Jump to content

VampireHunter_D

Members
  • Content count

    64
  • Joined

  • Last visited

About VampireHunter_D

  • Rank
    InsanelyMac Protégé
  1. @caleidosCope I ended up getting it working. I think a kext in my 10.12 folder might have been causing a conflict the the display port. I am working on putting together a little snapshot of my EFI folder and other kexts and will attempt to pull together a short guide. It would still be useful to know what kexts you have in your 10.12 folder since I think some from the original El Capitan package may no longer be needed anymore. Thanks again for the response.
  2. @caleidosCope So do you have the external display working with Sierra? Using your AppleIntelFrameBufferCapri.kext from post #176 and "AAPL,ig-platform-id", Buffer (0x04) { 0x04, 0x00, 0x66, 0x01 }, in the DSDT only the internal display works. Setting the ig-platform-id to 0x09, 0x00, 0x66, 0x01 allows the external display to work but causes the internal to fail with a severe backlight problem in which I feel left running too long might actually damage the display. (I get a halo affect on the screen on reboots if I do not let the device cool down a bit). So if it is working for you, a few questions., 1.) What is the ig-platform-id set to in the DSDT you are using? 2.) Can I confirm the AppleIntelFrameBufferCapri.kext you are using is the one from post #176. 3.) I see you mention you are using IOGraphicsFamily.kext (ver 2.4.1). This appears to be the same version I see in my install of Sierra (even though there is a size different form El Capitan). So assuming yours is different and you can not post it, do you know what modifications were to it to make it compatible? Do you know if it is vanilla which version of OSX it came from? 4.) Which kexts do you currently have in your 10.12 folder for CLOVER? 5.) In your config.plist do you have any binary patching settings implemented that might be related to this? Any useful information you can provide would be appreciated. Thanks!
  3. I did some testing and managed to modify JahStories DSDT from looking at myr1s to get BlueTooth working under El Capitan since when I used myr1s it seemed the display panel was not found correctly among other issues. It is counter to most DSDT edits that I have usually done but it appears simply renaming the XHC and EHC* to USB* solved the issue. Obviously it is somehow related to the new USB stack in El Capitan, and while it does not make any sense it seems to work. Attached are two modified DSDTs, DSDT_no_PWR.aml and DSDT_w_PWR.aml. DSDT_w_PWR.aml includes the _PWR Methods in the USB* devices, I am not sure if this will break sleep so if it does try DSDT_no_PWR.aml. Good luck! FixBtTestDSDT.zip
  4. VampireHunter_D

    [Guide] Surface Pro 1st gen Yosemite clean installation

    Got rid of the glitchyness. I ended up installing all the kexts from the 5.1 surface package to S/L/E except for AppleFrameBufferCapri which I rolled back to the default. At startup there is a quick glitch during the Apple Logo and progress bar, but nothing after log in. So macray, try installing the original AppleFrameBufferCapri and see if this helps. Note I have everything in S/L/E now so I do not need to do kext injection.
  5. VampireHunter_D

    [Guide] Surface Pro 1st gen Yosemite clean installation

    I am having the same problem. Oddly enough it only seemed to start happening after I added the modfied framebuffer, backlight, etc. kexts to S/L/E . Before that however even with kext injection from Clover my boots were spotty at best with the video not loading at all. If I figure anything out I will let you know. Please share if you find anything as well.
  6. VampireHunter_D

    IntelHD3000 - ML - White tint Internal Display

    Yes you can edit an EDID. I have been attempting different edits to mine (like the gamma) to try and resolve this. There is a Windows program called "Phoenix EDID Designer" which I use for this. It can be a little tedious since you need to format the hex string in a way which the application will load it, then back to the format for DSDT injection or Base64 for a boot.plist, but it works. I did not realize people with external Samsung monitors were also having this issue. Have they attempted to file a bug report with Apple? Obviously if I tried I would be ignored based on the situation, however if I had a monitor I was attaching to my MBP and it had this problem I would definitely be seeking support.
  7. VampireHunter_D

