Does showing seconds in the system tray actually use more power?
111 comments
·July 13, 2025layer8
Raymond Chen recently wrote about the history of seconds on the taskbar: https://devblogs.microsoft.com/oldnewthing/20250421-00/?p=11...
Jaxan
I hate these kind of “saves power” things in windows settings. The OS itself pings home so often, sends network request for everything you do, shows ads on the login screen, makes screenshots (for Recall), Edge sends contents from web forms for “AI”. And now it is my responsibility to disable showing seconds in the taskbar??? If microsoft really wants to be green, windows shouldn’t do all these wasteful things!
ctoth
I had some very technical friends be incredibly surprised by the Edge form thing, I think that is not sufficiently called out!
They send any text you type in a form to their AI cloud and hold on to it for 30 days.
Any form.
On any website.
What the actual fuck?
smokel
This is only true if you enable extended spell checks, which makes some sense. By default, no form data is sent to Microsoft AFAIK. Note that the same holds for Google Chrome.
perching_aix
Reminds me to a video I saw on YouTube from the "PC Security Channel", who was utterly flabbergasted that the Start Menu would send all keypresses inputted into its search bar to MS.
They had searching on the web enabled... Pretty hard to search the web using Bing without sending along a search term.
atq2119
In what world does holding the user's private data for 30 days make sense for a spell checker? Even sending the data at all is sad. We've had offline spell checking for decades.
foolswisdom
What setting is this? I can only find "Enable machine learning powered autofill suggestions" which seems to have defaulted to on.
null
null
IgorPartola
Whoa how is this not all over the news at all times?
NewJazz
People are tired of hearing about it. They don't feel like they can do anything about it.
saparaloot
The caring cohort has mortages and kids
tspivey
I was surprised by this. I don't use Edge much, and I don't remember being asked about it.
IlikeKitties
> Use Windows
> Expect Privacy
> Don't get Privacy
SuprisedPikatchu.jpg
Teever
Something I heard a while back but have never had confirmed is that the Nvidia driver sends the content of every window title to Nvidia.
Does anyone know if that is true?
svnt
This was when you had to create an account for GeForce Experience and they started sending crash stats.
Some people checked it with wireshark at the time and didn’t find anything other than what was stated. [0]
0: https://gamersnexus.net/industry/2672-geforce-experience-dat...
morkalork
There was a smart tv that did that with the titles of any media played too wasn't there?
Lu2025
Any form meaning passwords too?
perching_aix
Looked into it, the answer seems like it can be both a yes or a no, depending on the website and user actions.
By default, when you implement a form that takes a password, you (the developer) are going to be using the "input" HTML element with the type "password". This element is exempt from spellchecking, so no issues there.
However, many websites also implement a temporary password reveal feature. To achieve this, one would typically change the type of the "input" element to "text" when clicking the reveal button, thereby unintentionally allowing spellchecking.
You (the developer) can explicitly mark an element to be ineligible for spellchecking by setting the "spellchecking" attribute to "false", remediating this quirk: https://developer.mozilla.org/en-US/docs/Web/HTML/Reference/...
You (the developer) can of course also just use a different approach for implementing a password reveal feature.
As the MDN docs remark, this infoleak vector is known as "spelljacking".
ape4
I would check:
- Don't show ads (saves power)
- Don't call home (saves power)bee_rider
This might be considered if they ever find out how shitty Windows can get before people actually stop buying computers with it.
mouse_
As long as Red Hat keeps embracing and extending free desktop, and Apple keeps disallowing standard features like native Vulkan (Mac is not for games I get it but come on, please?), people will either keep using Windows or, more likely, switch to Android devices for their home and business needs.
lxgr
Both things can be true/desirable at the same time.
If, as tested, this setting makes a double-digit percentage difference, I'm glad Microsoft exposes it in the UI. I'd also be glad if they didn't do as much weird stuff on their user's devices as they do.
pavel_lishin
> If, as tested, this setting makes a double-digit percentage difference, I'm glad Microsoft exposes it in the UI.
I'd rather them write more performant code. This feels like your car having the option to burn motor oil to show a more precise clock on the dash; you don't get kudos for adding an off-switch for that.
minitech
> I'd rather them write more performant code.
In keeping with the theme of the comment you're replying to, writing better-performing code and providing performance options are not mutually exclusive. Both are good ideas.
> This feels like your car having the option to burn motor oil to show a more precise clock on the dash; you don't get kudos for adding an off-switch for that.
(Sounds more like you're arguing that it should be forced off instead of being an option? Reasonable take in this case, but not the same argument.)
criddell
> I'd rather them write more performant code.
My expectations of Microsoft software aren't terribly high. I'd say Windows is performant (ie it works about as well as I expect).
daveoc64
It's really an on switch.
