standby battery tests after 3.06 BIOS upgrade

1. hardware
2. laptop
3. framework-12th-gen

Result summary:

Device	Wattage	Amperage	Days	Note
baseline	0.25W	16mA	9	no cards, same as before upgrade
1 USB-C	0.25W	16mA	9	same as before
2 USB-C	0.25W	16mA	9	same
1 USB-A	0.80W	62mA	3	+550mW!! worse than before
2 USB-A	1.12W	73mA	<2	+320mW, on top of the above, bad!
Ethernet	0.62W	40mA	3-4	new result, decent
1TB SSD	0.52W	34mA	4	a bit worse than before (+2mA)
MicroSD	0.51W	22mA	4	same
DisplayPort	0.52W	34mA	4+	upgrade improved by 300mW
1 HDMI	?	38mA	?	same
2 HDMI	?	45mA	?	a bit worse than before (+3mA)
Normal	1.08W	70mA	~2	Ethernet, 2 USB-C, USB-A

Dig below for the procedure and raw results.

Procedure

Basically, the test procedure is:

1. resume
2. add/remove modules, change configuration
3. suspend for 60 minutes
4. collect results

The command, as root:

journalctl -b | grep charge_now | tail -2 ; date ;\
rtcwake -m no -s 3600 && systemctl suspend ; date ;\
sleep 10 ; date ;\
/opt/batterylog/batterylog.py ; \
journalctl -b | grep charge_now | tail -2

Note that for the mA calculations, I just subtracted the before/after stats as I have found out that the few extra seconds typically do not change the result, it's generally a rounding error and within the margin of error. This runs the risk of expanding the margin of error by an extra mA of course. But a quick spot check on a few calculations showed the result typically do not diverge even after rounding.

Research hypothesis and bias

The theory is that the 3.06 BIOS upgrade improved power usage. I am biased towards feeling the Framework uses too much power and I am doubtful of the claims. Let's see!

Baseline: 0.25W, 16mA, ~9 days

root@angela:/home/anarcat# journalctl -b | grep charge_now | tail -2 ; date ;\
    rtcwake -m no -s 3600 && systemctl suspend ; date ;\
    sleep 10 ; date ;\
    /opt/batterylog/batterylog.py ; \
    journalctl -b | grep charge_now | tail -2
jan 21 22:22:20 angela systemd-sleep[65127]: /sys/class/power_supply/BAT1/charge_now                     =   1692 [mAh]
jan 22 14:20:05 angela systemd-sleep[65404]: /sys/class/power_supply/BAT1/charge_now                     =   2679 [mAh]
dim 22 jan 2023 14:55:21 EST
rtcwake: assuming RTC uses UTC ...
rtcwake: wakeup using /dev/rtc0 at Sun Jan 22 20:55:22 2023
dim 22 jan 2023 14:55:21 EST
dim 22 jan 2023 15:55:31 EST
Slept for 1.00 hours
Used 0.26 Wh, an average rate of 0.26 W
For your 52.88 Wh battery this is 0.49%/hr or 11.88%/day
jan 22 14:55:22 angela systemd-sleep[73362]: /sys/class/power_supply/BAT1/charge_now                     =   2352 [mAh]
jan 22 15:55:23 angela systemd-sleep[73654]: /sys/class/power_supply/BAT1/charge_now                     =   2335 [mAh]

again:

root@angela:/home/anarcat# journalctl -b | grep charge_now | tail -2 ; date ;    rtcwake -m no -s 3600 && systemctl suspend ; date ;    sleep 10 ; date ;    /opt/batterylog/batterylog.py ;     journalctl -b | grep charge_now | tail -2
jan 22 16:02:38 angela systemd-sleep[75222]: /sys/class/power_supply/BAT1/charge_now                     =   2298 [mAh]
jan 23 10:45:04 angela systemd-sleep[75517]: /sys/class/power_supply/BAT1/charge_now                     =   2822 [mAh]
lun 23 jan 2023 11:38:30 EST
rtcwake: assuming RTC uses UTC ...
rtcwake: wakeup using /dev/rtc0 at Mon Jan 23 17:38:31 2023
lun 23 jan 2023 11:38:30 EST
lun 23 jan 2023 12:38:40 EST
Slept for 1.00 hours
Used 0.25 Wh, an average rate of 0.25 W
For your 52.88 Wh battery this is 0.47%/hr or 11.18%/day
jan 23 11:38:31 angela systemd-sleep[83629]: /sys/class/power_supply/BAT1/charge_now                     =   2915 [mAh]
jan 23 12:38:32 angela systemd-sleep[83909]: /sys/class/power_supply/BAT1/charge_now                     =   2899 [mAh]

This is a good baseline, fairly reliable as it's mostly reproducible (within the 1mAh margin of error). Using 16mA or 0.25W power usage as a baseline.

