it apparently can work for desktops someone posted earlier that it helped his desktop core i7 on page 3
-
-
idk look back on page three and pm the guy to see what his is.
-
I thought people already knew about this. thanks for the share though, I never noticed that much of a difference but any performance gain is a good gain in my books.
-
It's not so much the performance increase (which is minor) it's in the snappier response. Sometimes this mobile 720QM processor was performing like a 486 in terms of visual latency and now it's snappy and much more instant-feeling..
It was initially posted on the Alienware forum and got moved to a general Win7 spot later, I didn't realise at the time that it affected more people and am surprised that it isn't better known (I found it by chance).
Still, my computer feels like new since I did this -
That you have an alternative solution is better is fine, but try to stop acting so damn elitist next time and just post explaining what is going on without all the hyperbole.
If you knew all about this wonderful fix you should have spoken up instead of letting someone else discover the approach from a different perspective. -
I'll agree with Newsposter here. If there are command line options, or better yet GUI, you should go there way before regedit.
You can never assume the competence level of the user reading your posts. What may seem fool proof to you will get someone else into trouble. Also do not assume just because this is an enthusiast site there are not inexperienced users reading the posts. -
In my opinion, there's no extra risk by using Regedit than using the command line. In this case the regedit is much simpler, both to set up and to revert.
The regedit approach is better for inexperienced users in my personal opinion. Command line options in this case are more confusing. Plus the registry editor is much easier to use than cli I think. -
-
Command line utils, including all of the new ones that MS provides in Win7 and Server 2008, do consistency checks BEFORE applying changes to the registry.
It's only confusing to those who are deliberately ignoring the long-established dangers of manually editing the registry.
Again, the key to being successful in making config changes that are not already set up via the control panel GUIs, is to read and understand the damned documentation. Which is freely available.
Yah, it takes more EFFORT to understand a command line util than it does to blindly cut and paste from someone elses web posting. But at the end of the day, the user will UNDERSTAND what they are doing and be able to make their own decisions based on that knowledge.
Don't confuse the effort needed to learn and understand new technology (it's called education!) with the so-called 'confusion' caused by a lack of exposure to that new technology. Knowledge can be gained by reading and applying ones self. We can all do it.
If a user is unable to understand (confusing??) the documentation, they shouldn't even consider making system changes they don't understand. -
-
HOLY.........My mouse feels like its on steroids now... +rep.
PS: my value max was 64, did anyone else have the same as me? -
My software scans the user system, detects which graphic card you have, and which driver you have, and applies the correct tweaks to your system based on the settings you set. And I don't say this as B.S. My software is free despite me testing every option on about all Nvidia GPU architecture at every driver release. And I can tell you, that it happened where my tool tweak made my test computer BSOD at startup because the wrong tweak was done (ie: new driver, different tweak) -
Yeah im aware of you not being a noob or anything, but messing with drivers, especially nvidia is a complete other thing, this tweak is so extremely simple that i dont even get why we are having this talk
-
it's not the tweak, it's the **method** that some are advocating that implements the tweak.
-
@Newsposter.. Nobody on this thread has had a problem with it except you. Are you arguing for intellectual superiority or for what is best for the noob?
I'm not a noob, but the method worked fine for me. If I want to set it back, just as quick and easy. Who cares whether you use the correct utility or change a value in the registry?
Nobody has had a problem and I doubt anyone will, so what are you bothered about?
There is nothing wrong with the method at all. The program (REGEDIT) was designed to access the registry. It's the same thing that the CLI tool does.. I think you have to ask yourself what the problem is.
You've expressed that you think there is another method. Thanks for that. You keep on with your elitist comments, as if we are advocating some kind of incorrect method. I'm not sure but I think it's the most anal commentary I've seen for a while! Thanks though, it kinda makes me laugh -
ok ok ok let's sedle this.
You are both right and both wrong.
newsposter is taking the safe approach, in insuring the best experience for all while performing the tweak.
Mr Pras, is trying to the helpful, which he is, but luck kinda plays inside his suggestion, in the sense that no one had or reported to have a problem after applying the tweak.
During the week-end, I'll look into making a utility that will scan the user system ensure that he has a Core i7 and offer the user to tweak and uintweak with a press of a button.
I invite Mr Pras to PM me the maximum information on the issue. And I want to know if it affects Core i5 as well, or if it's exclusive to i7, or even i7 for laptops only. Mr Pras and the person where you found the tweak will be credited in the program, of course. However, I dont' know how to check BIOS feature status information. -
Not to change the subject and get everyone back on track, but my question concerns possible repercussions of this fix. Worst case scenario, what happens? Increased temps?
-
ahsan.mughal Notebook Evangelist
-
-
-
- High Performance mode, when On Battery
- High Performance mode, when Plugged In
- Balanced mode, when Plugged In only
- PowerCFG -SetDCValueIndex 8c5e7fda-e8bf-4a96-9a85-a6e23a8c635c 54533251-82be-4824-96c1-47b60b740d00 0cc5b647-c1df-4637-891a-dec35c318583 100
- PowerCFG -SetACValueIndex 8c5e7fda-e8bf-4a96-9a85-a6e23a8c635c 54533251-82be-4824-96c1-47b60b740d00 0cc5b647-c1df-4637-891a-dec35c318583 100
- PowerCFG -SetACValueIndex 381b4222-f694-41f0-9685-ff5bb260df2e 54533251-82be-4824-96c1-47b60b740d00 0cc5b647-c1df-4637-891a-dec35c318583 100
And then reboot. The first GUID is the power plan (The ones I've shown are the default built-in power plans of Windows 7. You can run PowerCFG -List to see which plans you have on your system), the second GUID specifies the Processor Settings, and the last GUID (the one on the first page of this post) specifies the "Processor performance core parking min cores" setting. In english this means "The minimum percentage of active cores". The last value, 100, specifies that 100% of cores should always be active. (By default on Windows 7 it's set to 10%, if you're curious.)
