Hervé

Assuming you've partitioned your disk under GPT/GUID scheme (not MBR), format your target Windows partition as exFAT from Disk Utility. Then boot your Windows installation DVD and install Win on that partition, reformatting it NTFS. If, at reboot, you seem to have lost your Lion partition, simply boot it via your USB installer and re-install Chameleon bootloader on it. Chameleon should then see both the Lion and Windows partition and offer them for potential selection at boot time.

You can only achieve max resolution if you have full graphics support/acceleration, otherwise OS X will limit the screen res. 1333x720 is not what I call a normal/regular screen resolution, so one can presume you do not actually have full QE/CI. Re: SSDT, you would not normally patch the extracted files. That's not normally required. You would only use a patched DSDT and generate a SSDT via the generator script tool to manage CPU SpeedStep. As such, you would not re-use or patch the extracted DSDT files.

'could be, that's the problem with generic distros: you don't know what's been modified... But Jake's DSDT also injects the wrong device id, so that may be the problem too. I'd change that back to 0x0416 in Mavericks since it is fully supported. For Mavericks, with a vanilla installation, you could experience booting your installation with various values for boot option IntelAzulFB until you reach the framebuffer that gives you full graphics acceleration with the HD4600 (I'm pretty sure your discrete AMD Radeon HD 8790M GPU is unsupported, just like the 8690M I have in my E6440). You would then be able to patch your DSDT accordingly. I would assume that the same framebuffer I used for the HD4600 of my i5-4300M (FB=12, i.e. ig-platform-id 0x0600260a) would apply to the HD4600 of your i7 but I can't be 100% certain. You'll easily find out by trial and error... Try to boot Mavericks with just the following flags and options (respect syntax and casing): DSDT=none IntelAzulFB=12 -v The Azul FB range you can use is 0 to 15. Look at RampageDev's blog for details.

Did you rebuild your cache after AppleHPET kext removal? Until you do that, it's as if you never removed it...

If you read my E6440 guide (linked above), you'll have all details about getting the HD4600 fully functional. Jake has patched your DSDT to inject the correct platform id and that'll get you full QE/CI in Mavericks. The alternative is to boot with Chameleon option IntelAzulFB=xx (it was 12 in the case of my i5-4300M based E6440). It's all mentioned in the guide. Jake's patch injects incorrect model info though (but that's just cosmetic) and desktop HD4600 device id (0x0412), not mobile (0x0416); this is not required with Mavericks only with Yosemite (since Yos has no native support for mobile HD4600) for which this can be circumvented through binary patches of HD5000+Azul framebuffer kexts. True, DSDT injection of desktop HD4600 id saves the hassle of re-patching those kexts after each Yosemite update. Your boot plist is very basic and misses a few options & flags: kernel cache, built-in Ethernet, graphics enabler, DropDSDT if you've done SpeedStep tuning properly (i.e. generated your CPU SSDT). No need for the npci=0x2000 and PCIroot=1 though, you can get rid of that. Have a look at the boot plist posted in my guide. Re: your KP screenshots, it seems you used weird boot args combination...

Try it out. Layout-id and port numbers could be different from the E6440 of course (sorry, no personal experience with HD4400 on the E7440 mode) but the HD4400 is more or less a low-energy tuned-down version (reduced frequencies, no VGA support, etc.) of the HD4600 for ULV Haswell CPUs. Nevertheless, they're brothers of the same Haswell GT2 family. Have a look at RampageDev's blog pages for HD4400/HD4600 support.

Tried removing the AppleHPET kext?

4k sector-related boot0 Chameleon error can be fixed by booting from a different device/partition (for instance your USB installer), unmounting the unbootable disk/partition and, through Terminal app, use the dd command to overwrite the boot file to install the boot1h file. You'll get the complete set of boot files (i386 folder) for your target Chameleon version from InsanelyMac repository. This has been widely documented so you'll find the exact command and process details through a little Google search.

Look here for the patch to enable HDMI output on HD4600. Could be the exact same patch on HD4400.

You must have patched incorrectly; I had no problem activating on/off control from OS X for my DW375 whether in Mavericks or Yosemite. All that's required is to add the Broadcom card PCI ids to the BroadcomBluetoothHostControllerUSBTransport PlugIn of the IOBluetoothFamily kext: <key>Dell DW375: PID 33159/0x8187, VID 16700/0x413C</key><dict> <key>CFBundleIdentifier</key> <string>com.apple.iokit.BroadcomBluetoothHostControllerUSBTransport</string> <key>IOClass</key> <string>BroadcomBluetoothHostControllerUSBTransport</string> <key>IOProviderClass</key> <string>IOUSBInterface</string> <key>LMPLoggingEnabled</key> <true/> <key>IOProviderClass</key> <string>IOUSBDevice</string> <key>idProduct</key> <integer>33159</integer> <key>idVendor</key> <integer>16700</integer> </dict>

Check your Energy Savings PrefPane settings... They're clearly incorrect.

Patch it for what? What's not working?

