One of the most striking trends in smartphone designs is the push towards ever-larger screen sizes. Not only has average screen size increased steadily since 2008, the so-called “phablet” market is poised to explode. Nearly 26 million crossover devices shipped in 2012, with an estimated 60 million shipping in 2013 and 146 million by the end of next year. That’s a huge growth rate for a market at a time when both tablets and smartphones are growing more slowly than in past quarters.
But what’s really surprising about phablets is that manufacturers continue pouring money into the category even as the devices grow increasingly unwieldy for many of the basic functions of a smartphone. We’ve dug into the trends behind this push – why are so many companies dead-set on selling you a device that can double as a dinner platter or indeed operating table?
Balancing performance and power consumption
One of the most straightforward explanations for the increase in screen size is that modern smartphones have to be larger to dissipate the heat their own components generate. There’s plenty of good baseline evidence for that conclusion.
Smartphone performance has increased far more rapidly than Li-on battery capacity, and high density screens require more backlighting, which increases temperatures and draws more power. The reason we haven’t seen an even stronger link between increasing smartphone capability and increasing size is because manufacturers have fought back with better power management and more advanced hardware.
Take Wi-Fi and 4G LTE. For any given radio, higher bandwidth means higher power expenditure, but higher bandwidth can also mean that a page transfers to the device more quickly, thus allowing the radio to spend more time in sleep mode. If you surf 3G and 4G LTE at the same speed and read the same websites, 4G may well deliver better battery life. If, on the other hand, you use your 4G connection to start streaming video, or you surf more websites within the same period of time, 4G ends up costing you battery life relative to 3G. The difference lies in how you use the device. Another reason why 2013 devices were able to show improvements over early 2012 hardware is the availability of newer, 28nm lower-power radios.
Then there’s battery density. Battery energy density has improved in recent years, but the rate of increase has slowed significantly. We calculated these figures by taking the watt-hour ratings on each smartphone battery divided by the battery volume in square millimetres. This gives us a normalised rating of battery efficiency. The iPhone 3G used a lower density lithium-polymer battery (Li-poly) rather than lithium-ion; the Galaxy S2’s battery packed significantly more power.
The rate of increase slows substantially as we approach the present day. The S2 is 53 per cent more dense than the iPhone 3G, the S3 is 27 per cent more dense than the S2, and the S4 improves on the S3 by around 10 per cent. The iPhone 5 and S4 batteries pack roughly the same amount of energy into their respective volumes. Historically, the performance of lithium-ion batteries increases at about 10 per cent a year, which means Apple and Samsung may be bumping up against the material limit of the underlying tech.
Heat dissipation is the other major factor to consider. A quick check of reviews and coverage around the web confirms that while the S4 has improved battery life over the S3 thanks to features like a 28nm modem, one fact holds true – run these phones up to full power, and they get hot (and chew through battery life) very, very quickly.
Given that tablets and SoCs regularly use the same components, the limiting factor for how much time a system spends at its “Turbo” setting is often limited by form factor. A tablet can dissipate more heat than a cell phone, which translates directly into longer burst operation. While neither the Galaxy S4 or Google Nexus 7 have an active cooling system, it’s not hard to see that the S4’s “motherboard” packs far more components into a much smaller space.
Data suggests that larger devices are increasingly important to manufacturers that want to showcase products that are meaningfully faster than the hardware they ostensibly replace. Smaller process geometries and more efficient software may have pushed back against power consumption, but without historic Dennard scaling, we’re only blunting the power consumption trend, not reversing it. If you want big performance and acceptable battery life, you’ve got to increase the form factor. Meanwhile, our obsession with thinness leaves the mobile form factor with only two dimensions to reasonably increase.
This means that there may well be a trade-off between buying the most powerful devices and buying devices that fit comfortably in the hand – but the performance of even modest handhelds is high enough now that consumers may simply regard this as a form factor question rather than an issue of performance. We already see in this in the laptop market, where devices with 11in to 13in screens are just as fast as 15in to 17in models for mainstream workloads. A larger chassis only become necessary if you specifically want a large screen, or if you need serious gaming/workstation-level horsepower.
