The "Undervolting" Guide

[email protected], XP Pro, stock voltage 1.3560v, temps w/stock voltage: idle 39C, load 59C, undervolted 16x FID to 1.1000v, u/v temps: idle 33C, load 44C. Nice.

 

Has anyone gotten RMClock to work in Win7 on battery? I've been trying everything and no results yet.

 

I thought it was clear that you're not allowed to have more or better toys than me, eh

Gawd, what a stock voltage! Direct connect to Grand Coulée is it ?

 

Lots of people have. What's your exact issue ? "No results" is not helpful.

 

Sorry Scott, I must have missed that post. Besides, my collection is nothing special. Besides the sig unit, all I have are 2 M1710s, 2M170s, an E1705, a Gateway ARC200, a few docking stations, 3 24" LCD monitors and 1 19" LCD monitor. Fairly insignificant, in the overall scheme of things.

Yea, I had a couple of 240V/110A lines feeding the two M170s, before I U/Ved them.

 

It all pales in comparison to one single, highly polished, 8 inch DeLorean belt buckle

 

The FID/VID's don't stick when it's not wall powered. No matter which power profile I choose, or setting I change.

"High performance" with min cpu set to 100% still drops to 1.2GHz as soon as I unplug it.

 

Do you have a processor cap set in Vista/7 processor management?

 

Cap as in max setting?

 

Here's a flash, Johnny: Only the balanced plan in W7 will access all FID's for both battery and AC. Modify that plan to cover all modes and let RMC do only the voltages.

 

Has anyone figured out how undervolt a Core i7 yet?

 

Undervolting is amazing. My 3810t with and SU9400 is at 49C while recording in WMC with my tv tuner.

 

I think CPUgenie works but its only a 30 day trial and u need to buy it afterwards...

 

Ah that's too bad. I'll try it out anyway. If it successfully reduces my temps, it's worth every penny!

 

It does not work with i7's.

 

I'm using a studio xps 13 with P8600 cpu, and I've set RMmonitor to adjust the max mult to 1.000V. similar to what other people have reported, adjusting the max mult to a voltage below 1.000V doesn't actually lower the voltage. I've read that the studio xps 13 cpus are locked so that they can't be set below 1.000V, and i've accepted this (although somewhat disappointing).

However, I'm seeing two things that I don't think i should be seeing:

1. every once in a while i'll see RMmonitor and CPU-Z indicate a voltage spike back to the original 1.125V setting that the CPU is defaulted to.
2. both RMmonitor and CPU-Z never show the voltage dropping below 1.000V, even though they both indicate the multiplier dropping to 6.0x (the lowest muliplier). The multipliers 7x and 8x are also never seen.

Can someone explain this to me please? Maybe if the cpu is locked from going under 1.000v at max mult, this also applies to the lower multipliers too? Even though that sounds really dumb.

 

OK, but you need to understand that transistors don't dim down like a light bulb or slow down like a motor when voltage is decreased. At a certain point they just quit functioning. It's not negotiable. You're lowest allowable voltage is locked in your BIOS chip so the owner (you) won't drop the voltage below the BSOD point, and return the unit as defective.

1. Do you have IDA enabled? When it spikes the multiplier up, the voltage spikes up too. It's supposed to do that. W7 does it even if you turn IDA off.

2. Like I said, the lower limit is locked. You can ask RMC for 0.500v too, but you won't get it.

3. Intermediate multipliers are never used in transitions. It would be a ridiculous waste of processor time going from 6x to 9x and having to spend time at 7x and 8x.

 

thanks, that really helped! is it weird that the default voltages for the multipliers drop below 1.000v then? so dell set the voltage for the 6x mult to .8750v and also locked the cpu voltage from dropping below 1.000v?

 

The voltage table that pops up by each multiplier is just a list. It doesn't mean that all of those selections are really available to you. Some will work, others are forbidden and won't stick.

I don't understand this sentence: "so dell set the voltage for the 6x mult to .8750v and also locked the cpu voltage from dropping below 1.000v" Tell me how Dell selected 0.875 and are you using SLFM?

 

yes, i'm using SLFM, and i meant that in the step of voltage calibration where you go into the profile page and click on "default" to get the computer's default voltage settings for each multiplier, the voltages that are set for the 6x and 7x multipliers are both below 1.000v. but if the cpu is locked at 1.000v, then why would these multipliers show a voltage below this value BEFORE i've made any changes?

 

Because RMC has no way of knowing what's locked out by your Dell bios, so it offers up the entire known range for your CPU model.

Why does Dell lock out low voltages? Because the cost of replacing a "defective" unit that BSOD'd due to reckless undervolting wipes out the entire profit they made by selling it in the first place. They're protecting themselves.

Don't use SLFM. After undervolting your CPU, it serves no purpose regarding heat or battery life.

 

Windows 7 UAC is stopping rmclock startup on boot, any ideas on how to fix without disabling UAC ????

 

Try this:

http://blogs.techrepublic.com.com/window-on-windows/?p=616

 

I tried that a few time with Vista and it did not work.

 

I know it can be done but I didn't pay attention. I just turned it off. Do a "Search This Thread" for UAC and shorten your reading list.

