Jump to content
InsanelyMac Forum


  • Content count

  • Joined

  • Last visited

About DKMN

  • Rank
    InsanelyMac Protégé
  1. DKMN

    P9X79 compatibility ?

    P9X79WS works great; see RampageDev's site for packaged files and install instructions that are copacetic with the vanilla method but require less Googling... ps. agree re: Clover for these UEFI-capable boards. Older stuff seems a slog right now.
  2. Sorry for the delay, the notification ended up in my Spam folder. Here is a link to his Mavericks install guide: http://rampagedev.wordpress.com/os-x-10-8-mountain-lion/ (I know it says Mountain Lion ) He has .DMGs of files posted for X79 and Haswell motherboards, including all the other kexts needed (e.g. an appropriate FakeSMC.kext). I do not know the SuperMicro mobo chipsets and the dualie board might raise other issues. But, if I were doing this, RampageDev's site is the first place I'd look, in addition to here. I think there is another user called neuromac on both sites who uses a dual Supermicro board. HTH, David
  3. BTW I got throttling to work. Using the RampageDev method and patched AICPM.kext of 3/18/2014. Note that he has a patched FakeSMC as well, which I have installed. Turns out there was a slightly older AICPM.kext hiding in myHack.kext (if you use kext Utility it will place it there)... cleaned that out, along with /Extra/Extensions. Hadn't even thought of that, only looking in /S/L/E Got basic throttling of turbo states with Pike's script v12.7 and "-p 'E5-2620'". Seeing states 12x, 23x, 24x, 25x (anything above 20x is turbo). Still don't have most of the lower states active, but a big breakthrough nonetheless. Anyone have a more sophisticated SSDT-1.aml for this series of chips? (Sandy Bridge E5's) Here's my current basic SSDT-1.aml: http://pastebin.com/VeZzteBv
  4. Hi All, I've been struggling for a while trying to get any throttling of my Xeon E5-2620 (v1 thus Sandy Bridge). Using Asus P9X79WS, 10.9.2 (13C64, the release version), and RampageDev's clean SSDT-based installation with updated FakeSMC. Using PikeRAlpha's ssdtPRgen script (newest revisions), with various settings. Pike kindly added the v1 definitions for me, but I'd been hand-jamming them before. Tried without and with explicitly forcing SB cpu (-c 0) and xcpm (-w). Nothing I do budges the speed from 20x. Nothing lower unless I use NullCPU (then it naturally throttles between 12x and 20x). No turbo no matter what I do (yes, I specified -turbo 2500 with ssdtPRgen). 1. Has anyone got this chip (or very similar) to work? Esp. with the new PM methodology in Mavericks? 2. Any general hints in troubleshooting? I've re-done the installation and script gen fairly carefully, but I can't help but feel I'm missing something. Perhaps SB has just fallen into a gap in support and everyone is working only on the IB models as they're more focused on enthusiast home use (I do understand why, just that my use case is different and I bought back when V1 was out). Thanks much for any tips! David PS. Here is my SSDT-1 as of today: SSDT-1 as of 20140315
  5. DKMN

    [UEFIPatch] UEFI patching utility

    CodeRush, Back in town and working on this. I booted to Win7x64 native and downloaded RW-Everything. Under the MSR Registers listing, I clicked to add a User value and put in "0xE2" (verbatim, that seemed most correct with respect to endian-ness). I got "1E000407" back as a result, which when converted to binary has bit 15=0. Does this procedure sound correct to you? This would indicate that this MSR is not locked under BIOS 4404 for P9X79WS, if I understand correctly... If so, back to the drawing board as a "Sandy Bridge v1" refugee, LOL. Thanks again! David
  6. DKMN

    [UEFIPatch] UEFI patching utility

    CodeRush, No, I am not sure. I'd heard that closely-related boards were not locked, but have not been able to get power management working. I also read from your post there were different codes that needed to be checked for, and wasn't sure how to do this myself. One of the folks who was helping me troubleshoot (knowledgeable) suggested it could be a locked MSR code... So I figured it might have been a newer pattern or something. Anyway, if you don't see a lock that's reassuring. And helpful, though I'm still not sure why I haven't been able to get something working. Might be the processor type (Xeon Sandy Bridge E5-2620 v1... most the hacking work so far is supporting the Ivy Bridge chips, so maybe the hacked AICPM I'm using isn't handling it correctly). Thanks again! David
  7. DKMN

