Firmware Update 5.13.6 drains battery much faster than 5.12.4 on voyage

[Hzj_jie]

May be not important anymore, it's an old device. But I do notice 5.13.6 drains battery much faster than 5.12.4 on voyage, roughly 30% more when screen is on and 3x when screen is off.
I compared my three voyages and noticed the difference. I downgraded one from 5.13.6 to 5.12.4 and will report back with detailed logs.

Considering 5.13.6 is the latest voyage version, I think amazon is playing apple dirty tricks.

[Frogm4n]

It's not dirty tricks. I'm assuming you're referring to the battery issue on old iPhones. Even what Apple did wasn't harmful to users, it kept phones running longer. People have grossly misunderstood what that was about.

As for your Kindle, have you added a lot of content recently? Have you done mods to it? What has happened aside from the firmware update?

[Quoth]

Unless you replaced the battery it can be co-incidence.
1) The battery has less capacity as it ages.
2) Always keep connected to power while it indexes when you add new ebooks.

[Hzj_jie]

Thank you for your reply, I do know what I am talking about. I need extra weeks to get the statistics for a meaningful comparison.

But for the questions here, slowing the device or draining battery unnecessarily are both unacceptable, no matter how good the excuse is, at least there should be a config to flip back.
Also, I am talking about using exactly the device but only downgrading the firmware, no other changes.
My oldest voyage is over 10 years old with at least several hundreds cycles, and it's the one lasting longest. Unlike the phone batteries, Kindles are charged once per month or even longer, the battery wearing should be much smaller and won't be a meaningful fact here.

[Quoth]

Originally Posted by Hzj_jie

Thank you for your reply, I do know what I am talking about. I need extra weeks to get the statistics for a meaningful comparison.

Impossible without a control device or fresh battery to swap.

Originally Posted by Hzj_jie

But for the questions here, slowing the device or draining battery unnecessarily are both unacceptable, no matter how good the excuse is,

I only ever heard of Apple doing that. There has been no suggestion of Amazon doing it. Apple makes 30% to 70% margin on their products. Amazon wants you to buy the ebooks and/or have a Prime sub or a KU sub. They make little selling their Fire or Kindle models. The profit is on the content.
I've a Paperwhite 3 and I know someone with a Paperwhite 1 and also people with several generations of Basic. I also have other older Kindle models that all needed new batteries (cells). One needs a 2nd battery replacement now. I see battery life suddenly get much poorer, but never due to a FW upgrade. I've two PW 3 models. One is on last FW as are my other non-PW Kindles. No effect on run time from firmware.

Originally Posted by Hzj_jie

at least there should be a config to flip back.
Also, I am talking about using exactly the device but only downgrading the firmware, no other changes.

I agree there should be, but AFAIK there has never ever been the ability to go back on FW on a Kindle like on a Kobo, or updates on Linux and Windows. I've no idea how to "go back" on Android either. It seems vaguely possible on Chromebook, if you did a FW backup first, as it can reload FW from a USB device at boot time.

I've been using Kindles since before the Voyage was released and the most likely thing is your battery (really an LiPoly cell).

The Voyage came out in late 2014 and ceased about Q3 2016, so the battery/cell is about 8+ to 10+ years old. That also is about the lifetime of an LiPoly cell, though reduced by number of complete charge cycles (10x 10% = 1 cycle) and elevated temperature.

The PW3 in mid 2015 was a budget alternative, the 2nd 6″ model with a 300 dpi screen. The Oasis 1 (6″) is really the mythical Voyage 2 in mid 2016 with Voyage discontinued a few months later. The two later Oasis models are 7″.

Occam. No-one else seems to be claiming that a FW update makes a significant difference and last Voyage update was Version 5.13.6 – April 2021 That means almost FOUR years for people to complain.

Firmware Versions for Kindles. From Amazon.

Timeline of Amazon Kindle Models. Probably USA dates. Some models are months later for rest of world, or EMEA.

An LiPoly cell can last for even 15 years or as short as 5. There are a lot of variables including materials quality. The 8 to 10 is most common and they do degrade faster beyond a certain point.

[DNSB]

Originally Posted by Quoth

I agree there should be, but AFAIK there has never ever been the ability to go back on FW on a Kindle like on a Kobo, or updates on Linux and Windows. I've no idea how to "go back" on Android either. It seems vaguely possible on Chromebook, if you did a FW backup first, as it can reload FW from a USB device at boot time.

The ability to go back to an older firmware revision seems to be one of the common reasons people give for jailbreaking their Kindles.

[Quoth]

Originally Posted by DNSB

The ability to go back to an older firmware revision seems to be one of the common reasons people give for jailbreaking their Kindles.

I'd got that impression. KOReader is the other common reason. On those without WiFi, the USB Networking or other methods all setting time and date. Totally stupid that Time & Date is set only by successful WiFi or Cell/Mobile connection on many Kindles.

[chronoreverse]

FWIW, my Voyage on 5.13.6 has great battery life still but I always keep WiFi off and set my backlight at 10.

[j.p.s]

Originally Posted by chronoreverse

FWIW, my Voyage on 5.13.6 has great battery life still but I always keep WiFi off and set my backlight at 10.

Well, that is at least one nail in the coffin of the claim that 5.13.6 in and of itself greatly increases battery drain. But kindle firmwares have been getting more and more aggressive about wanting to phone home. Keeping wifi off can help with that, but there have often been bugs where it keeps checking for wifi or something and drains battery even faster than when wifi is on. With any wifi enabled device there is always the problem poor reception causing battery drain.