 

I have a problem with the rmclock...

Mine is x61t L7500.

Under xp ,rmclock shows a lowest voltage at 0.9000v
but last time I installed win7, rmclock shows 0.8500v....

After I sent back for reparing, now under win7,it agains shows 0.9000

all the voltage have been checked by cpuz...

and under 0.8500 6x is stable...
so....what can I do in order to let the rmclock show 0.8500 again?

THX

 

Read #4271

Looks like you got a new bios flashed at the repair shop. BTW, how much battery life and operating temp. is the extra 0.050 volts costing you ?

 

Mine is a 8 cell battery.
not very accruate,0.05 gives me about 20 minutes.
It now can reach 3.5 hours easily(you know,touch screen costs lots of power)
before that,under 0.9 volt,I can hardly reach 3 hour...
All are under SuperLFM w/ 0.85v or 0.9v

You know,the deafult range is just 0.98xx ~ 0.90

 

Hey, for some reason when I run RMclock, the "Advanced CPU Settings" option is not available and neither is the sub-options under "Profiles" (except for "No Management"). I have downloaded the program a few times already with the same problem. I am running it on 32-bit Windows seven. Any ideas where I am messing things up?

 

Is RMC not loading correctly because of UAC ?

 

Close enough. The exact mathematical prediction is 25 minutes. SLFM doesn't give any extra help over the reg. FID at the same voltage.

 

I have downloaded RM clock several times, but cant get it to work. says it needs a 64bit driver

im running windows 7

 

Well as you can see I've done that, and it still defaults to 1.20GHz.

 

@ Derek1387, See the first post of this thread for the 64-bit driver.

 

That's on the very first post of this thread.

 

Using thumbnails, show the OS settings tab for Max. Performance.

 

Uncheck "Use OS Power Plan". RMC must not be involved in anything but voltages.

 

I am not sure, how do I check the UAC?

 

I followed the guide, reducing the voltage of my highest multiplier and allowing my m1530 to run for a while whilst having the CPU stress-tested but the temperature reached the same maximum. So I continued to lower the voltage and there was still no decrease in temp and then eventually the laptop just shut down. I assume the voltage was too low, but I'm surprised the max temperature was consistently the same with each lowering. Any ideas?

 

Probably the fan slowed down.

Do you hear the fan before/after undervolting?

 

Can anyone help me??



Im getting BSOD at 4.0x but in 8.0x it is stable, I stress test the 8.0x for about a hour and I didn't get BSOD, but at 4.0x in idle mode (not using ORTHOS), my laptop randomly shutdown or getting BSOD, I also try increasing the voltage up to .850v(which is my 8x) and Im getting same results...

I also check CPU-z to know what voltage that the 4x is using and it only using .800v, so I dont know why it is getting BSOD or shutdown, pls help

My netbook
MSI u210
AMD Athlon Neo x2 Dual Processor L335
(1.6GHz, 512Kb L2 Cache)

 

I have the same problem, same cpu L335, I tested each multiplier for at least 45mins each with orthos, and it only seems to crash when the notebook is idle or doing very little.

 

I just backread some couple of pages and I will disable the 4x and I will only check the 8x, it seems that 4x and 8x have same power consumption even it changes it core speed, well correct me if Im wrong ^^

 

The temps probably did not drop because the heatsink/fan is dirty/plugged. Take the laptop apart and clean the heatsink and fan. Just blowing air into the exhaust port will not do it.

If your unit is still under warranty, check with Dell as to whether the warranty will be affected. If so, send it in to be cleaned under warranty.

 

I ran a test to see the difference in battery life between the min FID and the max FID. The difference was a couple of seconds. I run all my laptops undervolted on max FID only. If the 4x FID is giving you problems, uncheck all the FIDs except the max and you should be golden.

 

Windows User Account Control

 

wow thanks for another info ^^, thanks for the guide threadstarter, been doing undervolting last night, need to sleep hehe...

 

hi to all.

dows anybody know how to undervolt beyond the available steps in rmclock?
I can run my first 5 ( !) p-states (and therefore multipliers) with the lowest available voltage step which is 0,9 V by the way.

I ve done the following things,

I 've run RM_clock TWEAKS.reg which is found on the rmclock folder in program files,
and then gone to regedit,
i ve navigated to
hkey current user,software, rightmark,rmclock,
found the "UNLOCK VID" entry, changed the value it so i could select lower voltages in Profiles tab in main rmclock menu
.
I tried to use lower voltages in profiles tab, but they don't seem to take effect. the voltage used remains 0,9V as i see from other programs like cpu-z

p.s.

I am not sure if that works or has any relevance at all after all, but i also tried to choose custom p-state from Management page in rmclock main menu. but this won't let me choose a different voltage. the voltage is like a read-only reading depending on what p-state you choose in the drip down menu next to it.

thanx for your time. as you see i did quite a searching on my own but just didn't find any solution.

p.s2. i don't know it is any help but i have a vaio notebook fw series, with a t9900 cpu @3,06 ghz

 

See posts #4266 to #4271 from less than a week ago.