The "Undervolting" Guide

Many Thanks Flip, i battled this issue for more than a year at last it ends with cooler temps without crashing....

you owe me a reputation...you will get it....

 

Installed CPUGenie v1.5 x64 on Windows 7 and on max load i gone from +95c to 80c.

 

After reinstalling the OS back to factory settings, do I need to redo the process of undervolting?

 

you just need to save these settings and then apply again. A quick stable test is preferred, just to be sure.

 

Excellent guide. Just did this on my work computer and dropped down 0.150V which translated to a 16C drop in temperature! I'm going to continue to run a stress test for the rest of the work day to ensure that this is stable.

Go green! xD

 

Hi and firstly thanks a lot for this amazing guide ; i am a real beginner to all this and could do it without any issue.

I just have one little question : Once my optimal settings found, do i have to open RMClock everytime i start up my computer or do the settings apply automatically by checking "Performance on Demand" in the "Startup" field ?

Thanks a lot

 

RMClock must be running all the time for its settings to be used.

 

Thank you !

 

What should i do if the "Advanced CPU Settings" tab is not showing up in the selection area

 

I have a Dell latitude E4300. I've had previous bad experience with RMclock because it causes my computer to crash. At first, I assumed it was the instability at lower voltages. So I put of undervolting until now.

I've found another program called CPUgenie, which most of you probably have heard of or use. It's pretty similar to RMclock anyway.

CPUgenie works, however, I can't seem to to undervolt my SP9400 processor below 1V.

My processor supports SuperLFM and it's FSB multipliers ranger for 6-9, increasing by 0.5 each step.

I use both CPU Z and Hwinfo 32 to monitor my voltage, FSB and multiplier.

In SuperLFM, the voltage is 0.875v and even if I try to set it below that, it won't change. That's fine since 0.875 is the lowest voltage specified by Intel, and I head you have to do pin modding to overcome this.

In Normal, the voltage cannot be set below 1V regardless of what the multiplier is. Is there a way to get below 1V? I can already undervolt the higest multiplier, 9x from 1.135v to 1v and it runs stably. So I do wish to undervolt the other lower multipliers below 1V to get a cooler system since I hardly ever need a full 9x.

 

Edit: RM Clock lets you use the SLFM voltage even when you are not using the SLFM multiplier.

In the Advanced CPU Settings you need to have Dynamic FSB Frequency Switching turned on and then in the P-States editor you can use the SLFM voltage with the 6.0 LFM multiplier.

If the Advanced CPU tab does not show up that means your CPU or motherboard is not supported. What CPU do you have? The CPU-Z program can show you that as well as what chipsets are on the mainboard.

 

CPU - Z says its a AMD Turion X2 RM-70

 

Got it to work on a 4years old AMD Turion 64 Mobile Technology ML-32 with the Cool'n'Quiet tech....I'm still testing it, some freeze appeared....(min 4X 0.7000V - max 9X 1.4000V).

 

Thanks for your reply. I have Intel's C2D SP9400 processor.

Assuming that CPU-Z and Hwinfo32 are not giving false readings(although I heard that the voltages can be a little off) I've now realized what the problem is.

The first time I tried undervolting with CPUgenie, there must have been some conflict between two things controlling the CPU voltages at the same time. So when I tried to set the voltages between 1-1.375, the processor obeyed CPUgenie. But when I tried to set it below 1 it didn't.

Now after a few restarts, the processor seems to obey CPUgenie and the I can set the voltage down to 0.875 for for any state, SLFM, Normal and HFM.

The problem is actually that I ran ThrottleStop to monitor the load at the same time. Even with throttle stop turned off, I notice this peculiar thing. Lets say I have CPUgenie, Hwinfo32 and CPUZ running, I can set the voltage to 0.875 and it shows correctly on CPUZ and Hwinfo32. However the moment I launch ThrottleStop, the voltage jumps back to 1 and even if I close ThrottleStop, I won't be able to set the voltage back below 1 until I shut down and boot up my computer again. Simply restarting won't work. Another thing that ThrottleStop does is prevent my processor from going into SuperLFM, even after I close the program, my computer will not go into SuperLFM (half FSB). Shutdown and boot up fixes this and I will be able to enter SuperLFM again

 

fatedquest: Can you post a screen shot of CPU-Z and ThrottleStop?

If ThrottleStop recognizes your CPU correctly, it will provide you with an option to toggle SLFM. The default when you start ThrottleStop is for SLFM to be turned off. If the SLFM box is cleared and you shut down ThrottleStop then you would run into the problems that you had. If you enabled SLFM in ThrottleStop before you shut down then everything should be fine without having to reboot.

