Jump to content

maleorderbride

maleorderbride

Member Since 17 Jan 2009
Offline Last Active Yesterday, 10:35 PM
-----

#2281969 Haswell-E 5820k 5930k installation guide with full speedstep working

Posted by maleorderbride on 22 September 2016 - 06:18 AM

Yes, mine is also fine with Sierra. This is an i7-6850K @ 4.3GHz. The second picture is an i7-6900K @ 4.0GHz. Attached Files  Screen Shot 2016-09-21 at 11.20.50 PM.png 1006.99KB 6 downloads  Screen Shot 2016-09-21 at 11.30.22 PM.png 1.11MB 5 downloads

#2272543 Haswell-E 5820k 5930k installation guide with full speedstep working

Posted by maleorderbride on 01 September 2016 - 06:03 PM

Ergot, if you are using the method I wrote out, then you are not using AICPUPM, so Null is doing nothing. You can keep it removed.  If you also unlocked the MSR, then you do not need VoodooTSCSync to wake from sleep (VodooTSCSync has not been necessary to boot since 10.10.3).    I have not been using my other build lately, but if anyone wants to test the following patches to kernel in 10.11.6, then feel free. Please let me know if it helps with their reboots. I could not have found what to patch without Pike, of course. Keep whatever patches you currently have, but just add these to them.  **I have not tested these yet, no idea if they really work or cause any problems. Have a way to recover if this renders your install unbootable.** <key>KernelToPatch</key> <array> <dict> <key>Comment</key> <string>nop MSR #1</string> <key>Disabled</key> <...

#2272281 Haswell-E 5820k 5930k installation guide with full speedstep working

Posted by maleorderbride on 31 August 2016 - 10:41 PM

Sorry, I have a Haswell-E CPU. I have not tried this on Broadwell-EP.

#2270995 Clover General discussion

Posted by maleorderbride on 28 August 2016 - 09:09 PM

Got it--thanks. If only it was, it would be easier to track down ;p

#2270988 Clover General discussion

Posted by maleorderbride on 28 August 2016 - 08:59 PM

0xA8 (1010 1000)Thanks! I will try that. Wait, I want it to show me the panic, so it sounds like it was already set to that with 0x28. Which indicates what, that this is not a kernel panic that is causing the restart?

#2270965 Clover General discussion

Posted by maleorderbride on 28 August 2016 - 08:11 PM

0 = disabled, no reboot on panic1=enabled, reboot on panic.0x28 has (bit 0) = 0, so disabled. I have BooterConfig set to 0x28, but I have random restarts sometimes (some under idle, some load) but without a panic. Assuming my below RtVariables are correct, is kBootArgsFlagRebootOnPanic enabled? If so, then is my reboot not a kernel panic? Thanks! (I have swapped motherboard, RAM, PSU, GPU, and done lots of hardware tests already. This only happens in in MacOS, never in Windows.) <key>RtVariables</key> <dict> <key>BooterConfig</key> <string>0x28</string> <key>CsrActiveConfig</key> <string>0x67</string> </dict>

#2270304 Haswell-E 5820k 5930k installation guide with full speedstep working

Posted by maleorderbride on 27 August 2016 - 01:53 AM

No worries, I've been wrong before. Asus stopped locking MSR with the Z87 chipset and I've never done a X99 build with anything but Gigabyte so maybe there's a surprise here... I haven't seen a Asus X99 post anywhere stating that MSR was locked... Throw out all Pikes patches, load null and show me one ounce of difference in GB scores, sleep/wake or anything else, there isn't any... (Random reboots will probably stop as well). I'm not looking for an argument here so please continue on the path you're on, I don't have a horse in the race. IMO, X99 issues will get resolved only if/when Apple decides to support it... Surely there's more to running X99 correctly in OSX than Pikes XCPM patches. Starting with 10.10 my ASUS X99 builds started having a random reboot problem (with Null). It lessened in 10.11, but still remains. I just want to track that down. I am not interested in looking for problems, just solutions! (FYI, most GB X99 boards have E2 unlocked) Performance-wise, I h...

#2270280 Haswell-E 5820k 5930k installation guide with full speedstep working

Posted by maleorderbride on 27 August 2016 - 12:56 AM

 ASUS X99 aren't locked. If they were you could still patch in Clover: <key>AsusAICPUPM</key> <true/> What is your support for this statement? This is the output I get from Clover on both X99 Deluxe and an X99-A USB 3.1: 0:104  0:000  MSR 0xE2 before patch 000084000:104  0:000  MSR 0xE2 is locked, PM patches will be turned on0:104  0:000  MSR 0xE4              000104140:104  0:000  MSR 0xCE              20080C3B_F3811E000:104  0:000  MSR 0x1B0             00000000 Does that not mean that 0xE2 is locked? Furthermore, Pike claims that his ASUS X99-E has a locked MSRs in the comments of his blog: https://pikeralpha.w...essor/#comments A locked MSR, accordingly to Pike, can lead to sudden reboots when the OS attempts to write to the MSR, hence his advice (and ker...

