The ThrottleStop Guide

Reset XTU To default values then *EXIT* it completely.
Reboot.

Run throttlestop, downvolt your core and cache, exit throttlestop, then run XTU.

Does it still say the core and cache are both undervolted?

 

So my issue is not a problem with Throttlestop, but rather a query about how it manages to solves an odd problem I'm having. I recently experienced a liquid spill on my laptop. It hit the keyboard and drained out the bottom. The keyboard is dead, but I'm able to boot the laptop with a USB keyboard. However, when I put a sustained load on the CPU temps suddenly spike to 80C+ and the CPU throttles to 800MHz. Checked the fans with Speedfan and they were at 0 RPM. Speedfan was unable to turn them on for more than a couple of seconds.

However once I opened up throttlestop, set the the multiplier to 30T and checked the "Clock Modulation" option, the CPU ramped up to 3GHz and the fans kicked on. I was then able to play Skyrim for 45 mins uninterrupted, where before it would turn into a slideshow. Once clock modulation was turned on and the fans kicked in, temps stayed at around 50C.

Anybody know what's going on? As it stands, the only way the laptop is usable is if I manually run Throttlestop after every boot.

CPU is an i5-2520M (2.5 GHz base clock, 3.0/3.2 turbo) in a Dell Precision M6600.

 

pinging @unclewebb
I wonder if you damaged something with the liquid spill. Might have affected something.
Did you have this 800 mhz throttle problem before the liquid spill? If not, then clearly something besides the keyboard got damaged by it.

 

So Intel has just released their 8th generation Intel Core processors, with the i7-8650U featuring a base clock of merely 1.9GHz, but a boost clock of 4.2GHz

But here's the thing: Intel has now doubled the number of cores on their ULV chips (Now 4 Cores/8 Threads) but kept the TDP the same. What does this mean? More throttling!!! Yaaaayyyy!!!!

I won't be buying a new laptop for many years to come, but it is interesting as to how power will be managed.

I know, I know, clock speed barely matters at all for CPUs nowadays, but the difference between 1.9GHz and 4.2GHz (My skylake CPU can go all the way to 0.4GHz) is huge.

I wonder how clocks will be managed.

For me, the difference between base clock speed (2.5GHz) and turbo speed (3.1GHz) is negligible. My CPU either throttles waaay down to 1.8-2.2GHz, or goes full speed at 3.0GHz all cores loaded, depending on whether the iGPU is loaded.

Maybe the newer CPUs will have some very quick (millisecond?) fluctuations in clock speed, which might wreak havoc on undervolts.

 

That was my original conspiracy theory but it seems that Microsoft is blocking any program that tries to load WinRing0 directly.

@0meg4 - What sort of testing are you doing to determine that the temperatures have not changed when using ThrottleStop compared to when using XTU. Make sure to completely exit XTU when testing ThrottleStop and exit ThrottleStop when testing XTU. Check the Task Manager to make sure XTU is not running at all. Both programs are writing information to the same registers in your CPU so best if these programs are not fighting each other for control.

For testing I would do something simple like run the TS Bench 1024M. While the benchmark is running, switch ThrottleStop profiles back and forth between a profile with no under volt to a profile with your under volt applied. The monitoring table in the top right corner of the FIVR window should show your changes in voltage. Look for the same offset voltage for core and cache. Post some screenshots of the FIVR window for each test as well as the main screen of ThrottleStop. With a consistent test like the TS Bench, the reported power consumption should also show a change if voltage control is working.

That's not so bad is it? At least your laptop is still usable. Just add ThrottleStop to the Windows start up sequence using the Task Scheduler and problem solved. Here's a guide how to do that and there is also a link to a guide in the first post of this thread.

http://forum.notebookreview.com/threads/the-throttlestop-guide.531329/page-514

I have no idea what but it seems like something has shorted out in your laptop. If you ever dump a drink in a laptop, do not turn it on or turn it off immediately. Pull it apart and dry it out before trying to turn it on. A lot of people turn it on right away just to see if it still works but that is usually a bad idea.

 

Yes, it stays undervolted. Recognizes the undervolting done by TS.
This is strange:
Setting to -140 with TS, running BF4 on ultra: 90°-97°
Closing TS, opening XTU, same -140. Runs BF4, same map, same server, same everything --> 75°-89°
Is this possible?
This is how my TS interface looks:



Maybe that 139.6 is not the same as 140 on XTU? Does that 0.4 really matters?