If the SLFM option in ThrottleStop is grayed out because it does not recognize your CPU correctly then that would be a problem. I will adjust ThrottleStop so it does not automatically disable SLFM when you start the program for the first time. Thanks for the feedback.

 

@unclewebb, here's the screenshot of CPU-Z. I'm lazy to take the screen of ThrottleStop(btw, thanks for the great program) cause I don't want to reboot now. I think it does recognize my CPU correctly. The SLFM option is not greyed out and as you said, I never ticked it. Also below is an undervolting table for a rough reference. It should be good for other SP and p9XXX series processors. The estimated energy saving is a very very rough estimate because it uses the values for full load.

 

Thank you for this easy, yet comprehensive guide.

I never thought it'd be worth it, that was until I bought Crysis for my SXPS & the CPU throttled down...

My prime loads were just under 90c, after a 0.15v under-volt they're just under 80 & more importantly, it doesn't exceed 80 with the GPU additionally loaded.

I'm left curious though, is it possible to under-volt the GPU on a SXPS too? I've hard of crazy people overclocking it with rivatuner, not something I'd contemplate with it hitting 95c...

Thanks again.

 

I'm already stuck at step iii:

Because when I go to Advance CPU Settings I don't get the options described or see my processor. It says "AMD K8 family advanced CPU settings" across the top and I get three tabs: CPU setup, P-states transitions and Misc. control.

On the CPU info page the graphic detects it as an AMD 64 processor, but the model displays "AMD Processor model unknown". CPU-Z correctly identifies the processor, which is an AMD Athlon Neo MV-40. HWMonitor doesn't display any information about the processor at all and RMClock doesn't detect any temperatures either.

How can I go about setting up RMClock with my processor and should I disable AMD's PowerNow feature (speed stepping) in the BIOS before setting up RMClock? With PowerNow enabled the processor runs at 800 MHz on the battery and 800 / 1,600 MHz as it needs it when plugged in. With PowerNow disabled the processor runs at 1,600 MHz at all times.

 

RM Clock is no longer being actively developed so it's not going to be able to correctly detect and be used on modern processors and on modern chipsets. That's why pages get grayed out.

I tried the AMD power saving features recently on a desktop Quad and it killed performance. The time required for a single Super PI benchmark increased by 50% when using this feature. It was obvious that the CPU was not being allowed to get back up to full speed like it should have been allowed to do. I'd never use this but do your own testing. Maybe it is more responsive on AMD's mobile CPUs. Intel SpeedStep works significantly better.

 

When running RMClock, my CPU Load maxes out at 95.65%. It happens for any voltage lower than default. Has anyone else experienced this?

 

What CPU are you using, what type of load are you using to test with and what does the task manager load meter say? Is your CPU fully supported by RM Clock? Prime95 Small FFTs is the best test for a consistent load.

 

unclewebb, thanks for the reply. Recently upgraded my GW P7805U from the stock P8400 a T9900 Running Vista 64bit. Using Orthos (v0.41.110.18) as my load. The task manager load meter shows the processor load is 100% for the test I'm currently running (10X @ 1.0375V). I'm guessing that RM Clock doesn't support the T9900 and it isn't resolving the actual cpu load. I'll download Prime95 and give it a go.

 

Orthos is good enough for a Core 2 so Prime95 won't make any difference.

RM Clock does not support the half multiplier that your CPU uses and that screws up the load meter it reports.

The default multiplier for your CPU is 11.5 but RM Clock assumes that the default is 11.0. Grab a calculator and do 11.0/11.5=

Do you see a familiar number? 0.9565 or 95.65%

Have you tried to modify RM Clock yet so it can correctly support the half multiplier? I did a registry mod that I mentioned previously and now my T8100 can use its 10.5 default multiplier or the 11.5 IDA multiplier when using RM Clock and that fixes the load meter too.



Here's what you need to do.

http://forum.notebookreview.com/6130270-post4855.html

A T9900 has a default multiplier of 11.5 and the IDA multiplier is 12.0.

 

I'll make the switch to TrottleStop this weekend and re-test/adjust for the 11.5X multiplier voltage. Can you post a link to the latest version?

 

ThrottleStop doesn't offer the fine control of multipliers and voltage that RM Clock does so with a quick registry tweak, it's easy to get RMC working more or less correctly with the newer CPUs. If you need any help with that just let me know.

