Battery Life?

Framework Laptop 13
79

5.14.0, just tested bluetooth to my JBL Flip 5 which works on my end (connects and outputs sound). Booted into 5.13.12 and that also works.

I glanced at that thread before, but I never encountered any bluetooth issues (though I’m on Fedora, if that matters). AX210 no vPro
Read it again and found someone on Fedora that also didn’t have issues:

dmesg | grep hci0
[    4.727743] Bluetooth: hci0: Firmware timestamp 2021.28 buildtype 1 build 28502
[    4.798497] Bluetooth: hci0: Found device firmware: intel/ibt-0041-0041.sfi
[    4.798669] Bluetooth: hci0: Boot Address: 0x100800
[    4.798672] Bluetooth: hci0: Firmware Version: 86-28.21
[    4.798674] Bluetooth: hci0: Firmware already loaded

If there’s anything else that would help, lmk!

80

@Michael_Wu that power usage seems a bit low for youtube in my experience. Are you on chrome or firefox? With firefox/vaapi enabled/hardware on 5.14 I am averaging about 7.5W on a 720p youtube video which gives me about 7-8 hours on average.

81

Also, one thing to note, I would not recommend CPU_ENERGY_PERF_POLICY_ON_BAT=power unless you are really trying to save battery as it does significantly impact the responsiveness of the system. As a contrived reference, starting emacs takes ~3x longer with the aggressive power saver profiles (either via TLP or via power-profiles daemon).

82

Interesting, I haven’t noticed significant performance degradation. But, as always, YMMV!

83

Oh I should have probably mentioned before I have it set to Balanced Power (in Settings > Power). Probably should try some of these measurements in Performance and Power Saving mode as well.

84

I am unsure how everyone is getting down to such a lower wattage. Running just firefox for this forum and using tuned-adm powersave I am seeing around 9-10w consumed in powertop. I have never been able to get it below 6w or so. This is with 1 usb-c, 2 usb-a and an HDMI adapter. I tried removed the HDMI one but it didn’t do much. This is at lower brightness also.

Oh and Ubuntu 21.04 mostly stock.

1 Like
85

My results were all with tlp and the settings I described previously. Have you tried tlp and compared?

Browsing with Firefox I’m usually in the 6W range, though that can vary a lot by website (some sure love to spin the JavaScript…)

1 Like
86

I am also having a difficult time getting power consumption down, I see ~6.5w just from the screen backlight at %20. Can tlp and tuned be used in conjunction or should I use just one or the other? I have windows installed on the 250gb expansion, and powertop shows ~1.5w of power consumption and %100 usage, is there a way to minimize that?

87

I have pretty high “total system power” in hwinfo, usually 10w+. “cpu package power” seems to be around 4-5w.

Screenshot 2021-09-04 091542
Screenshot 2021-09-04 091542751×979 91.5 KB

88

You may need to set the cpu scheduler to schedutil or ondemand.

89

Interesting. I wasn’t able to get lower than C8 reported using the same settings, and that is with PSR enabled as well. I usually have it disabled because mine is stuttering with it on. I wonder what the difference in software and configuration is. Would you mind uploading the output of tlp-stat somewhere? I would like to compare it to mine when I get the chance.

90

How do I do that? Is that through tlp?

91

Sure thing! You can find it here:

https://blog.b-ark.ca/assets/files/framework-tlp-stat.txt

1 Like
92

Sorry, I meant CPU governor! Here: How to Permanently Set CPU Governor to Performance in Linux - PreProgrammer But use ‘schedutil’ instead of ‘performance’

93

Thanks! I was surprised at how similar our configurations actually were already. Even kernel arguments are more-or-less simlar. Turns out, though, that if I turn on PSR and remove the SD card module, I get C10 55+% of the time with a browser open. So those are what are really holding me from dropping down.

1 Like
94

That’s strange. I also have an SD card module and haven’t had trouble reaching C10 with PSR enabled in Arch. However, I’ve found that the SD card module sucks ~1W of power even when empty. I’ve seen other posts mention it potentially being an issue with the retimer on the mainboard not entering a low power state when the expansion card is empty. Hopefully it gets fixed with a future firmware update.

95

Strange indeed. I’m running Arch, so there goes my “it might be kernel differences” guess. The only configuration I’m passing to the kernel that could have any effect at all is nowatchdog. I have an i5 model. I’m not sure if that could be the difference.

96

From what I see so far, I am getting great battery life. I run 35% screen brightness (turned it up to 100% once and almost blinded myself LOL)
[i7-1165G7 W/10]

But I also only keep pass-through modules in the unit normally (USB-C > USB-C & USB-C > USB-A)

I consider any others (HDMI, DP, SD, storage expansion) to be nothing more than a USB-C conversion dongle that happens to not stick out of the unit when using it.

Anything converting is going to take more power IMO since it is not on the motherboard. Firmware is not always going to be able to change that. So I set my expectations for that before I ever ordered the laptop.

1 Like
97

I have the i7 (non vPro). I’m using the testing repo kernel (5.14.1) because I had PSR glitches with 5.13.x. I’ll try adding nowatchdog to my boot params and see if I’m able to enter C10. Still able to enter C10 with nowatchdog.

You could try using 5.14.x to see if it fixes your issue?

Add the following AFTER your [core] entry in /etc/pacman.conf:

[testing]
Include = /etc/pacman.d/mirrorlist

Then run sudo pacman -Syu testing/linux

98

@ImaxinarDM What kind of battery life are you getting?