The ThrottleStop Guide

Yes. L9400-1.86Ghz->2.64Ghz (+42%) is sweet performance boost on AC. I get idle < 5W idle on battery seeing 8.5hrs battery life from a 6-cell or 13.5hrs from a 93Whr 9-cell as measured by BatteryBar. See INFO: 2530P Overclocking for details on how I did it. Impressive if I do say so myself. A L9600-2.13, with the extra multiplier, would be getting up to 2.97Ghz with the same mods.

Could try a 200->266Mhz FSB overclock on your T61 as detailed in Lenovo/IBM T61 FSLx PLL pinmod. Theoretically could get your T9500 to a dual-IDA 3.7Ghz, giving you i5-5xx performance levels.

 

^ I saw what you posted there, but also saw people reporting problems with the SATA controller and/or with the Expresscard slot(and I do have eSATA ExpressCard), with resume/sleep stuff ... etc. Good info nonetheless, thanks.

but for now things look fine for my T61 One day I may boost it up again though, hehe

 

OT, briefly (sorry Unclewebb). You'd be doing a FSLx PLL pinmod which would either work on bootup or not. There would be no sleep/resume/expresscard issues since you wouldn't be software programming the PLL after bootup. Well worth a T61 owner pioneering the way to i5-5xx performance levels from 2-gen older tech.

 

changed my Avatar, TS ROCKS........................

 

......
.... .........
.... ...........
.... .................
.... ............

 

Hi Unclewebb - firstly thanks for picking up the mantle left by RMclock and producing a great tweaking tool for the community.

Secondly, my question (yes, you're the exact reason I signed up to these forums!). I'm looking to be able to set a VID for each multiplier setting on my P7450, whilst leaving the dynamic FSB switching abilities of Core 2 Duos intact. As far as I can tell so far, there is no way to set up "per multiplier" profiles on throttlestop, you have to manually select between modes?

If you are able to advise if I'm missing something I'd be hugely grateful!

Thanks.

 

Makhaira: I didn't bother to include in ThrottleStop the fine CPU control that RM Clock gives you over the Core 2 CPUs. If you feel you need that control then it is easy enough to modify the RM Clock registry to support the half multipliers that the newer 45nm CPUs use. Here's a guide I wrote for that.

http://forum.notebookreview.com/6534263-post5202.html

Based on my testing, it was difficult to measure any advantage to using the intermediate multipliers on the newer 45nm CPUs. The only important thing I found was controlling the maximum voltage so that's why ThrottleStop operates the way it does.

At idle, whether the FID and VID are high or low doesn't make any significant difference to power consumption because internally these CPUs go into the lower power C States like C3/C6 over 95% of the time where the VID they use can't be controlled anyhow. Here's an example.

http://forum.notebookreview.com/6369641-post5060.html

When lightly loaded, running the CPU fast with a high multiplier so it gets its work done quickly is more efficient than running the same CPU at part throttle by using a reduced multiplier. Running the CPU fast allows it to go back to sleep and spend more time in the low power C3/C6 state where the real power savings can be found. For a P7450, the Intel specification for core voltage when in C6 is between 0.35 to 0.70 volts.

I included power consumption monitoring and logging in ThrottleStop but I haven't seen much user testing yet. It's interesting to watch how power consumption varies while playing around with the multipliers and VID in ThrottleStop while watching a YouTube video or similar. Windows only updates power consumption data every 10 seconds so keep that in mind when testing.

If you find some big savings then use RM Clock or to maximize battery runtime, do what nando4 does. He shuts off ThrottleStop on battery power because he found that the less software running, the better. That's what the ThrottleStop DC Exit option is for.

 

So quick question (hopefully not a dumb one)...