You can download ThrottleStop 2.00 Build 25 here.
http://www.fileden.com/files/2008/3/3/1794507/ThrottleStop.zip

Any newer versions end up at that link. I'm just about ready to upload Build 26 within the next day. Just a few minor adjustments lately to better support QX CPUs.

What I found when testing is that the newer 45nm CPUs automatically spend most of their idle time in the C3 or deeper sleep state so any settings in RMC or ThrottleStop are not relevant when this happens. The CPU then uses its own internal settings for VID.

For this reason, ThrottleStop only concentrates on idle and full load so you don't need to spend hours coming up with the perfect VID / FID settings. My theory is to let the CPU run at full speed as much as possible so it can get its work done and then quickly go back into the deepest sleep states to save power. There isn't a huge difference in overall power consumption between RMC and TS when used on the same 45nm CPU and the ThrottleStop advantage is that it supports the newer CPUs with half multipliers without needing a registry adjustment.

If you have an Nvidia GPU then just add GPU=1 to the ThrottleStop.ini configuration file so you can monitor and log GPU temperature information too.

 

Regarding this. My BIOS shows that my CPU voltage is like 0.2V although CPU-Z reports 0.875V. Is it related to C-States?

 

I don't think so.

The Intel Spec Update for many newer Core 2 mobile CPUs like the T9500 and even many low voltage models, shows a voltage range of 0.350 - 0.700 volts when the CPU is in the C6 sleep state. I think the bios reporting 0.2V is most likely just monitoring error. I don't think the CPU can enter a deep sleep state when you are sitting in the bios.

 

I have replaced T7500 in T8100 in mine hp 8510p... (just a bit slower, but much cooler). It works stably on second lowest voltage (0.9 + one step up)... I have noticed that core 0 ALWAYS shows 30C in any conditions and core 1 changes temps normally, according to CPU load... Is it normal or, if no, how to

 

I'm actually booted up into Windows already.

I'm monitoring from the fn+shift 15234 Thermal Control Panel.

Btw, do you know any way to undervolt lower than the lowest voltage in SLFM. Right now my computer is running stably on the lowest SLFM voltage in normal FSB clock and multiplier. I guess my SLFM state would still be stable if undervolted further.

 

You can't, that's as low as the CPU will go.

 

My question is smiliar to the previous...

My laptop has an T9400 cpu (2.53ghz @ 9.5x, 2.66ghz IDA 10x). In SLFM mode this cpu goes to 800mhz (cpu-z reports it has 3x multiplier instead of 6x with half fsb).
I've been reading that this cpu in slfm has lower voltages
Core voltage in SLFM mode (V) 0.75 - 0.95
CPU-z is reporting 0.95v in SLFM, is there a way to make it 0.75v? (in slfm i am capable of getting the 8x multiplier with 0.95v) i think 0.75v is pretty stable with 6x multiplier and has a much lower power usage...

Keep up the good work with throttlestop, excelent software

 

CPU Z is unreliable to check your SLFM state, multiplier and voltage.

Use HWinfo32 instead. You can easily see when it goes into SLFM, the correct multiplier, correct voltage and even your actual RAM clock, before DDR and before Dual Channel.

 

ivanox1972: Core 2 thermal sensors were not designed to be accurate at reporting idle temperatures. They have several issues including the fact that many of them get stuck as the actual core temperature gets lower. There's no fix for this problem. All you can do is ignore what it says at idle. As the core temperature increases, sooner or later they all start moving but below the sticking point, they'll just sit there and keep reporting the same thing.

I still don't think the 0.2V number is accurate but anything is possible. That number seems significantly below the minimum voltage number listed in the Intel docs so I'm a little suspicious.

The minimum SLFM voltage is locked at the factory in each CPU and you can't go lower than that by any method that I know of.

What I found on my T8100 is that at idle, none of this matters. According to the Windows Performance Monitor, it spends 99% of its time in C3 or a deeper sleep state so the SLFM voltage I choose in RMC or in ThrottleStop doesn't change power consumption one bit. This value is ignored 99% of the time at idle or when lightly loaded in the 45nm Core 2 mobile CPUs so I see no point in getting super technical about the perfect SLFM voltage.

TigTex: A CPU can rapidly, hundreds of times a second, be entering and exiting SLFM mode. The most accurate way that Intel provides to monitor for this results in the numbers being reported like this:

3 X 266 MHz = 798 MHz

even though internally, this is what is actually happening.

6 X 133 MHz = 798 MHz

Same total MHz either way RealTemp, ThrottleStop and CPU-Z all report this the same way now.