 

Did that, and the undervolting is real.
I don't know what is wrong with bf4 then, that shows those 97°peaks

Here are my results using your suggestion:

 

Your undervolting is definitely working.
Power consumption lower, vcore lower.
However in both of your screenshots, you clearly are overheating and triggered #Prochot.

Have you compared the actual in game framerates in BF4, when doing the undervolt in TS, vs the undervolt in XTU?
The only way to do this properly is to go on an empty single player or custom map and find a place where you can stand at the same spot and get a consistent repeatable FPS. Then test that same spot with XTU after testing it with throttlestop. If TS undervolt is giving you higher FPS than XTU undervolt, then it means XTU is throttling your CPU for some reason.

One thing to remember is your CPU has an absolute 45W power limit. You also need to tackle those temps. a 7700HQ should not be reaching prochot. Check the fan speeds, or also consider a repaste with Grizzly Kryonaut or Phobya Nanogrease Extreme to deal with those thermals.

 

Well, it's not a big deal if I want to run Windows, but Linux is also glacial and I don't know of a comparable tool that runs under the other OS.

I found the laptop shut down with a small pool of water underneath it. I have no idea if it shut itself off after it got flooded, or if it was already off. My cat probably knocked the glass of water over onto the keyboard. In any event, I left it off for a couple of days before trying to power it up.

 

Yeah, did it, and XTU always showed a little lower temps.
I already orderer Grizzly Kyronaut. Will arrive in a few weeks and gonna do repaste.

Two things i like to ask, that i don't really know.

1) Is Prime95 the best benchmarking tool? Because no matter what i do, prime95 always pushes the CPU 95°+. When i used TS Stress Test, it did show "100% CPU usage" but the temps were lower than Prime95 using 100% aswell.

2) While stressing, the temps always stayed somewhat "stable" between 80-85.. and suddenly it pumps 10° to 95-97 for a second, then come back to the usual 80's.
Is this normal? Is this the expected undervolting result?

 

Prime95 is unrealistic load. If you game and do not throttle I would not bother.

 

Prime95 is not for benchmarking. It's only for absolute worst condition stress testing.
Prime95 will simulate the absolute highest, worst case load you will ever have.

If you run prime with AVX and FMA disabled (read the undoc.txt file for instructions on how to do that), then Prime load will equal the most heavy load any game or application would ever put on your system.
Running Prime with AVX/FMA3 enabled is nothing but a power virus, putting you on loads NO real application besides AVX number crunching apps would ever do, and is only good for testing absolute stability limits of your system (if AVX/FMA3 is stable, nothing will crash the CPU, except strange undervolting cases where you crash going from load to idle). Even for games that DO use AVX, NONE of them will load 8 threads to 100%--ever.

Anyway I think your absolute first priority is to do what I said and buy some Grizzly Kryonaut or Phobya Nanogrease Extreme (I think the Phobya is only available easily from newegg, or ebay. There is a shop on Amazon that is a third party, but pretty sure that shipping is going to be slow, but it's the same shop on newegg). You should not be reaching Prochot# at all on TS Bench, under any conditions.

 

Thanks for so much info.
Yeah, Grizzly is on its way.
My -140 undervolt successfully done 10hrs of prime95 with no issue, so i think I'll stick to that.
Going to repaste as soon as it gets here, and will try everything again.

About Prochot#, i'm not really sure of what it is, and how it affects or is being triggered is related to a succesfull undervolting.

About the high sudden temp peaks, is that even normal?

 

How is your -140mv doing at low multipliers?

 

Just a couple of things. When your CPU is idle at the desktop, what sort of C0% is being reported? An idle 4 Core / 8 Thread CPU should be down around 0.5%. If you are a lot higher than this when idle, it will be difficult to do a fair temperature comparison.



ThrottleStop shows that you have Speed Shift enabled but you did not check the Speed Shift - EPP box on the main screen. For consistent testing and maximum performance, I would check that box and set EPP to 0. Was Speed Shift enabled when you were testing with XTU?

I agree. You can use the CLR button in ThrottleStop to clear the maximum temperature between tests and HWMonitor has a similar option in its menus. When the TS Bench is running, your current core temperature reported by TS dropped from 96°C to 85°C. Less power consumption and less heat confirms that under volting is working and a difference of 10°C or more is a great result. Not sure if you noticed but your CPU is running cooler as well as running 100 MHz faster. That is a win win situation.

