Of course, we all hope for that. Who wouldn't want his skins to display consistent values?
![Confused :confused:](./images/smilies/bn.gif)
Of course, we all hope for that. Who wouldn't want his skins to display consistent values?
Yes, true. There is an article on the MS website that I excerpted on my Gadgets wiki HERE in part that explains how they calc core performance using "time-based counters" vs. "utility-based counters". This article concerns core performance and not process performance, but the end result is the same. Win 11 22H2 seems to have extended this "utility" concept to processes, so now that is "messed up" too.
I don't know if it matters here or if it happens in a consistent pattern all the time and for other machines, but in ptmax13's screenshot, when comparing Siberia and Rainmeter usages in the skins and Task Manager, the division between 2.4 and 0.1 is roughly the same as the one between 32.3 and 1.1 at 24..29, while the one between 32.3 and 2.4 is roughly the same as the one between 1.1 and 0.1 at 11..13 (the differences might be due to timestamp differences when the measurements were made, I reckon), so some sort of proportionality seems to exist. I didn't mention this before since I'm not sure how it relates to, for example, his core count or other values close to the results of these divisions. Plus, I'm pretty sure that even if it does, a potential formula (if it exists or makes sense in the first place), would be significantly more complex...SilverAzide wrote: ↑January 13th, 2024, 4:41 pm Yes, true. There is an article on the MS website that I excerpted on my Gadgets wiki HERE in part that explains how they calc core performance using "time-based counters" vs. "utility-based counters". This article concerns core performance and not process performance, but the end result is the same. Win 11 22H2 seems to have extended this "utility" concept to processes, so now that is "messed up" too.
From what I gather, this is never going to be "fixed" because MS doesn't consider it a bug. Lots of other folks disagree, which is why MSI AB and now HWiNFO implemented work-arounds. Their work-arounds only affect CPUs however, so process performance is still an issue.
![]()
Nice setup!
Code: Select all
UseLegacyRAMOverride=1
The above corrections have been completedSilverAzide wrote: ↑January 23rd, 2024, 4:08 pm Nice setup!
Are you saying you want to see the full 64GB of RAM shown in each CPU Meter? If so, I just want to point out that each of your CPUs cannot use the whole 64GB, they can only each see half. But it is up to you if want to show it differently. Fortunately, this is a simple change (I think).
Open the CPUVariables.inc file; it is normally in the C:\Users\<username>\Documents\Rainmeter\Skins\Gadgets\@Resources folder. In the file, search for UseLegacyRAMOverride. Change the line like so:Refresh the skins and you should see the RAM values based on the installed amount. MAYBE. I can't confirm this, so if this value isn't correct, let me know. These numbers are coming from Windows itself, and I'm not sure how it will report on a dual CPU machine.Code: Select all
UseLegacyRAMOverride=1
P.S.: I'm more concerned about why one of your CPUs is reporting it is running at 0 MHz. There must be a bug someplace.![]()
Well, you really should ask the developers; this forum is not the right place for this question. However, the Rainmeter devs usually release a new version when new features are added or bugs are fixed. There have not been many of either one lately. The app has been very stable and not many new features have been requested. Hope this helps answer your question.