Jump to content

NEW Intel P55 Mobos & Core i7 860 / i5 750 CPU LGA1156

Core i7
 Share
1554 posts in this topic

Recommended Posts

goldenaustin
goldenaustin
Posted
Posted
I definitely don't have speedstep. Thought it maybe would be dangerous with the OC. Is it ok the add it on?

 

Did you get your DSDT from tonymac's site?

 

Yes, I got the one from tonymac's site. I didn't check myself, but seems like it has speedstep to me since I saw my temps drop around 7-10 degrees after dropping it in and after enabling speedstep in the bios. I also tested it with the speedstep test utility. At fist I thought it wouldn't work well with OC'ing, but it's working fine for me and Bansaku and probably others too.

 

On another note, I just went into my bios, and realized I was limited to 3.5 not because it was causing my GB to tank, but it was because of a PCI-E firewire card I had installed. For some reason I can't post at anything greater than 166 for my BLCK so I'm stuck at 21x166 = 3.5GHz. However, I just tried removing it, and whatdya know...I was able to boot into SL at 3.8Ghz. GB bumped up to a tick under 10k, but temps raised up about 5 degrees. Not worth it for me and besides I need that FW card installed, so now I'm back to 3.5.

 

I would say try out the DSDT from TonyMac if you haven't already.

tasc
tasc
Posted
Posted

Ok, another brief update on the static/electronic noise issue. The noise does appear to be coming from the power supply when power management features are enabled. I did some more searching and came across this thread:

 

http://forum.corsair.com/forums/showthread.php?t=75330

 

It seems there are a lot of people that have similar problems with corsair power supplies. One suggestion is to disable EIST and C1E settings in bios, but that did not work for me.

 

I remember someone said that people with Seasonic PSUs were also having this problem. Turns out that some Corsair PSUs are OEM Seasonics.

profdre
profdre
Posted
Posted

So nobody can help me including Speedstep? I posted my error and wrote that I don't know from which line to which line I have to replace the CPU section in my DSDT with the Speedstep information in tony's DSDT fixes...the beginning is clear, but where ends the section?

Scope (_PR)
   {
       Processor (CPU0, 0x00, 0x00000410, 0x06) {}
       Processor (CPU1, 0x01, 0x00000410, 0x06) {}
       Processor (CPU2, 0x02, 0x00000410, 0x06) {}
       Processor (CPU3, 0x03, 0x00000410, 0x06) {}
       Processor (CPU4, 0x04, 0x00000410, 0x06) {}
       Processor (CPU5, 0x05, 0x00000410, 0x06) {}
       Processor (CPU6, 0x06, 0x00000410, 0x06) {}
       Processor (CPU7, 0x07, 0x00000410, 0x06) {}
   }

   Name (_S0, Package (0x04)
   {
       Zero, 
       Zero, 
       Zero, 
       Zero
   })
   Name (SS1, Package (0x04)
   {
       One, 
       Zero, 
       Zero, 
       Zero
   })
   Name (_S3, Package (0x04)
   {
       0x05, 
       Zero, 
       Zero, 
       Zero
   })
   Name (_S4, Package (0x04)
   {
       0x06, 
       Zero, 
       Zero, 
       Zero
   })
   Name (_S5, Package (0x04)
   {
       0x07, 
       Zero, 
       Zero, 
       Zero
   })
   Name (FLAG, Zero)
   Name (STAT, Zero)
   OperationRegion (SMOD, SystemMemory, 0x000FF840, One)
   Field (SMOD, ByteAcc, NoLock, Preserve)
   {
           ,   7, 
       SUSF,   1
   }

 

Is that too much or too less? What is the termination sign? I thought two brackets?

Thanks!

wlp5
wlp5
Posted
Posted
So nobody can help me including Speedstep? I posted my error and wrote that I don't know from which line to which line I have to replace the CPU section in my DSDT with the Speedstep information in tony's DSDT fixes...the beginning is clear, but where ends the section?

