the updated Tegra X1 thoroughly tested • Eurogamer.net

Discovered for the first time in the firmware 5.0 of Switch in March 2018, the Nintendo hybrid console evolves with a new Tegra X1, smaller and colder. Efficiency benefits are the foundation of the next portable Switch Lite, but Nvidia SoC's upgrade is also central to a revised version of the existing model, which has recently arrived on store shelves in Hong Kong. Eager to see how the new silicon works, we imported a unit and got to work.

On a superficial level at least, little separates this update from the original model switch. The outer packaging is essentially the same, apart from a new theme headed by Joy-Con on one side of the box. Inside the package, nothing has changed – Joy-Cons and the tablet rely on a top layer, with peripherals (docking station, HDMI cable, power supply, etc. .) arranged in their respective lower compartments, exactly as they have always done. A quick check confirms that, despite the reduced power requirements of the new processor, the AC charger is identical to the old one and it goes without saying that all the existing Switch devices that we had to work were working perfectly on the new machine.

So this is a kind of "silent update" for Switch, a replacement model that is coming out with a fanfare, so much so that Nintendo's chief, Doug Bowser, does not seem to think of it as a new hardware launch. On the one hand, this ensures that the entire marketing strategy of Nintendo can rely on the Switch Lite, but it means that the ability to even slightly improve the issues related to the original version n & rsquo; Was not taken. The unstable crutch still does not feel strong enough to count, while the wharf remains unchanged despite its rather "agricultural" build quality, not to mention the many complaints about it over the years.

The biggest missed opportunity – and the most annoying thing I've experienced with the switch since it was launched – is that the poor performance of Wi-Fi also seems to remain unchanged, as the Nintendo machine is less likely to connect to signals. weaker and more inclined to drop its connection if it does. Side-by-side tests with smartphones and even an Amazon Echo Dot did not reveal any problems, even though they probably had less internal space for a decent antenna. The updated switch has a new motherboard. There are revisions beyond the new Tegra processor – but improving the overall quality of the device was clearly not a priority for Nintendo.

However, the new Tegra X1 – codenamed Mariko – is indeed a big problem. The original chip was manufactured according to TSMC's 20 nm process, a "step-by-step" technology used primarily by smartphones and ignored processors, GPUs and gaming console systems-on-chip, alongside Switch. We know that the new chip is smaller and the announcement of Switch Lite has strongly suggested benefits in terms of efficiency. It is likely that TSMC's 16-nm FinFET is in the frame, a process identical to that used for the latest Microsoft and Sony consoles. In other words, the energy savings achieved by the new technology are enormous.

Using The Legend of Zelda: Wild Breath and Fast RMX to test out comparable gaming scenarios, the power meter delivers impressive results in terms of power consumption from the wall. Even in docked mode, when the high-end GPU frequencies of the switch come into play, we expect a reduction in power consumption of 40 to 50%, depending on the content. What is even more remarkable is that many games in Zelda consume only six to seven watts of consumption. Faced with the latest and most powerful mobile processors on 7nm technology, this is probably not a problem but compared to the original Tegra X1, it's a big win.

Nintendo retains the original battery HAC-003 used in the original switch with its capacity of 4310 mAh, which means that this reduced power consumption results in an impressive improvement in the battery life. battery life. At 50% brightness, Zelda lasts three hours and five minutes on the old model, then five hours and two minutes on the new model, an improvement of 64% of the game time. If you increase the brightness up to at 100%, the gap is widening more and more – two hours, 25 minutes of launch, a switch means four hours, 18.5 minutes thanks to the new processor, or 78% of additional play time (but note that in all the cases, different screens may different levels of brightness).

The use of less energy results in improved thermals – but perhaps not to the degree that you might expect. The original switch is always a device with low power consumption, especially when you play in mobile mode. When moored, the heat from the exhaust zone changes from baseball at 54 degrees Celsius to 50 degrees on the new switch, but it is even more remarkable that the fan noise of the unit is reduced (and he was hardly invasive). The machine is always warm to the touch when you play without being moored, but it has the same general feeling.

Until here, the story is quite simple. It's the same switch with a new processor, which means efficiency gains and battery life, but not much more. However, there are some curved balls. First of all, the display has changed. The Wall Street Journal has announced an agreement for Sharp to supply its IGZO display technology to Switch, but it is unclear whether this concerns the Lite model or this one as well. What I can say is that side by side, the difference between my old model and the new one is striking. Although my existing switch has a slight cyan / yellow hue, white is usually white. Meanwhile, on the new model, the magenta hue is more noticeable.

The above comparison photo (click on the thumbnail to get a higher resolution) is derived from the same photo, both switches operating at maximum brightness. The difference therefore does not lie in the changes to the settings of the camera. I would say it's exactly the two screens look me in the eye side by side. Given the choice between them, I prefer the old switch, although the battery life data suggest that the new display is significantly more efficient at maximum brightness.

The other curiosity concerns the performance. Nintendo has made it clear that the new Mariko processor offers no advantage over the existing model, and comparable tests on multiple titles suggest that this is the case. The new SoC is capable of performing faster GPU clocks based on Nvidia data, but Nintendo has decided not to use them – at least for now. However, the new switch does not only exchange the old Tegra X1, it also replaces the 4GB memory of LPDDR4X, providing further energy efficiency improvements over the existing model. This may explain some oddities in a minority of performance tests.

For example, the Mortal Kombat 11 playback mode offers the ability to re-execute battles multiple times in the game engine. Nintendo Online's cloud backup transfers allow us to compare the two units with the same content . The new switch offers a very slight performance advantage, which has remained constant over five separate runs on both machines. At the same time, testing the classic Korok forest bottleneck in Zelda: Breath of the Wild – a known memory bandwidth problem – again suggests a change in the performance of some descriptions on the new Switch model . The camera's point of view plays a crucial role in this test, and it's impossible to replicate it 100% between tests, but the results seem to indicate a benefit for the new Switch.

If the theory on the new configuration of the memory offering an advantage loses its effectiveness, it would probably only appear in rare scenarios and explain why the differences are not conclusive or are excluded for other games where the GPU is clearly the limiting factor of gaming performance. Anyway, here and now, it's more of a technical curiosity than anything that, even remotely, deserves to be taken into account in any decision of any kind. 39; purchase. In general, the new Switch plays like the old one, exactly as Nintendo promised.

In summary, this is the switch we know, but with a change of display and a considerable improvement in the efficiency and the life of the battery. Given the choice between the old and the new, I would take the revised model – simply because the extra battery life is highly appreciated when most of my own Switch gaming conditions are in mobile mode . However, for those who play mainly at the dock, the two machines are virtually interchangeable – the differences in fan noise are very minor. If you already have a switch, you do not need to upgrade unless you really need more battery and do not like to use USB-C power banks.

With new stocks likely to arrive in-store with the existing stock still on the shelves, it can be difficult to know which machine you actually buy? My advice here would be to ask to see the physical tablet itself. Look at the back and make sure the product code is HAC-001 (-01) – the older switches carry a simple HAC-001 designation (here is a comparison picture). It sounds pretty simple, but confusingly, Nintendo has just released new instructions suggesting that extended battery life models will have a HAD designation – so model numbers could adapt to the western market. In any case, these are the key things to watch for.

Otherwise, if the handheld game is your main goal, it may be worth the wait for the next Switch Lite. It's not as versatile as the standard model (it's a switch that will not change) and the battery life is not as good (according to Nintendo's numbers) – but it might well have its own charms and obviously it be cheaper. Our own John Linneman has seen the Lite at Gamescom and thinks it looks great and highly desirable. We will have a closer look at the exit approach.