The consumer electronics market is converging. Where you once had a work computer, a home computer, a game console, a TV, a DVD player, and a mobile phone, in 2013 it is not unusual to simply have a laptop and a smartphone, or perhaps a smartphone and a tablet. This has led many people to believe – myself included – that the PC will eventually die out, to be replaced by smartphones that are infinitely more portable and flexible.
One argument against the supersession of smartphones, however, is the small matter of desktop PCs having more processing power due to a larger power envelope (TDP) and better heat dissipation. There will come a time, though, probably in the next few years, when processors are so small and energy efficient that smartphones will have enough processing power to meet the needs of all but the most discerning consumers.
An even stronger argument against the rise of smartphones is their lack of customisation and upgradeability. As it stands, smartphones (and tablets) are designed, marketed, and sold as consumable devices. You buy a smartphone, chew through its storage and battery for a couple of years, and then throw it away when your two-year contract comes up for renewal.
Carriers love this, of course, because it keeps you locked into an incredibly lucrative contract. As far as manufacturers are concerned, the two-year upgrade cycle represents the most orgiastic piñata whacking ever devised – just look at Apple’s record profits, which stem almost entirely from monumentally massive iPhone sales. Do any of your other gadgets get upgraded every two years? Maybe your laptop, if you’re rich, but otherwise the consumer electronics upgrade cycle is usually nearer four or five years.
If smartphones were upgradeable, there would be significantly less reason to buy a new one every couple of years. If you could simply slide in a new processor, RAM, and battery, your smartphone’s useful life could be extended almost indefinitely – just like a PC. Likewise, if carriers and manufacturers didn’t leave older devices to languish with old versions of iOS, Android, and Windows Phone, then consumers would have very little reason to upgrade. In both cases, though, the consumer would end up buying less phones – and so it isn’t really surprising that neither the manufacturers or carriers are attempting to improve either the hardware or software situation.
The other problem with upgradeable hardware, of course, is that it’s hard to create a device that is small, powerful, efficient, and upgradeable. There are certain trade-offs intrinsic to the minimisation of computer hardware, and losing the ability to replace individual components is one of them. Even desktop PCs are not immune to this axiom of computing – in the olden days, almost any chip could be unplugged from a circuit board, but today, almost everything is soldered except for the CPU, RAM, and graphics card. (Do you remember when graphics cards had sockets for more expandable RAM?)
Laptops used to have replaceable CPUs and RAM, but that’s slowly becoming a thing of the past as they miniaturise, too. Smartphones sometimes have a replaceable battery and microSD card, but many have no replaceable parts at all. (The iPhone 5’s main logic board, pictured below, crams an entire computer into roughly the same volume as the CPU socket shown in the image above).
The simple matter of the fact is that it takes extra space to make something upgradeable. While there’s enough space in your desktop PC to provide a CPU socket and clamp, it just isn’t feasible in an ultra-thin laptop or smartphone. Even with the shift to smaller socket types, such as LGA, you’re still talking about an additional 5mm or more – and that’s before you add any kind of cooling solution, which will probably be fiddly to remove/replace in its own right.
RAM and flash storage, while only a couple of millimetres high when surface-mounted on your smartphone’s logic board (again, see the above pic), expand to maybe two or three times their original volume when they’re attached to their own removable PCB and slotted into some kind of socket. Replaceable mobile graphics cards do exist (and are used in some laptops), but again they’re a lot bigger than integrated GPUs; there isn’t really enough space in modern smartphones for a separate CPU and GPU.
I have seen the future
It might not always be this way, though. In the future, it’s possible that the components will be so small that replaceable parts again become a reality. After all, a smartphone can’t really get much smaller, otherwise we won’t be able to hold it or see the screen – but the components inside will continue to shrink. This means that there might eventually be enough spare volume to cater for the sockets, mounts, and extra layers of PCB that replaceable parts require.
Also due to continued consolidation and integration, it wouldn’t be a surprise if future mobile devices squeeze almost their entire hardware into a single chip. For now, the system-on-a-chip (SoC) in your smartphone or tablet has: A CPU and GPU; a memory controller; dedicated hardware for GPS, USB, still images and video, and digital signal processing (DSP); and radios for Bluetooth, Wi-Fi, and 2G/3G/4G.
When a smartphone is essentially a battery, display, and a single chip, it suddenly becomes rather easy to upgrade. In much the same way as you replace a SIM or SD card, you might just slot in a new SoC. With reduced complexity, it will be a lot easier to provide a replaceable battery. The display will probably be the only bit that isn’t easily upgradeable – but that’s okay; you can just buy a new phone if you need a new screen.
Vote with your wallet
The good news, then, is that upgradeable phones are a distinct possibility in the next few years. The problem is, much in the same way that the expensive-but-completely-non-upgradeable MacBook Pro with Retina display outsells upgradeable Windows laptops, is that there may never be enough commercial demand for an upgradeable smartphone. I would not be surprised if someone like Samsung or Asus produces an upgradeable smartphone in the next few years – but whether that device will actually sell is another matter entirely.
You see, the vast majority of the market is more than happy to keep buying non-upgradeable laptops, tablets, and smartphones. The cost of ownership might be higher, but that’s balanced out by improved aesthetics, reliability, and functionality as well (when you design a device that will never be opened up, you can cram a lot more in there). The idea of replacing your smartphone’s SoC is nice, but in practice I don’t know if there’s enough market demand.
When an upgradeable smartphone finally comes to market, then, be sure to buy it. If a million people buy an upgradeable Samsung Galaxy S6, Samsung will be forced to produce upgraded SoCs that you can slot in. Other smartphone makers, foreseeing the massive backlash of continuing to produce non-upgradeable devices, will be forced to follow suit. Voilà: We will finally have smartphones that last as long as the rest of your consumer electronics.
Finally, one last thought. There are two driving, diametrically opposed forces at play in the computing market: Commoditisation and differentiation. On the one hand, the internal parts that make up a computer are mostly commodities; there is some differentiation on the bleeding edge, but for the most part 8GB of RAM or a Cortex-A9 SoC is much the same wherever you get it from.
On the other hand, device makers are trying as hard as they can to differentiate their devices from the opposition. This is a hard task that can only be somewhat ameliorated by different/better software – and over time, as software features are copied by competing operating systems and every device has functionally the same hardware, the difference between any two smartphones will creep towards zero.
Just like the car market, just like the TV market, and just like the PC market, the smartphone market will be commoditised – and then, and only then, smartphones will become the PCs of the future.
Image Credits: CPU World, Hardware.info