PritongKandule Posted September 19, 2013 Share Posted September 19, 2013 GT630m shows wrong RAM. 1GB instead of 2GB how to fix this? ML 10.8 on Asus k43sm Thanks a lot GPU.tiff Link to comment Share on other sites More sharing options...
Pimentel Posted September 19, 2013 Share Posted September 19, 2013 HelloHere have a guide: http://olarila.com/forum/viewtopic.php?f=3&t=3014It is in portuguese, i don't find nothing about this in english so translate with google. Link to comment Share on other sites More sharing options...
Gringo Vermelho Posted September 19, 2013 Share Posted September 19, 2013 Could be just a cosmetic issue. Install CUDA drivers and CUDA-Z to check if all your VRAM is being used. Link to comment Share on other sites More sharing options...
PritongKandule Posted September 20, 2013 Author Share Posted September 20, 2013 @gringo installed the driver and cuda-z says "CUDA not found" @pimentelx86 I followed the guide and i have this edited my org.chameleon.Boot.plist i edited this part <key> GraphicsEnabler </ key> <string> YES</ string> i change it to <key> GraphicsEnabler </ key> <string>NO</ string> and put this part <key>device-properties</key> <string> (i paste the the device-properties-new.hex here)</string> my device-properties-new.hex file i put it in the /Extra folder. I dont know if its the right thing to do. after the restart. the screen stays at black.. do you think i just need my GraphicEnabler to stay at YES instead of NO? thanks Link to comment Share on other sites More sharing options...
The_Village_Idiot Posted September 20, 2013 Share Posted September 20, 2013 I'm assuming you put the converted hex VALUES,not the filename. Here's how to install CUDA. docs.nvidia.com/cuda/cuda-getting-started-guide-for-mac-os-x/index.html You may have to edit your dsdt. Go back to GraphicsEnabler=Yes. Remove device properties,and add <key>VRAM,totalsize</key> <string>0x80000000</string> Link to comment Share on other sites More sharing options...
PritongKandule Posted September 30, 2013 Author Share Posted September 30, 2013 I have already installed the cuda driver and the cuda-z and the cuda-z show 2047.56 MiB global memory.. its just a cosmetic thing.. question now do i still need to delete kext after the installation of the cuda driver?.. thanks. Link to comment Share on other sites More sharing options...
iFIRE Posted September 30, 2013 Share Posted September 30, 2013 How To Install Nvidia Retail Drivers http://www.insanelymac.com/forum/topic/288582-new-nvidia-retail-driver-install-solution/ Link to comment Share on other sites More sharing options...
Gringo Vermelho Posted September 30, 2013 Share Posted September 30, 2013 iFire, your first link points to the staff section! Users can't see it. Here's how to install the nvidia web driver: http://www.insanelymac.com/forum/topic/291929-1085new-nvidia-web-driver-available/?p=1946332 Link to comment Share on other sites More sharing options...
PritongKandule Posted October 7, 2013 Author Share Posted October 7, 2013 downloaded the web driver.. cant install it says my computer is not supported.. is it all right if i just use my cuda driver?.. Link to comment Share on other sites More sharing options...
Gringo Vermelho Posted October 7, 2013 Share Posted October 7, 2013 As it is, the nvidia web driver only installs on Mac Pros (or Hacks using a MacPro model identifier) because those are the only Macs that let you upgrade the video card. But there are ways around that. Follow the link I posted - notice how it says, in plain English: "Here's how to install the nvidia web driver". That's why I posted it. The CUDA driver doesn't do anything for you, unless you use software that perform calculations on your GPU through CUDA. Like certain video or image processing plugins. IIt is not required and not needed on most systems, but people like to install it anyway because its nice to have more stuff working on your hack. Personally I think most people install it because they see other people installing it, without knowing what it is or what it does. If you aren't actually using CUDA, installing it is as useful as installing PhysX on an office PC. I only suggested installing the CUDA driver because I knew CUDA-Z would correctly show how much video RAM you have, it can't do that without CUDA drivers installed. 1 Link to comment Share on other sites More sharing options...
Recommended Posts