Doublepranks Posted September 21, 2016 Share Posted September 21, 2016 I can confirm that the new web driver update received by autoupdate addresses the top bar transparency glitch.  Link to comment Share on other sites More sharing options...
oskr66 Posted September 22, 2016 Share Posted September 22, 2016 (edited) Unable to get graphics card working  it says " version no compatible"  GPU is an Nvidia  750Ti Edited September 22, 2016 by fantomas1 Link to comment Share on other sites More sharing options...
polyzargone Posted September 22, 2016 Share Posted September 22, 2016 (edited) Unable to get graphics card working  it says " version no compatible"  GPU is an Nvidia  750Ti  As usual : sudo nano /System/Library/Extensions/NVDAStartupWeb.kext/Contents/info.plist Then change NVDARequiredOS to this (assuming you're indeed on 16B2327e) : <key>NVDARequiredOS</key> <string>16B</string> Repair permissions, rebuild kernel cache and reboot.         Edited September 22, 2016 by fantomas1 2 Link to comment Share on other sites More sharing options...
oskr66 Posted September 22, 2016 Share Posted September 22, 2016 (edited) it works , thanks Polizargone, now graphic are working. Edited September 22, 2016 by fantomas1 Link to comment Share on other sites More sharing options...
Doublepranks Posted September 22, 2016 Share Posted September 22, 2016 Back on topic thenI'm still getting random flickering screen. It's so weird it happens from now and then and suddenly there seen to be nothing wrong with graphics again.Anybody else running into this? Link to comment Share on other sites More sharing options...
phi777 Posted September 22, 2016 Share Posted September 22, 2016 Off Topic:Too bad that I’m not a good programmer. Would be a fun project taking/reversing the maxwell driver and writing/patching a kext to support pascal. Linux support is here, the framebuffer mode is already working on OSX, FreeBSD drivers are downloadable from nvidia.Looking at the sizes of the kexts is interesting. Nvidias kexts are small, compared to AMDs...Maybe we have to buy a GTX1080 for Slice This are the binary drivers from FreeBSD, even some source code is included: NVIDIA-FreeBSD-x86_64-367.44 $ find . <...parts ommited...> ./obj ./obj/.DS_Store ./obj/libEGL.so.1 ./obj/libGL.so.1 ./obj/libGLESv1_CM.so.1 ./obj/libGLESv2.so.2 ./obj/libglx.so.1 ./obj/libnvidia-cfg.so.1 ./obj/libnvidia-eglcore.so.1 ./obj/libnvidia-glcore.so.1 ./obj/libnvidia-glsi.so.1 ./obj/libnvidia-gtk2.so.367.44 ./obj/libnvidia-ml.so.1 ./obj/libnvidia-tls.so.1 ./obj/libvdpau_nvidia.so.1 ./obj/linux ./obj/linux/libcuda.so.367.44 ./obj/linux/libEGL.so.1 ./obj/linux/libEGL_nvidia.so.367.44 ./obj/linux/libGL.so.367.44 ./obj/linux/libGLdispatch.so.0 ./obj/linux/libGLESv1_CM.so.1 ./obj/linux/libGLESv1_CM_nvidia.so.367.44 ./obj/linux/libGLESv2.so.2 ./obj/linux/libGLESv2_nvidia.so.367.44 ./obj/linux/libnvidia-cfg.so.367.44 ./obj/linux/libnvidia-eglcore.so.367.44 ./obj/linux/libnvidia-fatbinaryloader.so.367.44 ./obj/linux/libnvidia-glcore.so.367.44 ./obj/linux/libnvidia-glsi.so.367.44 ./obj/linux/libnvidia-ptxjitcompiler.so.367.44 ./obj/linux/libnvidia-tls.so.367.44 ./obj/linux/libOpenGL.so.0 ./obj/linux/libvdpau_nvidia.so.367.44 ./obj/linux/tls ./obj/linux/tls/libnvidia-tls.so.367.44 ./obj/linux/tls/tls_test ./obj/linux/tls/tls_test_dso.so[/b] ./scripts ./scripts/linux.sh ./scripts/setup.sh ./src ./src/common ./src/common/inc ./src/common/inc/cpuopsys.h ./src/common/inc/nv-gpu-info.h ./src/common/inc/nv-modeset-interface.h ./src/common/inc/nvCpuUuid.h ./src/common/inc/nvstatus.h ./src/common/inc/nvstatuscodes.h ./src/common/inc/nvtypes.h ./src/common/inc/xapi-sdk.h ./src/Makefile ./src/nvidia ./src/nvidia/ebridge_export.h ./src/nvidia/ebridge_freebsd.c ./src/nvidia/ibmnpu_export.h ./src/nvidia/ibmnpu_freebsd.c ./src/nvidia/Makefile ./src/nvidia/nv-freebsd.h ./src/nvidia/nv-kernel.o ./src/nvidia/nv-misc.h ./src/nvidia/nv-reg.h ./src/nvidia/nv.h ./src/nvidia/nvidia-sources.mk ./src/nvidia/nvidia_acpi.c ./src/nvidia/nvidia_ctl.c ./src/nvidia/nvidia_dev.c ./src/nvidia/nvidia_i2c.c ./src/nvidia/nvidia_linux.c ./src/nvidia/nvidia_modeset_interface.c ./src/nvidia/nvidia_os.c ./src/nvidia/nvidia_os_pci.c ./src/nvidia/nvidia_os_registry.c ./src/nvidia/nvidia_pci.c ./src/nvidia/nvidia_subr.c ./src/nvidia/nvidia_sysctl.c ./src/nvidia/nvlink_common.h ./src/nvidia/nvlink_freebsd.c ./src/nvidia/os-interface.h ./src/nvidia-modeset ./src/nvidia-modeset/Makefile ./src/nvidia-modeset/nv-modeset-kernel.o ./src/nvidia-modeset/nvidia-modeset-freebsd.c ./src/nvidia-modeset/nvidia-modeset-os-interface.h ./src/nvidia-modeset/nvkms-api-types.h ./src/nvidia-modeset/nvkms-ioctl.h ./src/nvidia-modeset/nvkms-kapi.h ./src/nvidia-modeset/nvkms.h <...parts ommited...> from the maxwell mac kext:Output from "nm NVDAGM100HalWeb" kext (lots of similarity if compared to NVDAGK100HalWeb kext) U _ADDRLIST_FBMEM_ONLY U _ADDRLIST_FBMEM_PREFERRED U _ADDRLIST_SYSMEM_ONLY U _ADDRLIST_SYSMEM_PREFERRED U _CliAddDeviceMemory U _CliCheckAdmin U _CliDelDeviceMemory U _CliDestroyDeviceMemoryMemDesc U _CliDispGetDispChannelInfo U _CliFindDeviceMemoryInfo U _CliFindDispChannelInfo U _CliFindDmaMappingInfoInClient U _CliFindVASpace U _CliGetChannelClassInfo <... lot's ommited ...> U _vpGetBinResInfoSigDbg_STUB U _vpGetBinResInfoSigProd_STUB U _vpGetBinResInfoUcodeHeader_STUB U _vpGetBinResInfoUcodeImage_STUB U _vpGetBinResInfoUcodePatchLocation_STUB U _vpGetBinResInfoUcodePatchSignature_STUB U _vpGetConfig_STUB U _vpGetExternalConfig_STUB U _vpGetUcodeImage_STUB U _vpHalIfacesSetup_VGPUSTUB U _vpIdle_STUB U _vpPatchUcodeSig_STUB U _vpProbeEnginePresence_STUB U _vpService_STUB U _vpSetBlockLevelClockGating_STUB U _vpStateDestroy_STUB U _vpStateLoadHal_STUB U _vpStateUnload_STUB U _vrrmgrCheckVrrReleaseAndSchedPolling U _wborConstruct_STUB U _wborDetermineCapabilities_STUB U _wborInitCapabilities_STUB U _wborReadCaps_STUB U _wborSetPropertiesList_STUB U _wborWriteCaps_STUB U _wbor_iGrp_ipVersions_getInfo_STUB U _windowConstruct_STUB U _windowDetermineCapabilities_STUB U _windowInitCapabilities_STUB U _windowReadCaps_STUB U _windowSetPropertiesList_STUB U _windowWriteCaps_STUB U _window_iGrp_ipVersions_getInfo_STUB Link to comment Share on other sites More sharing options...
crusher Posted September 22, 2016 Share Posted September 22, 2016 I wait publish in #1 post of @fantomas1 1 Link to comment Share on other sites More sharing options...