This does show that the BIOS upgrade did not improve power consumption with no cards, that said. Worse, my battery is slightly more depleted than the last time those tests were ran: it lost 1.65Wh or 3% of its capacity in a little over 3 months. At that rate, it will be completely depleted in 8 years or, more realistically, it will be at 50% capacity in 4 years.

1 USB-C: 0.25W, 16mA, ~9 days

root@angela:/home/anarcat# journalctl -b | grep charge_now | tail -2 ; date ;    rtcwake -m no -s 3600 && systemctl suspend ; date ;    sleep 10 ; date ;    /opt/batterylog/batterylog.py ;     journalctl -b | grep charge_now | tail -2
jan 23 11:38:31 angela systemd-sleep[83629]: /sys/class/power_supply/BAT1/charge_now                     =   2915 [mAh]
jan 23 12:38:32 angela systemd-sleep[83909]: /sys/class/power_supply/BAT1/charge_now                     =   2899 [mAh]
lun 23 jan 2023 12:59:04 EST
rtcwake: assuming RTC uses UTC ...
rtcwake: wakeup using /dev/rtc0 at Mon Jan 23 18:59:05 2023
lun 23 jan 2023 12:59:04 EST
lun 23 jan 2023 13:59:14 EST
Slept for 1.00 hours
Used 0.25 Wh, an average rate of 0.25 W
For your 52.88 Wh battery this is 0.47%/hr or 11.18%/day
jan 23 12:59:04 angela systemd-sleep[86439]: /sys/class/power_supply/BAT1/charge_now                     =   2829 [mAh]
jan 23 13:59:06 angela systemd-sleep[86741]: /sys/class/power_supply/BAT1/charge_now                     =   2813 [mAh]
root@angela:/home/anarcat#

Same as the baseline. Last time we did those tests, we had better results than the baseline with the USB-C card, which was strange.

2 USB-C: 0.25W, 16mA, ~9 days

root@angela:/home/anarcat# journalctl -b | grep charge_now | tail -2 ; date ;    rtcwake -m no -s 3600 && systemctl suspend ; date ;    sleep 10 ; date ;    /opt/batterylog/batterylog.py ;     journalctl -b | grep charge_now | tail -2
jan 23 12:59:04 angela systemd-sleep[86439]: /sys/class/power_supply/BAT1/charge_now                     =   2829 [mAh]
jan 23 13:59:06 angela systemd-sleep[86741]: /sys/class/power_supply/BAT1/charge_now                     =   2813 [mAh]
lun 23 jan 2023 14:01:20 EST
rtcwake: assuming RTC uses UTC ...
rtcwake: wakeup using /dev/rtc0 at Mon Jan 23 20:01:20 2023
lun 23 jan 2023 14:01:20 EST
lun 23 jan 2023 15:01:29 EST
Slept for 1.00 hours
Used 0.25 Wh, an average rate of 0.25 W
For your 52.88 Wh battery this is 0.47%/hr or 11.18%/day
jan 23 14:01:21 angela systemd-sleep[88077]: /sys/class/power_supply/BAT1/charge_now                     =   2801 [mAh]
jan 23 15:01:21 angela systemd-sleep[88313]: /sys/class/power_supply/BAT1/charge_now                     =   2785 [mAh]

Same as a single USB-C.

