Contribute
Register

New Gigabyte Z68 UEFI BIOS Betas Released!

Status
Not open for further replies.
I've been using this method for a while now but I don't really overclock too much. I'm only using the stock cooler so the highest I'll go is to 3.6ghz with my i5 so I don't use the overclocking SSDT. The thing is, I don't get any P-States between 22 and 35.

Should I be getting any p-ststes from 23-34? Also, it'll usually idle at 18 and momentarily go down to 16 for about a half of a second. I don't really know anything about p-states so any help would be appreciated!

Well the number and value of p-states will depend on several factors and is chosen by the appleintelcpupm driver. The number of states appears to depend mostly on the system definition, whereas the values seem to have something to do with your CPU clock speed and how many states are defined in your SSDT. if you want broader states you should try removing p-state definitions from your SSDT that are above the values you are actually using since they are never used but appear to skew the drivers choice of p-states. In my case i use macmini5,1 definition, have 16-43 defined in my SSDT and have an overclock of 4200mhz. I regularly observe p-states of 16, 29 and 42 and more occasionally hit 36 and 27. This seems like a reasonable spread to me.

Also please se below for more OC info that i am copying and pasting from my post in another thread as it is rather important for people trying to achieve higher speeds:

i had some interesting experiments RE getting my system back to full 4.2ghz OC. moving to 43x 43x 43x 42x caused the same p-state kernel panic at boot that running 42 42 42 42 does. So i had to add another entry to SSDT for the 43x state (even though this state never occurs in OS X) after that everything booted fine and i was hitting 42x max with all the normal in between states. pretty cool. Except my system starting locking up frequently. no kp, just totally frezzes/ locks up and usually within a minute or 2 of boot, so its happened fast and consistently. Looking back at the original post it mentioned gigabyte removed multi step load line calibration and that you had to adjust your vcore in order to achieve certain OC. In fact i think my system locked once even at 41x max but after hours of usage and i thought it was a fluke. but for some reason when using an SSDT and full p-states the system can become unstable even if it was working perfectly at 42x with only top and bottom p-states before.

Well the solution is simple enough. You have to adjust your voltage settings. change the vcore from "auto" to "normal". this unlocks the dynamic vcore setting below it which i put at +0.150v. It is possible you will need more or less depending on your CPU and overclock. the original poster suggested +0.200v for achieving 4.6ghz for instance. Once you have set an appropriate dynamic vcore setting and put an SSDT.aml that includes definitions for all the needed states you end up with full p-states and full overclock in OSX at 100% stability.

Cheers,
g\
 
I have the same board, GA-Z68X-UD3H-B3. Im new to this and wondering if I could get some advice on a from-scratch build. From what I'm reading here and other places the UEFI update is worth the effort and I do understand its still in beta. What I need to know is what process is best for a fresh build. Will I need to include any DSDT files when installing OSX if I have updated to UEFI beforehand or does the new BIOS totally eliminate the need for DSDT patches. Im using an i5 2500k, AMD Radeon 6770, Samsung 830 SSD. I don't plan to overclock. Any advice greatly appreciated.
 
Thanks for the reply genzai!

I have the same board, GA-Z68X-UD3H-B3. Im new to this and wondering if I could get some advice on a from-scratch build. From what I'm reading here and other places the UEFI update is worth the effort and I do understand its still in beta. What I need to know is what process is best for a fresh build. Will I need to include any DSDT files when installing OSX if I have updated to UEFI beforehand or does the new BIOS totally eliminate the need for DSDT patches. Im using an i5 2500k, AMD Radeon 6770, Samsung 830 SSD. I don't plan to overclock. Any advice greatly appreciated.

I don't have experience with your specific board but with UEFI on my Z68MX I don't use any added DSDT. The only things I needed were an HDAEnabler kext for audio and a SSDT.
 
Thanks for the reply genzai!



I don't have experience with your specific board but with UEFI on my Z68MX I don't use any added DSDT. The only things I needed were an HDAEnabler kext for audio and a SSDT.

And you wont need an SSDT if you don't overclock.
g\
 
Ok, Thanks for the reply's guys. I'll update my progress when I get my SSD and GPU on their way from Newegg.
 
Looks like the GA-Z68P-DS3 isn't getting an update; I wonder if it's a logic limitation or flash size? The 2.0 and 2.1 have downloads, but they're over 2mb, whereas the previous BIOS is only 1.7MB.

That's not the limitation, as you have a pair of 32Mbit chips, just like the later revisions. I'd try, I seriously doubt it won't work, but it's possible that you'll get an error message saying wrong board revision or something.
I can't see any technical reason why it wouldn't work, as the changes appear to merely be layout and cost related (as in they made the later revisions cheaper), but the tech specs are the same.
 
Well the solution is simple enough. You have to adjust your voltage settings. change the vcore from "auto" to "normal". this unlocks the dynamic vcore setting below it which i put at +0.150v. It is possible you will need more or less depending on your CPU and overclock. the original poster suggested +0.200v for achieving 4.6ghz for instance. Once you have set an appropriate dynamic vcore setting and put an SSDT.aml that includes definitions for all the needed states you end up with full p-states and full overclock in OSX at 100% stability.

Cheers,
g\

Thanks very much for this, caused me probs too.
 
I know it's early, but is there a general thought on whether or not it is worth upgrading? Things seem to be going fine with my system, so I'm curious how this can help...

Don't, just wait.
That is my advice.
ML is new, UEFI is new.
There are always bugs. If you need stable then stay with what you've got.
 
Don't, just wait.
That is my advice.
ML is new, UEFI is new.
There are always bugs. If you need stable then stay with what you've got.

And if all of us were like you, we'd still be using horses and carriages...
No offence, but as it's possible to revert back to the BIOS if it doesn't work, this isn't really the end of the world now, is it?
Admittedly the UEFI for these boards are Beta releases, but most of them appears to be perfectly stable. Sadly we only have access to so many different bits of hardware, so we can't guarantee that it'll be flawless, but that's also why we need the community try it out and report back.
 
hi all, i'm using the last official uefi bios on my ga z68xp ud3 rev 1.3 and cpu is i7 2600k @3,40 ghz score in ML 10.8.2 geekbench 64bit is 13740 while if i try to overclock cpu for example to 3,8 ghz score is around 6000 :banghead::banghead: whats wrong?

ps in extra folder is present the SSDT.aml that i install with multibeast 5.13

(Optimized SSDT supporting Sandy Bridge Core i5 or Core i7 CPUs overclocked up to 4.2 Ghz. Installs SSDT.aml in /Extra.)
 
Status
Not open for further replies.
Back
Top