Check the XTU graph for a consistent CPU speed while gaming. A drop in temperatures has me thinking that your laptop might be throttling the CPU MHz when you are only using XTU. For many games, a slower CPU is not necessarily a bad thing. Running a slower CPU can significantly reduce temps, especially in a laptop, without negatively hurting gaming performance. If that is what is happening, you can use TS to run your CPU at whatever speed you like. Fast for games that need a fast CPU while slower for games that do not.

PROCHOT stands for processor hot. As soon as your CPU reports that it is hot, it will start throttling and slowing down. This will change the MHz hundreds of times a second to keep your CPU from having a nuclear meltdown. Meaningful testing while your CPU is throttling and bouncing off of its thermal limits is impossible. Hopefully your repaste goes well. Another thing I noticed is a temperature difference from one end of your CPU to the other. This can be a sign of a heatsink not sitting square against the CPU cores so heat transfer is not optimum. Have a good look at how your heatsink sits when you have your laptop apart for service. Some people have been known to load the CPU while adjusting the hold down screws to try and equalize contact and improve cooling while others reach for the sand paper to square things up. You are on your own at this point.

The TS Bench test has always been a lighter duty stress test compared to Prime95. Both programs can load the CPU 100% but Prime95 is working parts of the CPU that create a lot more heat and use more power compared to ThrottleStop. The next version of the TS Bench test should be a little more demanding like Prime95 is without being a full blown torture test the way Prime95 is.

 

I guess it was and i didn't notice.
I'll do that in my next test and see what happens.

Yes! I'm really happy with the results, already 8hrs prime on -140. No issues. 80-92°

I agree with that statement. When i first started to worry about temps, the first thing i did was to disable turboboost, and running at 2.8MHz on BF4 wasn't that bad (90-110fps on ultra). So, didn't make a huge difference, while being significantly cooler.

I don't think I'll need that full top speed on my main games (BF4, Dota 2 [100+ fps on ultra]), so I'll keep the turboboost off option for now, until i repaste and get to see the heatsink.

This "Prochot 97°" that TS shows, is manufacturer related? I mean, who did set the 97 limit? Acer?
When you say "from one end of your CPU to the other", you mean from Core 0 to Core 3? Yeah, it was a little strange to me too, but i just tought thats the way is meant to be.
Do you have any input on why is it suddenly bouncing 10-12° at once? or is it normal?
I mean, i'm new to undervolting and don't know when people share their results they say "now is 80°", they mean as an average or as max temp. Because the temps were mainly on the 80's range, that 96° peak really freaks me out.

Sanding a heatsink i think is too much for me, tho. I think i'll pass on that one, haha.

Thanks so much for the huge help.

 

Yes. The Intel spec is 100°C but they left individual manufacturers with the option to set an offset in the bios. It looks like Acer is using an offset of 3°C which is reasonable. Some laptops use an offset of 15°C which is borderline criminal.

As far as I know, the 4 cores on these CPUs are in a line. It is possible that there is a heat source beside core 0 and this is adding extra heat to core 0 and to a lesser extent, core 1, core 2 and core 3 as the heat dissipates. If a heatsink is not perfectly flat, it is also possible that it is making better contact with one end of the CPU compared to the other end. When the heatsink was installed, it is possible that one side of the heatsink got snugged down first before the other side so the thermal paste and heat transfer might not be even. The temperature sensors that Intel uses are not 100% accurate so some reported temperature variation might be completely normal. It is always an eye opener when you remove a heatsink, clean it up and then lay a straight edge ruler on it from side to side and from corner to corner. Hold it up to a window and have a look for light bleeding in. If it is concave or convex, that can be a problem.

The bottom line is your heatsink is barely adequate. This is the norm. Barely adequate is the lower bar that most manufacturers shoot for. It is a competitive business. If the engineers recommend putting an extra dollar of cost into a heatsink, the accountants have a fit.

Intel says that 100°C is the "maximum safe operating temperature". Don't let peak core temperatures over 90°C scare you too much. The most important thing is keeping the temperature below the throttling temperature which for your laptop is 97°C. As soon as any core hits that temp, the performance of all cores will start to decrease. Keep it below 97°C for consistent and maximum performance.

It is normal for core temperatures to change rapidly. Most monitoring software is only sampling your CPU once every second and a lot can change in a one second interval. A huge amount of energy is flowing through something that is very tiny so when the heatsink is a piece of crap, temperatures will spike instantly.