1 USB-A: 0.80W, 52mA, ~3 days

root@angela:/home/anarcat# journalctl -b | grep charge_now | tail -2 ; date ;    rtcwake -m no -s 3600 && systemctl suspend ; date ;    sleep 10 ; date ;    /opt/batterylog/batterylog.py ;     journalctl -b | grep charge_now | tail -2
jan 23 14:01:21 angela systemd-sleep[88077]: /sys/class/power_supply/BAT1/charge_now                     =   2801 [mAh]
jan 23 15:01:21 angela systemd-sleep[88313]: /sys/class/power_supply/BAT1/charge_now                     =   2785 [mAh]
lun 23 jan 2023 15:08:48 EST
rtcwake: assuming RTC uses UTC ...
rtcwake: wakeup using /dev/rtc0 at Mon Jan 23 21:08:48 2023
lun 23 jan 2023 15:08:48 EST
lun 23 jan 2023 16:08:59 EST
Slept for 1.00 hours
Used 0.80 Wh, an average rate of 0.80 W
For your 52.88 Wh battery this is 1.51%/hr or 36.32%/day
jan 23 15:08:49 angela systemd-sleep[89636]: /sys/class/power_supply/BAT1/charge_now                     =   2749 [mAh]
jan 23 16:08:51 angela systemd-sleep[89930]: /sys/class/power_supply/BAT1/charge_now                     =   2697 [mAh]

Here we see that the USB-A card uses more power than the USB-C, which is similar to the result we had previously, so no improvement from the 3.06 BIOS upgrade.

2 USB-A: 1.12W, 73mA, <2 days

root@angela:/home/anarcat# journalctl -b | grep charge_now | tail -2 ; date ;    rtcwake -m no -s 3600 && systemctl suspend ; date ;    sleep 10 ; date ;    /opt/batterylog/batterylog.py ;     journalctl -b | grep charge_now | tail -2
jan 23 17:54:40 angela systemd-sleep[94541]: /sys/class/power_supply/BAT1/charge_now                     =   2558 [mAh]
jan 23 17:55:02 angela systemd-sleep[94770]: /sys/class/power_supply/BAT1/charge_now                     =   2558 [mAh]
lun 23 jan 2023 17:55:08 EST
rtcwake: assuming RTC uses UTC ...
rtcwake: wakeup using /dev/rtc0 at Mon Jan 23 23:55:09 2023
lun 23 jan 2023 17:55:08 EST
lun 23 jan 2023 18:55:19 EST
Slept for 1.00 hours
Used 1.12 Wh, an average rate of 1.12 W
For your 52.88 Wh battery this is 2.12%/hr or 50.99%/day
jan 23 17:55:09 angela systemd-sleep[95415]: /sys/class/power_supply/BAT1/charge_now                     =   2556 [mAh]
jan 23 18:55:11 angela systemd-sleep[95648]: /sys/class/power_supply/BAT1/charge_now                     =   2483 [mAh]

Even worse, those cards still add up, just like last time. Now your battery is down to two days.

3 USB-A: N/A

I didn't have a third USB-A card on hand to test at this very moment, might come later if requested.

Ethernet: 0.62W, 40mA, 3-4 days

root@angela:/home/anarcat# journalctl -b | grep charge_now | tail -2 ; date ;    rtcwake -m no -s 3600 && systemctl suspend ; date ;    sleep 10 ; date ;    /opt/batterylog/batterylog.py ;     journalctl -b | grep charge_now | tail -2
jan 23 17:55:09 angela systemd-sleep[95415]: /sys/class/power_supply/BAT1/charge_now                     =   2556 [mAh]
jan 23 18:55:11 angela systemd-sleep[95648]: /sys/class/power_supply/BAT1/charge_now                     =   2483 [mAh]
lun 23 jan 2023 19:10:54 EST
rtcwake: assuming RTC uses UTC ...
rtcwake: wakeup using /dev/rtc0 at Tue Jan 24 01:10:55 2023
lun 23 jan 2023 19:10:54 EST
lun 23 jan 2023 20:11:05 EST
Slept for 1.00 hours
Used 0.62 Wh, an average rate of 0.62 W
For your 52.88 Wh battery this is 1.16%/hr or 27.94%/day
jan 23 19:10:55 angela systemd-sleep[97482]: /sys/class/power_supply/BAT1/charge_now                     =   2419 [mAh]
jan 23 20:10:57 angela systemd-sleep[97793]: /sys/class/power_supply/BAT1/charge_now                     =   2379 [mAh]

Interestingly, the Ethernet card uses less power than the dreaded USB-A cards. This is just beyond comprehension for me, as the electronics in the Ethernet card surely must be more complex than the USB-A card.