P.S.
The M15x rocksAttached Files:
-
Thanks for the info!!
-
"The value specified is mal-formed, or is not within the range of the target power setting."
any ideas? -
I'd try using the method described in the first post. It's simple and it works.
-
wow, this tweak really works.
i notice my startup is faster loading up.
great tip!!!! -
Can I apply the fix when using Core i5? It's DualCore, but comes with HT technology, so is technically 4 thread...
-
when i first did the registry tweak i had 3 cases where i had to change the valuemax to 0. in between each of these i came across the ACSettingIndex and DCSettingIndex for that particular powerplan. the values for these show 0x0000000a(10).
this seems to be what is being changed above as mentioned by 'Identifier'.
As i said before though i get an error trying to run PowerCFG from cmd prompt. shall i just leave it as it is? -
-
Hmmm...if anyone's curious, this is what MSFT documentation says (I think this is from the link news posted):
Processor Performance Core Parking Maximum and Minimum Cores
Core parking is a new feature in Windows Server 2008 R2. The processor power management (PPM) engine and the scheduler work together to dynamically adjust the number of cores that are running threads. The PPM engine chooses a minimum number of cores on which threads will be scheduled. Cores that are chosen to be parked do not have any threads scheduled on them and they can drop into a lower power state. The remaining set of unparked cores are responsible for the entirety of the workload (with the exception of affinitized work or directed interrupts).
Core parking can increase power efficiency during lower usage periods on the server because parked cores can drop into a low-power state.
For most servers, the default core-parking behavior provides the optimum balance of throughput and power efficiency. If your server has specific core-parking requirements, you can control the number of cores available to park by using either the Processor Performance Core Parking Maximum Cores parameter or the Processor Performance Core Parking Minimum Cores parameter in Windows Server 2008 R2.
The values for these parameters are percentages in the range 0100. The Maximum Cores parameter controls the maximum percentage of cores that can be unparked at any time, while the Minimum Cores parameter controls the minimum percentage of cores that can be unparked. To turn off core parking, set the minimum cores parked to 100 percent by using the following commands:
Code:Powercfg -setacvalueindex scheme_current sub_processor bc5038f7-23e0-4960-96da-33abaf5935ec 100 Powercfg -setactive scheme_current
Code:Powercfg -setacvalueindex scheme_current sub_processor bc5038f7-23e0-4960-96da-33abaf5935ec 50 Powercfg -setactive scheme_current
Now, on my i7-720QM, only the HT cores are parked. I may try this, anyways. -
ScuderiaConchiglia NBR Vaio Team Curmudgeon
Gary -
NotEnoughMinerals Notebook Deity
Hmmm what do you think uses more power/creates more heat:
Cores that are regularly parking and unparking?
or
Cores that never park? -
cores that never park use more power and generate more heat
when they are parking they are saving energy but at the expense of some responsiveness since the cores have to "wake up" each time they park and this slows things down just a little bit, but at an essential moment that can cause noticeable lag. -
NotEnoughMinerals Notebook Deity
Well does the wake up phase use much power?
I'm thinking along the lines of the leaving a lightbulb on for a minute vs flickering it on and off every 5 seconds -
Thanks -
I think 45 and 35nm chips don't need so much to get going again (from off to on) when compared with something like a light bulb for example.
It might take some extra power to wake up from core sleep but it's probably negligable or at least is offset by the power savings from turning the cores off.
Don't take my word for it though, I'm not a microprocessor architect! -
So this works only for i7? What about i5-430m?
-
-
I think yes it does help. If your mobile chip is having it's cores parked then this should make a difference.
I don't have an i5 but someone else in this thread reported it was working. -
NotEnoughMinerals Notebook Deity
+rep for the solid find -
Hey sweet find man but im worry about burning up my cpus or motherboard lol we should get some benchs or temps on and off with intense gaming please my m17x coming in a week
-
no it shouldn't cause any kind of problems only a slight increase in temp and battery use. nothing to really worry about its a great adjustment to use
-
I for one could care less about if he used cut/paste for the information as from the very beginning he did not try to pass it off as his own. It was helpful, informative, & he gave props to where he found the information. There are many people in this thread that appreciate his findings & I am one of them. My system feels much more responsive now!
+Rep for you Mr Pras, keep the positive posts coming. -
Hi everyone, Someone PMd me who couldn't find the registry value, so I suggested he search for
"Specify the minimum number of unparked cores/packages allowed (in percentage)." (WITHOUT QUOTES)
Which also works! -
nice glad to see there is another way for this.
-
did this to my core 2 duo system cant really tell difference but hey was worth a try
-
Do Core2 chips do core parking? I don't think they do, but not sure.
Also I think the sluggishness is caused by so many threads (8 in my case) combined with the core parking and the shifting around of the threads by win7.. So with the Core2 dual cores it's unlikely they ever actually park anyway. -
So far so good. Thanks Mr. Pras ! Helped my stuttering in Fallout 3 immensely.
-
Have any of you tried setting the max to say 25% or 50% and see how that works? My thought is your dual threaded games will not have their cores park on them because the value limits how many can. But you still might get the power saving core parking has to offer?
Just a thought. -
Hmn, I've read through the thread and I'm interested in trying stuff that has been said. But before I do that, I have a few questions.
What is 'sluggish' performance to 'non-sluggish' performance? How to I determine this?
How can I access powerCFG? I've tried typing it in command prompt but it appears then, in a blink of an eye, disappears.
Thanks!
FIX for Win7 i7 sluggish performance
Discussion in 'Windows OS and Software' started by Mr Pras, May 25, 2010.