[Hzj_jie]

Sorry for missing the background here. I do not use native kindle os. I am using koreader 2024.11. Wifi is always on; background light is set to 10. But if the environment brightness is high enough, the background light is automatically turned off, i.e. the autofrontlight plugin.
I also always enable http inspector to remotely control the page turning on my phone. I do not change my reading patterns, and I didn't add too many new books recently - at least during the cycles below.

The analysis is definitely not accurate, but considering the difference I have noticed is very significant, I do believe the number indicates something interesting in 5.13.6.

In 5.13.6, I had three full cycles logged. I got the unit early this year to replace my 10-year-old voyage, and started the full cycle logging after noticing the excessive battery drain. The device is old, and the battery level reported by the OS is not monotone decreasing. It causes the total percentage to be over 100%.

231 Dump at Sat Feb  8 00:29:25 2025
232 Awake since last charge
233    Total time:  13:19
234    Consumed:    101.00%
235    Change per hour: 7.58%
236    Estimated remaining time:    01:03
237 Sleeping since last charge
238    Total time:  8d09:26
239    Consumed:    33.00%
240    Change per hour: 0.16%
241    Estimated remaining time:    2d00:50
242 During last charge
243    Total time:  00:11
244    Charged: 0.00%
245    Change per hour: 0.00%
246    Estimated time for charging: N/A
247 Since last charge
248    Total time:  8d22:45
249    Consumed:    134.00%
250    Change per hour: 0.62%
251    Estimated remaining time:    12:49

300 Dump at Mon Feb 17 01:03:57 2025
301 Awake since last charge
302    Total time:  14:08
303    Consumed:    91.00%
304    Change per hour: 6.44%
305    Estimated remaining time:    01:52
306 Sleeping since last charge
307    Total time:  7d20:06
308    Consumed:    26.00%
309    Change per hour: 0.14%
310    Estimated remaining time:    3d14:49
311 During last charge
312    Total time:  09:36
313    Charged: 93.00%
314    Change per hour: 9.69%
315    Estimated time for charging: 09:05
316 Since last charge
317    Total time:  8d10:14
318    Consumed:    117.00%
319    Change per hour: 0.58%
320    Estimated remaining time:    20:44

346 Dump at Thu Feb 27 11:01:33 2025
347 Awake since last charge
348    Total time:  15:21
349    Consumed:    85.00%
350    Change per hour: 5.54%
351    Estimated remaining time:    01:05
352 Sleeping since last charge
353    Total time:  9d09:58
354    Consumed:    30.00%
355    Change per hour: 0.13%
356    Estimated remaining time:    1d21:12
357 During last charge
358    Total time:  06:31
359    Charged: 86.00%
360    Change per hour: 13.18%
361    Estimated time for charging: 07:08
362 Since last charge
363    Total time:  10d01:19
364    Consumed:    115.00%
365    Change per hour: 0.48%
366    Estimated remaining time:    12:35

The usage per hour awaken is well over 5% and 0.13%-0.16% sleeping. (Un-)surprisingly, the numbers were also very inconsistent.


After downgrading to 5.12.4, I have only got one log so far,

392 Dump at Wed Mar 12 23:05:15 2025
393 Awake since last charge
394    Total time:  1d03:53
395    Consumed:    91.00%
396    Change per hour: 3.26%
397    Estimated remaining time:    01:32
398 Sleeping since last charge
399    Total time:  12d01:22
400    Consumed:    33.00%
401    Change per hour: 0.11%
402    Estimated remaining time:    1d19:51
403 During last charge
404    Total time:  06:40
405    Charged: 92.00%
406    Change per hour: 13.81%
407    Estimated time for charging: 06:53
408 Since last charge
409    Total time:  13d05:15
410    Consumed:    124.00%
411    Change per hour: 0.39%
412    Estimated remaining time:    12:48

I can attach more logs after several cycles.
The usage drops to 3.26% per hour. Coincidentally, the values almost match another unit which has never been upgraded to 5.13.6.


Anyway, these are several samples on my units; it may not be a real issue at all. I do see units behave slightly differently. For example, one of my voyages always maintains 0.04% battery consumption in sleep. It is much better than my other units. I have zero clue - maybe a different batch matters. I previously thought downgrading to 5.12.4 could drop the battery consumption in sleep to the same level, but it seems not.

TBH, in 5.13.6, reading ~15h is still a decent battery life. It may also explain why people did not complain about it. But I do not mind if my kindle can last 27h.

[chronoreverse]

I use Koreader almost all the time too, consistently 1.5% in active reading for my last 90% to 20% cycle, 0.05% in sleep. I do not have WiFi on though. Could maybe the newer firmware be less wifi efficient?

You do need to take the same unit, clean install and then do exactly the same thing to be sure it's the firmware.

[janrey]

Originally Posted by Hzj_jie

May be not important anymore, it's an old device. But I do notice 5.13.6 drains battery much faster than 5.12.4 on voyage, roughly 30% more when screen is on and 3x when screen is off.
I compared my three voyages and noticed the difference. I downgraded one from 5.13.6 to 5.12.4 and will report back with detailed logs.

Considering 5.13.6 is the latest voyage version, I think amazon is playing apple dirty tricks.

Turn off wi-fi when not using. Seriously, the searching for wifi signals can greatly reduce battery life. I have an old Kindle 3 sitting around and if I leave wifi on battery will drain to nothing in less than two days even though it is not being used. With wifi off it will last over a week. I always have my Voyage wifi off, especially when reading library books so that they aren't taken away before I finish.

I don't think they are updating the voyage anymore anyway.

[Frogm4n]

Originally Posted by Hzj_jie

Sorry for missing the background here. I do not use native kindle os. I am using koreader 2024.11. Wifi is always on; background light is set to 10.

The Kindle OS is still running, even when you are using KOReader. It's running as an app on top of it, not replacing it. You'd be using KOReader in place of the native Kindle reader, not the whole OS.