1TB SSD: 0.52W, 34mA, ~4 days

root@angela:/home/anarcat# journalctl -b | grep charge_now | tail -2 ; date ;    rtcwake -m no -s 3600 && systemctl suspend ; date ;    sleep 10 ; date ;    /opt/batterylog/batterylog.py ;     journalctl -b | grep charge_now | tail -2
jan 23 19:10:55 angela systemd-sleep[97482]: /sys/class/power_supply/BAT1/charge_now                     =   2419 [mAh]
jan 23 20:10:57 angela systemd-sleep[97793]: /sys/class/power_supply/BAT1/charge_now                     =   2379 [mAh]
lun 23 jan 2023 20:17:28 EST
rtcwake: assuming RTC uses UTC ...
rtcwake: wakeup using /dev/rtc0 at Tue Jan 24 02:17:29 2023
lun 23 jan 2023 20:17:28 EST
lun 23 jan 2023 21:17:39 EST
Slept for 1.00 hours
Used 0.52 Wh, an average rate of 0.52 W
For your 52.88 Wh battery this is 0.99%/hr or 23.74%/day
jan 23 20:17:29 angela systemd-sleep[99542]: /sys/class/power_supply/BAT1/charge_now                     =   2335 [mAh]
jan 23 21:17:31 angela systemd-sleep[99830]: /sys/class/power_supply/BAT1/charge_now                     =   2301 [mAh]

This is similar to last time results, but a little (+2mA) worse.

MicroSD: 0.51W, 33mA, ~4 days

root@angela:/home/anarcat# journalctl -b | grep charge_now | tail -2 ; date ;    rtcwake -m no -s 3600 && systemctl suspend ; date ;    sleep 10 ; date ;    /opt/batterylog/batterylog.py ;     journalctl -b | grep charge_now | tail -2
jan 23 22:23:32 angela systemd-sleep[106217]: /sys/class/power_supply/BAT1/charge_now                     =   1980 [mAh]
jan 24 07:16:37 angela systemd-sleep[106523]: /sys/class/power_supply/BAT1/charge_now                     =   1703 [mAh]
mar 24 jan 2023 07:17:20 EST
rtcwake: assuming RTC uses UTC ...
rtcwake: wakeup using /dev/rtc0 at Tue Jan 24 13:17:21 2023
mar 24 jan 2023 07:17:20 EST
mar 24 jan 2023 08:17:31 EST
Slept for 1.00 hours
Used 0.51 Wh, an average rate of 0.51 W
For your 52.88 Wh battery this is 0.96%/hr or 23.05%/day
jan 24 07:17:21 angela systemd-sleep[108096]: /sys/class/power_supply/BAT1/charge_now                     =   1693 [mAh]
jan 24 08:17:23 angela systemd-sleep[108339]: /sys/class/power_supply/BAT1/charge_now                     =   1660 [mAh]
root@angela:/home/anarcat#

Within the 1mA margin of error from last test.

normal setup (Ethernet, 2 USB-C, 1 USB-A): 1.08W, 70mA, ~2 days

Then I switched to my normal setup to go home, and ran the test by hand:

root@angela:/home/anarcat#  /opt/batterylog/batterylog.py ;     journalctl -b | grep charge_now | tail -2
Slept for 3.11 hours
Used 3.37 Wh, an average rate of 1.08 W
For your 52.88 Wh battery this is 2.05%/hr or 49.17%/day
jan 24 08:19:10 angela systemd-sleep[109724]: /sys/class/power_supply/BAT1/charge_now                     =   1646 [mAh]
jan 24 11:25:57 angela systemd-sleep[109965]: /sys/class/power_supply/BAT1/charge_now                     =   1427 [mAh]

So basically, with my normal setup, I have a two-day standby battery. Not great. It should also be noted that a lot of that power (~34mA, almost half!) usage is due to the USB-A card. The Ethernet card takes another good chunk (~25mA) but that's understandable and I expected it to take more power. The USB-A card, though, is really annoying.

