With those words you really should supply numbers of your own. No?Just lazy.
With those words you really should supply numbers of your own. No?Just lazy.
I doubt such a test would have accurate results, though, as it needs time to index things after updating. When running one test before another right after updating, it might make the first test seem slower if it still needs indexing.
I don’t know.
Maybe the test should be longer, so that possible trends aren’t just a single percentage point off of each other.
Also, like others mentioned, most apps that were tested don’t have any Liquid Glass effects anywhere anyway, so maybe you could test with first party apps and updated third party apps with more animations over a longer timeframe.
It's not a really well designed experiment. You only used 2/5 apps which use Liquid Glass in iOS 26 natively and a few other apps which run on React Native. It doesn't actually accomplish anything. Just lazy.
If you spend every minute of the day swiping on your phone battery life is not the life you should be worrying about.It's possible that a full 16 hours of swiping in and out of apps and features
I think something like this would be perfect to gauge its true impact:I don't think I'd need to run it back to back. I'd update and then wait a few days. It'd just use the same parameters I used here.
Yeah, TikTok and Instagram specifically don't have much. That's why I added in Safari and Maps, plus did all the Control Center/Notification tests. I can repeat, but I'd need some specific app recommendations. I mostly wanted to mimic how I/lots of other people are actually using their phones.
I wanted to do a test that would be closer to real usage conditions vs. just testing apps with Liquid Glass, but I added in all the Control Center + Notification Center access to try to get more Liquid Glass in there. I also had to have a sequence I could repeat exactly every time.
I'm open to more testing if you have a series of apps you think would be more representative.
My recommendation based on this video is to not constantly scroll the lock screen up and down as well as similar actions. Instead, just do them as needed.Watch this video, he is showing the actual power draw
This is a shame.Some people on Reddit the other day were guessing that even in tinted/disabled mode, Liquid Glass effects were still being rendered and then just masked after-the-fact. Meaning the visibility, for those who needed it disabled, is helped, but there was no compute savings from it.
Many smartphone screens were LCD before OLED came along (I don’t think miniled factors in yet, are there any smartphones with miniled?). Like all tech, the first OLED’s had some tradeoffs that had to be accepted. These days, the tradeoffs are fewer and some companies have taken steps to ensure their devices avoid burn-in and other downsides. But, with increasing SoC efficiency, the OLED will always highest potential for draining the battery in a smartphone.Last I checked oled is more efficient than mini led
The Power Draw of the ENTIRE system. Which includes the OLED. Which is the biggest power draw in a phone. There is literally no way that a power strip with numbers on it can say anything about what, inside the phone, is leading to those numbers. It was excellent at taking advantage of people wanting to believe what they want to believe, but that’s it. And, with people copying the link to the video hither and yon, I hope they’re getting some dollars in their pocket for the effort!Watch this video, he is showing the actual power draw
Spoken like a true ableist. Must be nice to have the option to have a life outsideIf you spend every minute of the day swiping on your phone battery life is not the life you should be worrying about.![]()
Are there same tests but using iOS 18 for comparison?Watch this video, he is showing the actual power draw
That makes no sense at all. That’s not how the rendering pipeline works .Some people on Reddit the other day were guessing that even in tinted/disabled mode, Liquid Glass effects were still being rendered and then just masked after-the-fact. Meaning the visibility, for those who needed it disabled, is helped, but there was no compute savings from it.