The feature is off by default in Windows 11 and was not offered in any previous non-beta Windows version.
orangecat
This feels like your car having the option to burn motor oil to show a more precise clock on the dash
I actively don't want to see seconds; the constant updating is distracting. It should be an option even if there were no energy impact. (Ditto for terminal cursor blinking).
aksss
Better analogy would be reducing your MPGs (fuel efficiency) to show a more precise clock, and arguably we all make that sacrifice to get CarPlay.
Energy isn’t free.
Even if they wrote more performant code, it would just mean less relative loss of energy to show seconds but still loss compared to not showing seconds.
Delk
Mentioning that some setting uses more power can be useful and desirable. I think Jaxan might be irked by "energy recommendations" Windows gives you in power & battery settings, though. It suggests applying "energy saving recommendations" to lower your carbon footprint, and while I absolutely support energy saving, I also find those "recommendations" obnoxious.
The recommendations suggest, among other things, switching to power-saving mode, turning on dark mode, setting screen brightness for energy efficiency, and auto-suspending and turning the screen off after 3 minutes.
Power-saving mode saves little at least on most laptops but has a significant performance impact, dark mode only saves power on LED displays (LCDs have a slight inverse effect), and both dark/light mode and screen brightness should be set based on ergonomics, not based on saving three watts.
When these kinds of recommendations are given to the consumer for "lowering your carbon footprint", with a green leaf symbol for impact, while Microsoft's data centres keep spending enormous amounts of power on data analysis, I find it hard to see that as anything more than greenwashing.
Xylakant
The test setting is important here - the test is on an otherwise idle machine. This means that the update ensures that some thread wakes on a timer every second which may explain the large drop. This test is interesting, but not very representative of a real world usage scenario. It’ll be interesting to compare it to the results of the other test they running, where they keep a video running in the background.
Delk
I'm still a little curious of what's causing the increase in power use. A single additional wakeup per second should not have a two-digit percentage impact on power use when even an idle machine is probably going to have dozens of wakeups per second anyway. I wonder if updating the seconds display somehow causes lots of extra wakeups instead.
cheema33
> And now it is my responsibility to disable showing seconds in the taskbar???
It is not. This "feature" is disabled by default.
Google "manufactured outrage".
zozbot234
This is not Windows-specific, it has been shown wrt. Linux systems also. It's why recent Linux desktop environments have gotten rid of the blinking cursor in command prompt windows (that also causes frequent wakeups and screen updates) and why it probably makes sense to disable most animations too.
userbinator
It's why recent Linux desktop environments have gotten rid of the blinking cursor in command prompt windows
This used to be done entirely in hardware (VGA text modes), and I believe some early GPUs had a feature to do that in graphics modes too.
IgorPartola
Wonder how much an OS that focuses on battery life can extend working time on a laptop. Would be a killer marketing point I think.
HPsquared
And Windows Update burns through an ungodly amount of CPU.
anonymars
To be fair, does it do all those things every second? https://learn.microsoft.com/en-us/windows-hardware/drivers/k...
(For the record, I abhor Windows 11)
gleenn
13% less battery time is pretty wild just from updating the screen once per second but interesting to understand why.
lxgr
That's quite surprising. I wouldn't have imagined Windows (or any other "desktop OS") to go to great lengths to optimize for static screen content in the way that e.g. smartphones or wearables do, which as I understand have dedicated hardware optimized for displaying a fully static screen while powering down large parts of the display pipeline.
layer8
Desktop OSs idle most of the time, and the comparison is with respect to an idle desktop. Forcing context switches and propagating updates through the GUI stack every second isn’t free in that situation, it means that at least one CPU core can’t stay in a lower-power state. In contrast, you probably won’t see much of difference in battery life for the seconds display when simultaneously watching video or running computational tasks.
pdw
The decision to now show seconds dates back to Windows 95. Back then the motivation was not power saving, but rather to allow the code related to the clock and text rendering to be swapped out to disk on a 386 with 4MB RAM... Raymond Chen: https://devblogs.microsoft.com/oldnewthing/20031010-00/?p=42...
jayd16
Windows runs on laptops and tablets and such. At this point they probably do a fair bit of that sort of thing.
rwallace
I hope so, because I actively want seconds absent from the system tray. Attention is a scarce resource; the fewer things on the screen constantly changing and thereby consuming my attention, the better. If saving power means we remain free from that anti-feature, great.
userbinator
Ideally the clock display should be customisable to display whatever level of precision you want; I believe at least one Linux application lets you specify it via a strftime() format string.
aksss
I, for one, love it for casual and incidental benchmarking. Of everything - not just a process I run, but also how long between bird chirps outside my window. But I also find it very easy to ignore, too. Glad it’s optional.
Asraelite
Does nobody care about just being able to tell the time accurately? 59 seconds makes a big difference for joining online meetings and things.
rwallace
Beware of concentrated benefit and diffuse cost. Sure, let a seconds clock be available to call up the 0.1% of the time when you want it. But it shouldn't be in the system tray presenting a small but ongoing attention drain the other 99.9% of the time.
bigstrat2003
Approximately zero people in the world care if you join a meeting at 1:00, or 1:01. It's good to aim to be punctual, but if you're off by a minute there is no consequence.