I've been trying to use TS on my Dell Studio 14z ever since I upgraded to a P9500 with limited success. It seems to be stable under "normal" use, until I pull it out to watch movies (MKV's of DVDs and Blu-rays). It'll run just fine for a while, then BSOD. The really odd thing is I can run Orthos overnight (8+ hours) at a given VID setting and not have any problems. Movies will sometimes crash it within minutes. Sometimes they'll play just fine. I usually try to increase the voltage a couple of times (one notch per time), get a couple more BSOD's, then end up just disabling TS and watch my movie in peace.

I'm up to a VID of 1.1250 currently (if I remember right, Orthos became stable somewhere around 1.0000).

For watching movies I'm using MPC-HC and ffdshow DXVA (Mpeg4 and VC1) and Microsoft DVD-DTV DXVA (Mpeg2). I tried a number of different codecs and these were the ones that had the least CPU% when in use (trying to eek more out of my battery). ffdshow audio decoder is always used.

Also, TS reports the actual voltage as 1.0V the vast majority of the time (so does CPU-ID and HWinfo).

Does this make sense? It just seems really odd to need the VID this high (not that I'm an expert...).

 

Near the top of the ThrottleStop docs it says this:

If you don't do this then sometimes ThrottleStop will be in control of your VID voltage and sometimes Windows will be in control of it. The VID will flip flop back and forth and could be the cause of some of the instability that you are seeing. If you set a VID of say 1.1000 V then ThrottleStop and other software should be reporting that VID.

If you watch a movie for an hour or two why not create a Movie profile with a lower multiplier and lower VID. You probably don't need to run your CPU at 2.53 GHz to watch a movie smoothly. Run a TS log file while you are watching a movie and it will show you the C0% or CPU load. You could also use the Disable Turbo feature since sometimes this needs significantly more core voltage to be stable for only a slight increase in speed.

When Orthos testing, have you tried running a single thread of it or Prime95 and then locked it to a single core of your CPU? Use the Task Manager Set Affinity function to do this and you could lock it to each core, one at a time, to see if either core is weaker than the other. This will allow your CPU to use Intel Dynamic Acceleration (IDA) which might show you that your CPU isn't stable with the VID voltage you are trying to use.

Some LinX testing is also helpful and doesn't take nearly as long as Prime95 to show you if there is a stability problem with your settings. Test both 1 and 2 threads.

Is your P9500 a retail CPU or an ES Engineering Sample? Hopefully I've listed enough things for you to try so you can get this problem figured out.

 

unclewebb,

Thanks for the insight...I must admit I hadn't read that about the power profiles. I'll give all the things you suggested a shot and see where I get...the idea about a different profile for movies is a great idea.

My P9500 is not ES, is a normal production CPU.

You've gven me some things to try...once I get a little time I'll give it a shot. Thanks again.

 

Somewhat-silly question: Is there any way to create keyboard shortcuts for TS profiles?

 

That's a good idea but it's not yet possible to use keyboard shortcuts to change profiles. Maybe someday.

At the moment all you can do is assign ThrottleStop to one of the function keys ( F1-F24 ) which allows you to toggle ThrottleStop on and off.

 

Question 2: Could an "anti-oopsie" provision be added? That is, if ThrottleStop runs at logon/boot and has settings which cause the computer to crash (say, voltage too low and the profile being loaded is the IDA--happened to me today), would it be possible to prevent TS from running next boot? Or perhaps run it in monitoring-only mode to avoid some kind of loop?

Example: TS runs with a "safe flag." If TS starts and 15 seconds pass, it will run next boot as if nothing happened. But, if TS starts and there's an unexpected shutdown before those 15 seconds (like with mine as I was trying to drop the voltage used for IDA), TS runs in read-only mode and notifies the user "Hey you might have done something wrong--check your settings."

 

I didn't get any sleep last night because I was seriously considering your first feature and now you've given me an idea for a second feature that would also be quite useful. Give me some time and I'll see what I can do.

 

unclewebb,

I've just noticed a peculiur bug in 3.00. In 2.99, if I tried to open TS when it was already open (minimized in the taskbar), it would come up with a message saying it was already open. 3.00 doesn't do this...I just opened three seperate instances right in a row. I'd be you could open many more if you tried.

