1945 replies to this topic

[mitch_de]

Chrysaor, on Sep 9 2009, 10:13 PM, said:

Modified MSR Tools to run on Snow Leopard. You can see the steppings if you have P-states support.

[size=4MSR_Tools.app.zip[/size]]

Thanks !

Question: If i use the MSR (old one) with 10.5 i will get extrem high cpu load of the kernel after some time.
Special if i run an app that needs lot of cpu like benches. System (finder) gets first slow then freezes - no more reaction to keyboard and mouse.
I seen that some others have same problem (kernel runs amok after some time). I use vanilla and have C2D with normally no such problems.
Also voodoopower or mark-i work without problem.
Are you the dev and can post source that some code heros can look in (not me; ) ?

[Beerkex'd]

ffbadkill, on Oct 23 2009, 11:35 PM, said:

After the modification, I cannot compile it

Nor can anyone else, until you edit your post and attach your dsdt.dsl in a zip file.

[ffbadkill]

i have changed the format
sorry about it..

[mitch_de]

Question about that 8 Pstates Stepping
   Package (0x06)
                {
                    0x0875, // 2165 MHz  ?
                    0x26,
                    0x0A,
                    0x0A,
                    0x04618, // multi = 6,5 ?!
                    0x04618
                },

                Package (0x06)
                {
                    0x07CE, // 1998 MHz = 6* 333 FSB
                    0x22,
                    0x0A,
                    0x0A,
                    0x0616, // 6* FSB(333), 16hex VID = 1070 mV
                    0x0616
                }
            })


Are you sure about 8 Pstates with E8500 CPU ?
Would mean that you have multipliers 6*,7*, ...13*  FSB, or can we use that 0.5 steps in multi also ?


0x0616 means : multiplier 6* (FSB) , 16hex VID (around 1070 mV)
But whats that  0x04618 ?
Does it mean multi 6,5 , because 6,5 * 333 would give that 2165 Mhz ?

I have E7300 and would also add those 0.5 multi steps. Does it depends on mainboard ?
I have Gigabyte and can also do 0.5 steps in BIOS, but never used that.

[Beerkex'd]

Finally a question I can answer in this thread, thanks Mitch!

mitch_de, on Oct 27 2009, 01:24 AM, said:

Are you sure about 8 Pstates with E8500 CPU ?
Would mean that you have multipliers 6*,7*, ...13*  FSB, or can we use that 0.5 steps in multi also ?

Yes, you can use fractional multipliers.

mitch_de, on Oct 27 2009, 01:24 AM, said:

0x0616 means : multiplier 6* (FSB) , 16hex VID (around 1070 mV)
But whats that  0x04618 ?
Does it mean multi 6,5 , because 6,5 * 333 would give that 2165 Mhz ?

Yes, that's correct.

mitch_de, on Oct 27 2009, 01:24 AM, said:

I have E7300 and would also add those 0.5 multi steps. Does it depends on mainboard ?
I have Gigabyte and can also do 0.5 steps in BIOS, but never used that.

The "default" multiplier for my E8500 is 9.5 (333x9.5 = 3163)

The multiplier goes from 6.0 to 9.5 = 8 P-States!
The P-state calculator doesn't support fractional multipliers so I added the 4 by hand - Master Chief explained it somewhere in the thread.

[mitch_de]

Thanks, I found the answer also reading to all pages here ;
Also my Pstates 6 ... 9 with 0.5 between are working !

What i found out is, that CST Adresses 0000814 / 000815 here found maybe 414 / 415 like the Processor Adresses 410 insted of 810).
Must be mainboard dependend. My GA-EP45-DS3 (no P, no L) has 410.

But i cant "see" if also C-State is working. Pstates worked with 4 Pstates and that with 0.5 too.
With voodoopower and AppleIntelCPU (each alone )
My problem was with AppleIntelCPU that the level of CPU load needed to switch faster is too high.
I want less powersaving but faster Cpu on less cpu load. Voodoopower can be configured that way be its threshoot key setting. But how can i change that level of cpu loads using AppleIntelCPU ?

I now disabled C-State by simple not use AppleIntelCPU and use voodoominipower again (with new 0.5 pstates)

[FKA]

Beerkex'd, on Oct 27 2009, 04:56 AM, said:

Finally a question I can answer in this thread, thanks Mitch!

I know how you feel - ever start something and then see it run rapidly away from you!

D

[mitch_de]

Anyone knows dev of MSR_Tools ?
For me, MSR_Tools seems to have some timing / communication problems with his .kext.
Specially on high CPU load System freezes after short time. CPU load 100% and only cursor can be moved. Finder and all other no reaction anymore.
I ask because its the onyl tool i know which can independed of other throttling .kext show Volts.
CPU-X only shows MHz.

[xopher]

Beerkex'd, on Oct 27 2009, 06:56 AM, said:

The "default" multiplier for my E8500 is 9.5 (333x9.5 = 3163)

The multiplier goes from 6.0 to 9.5 = 8 P-States!
The P-state calculator doesn't support fractional multipliers so I added the 4 by hand - Master Chief explained it somewhere in the thread.

I have 4-working p-states atm, 6.0, 7.0, 7.5 and 8.0. These are all he multipliers I have available in the BIOS. VoodooMonitor gives me only 6, 7 and 8, so I added the 7.5 by hand.

