Older Can Be Faster

2026-05-11

---

A few weeks ago, while helping to clean my parents' house, I found my dad's iPhone 3GS. Everyone in the family owned a refurbished 3GS from early 2011 to late 2012, but I've long since recycled mine, so coming across his brought back some long-forgotten memories. I brought it back to my house to play with it.

One thing stood out to me almost immediately. The smartphone runs iOS 5.0, which released almost fifteen years ago in October 2011. However, the Maps app still loads current data from Google Maps--and the tiles load faster than they do on my wife's iPhone 16.

The curse of powerful modern computing devices is that the software they run usually swells to take up all the extra power. Web browsers that initially used a few megabytes of RAM now need several gigabytes. Word processors add bloat by orders of magnitude. Files grow exponentially in size and complexity, requiring more processing power to decode in real time, more bandwidth to transfer, and more storage space to save. Code libraries, designed for efficiency, are piled one on top of another. Developers pull in hundreds of dependencies for simple actions, secure in the knowledge that if their programs are slow, improved performance in the next hardware iteration will cover for them. Vibe coding with AI makes the situation even worse.

Older devices, subject to very real and unavoidable hardware constraints, needed more efficient software. Limited memory had to be used wisely. Slower processors necessitated efficient computation and the establishment of strict priorities. Small storage volumes meant that programs had to choose carefully what to store locally. Small, low-resolution displays required simpler visual layouts. Ads in apps were few and far between--and fullscreen ads were completely unheard of outside of malware. Programmers couldn't waste precious resources on needlessly-flashy animations, extra sound effects, huge Javascript libraries, or deep learning systems. That all led to good programming practices--or, at the least, better programming practices than we tend to have now.

When actually using older devices, usually the hardware's age and slowness counterbalance the efficiency of the software, so we don't see an improvement in speed. Thus, when I saw how quickly Google Maps loaded on this old iPhone, I couldn't stifle a laugh. I love it when old computers genuinely perform better than new computers simply because they're programmed better.

The funniest aspect of it to me is that I remember the smartphone climate of the early 2010s. The iPhone 3GS was considered old and slow by the standards of the day, yet at least one of its apps performs faster than equivalent apps today.

---

Up One Level

Home

[Last updated: 2026-05-11]