10.10.15.120 --> build 17A405 (thanks to phi777)
      Nvidia Web Driver - 378.10.10.10.15.117 --> build 17A405 (thanks to phi777)
      Nvidia Web Driver - 378.10.10.10.15.114 --> build 17A365 (thanks to Asgorath).
    • By thehexabyte
      Hey all.
      I'm trying to install macOS Sierra on a partition of my main PC (specs below), and I keep running into the same problem. Once I get past the Apple logo and loading bar, it will stick. Whether that's while choosing a language or even before the installer shows up.
      Sometimes I get to the language picker, and when I pick the language and click the next arrow, the words will stop and the button will become greyed out. I can still move my mouse and select languages but nothing else. Or I don't even get to that. Sometimes it just displays the background of the installer with a cursor that I can move. I've waited for ages and nothing happens. Also, sometimes it'll reboot and when it does in verbose mode, it reboots too quickly to see the panic.
       
      Sorry I'm very new to hackintoshes.
       
      Hayden
       
      Specs:
      Intel Core i5-6400
      Intel HD Graphics 530
      8GB DDR3 RAM
       
      The computer is a prebuild HP Pavilion 550-251na Desktop.
×