Core 2 CPUs have a problem that their internal timers shut down when they enter deeper sleep states. This makes calculating the bus speed accurately difficult to do without having to constantly wake the CPU up. Intel fixed this limitation in the newer Core i CPUs so even in deep sleep states the timers continue to run at their regular speed.

Constantly waking up and loading a sleeping CPU with brief bursts of full load isn't recommended so ThrottleStop avoids that. Other software might be doing this.

If the adjuster in ThrottleStop only goes as low as 0.95 volts then that is all your CPU supports. That is your minimum SLFM voltage that was read from your CPU. If you enable the SLFM option in ThrottleStop then you can use this voltage at full load without the CPU cutting the bus speed in half. It can run the SLFM voltage at whatever multiplier you set it to and the bus speed will be the normal 266 MHz. My CPU only runs reliably at 1400 or 1500 MHz when using the SLFM voltage at full load but it's a great way to make a low power chip. At idle, the actual voltage will drop even lower when it enters C3/C6, etc.

 

Ok, looks that 0.95v is the lowest that i can get...

I have other question. How does throttlestop make cpu speed transitions?
Controlled by os?
Controlled by acpi?
Controlled by software?
because i see my cpu going from 6x slfm to 10x and 6x slfm a lot of times. acpi states are 6x slfm, 6x and 9.5x.
Using intermediate states maybe can decrease power usage no? On my pc i'm capable of running a 8x multiplier with 0.95v, better then the default 6x (normal mode) with 1.000v.

and one other question... under windows performance monitoring -> processor c states I see that cpu it's always on C0 to C2 states when on AC and goes to C3 on battery only. Is that normal?

 

All monitoring software has some flaky moments, even HWiNFO32 but I also prefer HWiFNO32 to CPU-Z.



ThrottleStop is designed to work at two speeds. Full speed or low speed and it ignores the OS and ACPI. You might get better power savings using RM Clock. It depends on how you use your computer. On a 45nm CPU that uses C3 at idle, I have not been able to measure any significant difference because most of these settings are ignored. At idle, my laptop consumes exactly the same amount of power whether I have it using the highest multiplier or the lowest multiplier or the highest VID or the lowest VID. All of this is ignored when in C3.

Maybe that is true but I don't have accurate enough equipment to test that. When lightly loaded, some people think it is better to run a CPU really fast so it can get the work done quickly and then when the work is finished, it will be able to spend more time in the C3 state and get better overall efficiency. Microsoft must also believe this because they changed how Windows 7 operates. Vista used to stay with the low multiplier for a bigger percentage of time but Windows 7 increases the multiplier much more quickly when lightly loaded. Use whatever method you think is best.

RM Clock still works great for many users and if you need fine control over VID and FID then I still recommend it. I will also help you get it working better with the newer 45nm CPUs and their half multipliers. For me, ThrottleStop is good enough as is. My laptop spends most of the time plugged in so saving half a watt here or there doesn't mean as much to me as it might to other users.

My Dell laptop spends most of its idle time in the C3 state whether it is plugged in or on battery power. I'm not sure what is normal. Your laptop method is probably better and might give you slightly better response when plugged in but it would be hard to notice a difference.

 

RMclock is too old and uses too much cpu time to be eficient. Throttlestop is running for more than 4 hours now and cpu time is 0:00:00 (perfect!). Rmclock, when i watch a video, cpu usage goes to 5-10% with no reason at all.

I'm testing here right now with a powermeter.
My laptop at slfm 800mhz 0.95v uses 25watts (no load).
Also no load but at 2.53ghz 1.075v uses 25watts, 1.250v uses also 25watts...
Looks like it's useless running my cpu at lower speeds, but somehow, maybe it's just me, but i think that cpu is hotter on idle even with the same power usage (fans start spinning more times)

ps. i saw your printscreen... your C0% time is almost 0%, on my cpu, on idle it's between 30-35% on core0 70-85% on core1.
by the way: asus G71v laptop.

 

That's exactly what I discovered too.
VID, FID, C States, etc., doesn't make any difference at idle on the 45nm Core 2 mobile CPUs. It used to make a difference on the original 65nm CPUs like the T7500, etc. I thought maybe my watt meter was not accurate but it goes up and down just fine at idle when I adjust the screen brightness. It just doesn't do anything when I use software to adjust the CPU. Monitoring software will report that you have significantly changed your CPU but in terms of power consumption, you haven't changed anything.