    IntelHD3000 - ML - White tint Internal Display

    I have the same problem with a Surface Pro and a HD 4000. In my searching it appears that this can affect both HD 3000 and HD 4000 in either AppleIntelFrameBufferSNB or AppleIntelFrameBufferCapri respectively. However there is one common thing it appears and it is that all the devices that exhibit the problem are using Samsung displays as previously mentioned. It appears it might be related to something in the EDID but I have not been able to determine what it is. If you reduce the gamma in the EDID it helps a little bit, but not enough and appears to be the wrong path (white point data?). It also appears that it might be related to the loading of AppleBackLightDisplay, as if I fool my device into thinking the monitor is a desktop monitor and load just AppleDisplay the problem does not appear, however I lose the ability to sleep the computer due to the monitor, and I am unable to reduce the brightness. In my case I can do this from using platform-id 0166000A on a hd 4000 and injecting my EDID. More or less you need to make the port configuration in the FrameBuffer kext match a DisplayPort and not a LVDS for the device id you are using. I would really like to figure out what the proper way to fix this is. Maybe if there is some EDID guru out there they might know.
  8. VampireHunter_D

    Intel HD 4000 On Surface Pro Washed Out Screen From AppleIntelFrameBufferCapri

    I made a clean install and update to Mavericks, and there has been a slight change. Using Platform IDs of 01660000, 01660001, 01660002, 01660008, and 01660009 I am able to boot and right when the display initializes it freezes at the apple logo. All off these ids use this port configurations in AppleIntelFrameBufferCapri 01 00 00 00 02 00 00 00 30 00 00 00 Using platform IDs 01660004 still results in the white tinted washed out screen with full sleep support and 0166000A boots to a normal screen with no Power Management since it thinks the display is a desktop monitor. Attached are the latest DSDT I am using. DSDT.aml.4 is using the native EDID from the Suface Pro display that has been run through FixEdid, and DSDT.aml.a is using the EDID from Clover that has been run through FixEDID. So the problems to solve are fixing a HD4000 with a white tint or washed out screen, OR, fix the issue when the display freezes on initialization, OR, get power management to work with a monitor the OS thinks is a desktop display (ACPIBackLight.kext does not seem to work). I am open to trying any suggestions, Thanks. DSDT_4_A_fedid.zip
  9. VampireHunter_D

    [HELP] HD 4000 Increase VRAM size

    Clover has a VRAM option in the Graphics section. <key>Graphics</key> <dict> <key>VRAM</key> <integer>512</integer> <key>DualLink</key> <string>1</string> </dict> You could give that a try. As for the perl script patch, what version of Mac OS are you on? You might need to verify that the hex string you are trying to modify is actually in you AppleIntelFrameBufferCapri.kext.
  10. VampireHunter_D