Scope (_PR)
   {
       Processor (CPU0, 0x00, 0x00000410, 0x06) {}
       Processor (CPU1, 0x01, 0x00000410, 0x06) {}
       Processor (CPU2, 0x02, 0x00000410, 0x06) {}
       Processor (CPU3, 0x03, 0x00000410, 0x06) {}
       Processor (CPU4, 0x04, 0x00000410, 0x06) {}
       Processor (CPU5, 0x05, 0x00000410, 0x06) {}
       Processor (CPU6, 0x06, 0x00000410, 0x06) {}
       Processor (CPU7, 0x07, 0x00000410, 0x06) {}
   }

   Name (_S0, Package (0x04)
   {
       Zero, 
       Zero, 
       Zero, 
       Zero
   })
   Name (SS1, Package (0x04)
   {
       One, 
       Zero, 
       Zero, 
       Zero
   })
   Name (_S3, Package (0x04)
   {
       0x05, 
       Zero, 
       Zero, 
       Zero
   })
   Name (_S4, Package (0x04)
   {
       0x06, 
       Zero, 
       Zero, 
       Zero
   })
   Name (_S5, Package (0x04)
   {
       0x07, 
       Zero, 
       Zero, 
       Zero
   })
   Name (FLAG, Zero)
   Name (STAT, Zero)
   OperationRegion (SMOD, SystemMemory, 0x000FF840, One)
   Field (SMOD, ByteAcc, NoLock, Preserve)
   {
           ,   7, 
       SUSF,   1
   }

 

Is that too much or too less? What is the termination sign? I thought two brackets?

Thanks!

 

You need to copy from Tony's file from the beginning down to (including)

 

Scope (_PR.CPU3)

{

Method (_PSS, 0, NotSerialized)

{

Return (^^CPU0._PSS ())

}

 

Method (_PSD, 0, NotSerialized)

{

Return (^^CPU0._PSD ())

}

 

Method (_CST, 0, NotSerialized)

{

Return (^^CPU1._CST ())

}

}

 

 

And paste it to your DSDT replacing just this

 

Scope (_PR)

{

Processor (CPU0, 0x00, 0x00000410, 0x06) {}

Processor (CPU1, 0x01, 0x00000410, 0x06) {}

Processor (CPU2, 0x02, 0x00000410, 0x06) {}

Processor (CPU3, 0x03, 0x00000410, 0x06) {}

Processor (CPU4, 0x04, 0x00000410, 0x06) {}

Processor (CPU5, 0x05, 0x00000410, 0x06) {}

Processor (CPU6, 0x06, 0x00000410, 0x06) {}

Processor (CPU7, 0x07, 0x00000410, 0x06) {}

}

kalamaris
kalamaris
Posted
Posted
@wlp5

Like I said this normal that you can't read the same UUID because on recent systems, the way to go is to change the system-id, then the system calculates a *new* IOPlatformUUID from the system-id which is not visble other than in your com.Boot.plist file, still this one is not generated from a random uuid but from the fixd one that you put in your plist file.

So it does not matter because with this (now fixed) UUID, you should get the same platformuuid forever, whatever it is.

 

This modified boot is definitely a step forward as it makes it possible to remove PlatformUUID.kext without getting the Error 35 messages in system log. (In my case the use of unmodified RC4 with no PlatformUUID was working OK, but gave those errors).

 

Regarding the IOPlatformUUID number I'm confused:

I have three installations of SL, two in the same disk and one in a second disk. I don't have any entry in boot.plist for UUID, but I have an SMUUID string in smbios.plist. In one of the installations I have ProductName=Mac4,1 and in the others iMac11,1.

Since I've installed (original) RC4 boot, I get the same UUID with all of them and this haven't change since then. I've installed rekursor's modified boot in two of them and I still have the same UUID. I've changed the SMUUID in smbios and the UUID didn't change.

Another thing I've noticed is that with PlatformUUID.kext there's an entry in ioreg which contains the old 00000000-00000-8000 etc UUID number (that used to appear in System Profiler with RC3), but after removing it, the whole entry dissapeared (the 0000000 UUID number too). It seems that this 0000 number was generated by the kext.