 

@unclewebb any ideas what all of these mean? CPU is fine at 4.9ghz then drops to 4.2 (Clevo 1080 sli, gpu+cpu load). If run single card the CPU will stay at 4.9ghz.

 

Are you using the prema Bios?

 

@Mobius 1 - Red means throttling is in progress. Yellow is a record of previous throttling since you booted up or last cleared this data from the CPU. The important one in red is VR THERMAL which means your CPU is throttling because the voltage regulator is reporting that it is too hot. I believe it is separate from the CPU on the motherboard, usually near the CPU socket. THERMAL in yellow means your CPU reached the thermal throttling temperature limit. PL1 and PL2 are your turbo power limits and EDP OTHER means your CPU reached its Electrical Design Point which is usually the current limit. Some of these throttle reasons will trigger and will be shown in yellow just after you boot up or resume from stand by. It is a good idea to click on the headings in Limit Reasons to clear this information from the CPU before you start doing any testing.

 

This seems more like an EC problem if none of these flags happen with single card but do happen with SLI. SLI doesn't put any extra load on the CPU VR regulator. The EC can override any Bios or CPU Register it wants to do (PECI?), just to force throttling.

 

Yes.

But strangely the voltage regulator don't overheat on single card mode?

 

OK. Finally getting the 'Could not open WingSomething' on Throttlestop after build 16281

I opened OHM and throttlestop would open after that,

Oh well I'll just have to add OHM to the list of my startup programs. I need TS that badly.

 

Time to kill a day or two or three and try to find out what is going on with ThrottleStop and WinRing0.

Edit - I found a trick. I cobbled together some code so double clicking on the ThrottleStop icon will first start Open Hardware Monitor quietly in the background, ThrottleStop will then be able to open up the WinRing0 library and after that, TS will tell Open Hardware Monitor to close and most users will barely notice what quietly just happened.

Don't tell Microsoft! Maybe this trick will keep TS going for another few months.

 

Awesome! Do you have plans of releasing this code anytime soon?

Thanks for all your help!

 

Nope. It will probably be a few weeks or a month before TS is ready for the next release. At the moment, the code for TS is a mess. A few months ago I started moving some things around and adding some new features and then I decided to take a break. My memory is terrible so it will be a while before I remember what I was working on and get back up to speed. I also plan to switch TS over to a much newer version of Visual Studio. I would like to get that sorted out before the next release.

 

Do you think there will be a new release by the time the current one expires?

 

Spoiler

Copy both TS & OHWM (openhardwaremonitor-v0.8.0-beta sub-folder) to the root of C drive, then copy task.bat to C:\ThrottleStop

Code:

@echo off
start /min "" "C:\OpenHardwareMonitor\OpenHardwareMonitor.exe"
timeout /T 2
start "" "C:\ThrottleStop\ThrottleStop.exe"
timeout /T 2
taskkill /F /IM OpenHardwareMonitor.exe
exit

Task scheduler -> Action -> cmd.exe -> Arguments -> /c start /min C:\ThrottleStop\Task.bat ^& exit.

....or just import Task (ThrottleStop.xml).

 

I sure hope so. Hopefully sooner. Too many complaints when I am late with a new version.

@j95 - I love simple solutions.

 

Thanks, but this isn't guaranteed to work if you're dealing with a laptop with a mechanical HDD that takes ages to boot. This is more reliable:

Code:

@echo off

taskkill /F /IM ThrottleStop.exe
start /min "" "C:\Storage\Software\OpenHardwareMonitor\OpenHardwareMonitor.exe"

:loop
timeout /t 2 /nobreak > nul
driverquery | find "WinRing0" >nul
if %errorlevel% neq 0 (
    goto loop
)

start "" "C:\Storage\Software\ThrottleStop\ThrottleStop.exe"
timeout /t 10 /nobreak > nul
taskkill /F /IM OpenHardwareMonitor.exe

exit

I scheduled the task to only run when the user is logged in, and triggered to run on user login trigger and on event:
Log: System
Source: Power-Troubleshooter
Event ID: 1

That will trigger on sleep/wake cycles. To be safe, the script deliberately kills ThrottleStop before relaunching to avoid multiple instances, but you can do a check at the beginning of the script just to see if ThrottleStop is an active process and skip the whole thing if you want...