    Intel HD 4000 On Surface Pro Washed Out Screen From AppleIntelFrameBufferCapri

    So I kind of gave up messing with the connector properties since that seems like somewhat of dead end especially with 10.8.5 and a new AppleIntelFrameBufferCapri coming. However after some poking I was able to create a DSDT that boots using the EDID from Clover's GFX injection. The issue is when using this EDID AppleBackLightDisplay fails to load and subsequently sleep and display brightness fail and the only resolution that works is 1920x1080. So here is where I am currently at. Using ig-platform-id of 4 and using the monitors original EDID I can boot to a washed out screen that is barely legible. All resolutions work and sleep and brightness work. Using ig-platform-id of a or b with the EDID from Clover's injection I can boot to a normal looking screen that only supports 1920x1080, brightness does not work, and sleep does not work. I have attached a zip file containing two DSDTs, one set to ig-platform-id of 4 and the original EDID for the surface and the other set to ig-platform-id of a with Clover's EDID , otherwise they are identical. Both have the common set of back light data in their IGPU/_DSM section. Also included in the zip are two ioreg text dumps from when booted with these DSDTs and the config.plist that Clover is using. For reference here is the hex and Base64 for both EDIDs Surface Pro EDID: /* 0000 */ 0x00, 0xFF, 0xFF, 0xFF, 0xFF, 0xFF, 0xFF, 0x00, /* 0008 */ 0x4C, 0xA3, 0x42, 0x32, 0x00, 0x00, 0x00, 0x00, /* 0010 */ 0x00, 0x16, 0x01, 0x03, 0x80, 0x17, 0x0D, 0x78, /* 0018 */ 0x0A, 0x87, 0xF5, 0x94, 0x57, 0x4F, 0x8C, 0x27, /* 0020 */ 0x27, 0x50, 0x54, 0x00, 0x00, 0x00, 0x01, 0x01, /* 0028 */ 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, /* 0030 */ 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x2A, 0x35, /* 0038 */ 0x80, 0xA0, 0x70, 0x38, 0x0B, 0x40, 0x18, 0x10, /* 0040 */ 0x35, 0x00, 0xEB, 0x84, 0x00, 0x00, 0x00, 0x19, /* 0048 */ 0x00, 0x00, 0x00, 0x0F, 0x00, 0x00, 0x00, 0x00, /* 0050 */ 0x00, 0x00, 0x00, 0x00, 0x00, 0x44, 0xDC, 0x03, /* 0058 */ 0x74, 0x00, 0x00, 0x00, 0x00, 0xFE, 0x00, 0x53, /* 0060 */ 0x41, 0x4D, 0x53, 0x55, 0x4E, 0x47, 0x0A, 0x20, /* 0068 */ 0x20, 0x20, 0x20, 0x20, 0x00, 0x00, 0x00, 0xFE, /* 0070 */ 0x00, 0x31, 0x30, 0x36, 0x48, 0x4C, 0x30, 0x31, /* 0078 */ 0x2D, 0x30, 0x30, 0x31, 0x0A, 0x20, 0x00, 0xEA AP///////wBMo0IyAAAAAAAWAQOAFw14Cof1lFdPjCcnUFQAAAABAQEBAQEBAQEBAQEBAQEBKjWAoHA4C0AYEDUA64QAAAAZAAAADwAAAAAAAAAAAETcA3QAAAAA/gBTQU1TVU5HCiAgICAgAAAA/gAxMDZITDAxLTAwMQogAOo= Clover EDID: /* 0000 */ 0x00, 0xFF, 0xFF, 0xFF, 0xFF, 0xFF, 0xFF, 0x00, /* 0008 */ 0x36, 0x7F, 0x03, 0x00, 0x0F, 0x00, 0x00, 0x00, /* 0010 */ 0x00, 0x0C, 0x01, 0x03, 0x80, 0x00, 0x00, 0xFF, /* 0018 */ 0x2F, 0x00, 0x00, 0xA0, 0x57, 0x49, 0x9B, 0x26, /* 0020 */ 0x10, 0x48, 0x4F, 0x00, 0x00, 0x00, 0x01, 0x01, /* 0028 */ 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, /* 0030 */ 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x02, 0x3A, /* 0038 */ 0x80, 0x18, 0x71, 0x38, 0x2D, 0x40, 0x58, 0x2D, /* 0040 */ 0x36, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x1E, /* 0048 */ 0x00, 0x00, 0x00, 0xFC, 0x00, 0x4E, 0x6F, 0x74, /* 0050 */ 0x65, 0x62, 0x6F, 0x6F, 0x6B, 0x50, 0x61, 0x6E, /* 0058 */ 0x65, 0x6C, 0x00, 0x00, 0x00, 0xFD, 0x00, 0x3C, /* 0060 */ 0x3D, 0x43, 0x44, 0x0F, 0x01, 0x00, 0x00, 0x00, /* 0068 */ 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x10, /* 0070 */ 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, /* 0078 */ 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0xBC AP///////wA2fwMADwAAAAAMAQOAAAD/LwAAoFdJmyYQSE8AAAABAQEBAQEBAQEBAQEBAQEBAjqAGHE4LUBYLTYAAAAAAAAeAAAA/ABOb3RlYm9va1BhbmVsAAAA/QA8PUNEDwEAAAAAAAAAAAAAEAAAAAAAAAAAAAAAAAAAALw= My thought is there might be some mod I can do to the EDID that might get the AppleBackLightDisplay booted when using ig-platform-id a. So if anyone has any thoughts or any good references about fixing the AppleBackLightDisplay from EDID issues please let me know. Thanks. Spro_Update.zip
  11. VampireHunter_D

    [HELP] HD 4000 Increase VRAM size

    What have you tried so far? Also what bootloader are you using?
  12. VampireHunter_D