 

I disabled the blocking code when testing version 3.00 and left it disabled. At the time I thought that it might be causing problems with ThrottleStop starting up using the Task Scheduler but that wasn't the case.

I'll put it back the way it was if enough people complain but so far, you're the only one to mention it. I was hoping some users might consider this new "bug" a feature.

 

would that not cause problems if you change settings in one of the open TS and not in another ?

 

Hi unclewebb,

Congratulations on the TS program. Ive been using it for a few days now and like it very much, its now part of my regular tool box. Im interested firstly in maximising battery life, secondly maximising CPU efficiency. I have some questions, open to yourself or anyone knowledgeable on this.

On maximising battery, so far Ive been using TS to limit the multiplier (to lowest) and disable turbo. Doing this Ive found the lowest W draw (but a relatively high C0 state). Is just slowing it all down the best way to prolong battery life?

Even slowed down like this, my PC is still very responsive, is there are way to go even slower (mhz)? Or is this the hardware limit imposed by Intel? Or perhaps there is another way to reduce C0%?

Edit: Below is my original question, since posting Ive experienced different results. Starting Windows in different power states, then changing the power state, yields different FID results (including the one described below).

Im a little confused about my Windows power states, and how TS is reporting FID and C state %'s. My laptop has a Toshiba eco power state. When this is on, FID sits on 8 and C6% is 50-40. When in High Performance (like I know I should have it for full TS control), FID fluctuates from 10-15, C3% is around 40 and C6% 0-5. Are the TS figures correct in eco mode, or is TS not getting the full picture as Windows is changing things under the hood? (in both cases my FID is set lowest and turbo disabled).

In one a previous posts, there was a screencap of a C0% of 0.2. How were you able to achieve this?

Above figures from an i5-2410m. Thankyou again for the program!

 

Hey Folks...

I would like a tiny opinion here...

I've used RMClock on my older laptop (Core 2 Duo) to underclock it, manage its power usage and keep an eye on the CPU temps. I really liked that app.

Now I would try it again on my new Sandy Bridge laptop...
But I remember seeing this thread on the top of the list and I wonder if this TS one wouldn't be more appropriate somehow. What would you say?
Is it able to do everything RMClock does? (except for undervolting, I know that people aren't doing that on newer processors)
But to manage the power consumption and keep a track on the temperatures, it does a good job? I really used to like RMClocks icons on the system tray...

I also wonder if RMClock works on newer processors as good as it does on the older ones...

Regards!

EDIT - I got Windows 7 64-bits, just FYI

 

I apologize for causing sleep loss. I'll keep my latest suggestion to myself until you either complete or discard both of my previous ideas.

 

It definitely could. That's why I will probably go back to blocking it so you can only run one instance of ThrottleStop at a time.

ursoouindio: RM Clock has not been updated for several years so doesn't work with newer CPUs. There's not a lot you can do to change power consumption in the modern CPUs but ThrottleStop can show you the core temperature in the system tray if you like to see that.

coletrx: The minimum multiplier for a CPU is set at the factory by Intel and varies depending on the CPU model. ThrottleStop reads the minimum value from the CPU so the multiplier adjuster won't go any lower than that.

In theory a CPU running at 800 MHz should consume half as much power as the same CPU running at 1600 MHz but an Intel CPU entering various C sleep states doesn't follow that theory.

One reason I included battery power consumption monitoring in ThrottleStop is because I found that a CPU can be set to some vastly different settings and the change in power consumption was so small that it was impossible to measure a consistent difference. Here's an example with a 45nm Core 2 CPU.

http://forum.notebookreview.com/6369641-post5060.html

In the Core i design, individual cores can go to sleep when not needed. This reduces the core voltage to zero so I don't believe that micro managing the multiplier is going to show any significant power savings during normal use. If running a CPU slow reduces the amount of time that it spends in one of the deeper sleep states then you might not be accomplishing anything. A low multi might cause power consumption to increase.