Or you patch your Bluetooth controller kext with your device id and you'll be able to activate/deactivate BT from OS X too.

The OP posted his Extra folder and that contains a DSDT file. The SSDT, you can always create yourself with the well-known generator script. Look it up at RampageDev's blog on the Net.

Make sure to have setup your E6510 for native speedstep. Process has been documented much widely. You may also want to verify AGPM support and possibly obtain it if not natively running. You'll find pointers at following URLs: https://osxlatitude.com/index.php?/topic/2673-performance-tuning-with-fakesmc/ http://www.insanelymac.com/forum/topic/298027-guide-aio-guides-for-hackintosh/?do=findComment&comment=2019604 https://osxlatitude.com/index.php?/topic/7807-latitude-d-series-nvidia-gpu-performance-tuning-with-agpm/

Having replaced the Intel SSD by a 256Go Samsung SM841N mSATA SSD, I can report that the issue lies with the Intel drive itself. Slow boots cannot be reproduced neither with the Samsung mSATA SSD, nor with the Samsung PM830 SSD whether fitted to my SATA-II Latitude D630 or my SATA-III Latitude E6220. However, slow boots are reproduced with the Intel SSD when fitted to the D630 and the E6220.

That card is supported. All details here.

Last update: 28 Jul 2016 Target OS X release: Yosemite 10.10.x Installation fully successful with usual manual procedure and attached E6220 bootpack on a 7mm SATA-III SSD partitioned GUID and formatted Mac OS X (journaled). Working: full QE/CI with HD3000 graphics OOB VGA output (with patched AppleIntelSNBGraphicsFB kext) HDMI output OOB full audio, including HDMI (with patched AppleHDA kext in /S/L/E. VoodooHDA v2.8.7 & AppleHDADisabler work too from /S/L/E w/o HDMI audio) built-in Gigabit Ethernet (with AppleIntelE1000 kext) bluetooth OOB (on/off supported with patched BroadcomBluetoothHostControllerUSBTransport PlugIn of IOBluetoothFamily kext) native SpeedStep at CPU multipliers x8/10/12/14/16/21/22/25 (with patched AppleIntelCPUPowerManagement kext, generated SSDT, tuned Kozlek's FakeSMC (SMC keys=1.69f4 (or 1.68f98)/SMC compatible=smc-huronriver) and MBP8,1 SMBios profile) Turbo boost to x32 3.2GHz at Turbo multipliers x26/27/28/29/30/31/32 system sleeps Ok through lid closure, power button, Fn-F1, Apple menu->Sleep and energy savings settings with hibernation mode set to 0 (sleep to RAM) and /var/vm/sleepimage file deleted. Wakes Ok through lid opening and power button wireless with any compatible card (having an unsupported Intel N-6205 in half-size WLAN port, I added a DW1390 in the full-size WWAN slot) battery management & monitoring (with ACPIBatteryManager or VoodooBattery or SmartBatteryManager kext) SD card reader (with DSDT patch for compatibility with Apple's default device 14e4:16bc) integrated Webcam OOB keyboard backlight control OOB (for backlit models) left combo eSATA/USB2.0 + right USB2.0 ports OOB ExpressCard slot OOB Not working Nothing Not tested: SmartCard reader USB installer creation procedure: Mount InstallESD.dmg found in Install OS X Yosemite.app/Contents/SharedSupport Then mount BaseSystem.dmg (hidden file) open /Volumes/OS\ X\ Install\ ESD/BaseSystem.dmg Use Disk Utility to restore the contents of BaseSystem.dmg on a USB key (or HDD partition) formatted Mac OS extended (journaled). Remove link for Packages in folder System/Installation of your USB key rm /Volumes/OS\ X\ Base\ System\ 1/System/Installation/Packages Copy folder Packages located in InstallESD.dmg to folder System/Installation of your USB key cp -R /Volumes/OS\ X\ Install\ ESD/Packages /Volumes/OS\ X\ Base\ System\ 1/System/Installation Copy the 2 hidden files (BaseSystem.dmg & Basesystem.chunklist) of InstallESD.dmg to the root of your USB key cp /Volumes/OS\ X\ Install\ ESD/BaseSystem.* /Volumes/OS\ X\ Base\ System\ 1 Copy the patched AICPUPM kext available below to /System/Library/Extensions of the USB installer to avoid Sandy/Ivy Bridge CPU-related KP (due to BIOS-locked MSR) or use NullCPUPowerManagement: 10.10.x patched AppleIntelCPUPowerManagement kext Repair kexts permissions sudo chmod -R 755 /Volumes/OS\ X\ Base\ System\ 1/System/Library/Extensions sudo chown -R 0:0 /Volumes/OS\ X\ Base\ System\ 1/System/Library/Extensions To boot with Chameleon (you can safely use versions r2395, r2401 or r2468): Download the appropriate package off InsanelyMac's repo and install Chameleon on the USB installer partition called "OS X Base System" Create a folder "Kernels" under /System/Library of the USB installer and place a copy of the vanilla kernel there as "kernel" Yosemite 10.10.x kernel available here Create a folder "Extra" at root of your USB installer using the following pack: E6220_i5-2520M_HD3000_Yos-Pack.zip E6220_i5-2520M_HD3000_Yos-Pack_#2.zip E6220_i5-2520M_HD3000_Yos-Pack_#3.zip E6220_i5-2520M_HD3000_Yos-Pack_#4.zip Copy the kexts from the pack to /System/Library/Extensions, repair permissions and rebuild kext cache sudo chmod -R 755 /Volumes/OS\ X\ Base\ System\ 1/System/Library/Extensions sudo chown -Rf 0:0 /Volumes/OS\ X\ Base\ System\ 1/System/Library/Extensions sudo touch /Volumes/OS\ X\ Base\ System\ 1/System/Library/Extensions sudo /Volumes/OS\ X\ Base\ System\ 1/usr/sbin/kextcache -Boot -U /Volumes/OS\ X\ Base\ System\ 1 -K /Volumes/OS\ X\ Base\ System\ 1/System/Library/Kernels/kernel With Chameleon, it's critical to rebuild the cache with above commands if you use the FileNVRAM.dylib module or the installer/Yosemite installation will be unbootable without cache. Ok without the module though. Reinstate the module to boot with cache. Bear that in mind any time you add or modify kexts in Yosemite. Yosemite installation procedure: Boot your USB installer and proceed with installation (if prompted for Bluetooth keyboard and mouse, just press [space]) Upon completion, reboot your freshly made Yosemite installation via your USB installer and complete 1st boot system configuration Install Chameleon bootloader on Yosemite partition Copy the patched AICPUPM kext to /S/L/E If applicable, patch the BroadcomBluetoothHostControllerUSBTransport kext of /S/L/E/IOBluetoothFamily PlugIns to gain enabling/disabling of DW375 BlueTooth through Finder bar icon Copy the Extra folder of your USB installer to your Yosemite partition From Terminal, repair kext permissions and rebuild cache sudo chmod -R 755 /S*/L*/E* sudo chown -R 0:0 /S*/L*/E* sudo touch /S*/L*/E* sudo kextcache -Boot -U / From Terminal, adjust Sleep mode sudo pmset hibernatemode 0 sudo rm -f /var/vm/sleepimage Again, E6220 laptops fitted with a CPU other than an i5-2520M will require a different and suitable SSDT table than provided in the above pack (can be looked up on the web or created with SSDT generator tool -> See RampageDev's blog for detailed guide). - - - - - - - - - - Edit #1: 14 May 2015 Added patched AppleSNBGraphicsFB kext to support VGA output - - - - - - - - - - Edit #2: 17 May 2015 Added patched BroadcomBluetoothHostControllerUSBTransport PlugIn kext to support DW375 enabling/disabling + revised patched AppleHDA to support HDMI audio - - - - - - - - - - Edit #3: 12 Apr 2016 Updated bootpack #3 with revised _PRW methods settings for DSDT devices EHCx/GLAN/HDEF in order to fix occasional/intermittent issues when waking laptop - - - - - - - - - - Edit #4: 28 Jul 2016 Revised bootpack #4 with updated DSDT devices to improve compatibility with MacBookPro8,1 and remove the need for VoodooSDHC kext which causes issues with lid sleep. wireless: renamed RP05.PXSX to RP05.ARPT SD card reader: renamed RP06.PXSX to RP06.SDXC + added compatibility with Apple's own pci14e4,16bc lid: renamed LID to LID0 Thread/post clean-up