So, although I don't have a com.boot.plist entry, UUID is stable after RC4 boot (modified or unmodified, PlatformUUID.kext or no-PlatformUUID.kext)

 

Anyway, rekursor's modification is worthy, as (in my case) eliminates the need of PlatformUUID.kext and speeds up the boot time by 1-2 secs (and maybe some other things in the direction of a more stable system).

rekursor
rekursor
Posted
Posted
../..

Since I've installed (original) RC4 boot, I get the same UUID with all of them and this haven't change since then. I've installed rekursor's modified boot in two of them and I still have the same UUID. I've changed the SMUUID in smbios and the UUID didn't change.

../..

I learned that SMUUID has been obsoleted since PCFIEFI 10.2, so does have any effect anymore since quite a while ...

So to sum it, Only the new RC4 (including my modded version) has a real effect on your system and platform UUID, as well as PlatformUUID.kext.

TR0LL
TR0LL
Posted
Posted
I learned that SMUUID has been obsoleted since PCFIEFI 10.2, so does have any effect anymore since quite a while ...

So to sum it, Only the new RC4 (including my modded version) has a real effect on your system and platform UUID, as well as PlatformUUID.kext.

rekurser,

i just tried to send you a PM but they're turned off. is there a way to contact you, or can you contact me? thanks...

kalamaris
kalamaris
Posted
Posted
So to sum it, Only the new RC4 (including my modded version) has a real effect on your system and platform UUID, as well as PlatformUUID.kext.

 

That's true (and the modifications that enable us to remove kexts from E/E are most welcome).

I just can't figure out why my UUID is stable without an entry in com.boot.plist.

tasc
tasc
Posted
Posted
That's true (and the modifications that enable us to remove kexts from E/E are most welcome).

I just can't figure out why my UUID is stable without an entry in com.boot.plist.

 

My UUID has remained stable ever since I switched to RC4 and removed platformUUID kext. I'm not using Rekursor's patched version and I have no entry in com.boot.plist.

rekursor
rekursor
Posted
Posted
My UUID has remained stable ever since I switched to RC4 and removed platformUUID kext. I'm not using Rekursor's patched version and I have no entry in com.boot.plist.

I believe that the difference will be more obvious if you use another OSX system partition or reinstall OSX and stills want a transparent behavior for your registered apple applications (like iTunes, Logic Pro,..)

So if you don't need it, don't use it, indeed.

profdre
profdre
Posted
Posted
Hey I'm back with some energy consumption measurements on my Hackintosh, if anybody is interested...

 

First OS X 10.6.2 @2,66 GHz with the DSDT.aml from tonymacx86's blog and two harddisks spinning:

 

idle (Desktop): 155 W

BOINC running on all cores: 210 W

prime stress-test maximum heat: 225 W

prime stress-test maximum heat+ Age Of Empires III: 290 W

just repairing permissions: 170 W

[...]

Okay I finally included the Speedstep code and relating to MSRTools it is working (clocks down to 1.2 GHz when idle and Vcore is reduced to 0.88 V) However, there is NO difference in power consumption, it still consumes 155 W when idling compared to 133 W in Windows 7 (at. max. I see 2 % load on the cores) The temperatures didn't change neither: I have had between 25 and 30 °C (which isn't really accurate btw because I heard the temp. diodes are calibrated for a much higher temperature) and I still have these temperatures. The Geekbench Score went down from 7700 to 7500 (okay, but when there's no less consumption, why include Speedstep?)

Do you think there's a reason why I can't compare the two OSes in such a manner? However it's still bizarre that the power consumption didn't change...

 

@ rekursor: Thanks again for your efforts for a fixed UUID. I appreciate it, seems to work so far, we'll see ;)

 

You need to copy from Tony's file from the beginning down to (including)

 

Scope (_PR.CPU3)

{

[...]