The ThrottleStop C3%/C6% data that is reported refers to the entire CPU package. You can block the CPU package from entering these deeper sleep states but individual threads and cores might still be going into C3/C6.

The ThrottleStop data should all be accurate including when you are in Eco mode. Most of the data is coming directly from high performance monitoring timers within the CPU. To minimize battery power consumption it is best to play around with the various ThrottleStop settings, do some thorough testing and closely look at a log file of the battery consumption results. Sometimes the difference is so small that it is within the margin of error. A typical Windows 7 PC has over 500 threads running in the background being constantly scheduled on various cores. It's impossible to control this load so it can take a lot of testing before you can come to any scientific conclusion about what setting is best.

I know TS is accurate when you are in Eco mode but without doing some hands on testing, I don't know if Eco mode is the best way to minimize power consumption. That can depend on what apps you are running. I couldn't see any of your screen shots. I'm always interested in learning more so post away. A TS log of battery power consumption with different CPU settings while watching a movie or something might be interesting.

I think the ultra low 0.2 C0% was my desktop CPU that is set to use the C1 sleep state but nothing deeper than that. On mobile CPUs that slow down and turn parts of themselves off at idle, the C0% won't be as low because the slowed down CPU will have to spend more time in the C0 state to keep up with all of the Windows background tasks.

 

I have a thinkpad w520 with i7-2920XM, I tried throttlestop 3.01, but it does not provide me with any of the "extreme edition" extras (no right click menu options for unlocking max multiplier/VID). I even tried to set Extreme=1 in the ini to no avail. In addition, all the options in the TPL dialog is greyed out.

it looks like thinkpad BIOS disabled it...

my cpu-z info: CPU-Z Validator 3.1

let me know if I can provide more info

 

If the BIOS took control, there's nothing TS can do.

 

Welcome to Sandy Bridge. The guys writing the bios get to decide what Extreme CPU features you get to use.

Does ThrottleStop show a TRL and TPL button? If the Turbo Power Limit settings are grayed out it is because the lock bit has been set in the bios for the register that controls this.

The TRL button should open up the Turbo Ratio Limits window. What does that show and can anything be changed in here and does it make any difference when you do make a change? The Extreme=1 INI option and the right mouse menu options have been replaced by these buttons.

I don't own a 2920XM so I'm going to need some feedback from users to learn more about these new CPUs. It's very likely that neither the turbo ratios or turbo power settings are going to be adjustable for you.

Posting some screen shots of these windows and ThrottleStop will help me learn if this is a problem with ThrottleStop or a problem with this new technology that is not very enthusiast friendly if a manufacturer decides to lock things up.

 

screenshot is attached

I can indeed see TRL and TPL, TRL is usable: I can change turbo ratio limit, but it does not actually affect anything. when I run "TS Bench", no noticable changes, and I can see the speed the CPU is running at is still the old turbo limit

let me know if you need more info

 

In the Power Limits window, ThrottleStop is correctly reading the 55W TDP limit for your 2920XM.

http://ark.intel.com/Product.aspx?id=52237

By default, Intel sets the higher turbo power limit to TDP + 25% so that is being read correctly too. Lock Power Limit Register and Lock Current Limit Register are both checked which means these registers have been locked by the bios. That's why nothing is adjustable in that window. The only way to unlock these limits is if someone writes a modified bios for your laptop. The turbo time limits are also contained in this locked register. If this window was unlocked it would be simple to uncheck Enable Limit for both power limits and away you'd go. No such luck at the moment.

For the Turbo Ratio Limits, I've left that unlocked at the moment so users can do some testing but obviously there is likely a lock bit set for this too. The lock bit is read only which means if it is set, ThrottleStop can't change it and once again you would need a modified bios to unlock this.

