Search the Community
Showing results for tags 'chipset'.
Found 2 results
AMD announced a new chipset at Computex in Taipei today which is said to challenge Intel's top of the line Core i9 9920X chipset. It's called the Ryzen 9 3900x chip that has 24 threads and 12 cores making it one of the top gaming processors today. Another impressive feat about the new processor is its price, AMD announced the Ryzen 9 3900x chipset will sell for $499, which is half the price of Intel's offering. © YouTube Currently, Intel's Core i9 9920X chipset costs t $1,189 and upwards. Even when adding import duties for Indian customers, AMD's offering is very much more affordable and may just become the preferred choice amongst gamers. Here's an exclusive #COMPUTEX2019 look at the newest edition to the Ryzen family, the 12 core/24 thread 3rd Gen AMD Ryzen 9 3900X processor. https://t.co/OgLHoqWv9T pic.twitter.com/75FzfpdiKx — AMD Ryzen (@AMDRyzen) May 27, 2019 The 3900x is clocked at 4.6 GHz boost speed and has a total cache of 70 MB. In fact, it even uses less power than Intel's i9 9920 x as it has a 105 watts of thermal design power whereas the i9 uses 165 watts. AMD claims that in a Blender demo the 3900x is 18 per cent faster than Intel's offering. To say the least, AMD has also launched more affordable chips from the same family i.e. a 6-core, 12-thread 3600 for $329 and an 8-core, 16-thread Ryzen 3700x (with 4.4 GHz boost, 36 MB of total cache and a 65-watt TDP); for $329 and an 8-core, 16-thread Ryzen 3800X (4.5 GHz, 32MB cache, 105w) for $399. The pricing on the new AMD processors are not only aggressive but makes it more affordable for PC gaming markets such as India. © AMD AMD also announced the next foundational gaming architecture i.e. RDNA. RDNA will power the upcoming 7nm Radeon RX 5700-series graphics cards, which feature high-speed GDDR6 memory and support for the PCIe 4.0 interface. Currently, the pricing has not been revealed yet but we expect it to be cheaper than Nvidia's offering.
In an effort to show off the AI computing power of the silicon in its Mate 10 Pro, Huawei chose an interesting challenge: it turned the device into a semi-autonomous driving system. RoadReader, which is being demoed at the Mobile World Congress in Barcelona this week, uses a Huawei Mate 10 phone's object-recognition AI to help a Porsche Panamera navigate without a driver. In particular, Huawei's demonstration is aiming for particular object recognition: cats, for instance, or dogs. © Reuters The Mate 10 Pro handset was able to 'see' obstacles - in this case, a dog - and could programme the vehicle to drive around the brave animal during tests. By using AI on the phone, the car was able to identify thousands of different objects including cats, balls, and bikes, and take action to avoid a collision. The tech giant said it was "the first mobile device manufacturer in the world to use an AI-powered smartphone to drive a car,". The technology was able to distinguish between thousands of different objects as well. Huawei was quick to assure us that this project in no way meant it was entering the connected car space — it was just a showcase of the Kirin 970's on-device neural processing capabilities, which currently grace the Mate 10 and are expected to touch down in the Huawei P20. © Huawei Limitations in cloud AI necessitated improvements across latency, stability, and privacy, Huawei said, with on-device AI providing this as well as adding sensor data to the offering. Huawei's AI push also saw it sign a strategic agreement with Chinese search engine giant Baidu in December to build an open mobile AI ecosystem that covers platforms, technology, internet services, and content ecosystems. The open ecosystem will be built using Huawei's HiAI platform and neural network processing unit (NPU), and Baidu's PaddlePaddle deep-learning framework and Baidu Brain, which contains Baidu's AI services and assets. It will allow AI developers to make use of the technology in better and faster ways.