The Microsoft Surface Laptop 2 Review: Surface Essentials

See the original posting on Anandtech

Microsoft’s Surface lineup was created to bring a spark of innovation into the PC industry at a time where much of the competition was slow to change, and slow to adopt new form factors and new technologies. Microsoft’s Surface Pro lineup has undoubtedly been a huge success in this respect, with the 2-in-1s providing plenty of flexibility coupled with great hardware.

Unsurprisingly then, Microsoft has taken this success and run with it, growing the Surface brand by fleshing out the product line with more models. However even as Microsoft expanded the Surface family, they have always tried to keep that same edge – always embracing a unique feature on their lineup to differentiate a Surface device from the competition. Surface Pro had the kickstand, of course. Surface Book is a laptop with a detachable display. Surface Studio is an all-in-one PC that can fold into a drafting table.

But even with the Pro as a successful template for how to build out the Surface family, Microsoft has one product that doesn’t really fit in with the rest, and that is the Surface Laptop. There are no tricks or unique chassis features here. Microsoft just set out to create a thin and light laptop to fill a void where people want to buy a Surface, but want to use it in their lap; and they don’t need the performance, heft, or price of the Surface Book. It’s a simple concept for a company that’s been more focused on distinctive designs, and, as we’ll see, one that helps tap an important segment of the notebook market.

Huawei Launches the P30 and P30 Pro: P is for Photography

See the original posting on Anandtech

The latest update in Huawei’s P series smartphones are the new P30 and P30 Pro handsets, launched today in Paris. These new devices hope to push the reason why the P series exists – to go above and beyond when it comes to photography. The P30 and P30 Pro include Huawei’s new rear SuperSpectrum camera that allows more light into the sensor, and a front facing 32MP camera for better selfies.

Google Announces Stadia: A Game Streaming Service

See the original posting on Anandtech

Today at GDC, Google announced its new video game streaming service. The new service will be called Stadia. This builds on the information earlier this year that AMD was powering Project Stream (as was then called) with Radeon Pro GPUs, and Google is a primary partner using AMD’s next generation CPUs and GPUs.

Stadia is being advertised as the central community for gamers, creators, and developers. The idea is that people can play a wide array of games regardless of the hardware at hand. Back in October, Google debuted the technology showcasing a top-end AAA gaming title running at 60 FPS. Google wants a single place where gamers and YouTube creators can get together – no current gaming platform, according to Google, does this.

Ultimately Google wants to stream straight to the Google browser. Google worked with leading publishers and developers to help build the system infrastructure. Google is one of a few companies with enough content delivery networks around the world to ensure that frame rates are kept high with super low latency.

Users will be able to watch a video about a game, and instantly hit ‘Play Now’ and start playing the game in under five seconds without any download and lag. The idea is that a single code base can be enjoyed at any stream. At launch, desktop, laptop, TV, tablets, and phones will be supported. With Stadia, the datacenter is platform. No hardware acceleration is required on the device. The experience can be transferred between devices, such as chromebook to smartphone.

One of the highlights of Google’s demonstration of Stadia was the platform working on Google-enabled TVs.

The platform allows users to have any USB connected controller, or mouse and keyboard. Google will also be releasing its own Stadia Controller, available in three colors – white, black, and light blue. The controller connects via Wi-Fi straight into the cloud, and also which device is being run (it’s unclear how this works).

The controller has two new buttons. The first allows saving and sharing the experience out to YouTube. The second is Google Assistant, using the integrated microphone in the controller. This allows game developers to integrate Google Assistant into their games. It also allows users to ask Google when they need help in a game – and the assistant will look for a guide to help.

Stadia uses the same datacenter infrastructure already in place at Google. There are 7500+ edge nodes allows for compute resources being closer to players for lower latency. Custom designed, purpose built hardware powers the experience. Interconnected racks have sufficient compute and memory for the most demanding games. The technology has been in development inside Google for years.

At launch, resolutions will be supported up to 4K 60 fps with HDR and surround sound. Future  plans for up to 8K streaming at 120 fps are planned. The platform has been built to scale to support this. While playing, the stream is duplicated in 4K for direct upload – you get rendering quality video rather than what you capture locally.

The platform is instance based, so Google can scale when needed. Game developers no longer have to worry about building to a specific hardware performance – the datacenter can scale as required.

Custom GPU with AMD, with 10 TF of power, with a custom CPU with AVX2 support. Combined they create a single instance per person. Uses Linux and Vulkan, with full Unreal and Unity support. Havok engine support as well. Tool companies are onboard.

(When Google says custom CPU and custom GPU – this could be early hardware of AMD’s upcoming generations of technology, put into a custom core configuration / TDP. We’re likely looking at a Zen 2 based CPU, based on AVX2 support listed, and a Radeon Instinct based GPU with tweaked settings specifically for Google.)

One of the first games supported will be Doom Eternal from id Software, which will support 4K with HDR at 60 fps. Every user will get a single GPU with no other users.

UL Benchmarks (3DMark) has been working with Google to help benchmark the systems and measure the power of the infrastructure. Developers if required can use multiple GPUs, it appears.

Multiplayer is also supported, at least between different Stadia players. Distributed physics becomes possible, which means up to 1000 players in Battle Royale titles. There’s also the advantage, according to Google, of getting around hackers and cheaters.

Developers can support multi-platform multiplayer, and transfer save files between platforms. Game developers have already been working on MP demos with destructive environments using real-time rigid body physics, allowing for perfect synchronization.

Google also points out that split-screen gaming has not been a priority recently because of rendering two scenes at once. With Stadia, that problem disappears, as each player will be powered by a separate instance, reviving the idea of local co-op and squad based gaming. This also allows for multiple cameras for a single player to navigate a single map, for better tactics in certain types of games. Google says that this ability allows developers to create new types of games.