If I ever find a solution for this liucougar, I'll post a ThrottleStop update and let you know. Thanks for your help.

If anyone wants to modify a bios, the Sandy Bridge Turbo Power Limits lock bit is in MSR 0x610 and it's bit[63] that needs to be cleared.

 

So regarding the multiple instances, what would be even nicer is if it checked to see if TS was already open, and if so, bring up the current TS's window.

Now I've gone through the testing and figured out that both cores seem to be satble at 1.025V. With that said, a few other things to bring up....

1) For some reason I can't get logging to work. I check the Log File option and nothing happens...no TrottleStopLog.txt file shows up anywhere that I can find. Tried exiting and re-entering the app and no luck. I've got it located in c:\ThrottleStop\ on a Windows 7 x64 machine.

2) I get much much worse battery performance with the "High Performance" profile Windows power profile and ThrottleStop. My un-scientific method...unplug power and wait until Windows starts reporting estimated run time. With TS turned off and using the "Balanced" Windows power profile, I get almost 4 hours. I change to "High Performance" and turn TS on and after a few minutes it updates to around 2 hours left. Switch TS off and go back to "Balanced" Windows power profile and it goes back to close to 4 hours. This is all while typing this post, doing work email, several browser windows open, etc (what I'd call typical usage for me).

In TS I've got Set Multiplier checked (9.5 for my C2D P9500), VID set to 1.0250, and Power Saver checked. Under Options I've PowerSaver C0% is set to 35.

Is this expected?

 

I failed to find any information on modifying MSR settings in thinkpad BIOS.

do you happen to know any such references? (or even a non-thinkpad BIOS, but a Phoenix brand UEFI BIOS)

 

Aegwyn11, Windows is atrocious at measuring remaining run-time. I'd use a program like BatteryBar (free) or some other program that applies statistical time estimation rather than just saying "If the conditions that exist at this nanosecond continue, the battery will last for this many hours." That's what Windows does.

 

Aegwyn11: If you have created a folder directly off of the C: directory, you probably don't have write access to that directory so ThrottleStop can't create a log file in it. Try right clicking on that and do a Properties and try adjusting the file permissions.

ThrottleStop can report battery power consumption by turning on battery monitoring and clicking on the battery button. What does that show at idle with different settings? I get between 13 W and 18 W depending on screen brightness but big CPU FID/VID changes don't change power consumption of my laptop any meaningful amount. The Windows Balanced setting you are using might be changing something else besides the CPU that also controls battery power consumption like screen brightness or maybe the GPU.

If you get better battery power consumption when not using ThrottleStop then don't use it.

 

ThrottleStop is also as susceptible to batteries reporting strange data as BatteryBar or any other program that purports to show "wattage." My battery reports data in mAh rather than mWh. I mean, I'm sorry, but using only 3.1w on my sig laptop while blasting music, screen full brightness, etc? No.

Lesson of the day: Batteries are weird.

 

To the best of my knowledge, Battery Bar and ThrottleStop are both getting their battery data from the same built in Windows function. On my laptop that is using an original Dell battery, the ThrottleStop power consumption and battery capacity data has always appeared to be accurate.

Do you have an original Dell battery or is it a made in China knock off from Ebay? I've bought some questionable batteries off of Ebay myself because these days, it's impossible to tell what is an original OEM battery and what contains some cells from some guys basement in China.

Can you post a screen shot of ThrottleStop at idle when it is reporting 3.1W? Can you also post what ThrottleStop shows for battery capacity? I haven't received much feedback about the battery monitoring code in ThrottleStop so it helps if I have some screen shots if it is showing data that is not accurate. Thanks.

ThrottleStop shows both battery discharge rate as well as the battery charging rate when plugged in. A positive number like 3.1W indicates a small trickle charge while -3.1W would indicate a small discharge which would obviously be wrong for your laptop.