Look up for the Service Manual on Dell's web site. Support section.

If Bluetooth module is present, it's listed as "installed" in BIOS Setup. With a cloned installation such as yours, BT icon will remain greyed out on finder's bar if BT device is disabled or absent. In addition, the BT Finder's bar icon activation is normally controlled from the Bluetooth PrefPane. Of course, no such PrefPane if there's no BT module on the system... Hence potential "ghost" BT icons in the case of your cloned drives.

Here's your answer: Link Speed: 6Gigabit Negociated speed: 3Gigabit You can Google the rest, especially for reviews of your particular SSD; but this is now off-topic.

If you seek support here for your next Hackintosh projects, I recommend you try and use one of the supported vanilla installation methods (myHack, Clover, manual), not distros for which the associated opacity too often renders support very difficult.

See my previous reply re: sleep. For Wake, yes, post your DSDT and we can have a look. IOReg always a plus but unlikely to be required in your case.

Some SSD run indeed much slower when TRIM is on than kept off. I've experienced that myself on my E6440 (TRIM off: boots in 7-8sec, TRIM on: boots in 45-50sec). Try and run the same SpeedTest with TRIM off and compare the results. I don't believe there is any other solution at this point. This being said, 240/260 MB/s rates aren't necessarily what I call bad results. Is the controller SATA-2 or SATA-3?

Check your CPU SpeedStep + AGPM if applicable. Some systems resume wake with GPU stuck at highest speed and require tuning of AGPM kext.