GLdRH
No and No, it doesn't.
erikpukinskis
I just say “one one thousand two one thousand…” under my breath.
seanalltogether
> Test Type: Idle desktop only (no applications or media playback, unless otherwise stated)
It's weird they didn't also include a simple web browser test that navigates a set of web links and scrolls the window occasionally. Just something very light at least, doesn't even have to be heavy like video playback.
rustyminnow
Not that weird. Idle desktop isolates the effects of the change to get a worst case scenario. Would be interesting to see a light activity test too though - see if you still get a noticeable difference.
viraptor
> We’re currently running the same test again on all three laptops to account for variance, but this time with a video playing to simulate a more active usage scenario. Once those results are in, we’ll update the relevant section with the new data.
delusional
What? "We are doing a second test to account for variance, but also changing the test setup" that doesn't make any sense.
viraptor
They account for variance between different laptops. The test is changing the same way for all laptops. It makes sense.
jasonthorsness
Yeah this is not meaningful due to the unrealistic workload. Sad thing is, I bet a web browser test would still show the difference, as long as a page is kept static on the screen for more than a few seconds before moving on.
Power consumption is incredibly difficult to benchmark in a meaningful way because it is extremely dependent on all the devices in the system, all the software running, and most power optimizations are workload dependent. Tons of time went into this in the windows fundamentals team at Microsoft.
0x_rs
I agree. My guess is the way this may be implemented could keep the system from entering a lower energy state in some way or another, something which would be far less noticeable during normal usage.
__MatrixMan__
Laudauer's principle (https://en.wikipedia.org/wiki/Landauer%27s_principle) tells us that you can't delete a bit without releasing some heat. As the new time digits come in and overwrite the old ones (in the framebuffer, in the LCD, likely other places too) this would occur as the previous digits were deleted. So the only case where showing the time would not take more power is one where other things are not held equal, e.g. some quirk of the software ends up doing more work to ignore the time than to show it (I'd call such a thing a bug).
This effect is likely vanishlingly small, definitely overshadowed by engineering considerations like the voltage used when walking pixels through changes and such. But still, it's a physics nudge towards "yes".
bee_rider
Landauer’s principle is an information-theoretic result about the fundamental cost of computations. With CMOS, every logic gate has multiple transistors, some of which just get charged and dumped to ground with every state transition anyway.
It is like worry about Carnot’s limit… for a motor boat.
ramraj07
When the start menu is a react native app that spikes up the cpu needing billions of flops just to do that, I doubt this number will make a difference.
__MatrixMan__
Agreed, the dominant effect would likely be which ads are being served to the start menu, or which user data is being exfiltrated to Microsoft at the time.
dlcarrier
Is that true? Was Active Desktop just a preview of what's to come?
internet2000
Wasn't that debunked already?
HPsquared
What if it's an OLED screen and the clock is in a dark font on a light background, so adding seconds means less light is emitted? (Light mode only)
__MatrixMan__
Yeah good point. With large enough pixels and pathological color choices you could almost certainly derive the opposite result.
It would be interesting to test it over a remote desktop session where the screen on the device under test is off. That would eliminate a lot of factors related to the display. Presumably you'd see that the network traffic is either larger to begin with, or doesn't compress quite as well, giving you another reason to say "yes, but what if..."
userbinator
There are two conclusions one can draw from this: either the idle power consumption of laptops is so low that something as trivial as updating the clock display on an otherwise idle system[1] is a significant amount, or their code is so shitty that it's taking an order of magnitude or more power than it should. Given this is Microsoft, I'm inclined to believe the latter, or that it was "deliberately" implemented in an inefficient way to "prove" their argument. It'd be trivial to write a tiny Win32 app that just has an incrementing seconds counter and use that to distinguish the latter two cases.
[1] The caveat is that the majority of the time the system will not be idle but doing something else possibly even more energy-intensive.
endorphine
I was wondering the same when configuring Polybar w/ i3 to show seconds on my Linux system. Even if it's marginal, I think I'll disable it.
renewiltord
42 minutes of battery life lost on 321 minutes of battery life is insane.
jambutters
Does this happen on linux? Polybar with i3 has an option to show seconds by clicking the date and time
pdw
It certainly does. There is for example a measurable energy cost for having a blinking cursor in a terminal, and there have been huge flame wars about efforts to move to non-blinking cursors.
The compromise for GNOME Terminal is that the cursor will stop blinking after a terminal has been idle for ten seconds.
dlcarrier
It's going to take power, no matter the operating system. What matter is how much power it takes. On most desktop environments and widgets, it's probably negligible.
wirybeige
It happens on GNOME at the very least, and I would expect every modern platform is the same way.
I would have been happy with no seconds in the tray, but showing the seconds if you click on the clock - technology that existed a decade ago in Windows 10, but is obviously technologically impossible for hundreds of PhD holding software engineers at the richest company in the world to figure out in 2025.