If power consumption is the same at idle then core temperature will also be the same. It takes energy to create heat. Same energy, same core temperature. Maybe your fan speed is determined by the VID setting.

Thanks. Someone finally noticed! When ThrottleStop is minimized to the system tray it is extremely efficient. It's also very efficient in terms of memory usage.

The C0% in ThrottleStop is based on the Intel approved method for Core i CPUs. This number is not accurate at idle on Core 2 mobile CPUs because of the issue with the timers going to sleep at idle in Core 2 Duos. As your CPU goes into a deeper sleep state, the displayed C0% number will significantly increase.

The screen shot I posted with a low C0% number at idle must have been with a desktop CPU which doesn't enter the deeper sleep states at idle on my motherboard.

 

throttlestop is really and excelent software for laptops users and even desktop.
I think all you need now is a website with your software, no? because it's difficult to keep us updated when new versions came out and a changelog (except the one included in help files).
I'm using your software and I never faced a bug.
Keep it up

back on topic

core 2 duo T9400
2.53ghz from 1.175 to 1.075v - less 7ºC. 55ºC full load, almost 10w less power usage. Success ^^

 

Thanks TigTex. I agree. ThrottleStop is a useful program for many users. Updating the documentation and creating a website is on the things to do list. Hopefully some day.

Good work with your under volting project.

 

How do you guys check when the processor is in a C1, 2, 3 or deeper sleep state? I can't find anything that indicates that from Windows Performance monitor or resource monitor.

 

just curious: how do you know that?

 

unclewebb, you can always use google sites.

 

Where can I find this throttlestop? My RMclock undervolt-setup doesn't work anymore, is this softwere better or is it something else?

 

a wattmetter atached to my AC adapter...
You can also see that in windows performance monitoring. There is a counter called voltimeter. It's shows the power usage in watts while you are on battery.
And there is other called processor, that shows current C states.
I'm portuguese, so maybe in a english windows the names aren't what i said...

Rmclock is old... i think it's better that you use throttlestop.
You can download it here http://www.fileden.com/files/2008/3/3/1794507/ThrottleStop.zip
Remember, you can only set 1 voltage with this software, instead of every voltage for every multiplier, but like I noticed... there is no diference at all in power usage and you will have a better performance (but probably not noticiable).

 

ThrottleStop 2.00 Build 26
http://www.fileden.com/files/2008/3/3/1794507/ThrottleStop.zip

ThrottleStop is not a competitor to RM Clock, it's an alternative. Some users don't need the fine control that is possible with RM Clock or some of its other options. When you discover how the 45nm CPUs operate at idle, a lot of the extra control that RMC provides isn't necessary anymore. You can use ThrottleStop and improve power consumption at full load with a couple of quick mouse clicks. ThrottleStop only supports Intel CPUs.

In the Windows Performance Monitor there is also an option to measure Battery Usage which is measured in milliwatts. The data only updates about every 5 seconds but it seems to be very accurate and tells you how many watts DC your entire laptop is consuming. This number is not the same as what a Kill-a-Watt meter shows. The KAW shows AC power consumption from the wall. That will be greater than how many watts DC your laptop is consuming.

You will need to adjust the scale in Performance Monitor to 0.001 to convert milliwatts to watts.



Just click on the green + sign at the top to add what you want to monitor.

 

Thanks unclewebb. You're right, my processor spends more than 90% of it's time in C3.

 

I think most Core 2 mobile CPUs try to do this if the bios allows them to. When this is true, coming up with some precise VID / FID settings at idle is pointless because the CPU totally ignores these settings when it is idle in C3 or C6. It's still important to keep your VID low at full load but you're better off learning ways to maximize the percentage of time your CPU is able to spend in the lowered power C3 state. RM Clock might interfere with this and decrease the C3% time. The graphs were an eye opener for me too.

Using a high multiplier at idle might not be such a bad idea if it allows the background tasks to be finished quickly so your CPU can spend more time in the low power C3 state.

 

I want to try to undervolt my AMD Athlon Neo MV-40 in my MSI U230 and don't really know where to start, but one thing I would really like to know is should I disable AMD's PowerNow feature before beginning any undervolting? (On battery PowerNow keeps the CPU at 800 MHz and when plugged in it runs at 800 MHz and then steps up to 1,600 MHz when needed.)

Will we ever see AMD support?

 

No. I don't have access to any AMD hardware.

 

I've been using process monitor (aka procmon) and I found that throttlestop access the hdd every 9 seconds for reading the .exe
Any way to avoid that?

hdd's doesn't spin down because of it...