#2264945 Haswell-E 5820k 5930k installation guide with full speedstep working

Posted by maleorderbride on 12 August 2016 - 10:30 PM

 how do I apply the following patches via Clover?  Can someone give me an example?  Again, sorry I'm a bit out of touch.xcpm_bootstrap 10.11.6 Broadwell-E © Pike R. Alphafind 83C3BB83FB09replace 83C3B383FB09_cpuid_set_info 10.11.6 © Pike R. Alphafind 5D7F288D48BB83F902replace 5D7F288D48B383F902reebot fix 10.11.6 © Pike R. Alpha(it´s the same patch for each CPU, just test your motherboard, some boards need this)find 554889E5415741564155415453504189D64189F74889FB4585FF0F84replace C34889E5415741564155415453504189D64189F74889FB4585FF0F84 Here is what I did:<key>KernelAndKextPatches</key> <dict> <key>AppleRTC</key> <false/> <key>AsusAICPUPM</key> <true/> <key>Debug</key> <false/> <key>KernelCpu</key> <false/> <key>KernelHaswellE</key> <false/> <key>KernelLapic</key> <false/> <key>Ke...

#2264587 Haswell-E 5820k 5930k installation guide with full speedstep working

Posted by maleorderbride on 11 August 2016 - 09:41 PM

Mystroe, try using a MacPro6,1 SMBIOS (and re-doing the freqvector) if you want the temp and idle speed to drop more. Are you on 5,1 just to avoid the GFX0/1 unloading from AppleGraphicsControl?

#2264476 Haswell-E 5820k 5930k installation guide with full speedstep working

Posted by maleorderbride on 11 August 2016 - 04:23 PM

Yes. With the newest Version.  Need this to get vectors load. Thanks, perhaps you can go back and edit your original post with a succinct numbered list of what needs to be done? i.e.1. Three patchesxcpm_bootstrap 10.11.6 © Pike R. Alphafind 83C3BB83FB09&nbsp;replace 83C3B883FB09cpuid_set_info 10.11.6 © Pike R. Alphafind 5D7F288D48BB83F902replace 5D7F288D48B883F902reboot fix 10.11.6 © Pike R. Alphafind 554889E5415741564155415453504189D64189F74889FB4585FF0F84replace C34889E5415741564155415453504189D64189F74889FB4585FF0F842. FrequencyVectorEdit.sh (on a MacPro6,1 SMBIOS targeting iMac14,2, I assume?) https://github.com/P...qVectorsEdit.sh3. Remove NullCPUPowerManagement & any FakeCPUID.4. Disable EIST in BIOS (Ai Tweaker, Internal CPU Management, Disable Intel Enhanced Speedstep)5. Oh, and have Clover 36xx? installed. What is the minimum version? I used 3688: https://sourceforge..../cloverefiboot/6. Have a CPU SSDT from SSDTPRGen: https:/...

#2264205 Haswell-E 5820k 5930k installation guide with full speedstep working

Posted by maleorderbride on 11 August 2016 - 04:25 AM

I added the three patches, removed Null, VoodooTSCSync, and it seems like the exact same performance as running just 10.11.6 without doing anything. I have a lower GB and random fluctuations in speed instead of sustained turbo. Can someone who has this "working" post their config.plist? Perhaps I am doing something wrong, but it also seems like there is something missing here. I would like to see GB/Cinebench scores with Intel Power Gadget readouts from anyone who says it is working. Don't get me wrong, it is possible to get good scores without doing anything in 10.11.6, but they are like 5% of the benchmarks and have wide variance.

#2240132 Haswell-E 5820k 5930k installation guide with full speedstep working

Posted by maleorderbride on 24 May 2016 - 10:11 PM

There is no always working method. If you add enough PCIe, TB, and other devices, then eventually you have the problem. You can always do a non-UEFI install of Clover (even to the ESP still) if need be. 

#2236169 El Capitan 10.11.1 versus the 5960X - A Collaborative Work-Thru

Posted by maleorderbride on 04 May 2016 - 10:38 PM

Can't you build again the other x99 build with El Capitan (10.11.4) on another HD and then have it booting again in your 14 core ? Boa sorte, Paulo Yea, I tried that. Still no go. There must be another variable here. I'll keep playing with it.

#2236107 El Capitan 10.11.1 versus the 5960X - A Collaborative Work-Thru

Posted by maleorderbride on 04 May 2016 - 05:06 PM

Here is the EFI, It think that is back in the exact state it was when it first worked: https://dl.dropboxus... E5-2697 V3.zip When it worked, I just assumed something had changed in 10.11.4 and I was out of the loop (haven't read this thread in awhile). I only came here after it stopped working =/

© 2016 InsanelyMac  |   News  |   Forum  |   Downloads  |   OSx86 Wiki  |   Mac Netbook  |   PHP hosting by CatN  |   Designed by Ed Gain  |   Logo by irfan  |   Privacy Policy