Now, does my BIOS know best, or can I force the 6.5 multiplier via the DSDT?

[DieBuche]

First of all, ty @ FormerlyKnownAs & Master Chief.
I followed the guide, speedstepping works...

But i got one question: Did anyone manage to modify their cstate tables in a way so that audio wouldn't stutter.
Right now i removed those tables. There were some posts about the cstate latency in facp.aml but i didn't understand what to do exactly.

I attached my current dsdt, i have no memory adresses in my ssdt for cstates

Attached Files

•  dsdt.dsl.zip   14.32K   7 downloads

[Beerkex'd]

Try removing/modifying IRQs in TIMR, RTC, HPET and IPIC devices. It worked for me. Look up those devices in Master Chief's DSDT from the P5K Pro thread:
http://www.insanelym...howtopic=188920

Or try THe KiNG's method here:

http://www.projectos...p?showtopic=564

[DieBuche]

Beerkex'd, on Oct 28 2009, 03:48 PM, said:

Try removing/modifying IRQs in TIMR, RTC, HPET and IPIC devices. It worked for me. Look up those devices in Master Chief's DSDT from the P5K Pro thread:
http://www.insanelym...howtopic=188920

Or try THe KiNG's method here:

http://www.projectos...p?showtopic=564

Thank you kindly for ur help, it worked!

[xopher]

DieBuche, on Oct 28 2009, 06:07 PM, said:

Thank you kindly for ur help, it worked!

Just wondering, what did you do exactly? I'm having the same problem, but with video. HD-video freezes for a second, then continues normally for a while. You barely notice it, but barely isn't good enough.

[DieBuche]

xopher, on Oct 28 2009, 07:09 PM, said:

Just wondering, what did you do exactly? I'm having the same problem, but with video. HD-video freezes for a second, then continues normally for a while. You barely notice it, but barely isn't good enough.

i attached a diff between my stuttering and non stuttering dsdt. Open it in Textmate: Green is new, red is old.
Try to add those changes to your dsdt, if it doesn't work could u post ur .dsl ?

Attached Files

•  dsdt.diff.zip   1.71K   6 downloads

[xopher]

DieBuche, on Oct 28 2009, 08:20 PM, said:

i attached a diff between my stuttering and non stuttering dsdt. Open it in Textmate: Green is new, red is old.
Try to add those changes to your dsdt, if it doesn't work could u post ur .dsl ?

I checked the site, and it seems like I already have all the fixes done, I now have made an edit to my CPU0 CST, so that it doesn't go down to the C2, C3, C4 states at all. Which appears to have fixed it. Have to revert to the old DSDT to confirm.

Here's my current, check it out if you like:  DSDT_28.10.09_MC_cst.dsl.zip   8.14K   15 downloads

[mm67]

xopher, on Oct 28 2009, 11:01 PM, said:

I checked the site, and it seems like I already have all the fixes done, I now have made an edit to my CPU0 CST, so that it doesn't go down to the C2, C3, C4 states at all. Which appears to have fixed it. Have to revert to the old DSDT to confirm.

Here's my current, check it out if you like: DSDT_28...._cst.dsl.zip

So you didn't see my edit about the typo in this one   Line 57 looks like this:
Package (0x04) {ResourceTemplate () {Register (FFixedHW, 0x00, 0x00, 0x00, ,)},0x01,0x9ED,0x3E8}

Should be like this:
Package (0x04) {ResourceTemplate () {Register (FFixedHW, 0x00, 0x00, 0x00, ,)},0x01,0x9D,0x3E8}

That was what made this one feel like everything was running in slow motion.

[xopher]

mm67, on Oct 28 2009, 11:15 PM, said:

So you didn't see my edit about the typo in this one   Line 57 looks like this:
Package (0x04) {ResourceTemplate () {Register (FFixedHW, 0x00, 0x00, 0x00, ,)},0x01,0x9ED,0x3E8}

Should be like this:
Package (0x04) {ResourceTemplate () {Register (FFixedHW, 0x00, 0x00, 0x00, ,)},0x01,0x9D,0x3E8}

That was what made this one feel like everything was running in slow motion.

Didn't notice that, thanks for pointing it out!

Edit: One more question, do you guys have different temps on different cores? My cores 3 & 4 are constantly a bit higher than 1 & 2. Eg. when 1 & 2 are at 33c, 3 & 4 are at 35c and 37c.

[DieBuche]

xopher, on Oct 28 2009, 10:38 PM, said:

Didn't notice that, thanks for pointing it out!

Edit: One more question, do you guys have different temps on different cores? My cores 3 & 4 are constantly a bit higher than 1 & 2. Eg. when 1 & 2 are at 33c, 3 & 4 are at 35c and 37c.

32,27,28,30 for me, b and c are always the coolest

[mm67]

DieBuche, on Oct 29 2009, 10:08 AM, said:

32,27,28,30 for me, b and c are always the coolest

Now in a warm room 36, 34, 38, 38. Same also in Windows.

[kdawg]

@FormerlyKnownAs

Question, when observing CPU performance with CPU-i pr VoodooMonitor how often does your CPU hit the x7.0 multiplier?  I see mine hit every once and while but not as often as you'd think. Does anyone else see short hits on the "in-between" multipliers? I have three x6, x7 and x8. The x6 and x8 get 90% of the activity.

I'm trying to see how effective my C & P states are.