 

Hi, I have some problems with wrong Turbo Ratio Limit settings after waking up from sleep. I have 7700HQ and 2 profiles:
- profile 1: max 4 cores turbo ratio is 28
- profile 2: max 4 cores turbo ratio is 34
After i put my laptop to sleep and then waking up, despite ThrottleStop displaying current profile was 2, it used the turbo ratio settings of 1 (i.e. max 28 ratio for 4 cores). I had to manually switch to profile 1 then back to 2, then the turbo ratio was back to be correct.

P/S: I already used a script to kill and restart TS after waking up from sleep.

 

Probably your computer Bios doing that. Throttlestop can't do anything if your Bios changes settings by itself. Sounds like the Bios turns off turbo boost if the computer goes into sleep. Or maybe it refreshes and read its internal settings. Have you tried changing it in the Bios to check that?

For example if you used XTU and set the max 4 cores to 28, with TS still open, TS won't magically change it back to 34.

 

Check profile settings on AC and battery in options. You must have set it to 1 for AC and battery which is the default value.

 

No profile 2 is set for AC. Profile 1 is my failsafe settings when cpu reaches 90c (automatically switch to profile 1).

 

Now the turbo profile is working correctly, seems like only at random waking up from sleep will make 4 cores turbo to be locked at 28

 

@kakashidinho - When you are having problems, it is always a good idea to post a few screenshots. You might not have something set up correctly. The more screenshots the better. So much easier than me having to ask, how is this set, and that and that and that and........

 

Here are my profile settings:

profile 1:


profile 2:


profile 4:


TPL:


Options:

 

Change clamp value to 45 and next value to 56 with Turbo time to 16 and uncheck Clamp option. Secondly, increase the primary power plane limit from 0.010 to 28. Enable Intel Power balance set CPU value to 31 and GPU to 9 or 13.

 

@kakashidinho - The Turbo Ratio Limits for Profile #1 and Profile #2 are both set to the 38 multiplier when 1 core is active and to the 36 multiplier when 2 cores are active. What sort of testing are you doing to show that this is not working correctly? If your turbo profiles are the same then for most loads you might not see any significant difference in the reported multiplier. A lot of tasks only use 1 or 2 cores simultaneously. Your Profile #4 with the very high EPP setting is probably going to interfere with your maximum multiplier. Thanks for your screenshots. I will do some testing but first you need to explain to me and post some pictures of what you are seeing that you think is a problem. It is possible that ThrottleStop is not setting the turbo multipliers correctly after you resume. I mostly use a single profile when using ThrottleStop so I might not have thoroughly tested this possibility. I always like chasing down bugs so a few more details will help me out.

 

I used Prime95 to test all 4 cores max speed. Actually my laptop was suffering from overheating when stress testing all 4 cores (rendering is one usecase). that why I need fail safe profile 1 which limits the 4 cores turbo to 28 (base multiplier), also that's why I reduced max Turbo power limit to 40 to reduce heat. For the multiplier after waking up, currently seems like it works correctly so far. I observed that after waking up from sleep, if I immediately run prime95, the multiplier will sometimes initially be 28 for a few minutes, maybe shorter, then revert back to 34 correctly.

One more question, does PROCHOT97 mean CPU will downclock if temp reaches 97c?

 

@kakashidinho - PROCHOT stands for processor hot and if your laptop is set to 97°C, it will start slowing down to prevent the CPU from exceeding that temperature. It might go a degree or two beyond 97°C but that is still less than 100°C which Intel says is the maximum safe operating temperature so no worries.

I did some testing and on my laptop, the turbo multipliers in TS seemed to be working correctly after resuming from sleep or when switching profiles. For your laptop, I would leave the turbo multipliers at their default settings. If the cooling system is not adequate, set the turbo power limit to 40 Watts, enable the Clamp option and let the CPU manage itself. It will automatically reduce the multiplier so power consumption and heat do not become a problem.

Prime95 is not a great laptop testing tool. It consumes far more power than most laptops are designed for so you usually get immediate power limit (PL2) or (PL1) throttling, or EDP throttling or temperature throttling. I like Prime95 but on a laptop, maybe only run 2 or 4 threads of it and not 8. Watch Limit Reasons when testing to see what is causing any throttling.

 

Wow I was researching why I had a BD Prochot flag on my MSI GT62VR 6RE and finally figured out. If I unplug/plug charger, the BD Prochot is flagged. Very weird.