Dunhillsmoker Posted September 22, 2016 Share Posted September 22, 2016 Apple has released 10.12.1 for devs. Do not install! The current Nvidia driver (...F01) is NOT WORKING! Â I have a flashed GTX960 and the flickering is terrible :-) So wait for Nvidia to catch up with Apple. Â Cheers, Â Marc Link to comment Share on other sites More sharing options...
fantomas Posted September 22, 2016 Author Share Posted September 22, 2016 Apple has released 10.12.1 for devs. Do not install! The current Nvidia driver (...F01) is NOT WORKING!  I have a flashed GTX960 and the flickering is terrible :-) So wait for Nvidia to catch up with Apple.  Cheers,  Marc  Hi,  It's not the first time... read post #390 Link to comment Share on other sites More sharing options...
sl0wtarget Posted September 23, 2016 Share Posted September 23, 2016 Latest driver adds option to enable G-SYNC, given a G-SYNC monitor, of course. Looks like 10.12.1 PB1 is already up as well. Link to comment Share on other sites More sharing options...
mitch_de Posted September 23, 2016 Share Posted September 23, 2016 I have GTX 980 and X99 platform. nvidia drivers working well dont use boot arg nvda_drv=1 on Mac OS Sierra . slice autor of clover write about that here http://www.insanelymac.com/forum/topic/304530-clover-change-explanations/page-2 last time when clover has been updated " Rev 3733 It was my mistake setting nvda_drv=1 flag as boot-args. No! It is NVRAM variable nvda_drv that can have a value "1". So I make new config value"     <key>SystemParameters</key>     <dict>         <key>NvidiaWeb</key>         <true/>     </dict>  thats all guys everything works   Hi, ony solution for CHAMELEON users to get the old  nvda_drv=1 work in SIERRA again? I cant get (like in older OS X) the drivers loaded /used anymore  I have luck, because my gou GT 640 also works with Apple drivers but want to use Nvidia Web drivers. Link to comment Share on other sites More sharing options...
miltonsj Posted September 23, 2016 Share Posted September 23, 2016 I updated to 10.12.1 Beta.  Followed polyzargone's instructions, and everything works again. Be sure you select the Nvidia driver again from the top menu to activate it.   Thanks! Link to comment Share on other sites More sharing options...
Asgorath Posted September 23, 2016 Share Posted September 23, 2016 Hi, ony solution for CHAMELEON users to get the old  nvda_drv=1 work in SIERRA again? I cant get (like in older OS X) the drivers loaded /used anymore  I have luck, because my gou GT 640 also works with Apple drivers but want to use Nvidia Web drivers.  I'm using Clover 3763 (latest from the website at the time of writing) and the normal nvda_drv=1 boot arg works fine for me. Link to comment Share on other sites More sharing options...
mitch_de Posted September 23, 2016 Share Posted September 23, 2016 Yep, Clover devs found an way to get  nvda_drv=1 working again. (Has to do that web driver not more using kernel flag , insted reading nvram value = chameleon hasnt nvram anymore). Link to comment Share on other sites More sharing options...
citydweller Posted September 24, 2016 Share Posted September 24, 2016 I always get a black screen after resuming Sierra from standby with the latest drivers. On El Capitan everything is working as expected. Link to comment Share on other sites More sharing options...
KernkraftX360 Posted September 24, 2016 Share Posted September 24, 2016 I have these graphics issues... does anyone know how to fix? Â @GTX 750Ti with web driver I have gotten the same transparency graphics issues with my GT730. Even with the latest f01 update of Nvidia web driver for Sierra. How did you managed to fix yours mate? Surprisingly, the previous Nvidia 10.11.6 update for El Capitan works wonders. The graphics glitches happened after I updated through system preferences pane of Nvidia web driver to the latest f01. Any idea of how do I roll back to the El Capitan nvidia web driver? I tried installing it but there's an error saying the software cannot be installed on 10.12 machines. Cheers. 1 Link to comment Share on other sites More sharing options...
phi777 Posted September 25, 2016 Share Posted September 25, 2016 Just wanted to tell, Cinebench R15 has the error "Strings not found" on startup for me, clicking the Icon, but works well in Sierra if started from terminal: Downloads/CINEBENCH_R15/CINEBENCH\ OSX.app/Contents/MacOS/CINEBENCH\ OSX Just FYI PS: (on Nvidia GTX970: Geekbench 4.0.1 exits on OpenCL Benchmark (just Nvidia related??) and CUDA-Z can't be started because of missing cuda (7.5.30 is installed) ??). Link to comment Share on other sites More sharing options...