DisplayPort: 0.52W, 34mA, 4+ days

root@angela:/home/anarcat# journalctl -b | grep charge_now | tail -2 ; date ;    rtcwake -m no -s 3600 && systemctl suspend ; date ;    sleep 10 ; date ;    /opt/batterylog/batterylog.py ;     journalctl -b | grep charge_now | tail -2
jan 24 11:26:41 angela systemd-sleep[111222]: /sys/class/power_supply/BAT1/charge_now                     =   1420 [mAh]
jan 24 11:27:16 angela systemd-sleep[111456]: /sys/class/power_supply/BAT1/charge_now                     =   1419 [mAh]
mar 24 jan 2023 11:27:24 EST
rtcwake: assuming RTC uses UTC ...
rtcwake: wakeup using /dev/rtc0 at Tue Jan 24 17:27:26 2023
mar 24 jan 2023 11:27:24 EST
mar 24 jan 2023 12:27:36 EST
Slept for 1.00 hours
Used 0.52 Wh, an average rate of 0.52 W
For your 52.88 Wh battery this is 0.99%/hr or 23.74%/day
jan 24 11:27:25 angela systemd-sleep[112157]: /sys/class/power_supply/BAT1/charge_now                     =   1418 [mAh]
jan 24 12:27:28 angela systemd-sleep[112388]: /sys/class/power_supply/BAT1/charge_now                     =   1384 [mAh]

Finally, a significant improvement! The previous results were pretty bad at 0.85W (55mA), now we're down to a more reasonable number, but it's still rather high for an inactive card during standby.

1 HDMI: W, 38mA, days

root@angela:/home/anarcat# journalctl -b | grep charge_now | tail -2 ; date ;    rtcwake -m no -s 3600 && systemctl suspend ; date ;    sleep 10 ; date ;    /opt/batterylog/batterylog.py ;     journalctl -b | grep charge_now | tail -2
jan 24 11:27:25 angela systemd-sleep[112157]: /sys/class/power_supply/BAT1/charge_now                     =   1418 [mAh]
jan 24 12:27:28 angela systemd-sleep[112388]: /sys/class/power_supply/BAT1/charge_now                     =   1384 [mAh]
mar 24 jan 2023 12:28:31 EST
rtcwake: assuming RTC uses UTC ...
rtcwake: wakeup using /dev/rtc0 at Tue Jan 24 18:28:32 2023
mar 24 jan 2023 12:28:31 EST
mar 24 jan 2023 13:28:34 EST
Slept for -0.02 hours
Used -0.12 Wh, an average rate of 6.93 W
For your 52.88 Wh battery this is 13.10%/hr or 314.50%/day
jan 24 12:28:32 angela systemd-sleep[113684]: /sys/class/power_supply/BAT1/charge_now                     =   1376 [mAh]
jan 24 13:28:34 angela systemd-sleep[113928]: /sys/class/power_supply/BAT1/charge_now                     =   1338 [mAh]

Here, batterylog seems to have gone a little cookoo and seems to be counting backwards (Slept for -0.02 hours). We can still calculate the mA used during that hour (38mA), which is similar to past results.

2 HDMI: W, 45mA, days

root@angela:/home/anarcat# journalctl -b | grep charge_now | tail -2 ; date ;    rtcwake -m no -s 3600 && systemctl suspend ; date ;    sleep 10 ; date ;    /opt/batterylog/batterylog.py ;     journalctl -b | grep charge_now | tail -2
jan 24 12:28:32 angela systemd-sleep[113684]: /sys/class/power_supply/BAT1/charge_now                     =   1376 [mAh]
jan 24 13:28:34 angela systemd-sleep[113928]: /sys/class/power_supply/BAT1/charge_now                     =   1338 [mAh]
mar 24 jan 2023 13:58:36 EST
rtcwake: assuming RTC uses UTC ...
rtcwake: wakeup using /dev/rtc0 at Tue Jan 24 19:58:37 2023
mar 24 jan 2023 13:58:36 EST
mar 24 jan 2023 14:58:39 EST
Slept for -0.50 hours
Used -1.85 Wh, an average rate of 3.69 W
For your 52.88 Wh battery this is 6.98%/hr or 167.55%/day
jan 24 13:58:36 angela systemd-sleep[117036]: /sys/class/power_supply/BAT1/charge_now                     =   1218 [mAh]
jan 24 14:58:39 angela systemd-sleep[117336]: /sys/class/power_supply/BAT1/charge_now                     =   1173 [mAh]