 

So looking at the actual discharge rate, it seems like there is very little/no difference between TS/High Perf and no TS/Balanced. Also, BatteryBar's indicator doesn't seem to care. Now for the wierd part....the Windows indicator doesn't seem to change much when I'm at home...only at work. At home, BatteryBar and the Windows indicators read within a few minutes of each other. At work with no TS/Balanced, same thing. At work with TS/High Perf, Windows says around 45-60 minutes less than BatteryBar. Odd, but whatever.

Interestingly I'm still fighting blue screens. I ran LinX with max problem size, single threaded for 500 minutes on each core individually, then ran it dual threaded for 500 minutes on both cores without any errors or crashes at a VID of 1.025. I've since gotten several BSOD's just doing normal random stuff...increased the VID by a notch each time. I'm currently back up to 1.05. I need to do some testing with video playback to see if I'll get much savings there.

Also, regarding the logs, I went in and totally removed the read-only flag from the entire TS folder and logging instantly started working. Wonderful, thanks!

 

Battery Bar keeps track of your average work load and tries to predict battery run time based on that. ThrottleStop and Windows are more instantaneous based. If you are gaming and putting a heavy load on your CPU then it will quickly adjust and lower the predicted run time. When you stop gaming, power consumption will go down and both of these will quickly correct and increase your predicted run time. There's no perfect solution. No software can predict exactly what you plan to do during the next hour so battery run time prediction is mostly guessing.

I didn't see the big power savings with some of the low power modes so I don't bother using them. Some CPUs seem to drop the voltage too low when dropping down to some of the deeper sleep states which can cause instability. The CPU might be OK the day it leaves the factory but after a couple of years of use, they do degrade slightly and might not be 100% stable anymore at the ultra low voltages that they use.

Glad to hear that you got the log file stuff working. Some of the default Windows permissions are frustrating when you don't know that Windows set them to something that you weren't expecting.

 

Unplugged, music going, half-screen brightness.


Same deal as above picture.

BatteryBar and ThrottleStop are only telling us what the battery is telling them. Note the stated capacity for my battery (4923mWh). The developer of BatteryBar was the one who informed me of the whole mWh vs. mAh thing. That's the battery's doing and programs like BB/TS can only display the data, as weird as it may look. Further down in the BB thread, you can see my suggestions for dealing with batteries such as mine that report information in non-mWh fashion.

This is an original Dell battery that shipped with my replacement system.

I'll chip in on power-saving settings: I don't use them so much to save battery as to reduce heat when my computer's sitting idle not doing anything. Cuts my idle temps from 45-50C (Balanced) to 35-40C (Power Saver, undervolt w/TS). The power plan changes when I lock my computer--I manually change the TS profile.

That's also where I got my wattages for the Unleash Your XPS 16 guide (BatteryBar mAh x Voltage = Wattage).

 

Thanks for posting that. I guess different battery types return different data to the Windows function that Battery Bar and TS use. I can't remember seeing a flag in that function to detect the difference but I'll have a look.

 

I don't know that there is one. The developer of BatteryBar seemed to suggest the onus was on programmers to make use of whatever data the battery chose to provide.

 

Hey uncle... stupid question but... would Ts work on an Atom n450 processor? I remember you saying its for C2D and up... but would some features work?

Thanks

 

I believe I added Atom support a while ago but I don't have any Atom hardware to test on. The Atom and Core 2 Duo are very similar. The Atoms don't consume a lot of power and can't be overclocked with the multiplier so beyond monitoring your N450 with ThrottleStop, I'm not sure how much you will be able to do with it. Post a screen shot and let me know how it works.

 

Any news on the sandy bridge front? Specifically for the M17x r3?

 

Just wanted to post my thanks for the program unclewebb. Installed an X3380 in my system finally a worthy CPU upgrade and spent hours trying to find a program to replace RMclock as it doesn't support half multi's and hey presto I remembered your program.