 

 

And paste it to your DSDT replacing just this

 

Scope (_PR)

{

Processor (CPU0, 0x00, 0x00000410, 0x06) {}

[...]

 

}

 

Thank you very much, wlp5, that's just what I wanted to know!

TR0LL
TR0LL
Posted
Posted

is there anyone out there who can talk on ichat or similar to help me troubleshoot why i cant get chameleon to boot without having the disc in? i'm on an i7 and UD3R and this is my first hack so i'm learning as i go.

 

trying to follow guides by tonymac and also help from basanku but after quite a few different configs we still can't get it to boot properly. must be missing something here. if there's anything i should check, verify, take a screenshot of, etc. then please let me know. you can PM or IM me if its better than going over in this thread.

 

thanks in advance...

lordduckx
lordduckx
Posted
Posted
Does anyone have DSDT for P55A-UD4P?

 

I have the board, and I'm working on it, but I can't get it boot without Tony's CD

 

Tony, what's in your CD that allows it to boot? could I take the dsdt from the p55a ud3 or p55 ud4p to get the SATA to work?

rekursor
rekursor
Posted
Posted
I have the board, and I'm working on it, but I can't get it boot without Tony's CD

 

Tony, what's in your CD that allows it to boot? could I take the dsdt from the p55a ud3 or p55 ud4p to get the SATA to work?

If I were you, I would first check on the gigabyte site which Mobo is the closest match from yours, it might be the p55-ud4p but it's not obvious so compare and try to match the audio chip, the optional peripherals (of course the usb3 won't match our currently working mobos).

 

Then I would take the closest dsdt and would give it a try.

 

The SATA drive is not a problem, it is automatically discovered by the system, this is why dsdt sata modifications are more about cosmetics than something else.

 

Now concerning the boot, I would give a try to the EFI Empire 1.8 for core i5/i7 boot CD and see if I can boot from that.

 

Then for the USB3 ports, I would pray and hope that the compatibility (2.0 and 1.1) modes work :rolleyes:

In case your are stuck with USB (though I know you're not there yet) you can always buy a compatible usb 2.0 pcie card (possibly with firewire as well) and shazam, you'll be ok !

vetters
vetters
Posted
Posted

I know some people in this thread were having trouble with getting automatic sleep (based off of the timer in the energy saver preference pane) working without using a 3rd party program/script such as PleaseSleep to get sleep to initiate on their rigs. All sleep functions work for me except for the automatic sleep. Did anyone find a solution for this? I am using a UD2 (bios f5) and i7. I appreciate all help. Thanks

aushack
aushack
Posted
Posted

jarst a quick qestion to see if this is widespread:

 

does anyone have garbled audio after waking from sleep with AppleHDA from 10.6.0 and LegacyHDA? music is garbled the first few seconds, but seems to play fine after and turning up the sound isnt smooth and only a few audio clicks play. for a quick test go to sleep, then wake, chuck something to the Trash and empty it (no, not literally :thumbsup_anim:) do the sound effects play smoothly? after rebooting audio is fine.

airecken
airecken
Posted
Posted

So I solved the low geekbench score without nullcpu kexts problem.

 

I ended up enabling all the advanced cpu features in the bios (instead of setting it at automatic) and now my geekbench score is 8900 on a core i5 OC'ed to 3.4 ghz.

 

With the custom dsdt, now I only have evoreboot.kext, fakesmc.kext, and jmicronata.kext in the extras folders. Have the hacked hda in the systems library and everything works perfectly (except sleep still). Then plus the netkas enabler I have 4890 functioning fine.

 

Also, my CPU temp is reading over 9000 degree C with the most recent fakesmc. Anybody figure out a way to use this with core i5 yet?

mulan
mulan
Posted
Posted
jarst a quick qestion to see if this is widespread:

 

does anyone have garbled audio after waking from sleep with AppleHDA from 10.6.0 and LegacyHDA? music is garbled the first few seconds, but seems to play fine after and turning up the sound isnt smooth and only a few audio clicks play. for a quick test go to sleep, then wake, chuck something to the Trash and empty it (no, not literally :thumbsup_anim:) do the sound effects play smoothly? after rebooting audio is fine.

