Next time do an ugly test for the UI - cos there is NO reason for all the glass nonsense and it just looks terrible
Got a tip for us?
Let us know
Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
iOS 26.1 Beta Liquid Glass Battery Drain Test: Tinted vs Clear Mode
- Thread starter MacRumors
- Start date
- Sort by reaction score
OMG, really? This has to be tested?
For all people who don‘t understand how things work:
We have a 32 bit color 8 bits for each channel r, g, b, a - a is the alpha value for transparency that applies when you mix layers or add a layer on top of each other. But it doesn‘t need more GPU power no matter if the alpha value is 0, 128 or 255. So just changing the transparency doesn‘t change anything.
Removing/reducing effects, if this is the case with this setting, it maybe helps a little bit. But those effects are baked into GPU shaders and are not that expensive.
A simpler Ui could be done with less complex operations and fewer operations in code, this is true. But only changing the transparency in a setting doesn‘t change the code, maybe it uses another code/gpu path but all of this doesn‘t lead to less battery consumption.
For all people who don‘t understand how things work:
We have a 32 bit color 8 bits for each channel r, g, b, a - a is the alpha value for transparency that applies when you mix layers or add a layer on top of each other. But it doesn‘t need more GPU power no matter if the alpha value is 0, 128 or 255. So just changing the transparency doesn‘t change anything.
Removing/reducing effects, if this is the case with this setting, it maybe helps a little bit. But those effects are baked into GPU shaders and are not that expensive.
A simpler Ui could be done with less complex operations and fewer operations in code, this is true. But only changing the transparency in a setting doesn‘t change the code, maybe it uses another code/gpu path but all of this doesn‘t lead to less battery consumption.
Last edited:
Excellent idea.
Cracking on with it, look forward to you sharing your results!
I absolutely agree that the battery page is utter garbage compared to iOS 18. Who thought "1hour longer" is a useful metric, and more importantly who the hell approved this for production.
What kind of proposal is that? Why is it so wrong to point out obvious oversights of a test?
You know, most of us dont have the time to do tests like these as we dont get paid playing around with Apple stuff.
Why would a reader do this kind of test in their free time when they could instead point out what should be considered when doing such a test to the author?
There is no shame in making mistakes.
I don’t think it’s intentional, unfortunately.
I think it’s just inefficient and badly written in general.
No conspiracy here, just bad software.
Except for the battery page in Settings.
That IS definitely bad on purpose.
This is gold!
Worth to mention, when I was using silicon Mac with widgets in Mac OS, I noticed the more I put them the more they use CPU, up to 50% in background, for minutes.
So this is nuts, because less you use all “beauty and comfort” the more juice you got.
With my 14 PM (83% capacity) I am maybe getting 5% less battery going from iOS18 to iOS26 with all the of the liquid glass kept on. I pulled that number out of thin air and just did some estimating compared to previous times on what I was ending up with at the end of the day with iOS 18.
I still end up with about 50% left per day so not the heaviest of users. Fine by me and I enjoy the liquid glass look. Probably keeping the 14 PM one more year and waiting for the 18 PM to make the jump.
I still end up with about 50% left per day so not the heaviest of users. Fine by me and I enjoy the liquid glass look. Probably keeping the 14 PM one more year and waiting for the 18 PM to make the jump.
In my not-scientific and entirely anecdotal experience, yeah, Tahoe reduced battery life on my 14" M4P MBP.
Using the method in that video, you could also come to the conclusion that increasing the screen brightness also taxes the CPU because there’s more of a power drain when the brightness is higher.
And that has GOT to be due to the CPU, right? Everyone knows power strip indicators are good at reporting out how much the CPU, specifically, is using.
Pointing out what wasn’t thought of to have more conclusive data IS constructive criticism.
You are complaining.
constructive would be to suggest tests that the poster thinks are more appropriate ... posting that post here again for you
Great useful information on this new feature. Many people on my iOS 26 battery life thread have been saying a lot about Liquid Glass and battery life/graphics performance so will be forwarding this article to them.
Not being rude but this toggle should not affect battery life because even though it’s more opaque, the glass distortion effects are still being used. Look closely and you’ll see it’s just less transparent, it’s not processing less data.
iOS18 forever!!! What this test doesn‘t mention - personally I really don‘t like operating an 40 degrees hot iPhone, it feels extremely uncomfortable to me.
At least i‘ll try to skip the complete 26 release cycle of all Apple OSs.
Apple Glass is crap. Apple should really scrap it and go back to the iOS18 UI. Some things like the lock screen are nice and should make it into the new version. But not only does Liquid Glass reduce usability, it also increases battery drain - for a couple of effects, really no one asked for.
Last edited:
I wouldn’t consider this complaining.
Pointing out obvious flaws of a test is far more constructive than telling that poster to do it themselves if they didn’t like the test.
People...
We are talking about Vista or Compiz/Beryl/Compiz Fusion like effects...we are talking about pre 2010 effects.
Do you seriously think that these effects have ANY performance impact on even 1st gen Iphone SE? And let's not even talk about fresh Iphones or Ipads.
If the graphics pipeline isn't horribly broken then both the new and old shouldn't have any noticeable impact. You have a supercomputer in your pocket, not a early 2000s Pentium II with 32 MB of RAM and Voodoo 2 almost exploding with a Windows XP and fancy Luna effects.
You honestly don't need tests to know glass or other graphical nothingburders use almost zero % of compute power considering how much of it is available.
We are talking about Vista or Compiz/Beryl/Compiz Fusion like effects...we are talking about pre 2010 effects.
Do you seriously think that these effects have ANY performance impact on even 1st gen Iphone SE? And let's not even talk about fresh Iphones or Ipads.
If the graphics pipeline isn't horribly broken then both the new and old shouldn't have any noticeable impact. You have a supercomputer in your pocket, not a early 2000s Pentium II with 32 MB of RAM and Voodoo 2 almost exploding with a Windows XP and fancy Luna effects.
You honestly don't need tests to know glass or other graphical nothingburders use almost zero % of compute power considering how much of it is available.