Easy product differentiation
There are multiple intersecting forces driving the adoption of larger screen devices. First, there’s the fact that screen size is an easy sell. Apple was the first smartphone vendor to leap for higher pixel densities, but the iPhone 5’s screen is only modestly larger than the iPhone 4/4S. That gave Android manufacturers an easy, highly visible way to differentiate their products from the ubiquitous iPhone family. Why buy a Galaxy S4 or Galaxy Note over an iPhone 5? Look at that screen!
Samsung has also used its AMOLED technology to create a particular, highly saturated look and feel to its flagship Android products that also adds to the feeling of visual distinctiveness. Large screens can also be argued as an immediate distinguishing characteristic between Android and Apple products. Apple’s lawyers rather famously squashed the aspect ratio on Samsung devices to make them appear more infringe-y, but it’s hard to imagine a lawyer holding up a Galaxy Note 2 in one hand, and Apple iPhone 5 in the other, and claiming: “Your Honour, consumers could easily confuse these products.”
Another potential reason for the rise in phablets is that for manufacturers like Samsung, they’re a much more lucrative business. The Samsung Galaxy Tab 10.1 (Wi-Fi only) had a materials cost (referred to as a bill of materials, or BOM) of $260 (£170) and an initial US list price of $500 (£320). The Galaxy S4 has a BOM of $236 (£153) and a retail price of $627 (£405). The 2012 Nexus 7, meanwhile, has a list price of $200 (£130) for the 8GB mode and $250 (£160) for the 16GB. BOM on those devices, including manufacturing? $159 (£103) and $167 (£108). I couldn’t find a BOM on the Galaxy Note 2, but the US retail price for the system – which sports a 5.55-inch screen – is $500 (£320).
The maths jumps out immediately. Customers are used to mobile phone prices that offer enormous margins. The success of the Galaxy Note series proves that they’re willing to pay for combined products. But the best part of all, from Samsung’s perspective, is that it doesn’t have to worry much about discounting the product. Contract phones with so-called “discounts” still dominate the US (and indeed UK) market, which means it’s the networks that pay the cost of discounting phone hardware – not Samsung.
Data from IHS Suppli on tablet margins puts some additional context on this. Low-end 7in tablets have the worst margins. The iPad mini is stronger, but its comparatively low price means Apple makes significantly less money. Assuming these margins are accurate, Apple earns $131.60 (£85) per iPad mini and $184.63 (£120) per iPad – a difference of nearly 40 per cent. Widespread phablet adoption is a “have your cake and eat it too” for Samsung and other vendors – these devices offer high margins, high absolute dollar/pound values, and, thanks to contracts and device discounting, very little downside.
The smartphone and tablet markets are cooling off
More specifically, the growth rate of the smartphone and tablet markets is cooling down. These markets are still set to explode as far as total sales expansion goes, and North America is expected to account for some 47 per cent of tablet sales worldwide in 2013. Nevertheless, if the cool-down continues, investors are going to start expecting companies to step back from the frenetic expansion and often overlapping products of the past few years and focus on a handful of designs that work.
Phablets are a way to sell a potentially price-conscious market on a combination device. They’re simultaneously conservative, in that they play to people’s desires for one device instead of two, particularly if the economy remains weak, but still daring in that they upscale the smartphone to form factors that really aren’t conducive to making calls. In fact, that’s one of the most striking things about phablets – their existence is a nod to the fact that for most people, phone calls are only a tiny percentage of what they do on a mobile computing device.
So why the push for phablets? It’s a play for margins. It’s an attempt to shoehorn more powerful silicon and the larger batteries such devices require into form factors that can use them. It’s a test to see if consumers want more crossover devices, and what sort of crossover devices they want. It’s a lot of things, and there’s no telling if we’ll look back at phablets as the logical combination of smartphone and tablet the way we look at the PDA/smartphone integration of years past, or if they’ll become the semiconductor equivalent of tail fins – a brief, enormously popular surge of interest followed by a sharp falloff as consumer tastes changed and people realised that the fins, while cool looking, offered very little practical benefit.
In absolute terms, this means that smaller devices are going to be less powerful than larger ones. The iPhone 5’s sales figures, and the fact that companies like Nokia are aggressively focusing on the mid-to-lower range markets, suggests that this will not be a problem going forward.