 

Hi

I have the same problem. It seems like it happens went the sound haven't been used for some time. But its only the first second you hear the "clicks".

wlp5
wlp5
Posted
Posted
So I solved the low geekbench score without nullcpu kexts problem.

 

I ended up enabling all the advanced cpu features in the bios (instead of setting it at automatic) and now my geekbench score is 8900 on a core i5 OC'ed to 3.4 ghz.

 

With the custom dsdt, now I only have evoreboot.kext, fakesmc.kext, and jmicronata.kext in the extras folders. Have the hacked hda in the systems library and everything works perfectly (except sleep still). Then plus the netkas enabler I have 4890 functioning fine.

 

Also, my CPU temp is reading over 9000 degree C with the most recent fakesmc. Anybody figure out a way to use this with core i5 yet?

 

By netkas enabler you mean evoenabler.kext ?

 

Yes, you need to enable all the features. I am a little concerned about turbo boost but so far, I'm not seeing any issues and geekbench goes up by 300-350 points on mine.

 

By latest fakesmc, I assume you mean 2.5. I just installed it and my temps look fine with istat. I now get two cpu readings, while with version 2 I was only getting one. But the readings look normal. UD2/i5 here and we have pretty much the same kexts.

RichBUK
RichBUK
Posted
Posted

Hi guys, thanks for the thread! great stuff

 

I've upgraded my E8400 system into an UD2/i7 860 pretty easily with this info. Everything working speedstep/sleep wise etc. Geekbench is at approx 8000 on the standard clock speeds which is about average i guess but still twice teh score the E8400 got.

 

Anyway, i have a couple of little issues, wonder if anyone has seen similar.

 

1) On boot into the chameleon boot loader and the initial stages of booting the system it appears that CPU is running at around 1Mhz! everything is very slow, booting into safe mode takes an age as the kexts loadings is about 1 a second. Plus the Chameleon menu is just dog slow. Once past the initial boot stage (i.e. when the screen clears on a -v boot) its fine. No real porblem but odd.

 

2) I have a Dlink atheros based wirless card that causes Kernel panics if its connected. This worked fine on the e8400 system, (same boot disk, just transferred over). I'm guessing that it must be a dsdt issue as i've just modded Tony's UD2 file. Anyone got any ideas?

 

 

Thanks again everyone in this thread!

aushack
aushack
Posted
Posted
Hi

I have the same problem. It seems like it happens went the sound haven't been used for some time. But its only the first second you hear the "clicks".

 

thanks good to hear someone else has the problem. so is the sound garbled when you empty your Trash? and does music stop playing while the Trash sound affects play?

mulan
mulan
Posted
Posted
thanks good to hear someone else has the problem. so is the sound garbled when you empty your Trash? and does music stop playing while the Trash sound affects play?

 

Jep i sounds like when you empty the trash bin.

I only hear the sound i the beginning of a song, or system alert sound. It only comes back when there haven't been played any sounds for a while.

arynjon
arynjon
Posted
Posted

Well, I haven't heard from anyone, but I've been working on the firewire problem I've been having. Firewire devices will only work if I have them powered on and connected when I boot. After any powerdown or disconnect, I can't get any firewire device to work unless I reboot.

 

Just to experiment, I decided to boot with my usb key, which doesn't have a DSDT and has a few different kexts. When I boot this way, firewire devices work the way they are supposed to with no problems whatsoever.

 

So, it comes down to the DSDT or one or more kexts in /Extra (although, I don't have many in my Extra folder). I'll work on it some more this weekend to see if I can pinpoint the problem.

 

If anyone has thoughts or ideas about this issue, please let me know.

 

Thanks!

 

BTW, I am using the DSDT from Tony's site, except I removed the graphics card information.

 Share
Go to topic listing
×
×
  • Create New...