wgroeneveld
/
brainbaking
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

typos

This commit is contained in:
Wouter Groeneveld 2021-07-15 14:26:59 +02:00
parent f3ab597215
commit 1e56a881fd
1 changed files with 3 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

6
content/post/2021/07/the-decline-of-battery-life.md
View File

@ -13,11 +13,11 @@ Sure, the Grey Brick is very old. Sure, it was based on what Nintendo's Gunpei Y
Take a look at the average battery life of Nintendo's handheld systems over the years:
![](../batterychart.png "Average battery live of devices.")
![](../batterychart.png "Average battery life of devices.")
Scrambling together these numbers required a lot of guesswork as my own playthroughs and reported numbers from various sources differed quite a bit. [Nerdly Pleasures](https://nerdlypleasures.blogspot.com/2014/10/battery-life-in-8-bit-game-boy-line.html) explains battery life in the 8-bit Game Boy line more thoroughly. [Wired](https://www.wired.com/2012/09/battery-test/) and [IGN](https://www.ign.com/articles/2019/07/17/nintendo-switch-how-the-new-models-battery-life-compares-to-other-handhelds) also conducted bigger tests.
Comparing battery life is much more involved than simply looking at the average play time. For one, there's the kilowatt/hour and electric charge/hour rates of the batteries, the composition of the battery (lithium, nickel-metal hybrid, ...) Secondly, there's the required current of the device itself, also based on a slew of different requirements (CPU kind, size, make, RAM, screen technology and intensity, and so forth). Thirdly, it depends on the load of the machine (the game itself, and the brightness of the screen). I tried adding these to the graph but the resulting mess made me even more confused.
Comparing battery life is much more involved than simply looking at the average play time. For one, there's the kilowatt/hour and electric charge/hour rates of the batteries, the composition of the battery (lithium, nickel-metal hybrid, ...) Secondly, there's the required current of the device itself, also based on a slew of different requirements (CPU kind, size, make, RAM, screen technology and intensity, and so forth). Thirdly, it depends on the load of the machine (how the game itself was programmed and the brightness of the screen). I tried adding these to the graph but the resulting mess made me even more confused.
In essence, as an end user of these machines, I do not care about `mAh`, `MHz`, `DDR3` or whatever. I care about the longevity. Will this thing last the long train ride? Can I still make a call after watching these YouTube videos on my smartphone?
@ -25,6 +25,6 @@ Remember the [Nokia 3310 cellphone](https://en.wikipedia.org/wiki/Nokia_3310) fr
Back to that [Evercade](https://jefklakscodex.com/articles/evercade/) I was fiddling with. It has a `1.2Ghz` Cortex-A7 processor, a `4:3` LCD screen (the same as a PSP), and runs a custom version of Linux (of course it does). That means it's basically a little computer I just happen to use for retro gaming. 8-bit, 16-bit and even 32-bit handhelds from the original GB to the GBA very simple embedded pieces of hardware. The Game Boy Color didn't even have another processor: it was still the `Z80`, only running at twice the speed of the original model (`8 MHz`). The GBA "finally" contained a proper CPU, an ARM7TDMI running at `16.78 MHz`. Except that its CPU cycles aren't clogged with needless Linux daemons or kernel programs: there's simply no OS! Pop in a cart and you're ready to go. The Evercade is also cartridge-based, but relies on emulation---hence, a software layer (and more power) is needed.
Twenty years later, the Evercade is almost eighty times faster, while its battery life has been divided by three. The original GBA's non-backlit screen is not that inferior to the Evercade's bigger LCD screen. The quality is very bad (see my review), and my modded GBA with backlit fares of much better, without having to give in that much on battery time. The upcoming [Analogue Pocket](https://www.analogue.co/pocket)---that uses hardware emulation with FPGAs instead of software emulation---reportedly does not perform that much better, with a battery life of six hours as a rough estimation. But that one comes with a high-quality screen. Or so they say.
Twenty years later, the Evercade is almost eighty times faster than the GBA, while its battery life has been divided by three. The first GBA model's non-backlit screen is not that inferior to the Evercade's bigger LCD screen. The quality is very bad (see my review), and my modded GBA with backlit fares of much better, without having to give in that much on battery time. The upcoming [Analogue Pocket](https://www.analogue.co/pocket)---that uses hardware emulation with FPGAs instead of software emulation---reportedly does not perform that much better, with a battery life of six hours as a rough estimation. But that one comes with a high-quality screen. Or so they say.
Power for battery life---a fair trade-off? I'm not so sure.