After reading this article about the Switch's battery life I got curious, how long my older and newer handhelds (and some of my other mobile devices) can still perform under different conditions (brightness, demanding games, simple games...) until the battery is drained. Many battery tests on Youtube just let the devices idle (and sometimes let them dim down the displays), but I wanted to simulate some worst case scenarios for the more demanding games.

I also wanted to find out, which range of energy the devices consume under these different conditions and how long the battery life is (with the most demanding settings) when connected to my power bank. Where is the sweet spot between battery life and display brightness?

My devices all still have their original batteries, which of course have degraded over the years and by recharging them again and again. To take that degradation into account, I reduced the original battery capacity of my handhelds by 5% per year. So if a new PSP-1000 battery has a capacity of 6.66 Watt hours (1800 mAh * 3.7 V), my 12 year old battery probably has lost 60% of its capacity (12 years * 5%), which results in a remaining capacity around 2.7 Wh (... of course that's just a rough estimate). In settings with a battery life of 2.7 hours, the device then needs an average of 1 Watt per hour.

Since my smartphone is active 24 hours a day and gets a lot more recharges per year, I estimated its degradation by 10% per year. My iPad also gets used much more than my individual handhelds, but due to its huge battery, it doesn't get recharged as often, so I left that at 5% per year.

The following results are no competition between the different devices (because of different games, different brightness ranges, different display sizes, different battery size and degradation...), it is just a current snapshot of my devices. Your mileage may vary due to diverging age of the devices/batteries, more or less recharges and varying battery quality.

So let's begin with my 12 year old PSP-1000, which I imported from Canada five months before the EU launch. To get a "worst case scenario" for the ancient battery, I let it run the demanding and still gorgeous "Burnout Legends" from UMD. With rubberbanding the right shoulder button, I put the pedal to the metal, which also stress the UMD drive by loading different parts of the track:

View on YouTube

I also tested the demanding "GTA Vice City Stories" with different brightnesses. I own the retail version as well as the PSN version and wanted to compare it running from UMD + memory stick... unfortunately the UMD didn't work anymore. I also wanted to constantly rotate the camera, but due to the lack of a second stick, the PSP-version doesn't support that... but later I found out with GTA Vice City on iOS + Android, that camera rotation doesn't matter for energy drain... just watch traffic with the same settings has the same results.

The simple 2D game "Puzzle Quest" idling around was the "best case scenario" for the battery and I also tested the energy efficiency of the PS1-mode:

The battery life on the right ("new battery") is just an extrapolation of my degradation estimate (Time / 40%)... perhaps I buy a new 1800 mAh battery to check that theory, but probably I go for a 3600 mAh battery pack instead.

Next is my DS lite, which I imported from the UK  9 years ago. To get a "worst case scenario" for the ancient battery, I let it run the demanding "Metroid Prime Hunters". Constant camera rotation wasn't possible via shoulder buttons, but with a bit of weight on the d-pad I kept the 3D engine busy:

View on YouTube

As "best case scenario" for the battery I chose the simple 2D game "Ace Attorney 3" and let it idle in the courtroom... I was surprised when the results were almost the same as "Metroid Prime Hunters" in the same brightness settings, so the SoC seems to run always with full power.

Totally different were the results running a GBA game with one of the displays turned off... the energy efficiency in GBA-mode is great!

4.5 years ago I bought my red 3DS XL. The "worst case scenario" for the battery was "Resident Evil: Revelations". I tested it in 3 brightness settings with auto-stereoscopic 3D turned on and off. The circle pad can't be rubberbanded like a stick to keep a constant camera rotation, but a paper clip did the trick.

As "best case scenario" for the battery I chose the simple 2D game "Ace Attorney 3 HD" and let it idle in the courtroom.

The battery life in both games is a bit better when turning off 3D, but I had expected a much bigger difference in "Resident Evil: Revelations", since only half the resolution has to be rendered that way. The minor differences in "Ace Attorney 3 HD" weren't a surprise... even in 3D these are still simple 2D-graphics placed more in the foreground or the background.

My two Vitas I always tested in tandem by rubberbanding their sticks. Even if the game don't have a function on the right stick (like Shovel Knight), it keeps the devices from dimming down the displays after a few seconds of no input:

View on YouTube

The Vita with its gorgeous OLED screen + 3G is 5 years old, I got in at launch day. Its brightness range is much more narrow than on any other of my devices, on minimum brightness it is still brighter than my Vita Slim or my 3DS XL on medium brightness. Unfortunately that also shows in the tests... you only gain a few minutes when setting the brightness to the minimum... "Killzone Mercenary" and "Uncharted Golden Abyss" as "worst case scenarios" both were around 3 hours, no matter which brightness ofr if Wifi and 3G were on or off:

"Limbo" and "Shovel Knight" as "best case scenarios" had much better results. Efficiency in PSP-mode and PS1-mode were also quite good. 

My 2 year old Vita Slim has a much wider range both in brightness and in battery life. I can play 5 hours "Uncharted: Golden Abyss" with all bells and whistles (or "Killzone Mercenary" with half brightness) and up to 10 hours Shovel Knight (Vita) or "Crash Bandicoot 3" (PS1) with half brightness or , "GTA Vice City Stories" (PSP) with low brightness:

My 3.5 year old iPad Air has a wide range of battery life depending on the game and the brightness settings. I chose "Real Racing 3" as "worst case scenario", since this racing game has detailed graphics and is rendered in the native resolution of 2048 * 1536. With brightness to the max (which is much brighter than any of my other devices) and Wifi on I managed to drain the battery in 4 hours. With half brightness (similar to full brightness on other devices) 6 hours are possible, up to 8 hours with low brightness + flight mode.

The iOS Version of "GTA Vice City" can run between 5 and 11 hours, "Ace Attorney 3 HD" as "best case scenario" 5 to 15 hours:

My 1 year old smartphone can play demanding games with half brightness and all connections on (cellular, LTE, WiFi and a BlueTooth-connected DualShock4) around 4 hours, no matter if it is a native Android game (GTA Vice City), an emulated PSP-game (GTA Vice City Stories) or an emulated PS1-game (Crash Bandicoot 3).

In flight mode probably more, but that's no viable solution for me since I don't want to miss calls or messages by putting my phone in flight mode for many hours.

Since I don't have a Switch (yet), I wrote down Digital Foundry's results for Zelda BOTW with full and half brightness; with and without different power banks:

This is only a place holder until I get my Switch, then I will test it with different games and more different settings.

Since one of the power banks of DF's test is similar to mine (both 15 aH / 55 Wh), I made long-term tests with most of my devices. No different settings or games... always the worst-case scenario (full brightness with the most demanding game):

Two exceptions: I chose "GTA Vice City Stories" for the PSP because I didn't want to kill my UMD drive. And I lowered the brightness for the 3DS XL and turned 3D off because that handheld transfered the energy too late while on maxed settings. 

So with a moderate power bank and if the cable to the handheld doesn't bother you (many play handhelds with a wired headphone), you can run any Sony or Nintendo handheld (besides the Switch) for 24 hours and more. A smaller sized power bank like the "Anker P'Core" recommended by Digital Foundry should still be enough to run any Sony or Nintendo handheld (besides the Switch) for over 20 hours.

Very cool experiment. I'd definitely like to see what you wind up with with a new battery if you ever get one; it's so hard to estimate as the "degradation" of the batteries is very difficult to predict.

I just ordered the 3600 mAh battery pack for my PSP-1000, but it will take a while for further tests.

Did Digital Foundry make mention of cord length when testing power banks? I imagine that'd affect the effectiveness.

That thread title...