I'm seeing high battery use with version 1.58.2. I've switched to full charging at night to get a better view of usage.
Just want to sure we cover the all the bases, can you check to see if there is maybe some rogue pending action that is just retrying forever? You can check by going to You > Your Actions > Pending actions. Make sure there isn't any actions there. If there is could you send me a screenshot (or if it's sensitive DM me).
I have two failed Mark as Read actions. They seem to occur regularly when I switch accounts. I'll DM you the image.
Ok I am now 99.99% sure we are both reading the battery usage graph wrong. It is not showing how much battery each app used as an absolute value. Its showing of the battery used, how much did each app use. Eg. If you use 1% battery but you spent all of that 1% in Summit then Summit will likely use close to 100% of that 1%.
As proof I let summit run for about 5 minutes on a pixel 7. Then i checked the battery usage screen (screenshot below). It showed that Summit used 66% battery despite my battery % being 100% and only letting the app run for 5 minutes.
So now that we know this, your screenshot just shows you spend a lot of time in Summit but it doesnt necessarily mean Summit is using a huge amount of battery.
That makes sense.
With that method of reading, it looks like Summit is using more battery than comparable apps. I'd expect Summit to use roughly the same amount of battery for each minute of SoT for Reddit or Sync (for Lemmy), but it seems to be higher. For example:
https://sh.itjust.works/pictrs/image/81bd9579-c4a5-4efa-b215-f6e4f8f736d7.png
shows
and https://sh.itjust.works/pictrs/image/c2ee348a-06a1-4298-9032-ea05a0fc4e5b.png?format=webp shows
It's possible that these values aren't comparable, or that more is going on (I'm not paying attention to background execution, for example). If there's another app I can use to measure battery consumption, I can do that.
I guess the most important metric is what is the absolute value of the battery drain and is that level acceptable?
As an example on my end. On average I notice that Summit is using about 10-13% of my battery on a S24 per hour and i think that is an acceptable amount of battery usage given what i do within the app.
I guess the most important metric is what is the absolute value of the battery drain and is that level acceptable?
That's a good way of putting it.
Last week I noticed that my phone needed to be charged before the end of the day. I was using Summit roughly the same amount as I used Sync before that. When I was using Sync, I didn't need to charge my phone before the end of the day. That's what prompted me to look at the battery usage chart.
One other question. Do you have mark as read on scroll in Sync as well?
Yeah, I have it enabled on Sync as well. But I don't think it works any more due to the API changes.
Wow that's insane. 40% in just 1 hour. Bleh back to the drawing board.
Hi, hijacking this thread I just noticed that with Summit 1.58.4 is doing a heavy processing in the background (I re turned the option to pre fecth posts, I just remembered that as we speak, I might as well disable it again and check).
Here you can see the process:
I kinda remember seeing a behavior like that in previous versions, but not sure how long it keeps working in the background.
Does Summit has a process to keep it alive in the background? That is not necessarily bad, as that way it means I won't lose my feed too often due to Android RAM management, but the CPU percentage used is concerning, especially for my aged battery ๐
@sbv@sh.itjust.works what do you think, how has your experience been these days?
Very good news! I made a breakthrough and figured out what is causing high CPU usage while the app is backgrounded. I should have a fix soon. I'll probably just tack this on to the previous release since it's sort of a big issue.
Well well, this is a big week for breakthroughs wouldn't you say? ๐
Is the prefetch option setting related to it? I haven't noticed since I disabled it yet.
No. Not at all. It was the inline image loader.
Well, then I am gonna turn said option again when the new update comes ๐
I did a 20minute test with your exact settings and unfortunately was unable to reproduce anything close to what you saw. I tried browsing both video heavy feeds and just "All" and neither feeds caused any noticeable battery issues.
Community to discuss Summit, a Lemmy reader for Android.
App (Play Store): https://play.google.com/store/apps/details?id=com.idunnololz.summit
APK: https://github.com/idunnololz/summit-for-lemmy/releases
Support the app
Website: https://summit.idunnololz.com/
