Sunday Debate: Are We Close to a Performance Plateau?
Join us in a fun Sunday Debate on Performance. Come with your opinions and feel free to read some of our thoughts, then pick your side or play devil’s advocate to get your voice heard and engage in friendly discussion. You can read our food-for-thought or jump straight into the fray below!
At XDA, we love optimizing every software bit to get the most out of our phones. And while we apply the grease, there’s only so much we can squeeze out of the oiled gears and cogs within our phones. Processors keep growing in power – number of CPU cores, upgrades in top frequency, new architectures coming to the scene and increasingly smaller processes are all the ones that immediately come to mind, but the advances in memory and storage speeds – and their clever implementations, such as ePoP – also are a big, relevant factor in the speed and fluidity of our phones.
Yet many users feel that performance has reached a plateau, and that increasingly fast processors and more RAM are no longer as needed as they once were. Mind you, this does not take away from the need for more efficient silicon packages — we can still use the power saving. But due to all the system optimizations of the Android platform plus the big increments in chipsets’ practical performance, theoretical maximums seem to be taking much more of a back seat in power user minds. Notable phones now come with year-old processors (like many Chinese phones) or newer ones without as much of a speed boost (like the LG G4) despite being considered premium flagships as well.
Did we really hit a plateau in performance? If so, what reasons do we have to focus on processing speed? Do consumers value tremendous performance less and less? Should we start caring about other things beyond smoothness now that most flagships and mid-rangers have solid performance? As always, we will put forth some of our ideas on the topic, but feel free to skip arguments and chime in if you feel like you have your mind made up.
As fluid as our user interfaces are, there is still a lot of untapped potential that OEMs want to touch on. While Android hasn’t needed insane specifications since KitKat (which had aimed to optimize operating system for low-end devices), not all of our smartphone tasks are necessarily low-intensity jobs. While it is true that users that do not have much gaming under their daily belts do not need the fastest CPU nor GPU to handle typical smartphone operations, future proofing is a big part of many users’ concerns when factoring in a purchase. What many ignore is that upcoming developments will need very efficient and powerful processors. Virtual Reality, for example, is becoming a predominant force in the upcoming tides of technology, and with mobile OEMs like Samsung and HTC attempting to grab the early slice, the extra power will be needed to drive the increasingly denser screens that the new technology will require.
But those of you who don’t want to spend much in Virtual Reality media, or don’t care for the platform at all, are not left out of the virtues of a strong processor. Regular application performance typically get optimized; if not by developers, they are taken care of by Google’s systemic platform enhancements. Things like games, however, require the extra performance to iteratively improve their graphics and physics. But outside of smartphones themselves, future developments such as the Internet of Things will require more efficient processors, as phones will act as one of the central hubs for many of the calculations that your personal and connected devices will need. And ultimately, more powerful processors can handle future operating system upgrades more reliably, as well as whatever new software developments the platform sees. A higher maximum theoretical performance with the proper optimization and scaling also means less power consumptions on the tasks that do not require a lot of power, and the “race to sleep” effect can also be very beneficial for an array of tasks.
Other Things, Please!
The other side of the argument is that new processors and their development and fabrication are typically costly, and when you factor in the fact that battery life did not substantially improve in 2015 smartphones (so far) it is clear that some of the things that consumers want the most are being overshadowed by the need for speed. In fact, a particularly crippling factor for this generation’s failures and stagnation was faulty hardware and performance from the Snapdragon 810, which arguably condemned a bunch of flagship phones to subpar performance and worse battery life. With Samsung investing billions in chip manufacturing factories, some users feel that those resources could be better spent in things such as battery technologies or even software advancements. Truth be told, no OEM has infinitely deep pockets, and a high upfront cost for a performance upgrade that might have very few noticeable enhancements may be considered an unworthy investment.
The fact that smartphones today are increasingly connected to cloud services is also an argument against the insane resource-focus of current smartphones. Google Now, for example, handles most calculations in the cloud, and with each passing year we hear more about this system approaching more vital smartphone functions. Videogame and media streaming in general can mitigate a processor’s power consumption, and services like that of Nvidia and the decrepit Onlive might make a comeback once better network speeds arrive (although on mobile, the spectrum crunch holds it back) and Virtual Reality becomes mainstream. Finally, as the display race ends with what most likely will be 4K (anything beyond that would be unpractical and only benefit VR consumers), the need for huge yearly performance gains could see a big deceleration (I personally run my 1440p Note 4 at 1080p for better battery life and performance, as 1440p is rather useless to me). The lack of resolution jumps would still leave gaming graphics evolution and the points featured above as possible necessities, however, but at this point we cannot predict the popularity that VR and IoT will amass in the short term (but to be fair, a lot of it depends on these processors).
On one hand, performance bumps in flagship smartphones do not amount to the same real-world user experience improvements that we saw in previous generations. If we did indeed hit a practical plateau, our devices could benefit from focusing on other key areas of the user experience. But on the other hand, these performance upgrade, while not as immediately or directly tangible as before, can unlock future technologies and pave the way to a more connected world and a new kind of realities.
- Do you think performance in today’s smartphones has hit a practical plateau?
- Do you personally need better performance out of your phone? What relative value do you give this when considering a phone?
- Do you believe OEMs should keep focusing so much on the latest chipsets, or on optimizing other aspects of the UX?
- How important are better chipsets are to your user experience, and how important do you think they will be in the future?