    [HELP] HD 4000 Increase VRAM size

    Here is the link again if it gets removed then I am not sure what else to do.
  13. VampireHunter_D

    [HELP] HD 4000 Increase VRAM size

    Strange, I thought I included the link. Maybe there is a cross-post issue I am unaware of. If this the case web search for Unified AppleIntelFramebufferCapri for Ivy ProBooks If this does not work you could also try injecting VRAM,totalsize or VRAM,totalMB into your DSDT _DSM section for GFX0 or IGPU.
  14. VampireHunter_D

    [HELP] HD 4000 Increase VRAM size

    Yes it appears that you can by hacking the AppleIntelFrameBufferCapri.kext. The link has a little perl script that will make the modification on all the ig-platform-ids, just scroll down to where it says "A little bonus discovery:".
  15. I am looking for any suggestions on how to resolve an issue I have where the screen is distored, blurred, hazy, or washed out with an intel hd 4000 on a surface pro. I have attached two DSDTs one with ig-platform-id 4 and the other with 9 and some screen shots. platform-id-4.jpg is what the washed out display looks like and gfxinject.jpg is an image taken if I use an older version of clover with a slightly patched DSDT and NullCPU (I can not get this combo to work on the newer versions of Clover nor do I really want to since there is no APM among other issues). The kernel options I am passing are -v npci=0x2000 slide=0 dart=0 To boot I am using Clover 1974, and using a MacBookPro10.1 SMBIOS definition. The Surface Pro appears to have two ports, 0 is the internal display and 1 is a mini displayport. I have tried every ig-platform-id. The best I get is using platform id 1660004 which seems to match the hardware pretty closely with 32mb (as Windows reports) and one port. With this platform id and no modification to AppleIntelFrameBufferCapri I get the washed out display as shown in platform-id-4.jpg. Sleep and speed stepping all seem to work fine, just the display is incorrect. Based upon what Windows reports as the card being a Rev.9 and when looking at the system profiler it sees it as a 1660009 I have also focused on this platform. If I hack AppleIntelFrameBufferCapri I still get the same results as using ig-platform-id 4 with booting to a washed out screen. Below are the mods I have tried on AppleInteFrameBufferCapri Defaults are AAPL,ig-platform-id: 01660004, Stolen RAM for FB: 32MB, Pipes: 3, Ports: 1, FBMem: 1 @0, 05 03 00 00 02 00 00 00 30 02 00 00 // (LVDS) AAPL,ig-platform-id: 01660009, Stolen RAM for FB: 64MB, Pipes: 3, Ports: 3, FBMem: 3 - 0mb VRAM @0, 01 00 00 00 02 00 00 00 30 00 00 00 // (LVDS) @1, 02 05 00 00 00 04 00 00 07 01 00 00 // (DisplayPort) @2, 03 04 00 00 00 04 00 00 07 01 00 00 // (DisplayPort) For 01660004 here are the mods I have tried on port 0 //Switch the port from 02 00 to 00 04. Boot is the same washed out 05 03 00 00 00 04 00 00 30 02 00 00 //Remove Dual Link?? It is necessary but I figured I would try. Boots to black screen 05 03 00 00 02 00 00 00 30 00 00 00 //Use LVDS from platform id-9. Boots to black screen 01 00 00 00 02 00 00 00 30 00 00 00 //Use LVDS from platform id-9 and add dual link??? Boots to black screen 01 00 00 00 02 00 00 00 30 02 00 00 For 01660009 here are the mods I have tried on port 0 //Add dual link??? to lvds. Boots to black screen 01 00 00 00 02 00 00 00 30 02 00 00 //Use LVDS from platform id 4. Boots to washed out screen 05 03 00 00 02 00 00 00 30 02 00 00 //Use LVDS from platform id 4 with port from 02 00 to 00 04. Boots to washed out screen 05 03 00 00 00 04 00 00 30 02 00 00 I tried a few other random combinations with no success, and honestly I was just guessing. I either got a black screen or no display. If anyone has any thoughts are suggestions I am open to anything. At this point I am not sure what else to try. If there is a piece of important information I forgot to include please feel free to ask. Thanks. Spro_hd4000.zip
×