 

my PROCHOT is set to 100c. When I am encoding with Adobe Premiere, gpu and cpu work together, and my cpu goes to 100c, even 101c, and Turbo Boost doesn't want to come down.

6700hq Max turbo is 3.1ghz on 4 cores, and hwinfo64 says there's thermal throttling, but it only drops to 3ghz, and it keeps staying between 99-101c.
One time, my laptop auto-shutdown while i was encoding, as it didn't throttle enough.

How do I limit Turbo Boost multiplier? Can throttlestop do that?

I want to limit to 28x or 29x

 

@plee82 - Some laptops use BD PROCHOT which forces the CPU to run like an 800 MHz slug when on battery power. I would feel ripped off if my shiny new laptop was slower than a 10 year old laptop that was sitting in the closet collecting dust.

@link626 - Intel designs their CPUs so they use as much turbo boost as possible, all of the time. I like that feature. Is there anyway you can improve your cooling? That is the real problem. If not then click on the FIVR button and try lowering the Turbo Ratio Limits a few notches. If your turbo power limits are unlocked, reducing the PL1 limit should prevent your CPU from running continuously at its full 100°C rated temperature.

 

It goes BD Prochot for like 0.5sec when I unplugged and goes back up to full performance.

 

in FIVR, reset turbo button does not work.
I have to manually reset each multiplier.

Is this a bug?

 

I've seen some cases where changing the 4 multipliers to the same value and hitting "apply" doesn't set the full speed bin, but instead sets it lower than it should be. This usually happens when going from a higher multiplier to a lower one then trying to go back to a higher one. I think some limit is being set in a MSR that is overwriting Throttlestop attempt.

To avoid this problem, uncheck "overclock" tab, change the multipliers, re-check overclock tab again, then hit apply. This has always worked.

 

Yes.

I did not know how to read the default multipliers from the CPU when I originally wrote the FIVR code. I am still not 100% sure. I do not have access to any relevant documentation from Intel. This is why the Reset Turbo feature was never finished. This is not really an end of the world kind of bug but it is a bug and I will try to remember to get this fixed up for the next release.

@plee82 - Dell used to use Clock Modulation throttling on many of their laptops when a user switched from AC to DC. Pulling the plug while running Prime95 would be pretty harsh on a battery so some built in throttling during this transition is probably a good thing. Temporary use of BD PROCHOT to slow the CPU down is OK but I think some manufacturers "forget" to disable BD PROCHOT so the CPU is left stuck at 800 MHz indefinitely. That's not good but it can be easily fixed by using ThrottleStop.

 

http://forum.notebookreview.com/threads/the-throttlestop-guide.531329/page-353#post-10086507

http://forum.notebookreview.com/threads/the-throttlestop-guide.531329/page-353#post-10086826

Hi, I still have this bug. Not every time but randomly, when i unplugged my computer, I see the icon in the icon tray sticking to "1" (when it's supposed to switch to 2), and when I get nearby it, it disappears. Then I can start again ThrottleStop and it goes to my profile 2. So it's exactly the bug the guy was having 2 years ago.

I have a Dell XPS 13 with these specs : i5 6200U, 8GB, 256GB, Full HD. I'm running Windows 10.
Here's my ThrottleStop config : https://pastebin.com/iVMnSr57

+ Screenshots :
http://prntscr.com/gkep5j
http://prntscr.com/gkepci

By the way, I've discovered this program only a few weeks ago and I'm loving it. It's an amazing job, so thanks a lot.

 

@Tomatot - I think I found the main source of ThrottleStop's icon issues. I have been programming with a very old version of Visual C++ from the 1990s so I have no one to blame but myself. I should have updated years ago. The icons being created are not fully compatible with Windows 7, 8 or 10. I am in the process of switching to a much newer version of Visual Studio so these long time icon issues should finally go away.

In the mean time, try cleaning the WIndows Icon Cache as explained here.

http://www.digitalcitizen.life/how-clean-notification-area-icon-cache

Download and run the .bat file from that site. This is mentioned in the ReadMe file that is in the ThrottleStop folder. I am using Windows 10 and my icon and profile are switching back and forth between 1 and 2 when I switch from AC to battery power without any issues.

Almost fall here. Once I get all the yard work done and watch some playoff baseball, I will have more time for TS.

Thanks for the words of encouragement. That will help get me back working on project TS.