nightgambler Posted September 25, 2016 Share Posted September 25, 2016 Just wanted to tell, Cinebench R15 has the error "Strings not found" on startup for me, clicking the Icon, but works well in Sierra if started from terminal: Downloads/CINEBENCH_R15/CINEBENCH\ OSX.app/Contents/MacOS/CINEBENCH\ OSX Just FYI PS: (on Nvidia GTX970: Geekbench 4.0.1 exits on OpenCL Benchmark (just Nvidia related??) and CUDA-Z can't be started because of missing cuda (7.5.30 is installed) ??). Â Your Cuda version doesn't support macOS Sierra. The next update should fix this. I have OpenCL problems with Adobe CC programs (no OpenCL preview or rendering). Don't know if its Adobe or Nvidia related... Link to comment Share on other sites More sharing options...
Frank Nitty Posted September 26, 2016 Share Posted September 26, 2016 Any known Web drivers that support Pascal cards yet? Link to comment Share on other sites More sharing options...
imatefx Posted September 26, 2016 Share Posted September 26, 2016 Can someone give me a hint of what could be wrong   My screen is all distorted with pixels but resolution is correct, even the model number is correctly identified by Nvidia  It is distorted only for some windows and Desktop background, finder, etc... chrome gives be correct rendering most of the time.  But the RAM is only 1023 MB instead of 1024 MB  Using NVIDIA Web Driver : 367.15.10.05f01  System Information :   Hardware Overview:   Model Name: Mac Pro  Model Identifier: MacPro5,1  Processor Name: Intel Core i7  Processor Speed: 1.73 GHz  Number of Processors: 1  Total Number of Cores: 4  L2 Cache (per Core): 256 KB  L3 Cache: 6 MB  Memory: 12 GB  Bus Speed: 533 MHz  Boot ROM Version: MP51.007F.B03  SMC Version (system): 1.39f11  Graphics :  NVIDIA GeForce GT 425M:   Chipset Model: NVIDIA GeForce GT 425M  Type: GPU  Bus: PCIe  PCIe Lane Width: x16  VRAM (Total): 1023 MB  Vendor: NVIDIA (0x10de)  Device ID: 0x0df0  Revision ID: 0x00a1  ROM Revision: VBIOS 70.08.19.00.5f  Metal: Supported  Displays: BenQ GW2760HS:  Resolution: 1920 x 1080 @ 60Hz (1080p)  Pixel Depth: 32-Bit Color (ARGB8888)  Display Serial Number: CAE00316SL0  Main Display: Yes  Mirror: Off  Online: Yes  Rotation: Supported  Automatically Adjust Brightness: No  Television: Yes Display:  Resolution: 1920 x 1080  Pixel Depth: 32-Bit Color (ARGB8888)  Mirror: Off  Online: Yes  Rotation: Supported  Automatically Adjust Brightness: No  Link to comment Share on other sites More sharing options...
sl0wtarget Posted September 26, 2016 Share Posted September 26, 2016 The last time I experienced glitches like that my GPU needed replacing. Link to comment Share on other sites More sharing options...
imatefx Posted September 26, 2016 Share Posted September 26, 2016 I was not using nvidia web drivers on el capitan Link to comment Share on other sites More sharing options...
phi777 Posted September 26, 2016 Share Posted September 26, 2016 you don't have two block of pins with nothing in the middle of your dvi cable? you have just one block of connector pins?? Good is 2nd or 4th of the http://cdn.overclock.net/f/ff/ffd2a16b_DVI.jpeg Link to comment Share on other sites More sharing options...
imatefx Posted September 26, 2016 Share Posted September 26, 2016 you don't have two block of pins with nothing in the middle of your dvi cable? you have just one block of connector pins?? Good is 2nd or 4th of the http://cdn.overclock.net/f/ff/ffd2a16b_DVI.jpeg I dont have dvi port, i use hdmi, I have only nvidia graphics card, no intel cards Link to comment Share on other sites More sharing options...
phi777 Posted September 26, 2016 Share Posted September 26, 2016 I dont have dvi port, i use hdmi, I have only nvidia graphics card, no intel cards You don't have a display port or dvi you could try?? maybe switching the connector, mine, next to mainboard, works best, some ports work better than others... Edit: Oh, I just saw you have a mobile grafic card, sorry Link to comment Share on other sites More sharing options...
Recommended Posts