Use it so that on startup the program kicks in automatically putting the voltage at 1.2V and 3.45ghz on performance mode with ability to go into battery mode at 1.1V and 2.2ghz Very happy. Although haven't yet pushed it only had it a day....

Btw is it possible to have it so that it will drop down to 6x multi and 1.1V when not under load. As currently I have to choose either one mode or the other.. I did this with RMclock with the old program. If possible would be amazing. Cheers!

 

King of Interns: Have you tried using the PowerSaver option? You can also fine tune that with the PowerSaver C0% value in the Options window to get a 6.0 multi at idle. I did some testing and didn't find any significant difference in heat or power consumption at idle because my T8100 spends 99% of its idle time in C3 where these settings have close to zero affect but a desktop CPU could be different.

http://forum.notebookreview.com/6369641-post5060.html

If you are saving power and your core temperature sensors are not stuck then you should be able to see a drop in the temperature or a decrease in power consumption when on battery power. Your laptop is a little different so maybe you will see some savings. You could also try enabling the C States option.

If you love RM Clock then you can always try my half multiplier registry fix for RMC.

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

technos: ThrottleStop can monitor Sandy Bridge but on most Sandy laptops, including the M17x-R3, all of the adjustments are locked by the bios so ThrottleStop is no longer useful for overclocking purposes.

 

Perfect Unclewebb!! Clicking the Powersaver makes it auto down to 6x multi and 1.1V on idle. When on load it auto returns to 1.2V and 9.5x multi. Just how I wanted it.

I don't think the clock speed or multi makes a difference to temps however 1.1V definitely shaves a few degrees off compared with idle 1.2V

It really is a very cool chip though with fans at lowest speed the chip runs between 30C-35C at idle! One sensor stuck at 40C but doesn't matter..

Thanks again you really are a legend for making this RMclock replacing program. It is much better in fact due to all the options it has and is being continually developed and improved upon. Keep up the excellent work!

 

I finally remembered my other idea that I'd said I would hold on to: command line method to change profile (or some other task-scheduler friendly method of doing so).

Basically, I'm envisioning ThrottleStop as a complete replacement for Windows/BIOS power management--that's what it has to be on a computer such as the 1640 where performance is often crippled by the BIOS for no good reason. Even cooler would be "Run as a service" but I feel like this is potentially heading off in a direction so focused on people like me (with fail BIOS power management) that it may not be worth the time.

 

More good ideas SovereignGFC but since ThrottleStop can no longer control Intel's new Sandy Bridge CPUs, further development is pointless. I have a few minor plans for TS when I have the time but nothing major.

Ask Dell to send you a replacement for your defective XPS 1640 since Dell never got around to properly fixing this model. Use the bosses email address which you can find on the internet.

 

The 1640 WAS a replacement for a m1530 which was an even bigger pile of fail. So I doubt I'm going to get anything else. And, to be honest, with the 130w adapter and ThrottleStop everything's under control--the stuff I'm referring to is mostly convenience moreso than anything.

As my guide in my signature suggests, the 1640 can be made livable, even for someone like me.

 

If we can remove the BIOS lockout bit, can we use throttlestop again for overclocking?

 

does throttlestop have expiration date ?

 

i have a question on your demo you explain that the c0% is the amount of time your cpu spend in C0 state how ever i have no values for the other states c3 - c6 what does it mean?

 

C3% and C6% is only reported if you have a Core i processor. The previous Core 2 generation is not supported so it shows --- for them.

Most Sandy Bridge CPUs have two TDP limits hard coded in the CPU so software can't change these values. It's possible that a moded bios and a 2920XM might be adjustable with ThrottleStop but without any hardware to play with, I can't continue development. Send me a M18x and a 2920XM and I'll see what I can do.

The very early versions of ThrottleStop did but none of the recent versions do. The only thing that might expire someday is the WinRing0 library that ThrottleStop uses. If Microsoft ever enforces the driver signing date, which expired long ago due to lack of support, ThrottleStop and many other utilities that use WinRing0 could stop working.