    [UEFIPatch] UEFI patching utility

    CodeRush, I have an Asus P9X79WS motherboard, and haven't had any success patching BIOS versions 4404 or 4505. I get: PMPatch 0.5.14 PowerManagement module at 004B7660 not patched: Patch pattern not found. PowerMgmtDxe/PowerManagement2.efi modules not found. AMI nest modules not found. Phoenix nest modules not found. CpuPei module at 0079F9E8 not patched: Patch pattern not found. CpuPei module at 007DF9E8 not patched: Patch pattern not found. as output, and the output file is never written. Do I understand correctly that that means these BIOSes are new and none of the existing patch patterns work? Thanks, David
  8. This looks very cool. I like the FireWire and Thunderbolt both in one board. Agree with the comment re: Target Disk mode. If this is a reliable, easy to setup OSX rig, I could see it tipping the scales in those situations when buying Apple hardware usually makes more sense (e.g. MacMini for a non computer-savvy family member, etc.) Like the idea of not skimping on components and having a little more control over the supply chain. Best of luck to the devs!
  9. That is *very* interesting. Thanks!
  10. Clever. Nice work-around if you don't have another working Mac.
  11. Thanks! Sorry, I missed your email for a few days due to a spam filter. I tried a number of settings (last was DropSSDT=Yes, SSDT=Yes, Generate P and C states, along with a custom SSDT for CPU states). Turns out I got it working after adjusting the BIOS settings (I think enable C1E). Still learning about power states so it's a little "empirical" unfortunately. I have to go back and clean up my configuration (Extra, etc) as I'm sure I've got redundant settings. Anyhow, your response helps with this, so thanks! DK
  12. Are other edits also required? E.g. explicitly adding state info in the DSDT, generating an SSDT, etc.? I have found a lot of "info" about power management but I am unsure what solutions overlap each other. I used the link you provided to patch the original AICPM from the installer package, and the system boots, but no throttling of CPU freq is noticeable (I am using HWMonitor via FakeSMC to check this... I assume it's valid). Thanks again for your help, David
  13. @PeterHaas: Thank you. That is very helpful. The only thing that doesn't work right now is power management. The version of AICPM I have causes a kernel panic at boot (I have generate C and P states = "Yes" in org.chameleon.boot.plist). Do you have power management running and if so did you have to do any 10.9-specific edits to your DSDT? David ETA: ameris_cyning just posted as I wrote the above... I'll check out that link. I assume this works with 10.8.5 and 10.9? Thanks!
  14. Yes, working now. After getting up-and-running with myHack, I just manually copied in Lnx2Mac's kext with KextHelper and it works. Thanks! I think a big part of getting things to work for me recently has been starting from know basics (e.g. minimal kexts from myHack), rather than using the all in one installers. I did keep the DSDT in place so there are hard references to the onboard ethernet, audio, etc... Do I understand correctly that you are saying it should detect correctly without any DSDT at all? (just curious) Do you know offhand if power management should work in a straightforward manner as well? (do I need a specific version of AICPM?) I've been out of the swim with this board for a while, so not sure what the expected status would be. David
  15. One workaround is to update on another Mac that works (say, actual Apple HW), then go back to App Store on the troublesome mac and hit update. That worked for me, anyhow. One of the reasons I did the above is that I have read that it will offer you the upgrade if iWork '09 is in place during an *upgrade* install. But moving apps over from a clean install and trying to update did *not* work (it did work for other apps, so it wasn't obviously a simple NVRAM, UUID thing in my case)... Not just on hackintoshes, I think, but most people aren't doing clean installs on Apple HW since update is so easy. This all assumes that you can log onto App Store in the first place. If not, then NVRAM, UUID, etc... Hope this helps, David