Again batterylog is unhappy and counting backwards, not sure why. We're again close to our previous results, if a little worse off (by 3mA), so that's not great.

s2idle, 1 USB-A: 0.42W, 27mA, 4+ days

After someone raised the possibility the poor USB-A performance might be due to deep sleep, I reran the USB-A test with s2idle sleep:

root@angela:/home/anarcat# journalctl -b | grep charge_now | tail -2 ; date ;    rtcwake -m no -s 3600 && systemctl suspend ; date ;    sleep 10 ; date ;    /opt/batterylog/batterylog.py ;     journalctl -b | grep charge_now | tail -2
jan 25 14:36:34 angela systemd-sleep[26685]: /sys/class/power_supply/BAT1/charge_now                     =   3460 [mAh]
jan 25 15:36:34 angela systemd-sleep[26914]: /sys/class/power_supply/BAT1/charge_now                     =   3460 [mAh]
mer 25 jan 2023 15:52:07 EST
rtcwake: assuming RTC uses UTC ...
rtcwake: wakeup using /dev/rtc0 at Wed Jan 25 21:52:08 2023
mer 25 jan 2023 15:52:07 EST
mer 25 jan 2023 16:52:16 EST
Slept for 1.00 hours
Used 0.42 Wh, an average rate of 0.42 W
For your 53.28 Wh battery this is 0.78%/hr or 18.73%/day
jan 25 15:52:08 angela systemd-sleep[30145]: /sys/class/power_supply/BAT1/charge_now                     =   3175 [mAh]
jan 25 16:52:08 angela systemd-sleep[30435]: /sys/class/power_supply/BAT1/charge_now                     =   3148 [mAh]

It seems like the power usage is lower in s2idle with a USB-A module! In my earlier tests, I stuck with deep sleep, because previous results seem to show it had better power usage. My baseline tests also confirmed this, but I did not test with expansion cards plugged in. So it could very well be that s2idle is actually more efficient in the general usage case.

Doing the test again shows it's reproducible:

root@angela:/home/anarcat# /opt/batterylog/batterylog.py ; journalctl -b | grep charge_now | tail -2
Slept for 2.25 hours
Used 0.75 Wh, an average rate of 0.34 W
For your 53.28 Wh battery this is 0.63%/hr or 15.14%/day
jan 25 16:53:26 angela systemd-sleep[31636]: /sys/class/power_supply/BAT1/charge_now                     =   3138 [mAh]
jan 25 19:08:09 angela systemd-sleep[31868]: /sys/class/power_supply/BAT1/charge_now                     =   3089 [mAh]
root@angela:/home/anarcat#

... in fact, this second test shows a lower wattage use, which is strange. But that test was done in rather exotic conditions: I was traveling home with the laptop, in a bag, in sub-zero temperatures, instead of just leaving it open on my desk... So to take with a grain of salt.

normal setup in s2idle: 0.88W, 57mA, 2.5 days

root@angela:/home/anarcat# journalctl -b | grep charge_now | tail -2 ; date ;    rtcwake -m no -s 3600 && systemctl suspend ; date ;    sleep 10 ; date ;    
/opt/batterylog/batterylog.py ;     journalctl -b | grep charge_now | tail -2
jan 25 16:53:26 angela systemd-sleep[31636]: /sys/class/power_supply/BAT1/charge_now                     =   3138 [mAh]
jan 25 19:08:09 angela systemd-sleep[31868]: /sys/class/power_supply/BAT1/charge_now                     =   3089 [mAh]
mer 25 jan 2023 19:45:41 EST
rtcwake: assuming RTC uses UTC ...
rtcwake: wakeup using /dev/rtc0 at Thu Jan 26 01:45:42 2023
mer 25 jan 2023 19:45:41 EST
mer 25 jan 2023 20:45:50 EST
Slept for 1.00 hours
Used 0.88 Wh, an average rate of 0.88 W
For your 53.28 Wh battery this is 1.65%/hr or 39.54%/day
jan 25 19:45:42 angela systemd-sleep[44135]: /sys/class/power_supply/BAT1/charge_now                     =   2770 [mAh]
jan 25 20:45:42 angela systemd-sleep[44426]: /sys/class/power_supply/BAT1/charge_now                     =   2713 [mAh]