58 billion transistors đ
AMD Navi 31, 2394 MHz, 5,376 Cores, 336 TMUs, 192 ROPs, 20,480 MB GDDR6, 2,500 MHz, 320 bit.
âThermal issues are currently one of the biggest bottlenecks that are plaguing any kind of microelectronics,â says team lead Srabanti Chowdhury, professor of electrical engineering at Stanford University. âWe asked ourselves, âCan we perform device cooling at the very material level without paying a penalty in electrical performance?ââ
Indeed, they could. The engineers grew a heat-wicking diamond layer right on top of individual transistorsâtheir hottest pointsâas well as on their sides. Heat flowed through the diamond to a heat sink on the back of the device. With this technique, the researchers achieved temperatures 100 degrees Celsius lower without any degradation of the deviceâs electrical properties. They will report their findings in San Francisco at the IEEE International Electron Device Meeting in December.
They demonstrated their technique on gallium nitride (GaN) high-electron-mobility transistors, or HEMTs. GaN is the go-to alternative to silicon for high-frequency applications, as it can sustain higher electric fields and responds more quickly to electric field changes. GaN also breaks down at a higher temperature than silicon. But not high enough: âIf you go by the physics of the material, you see what its potential is, and weâre nowhere close to that today,â says Chowdhury. Keeping GaN HEMTs cool as devices shrink and frequencies grow will allow them to live up to their physics-promised potential.
In September, Apple announced a new wearable called the Apple Watch Ultra, and one of the companyâs key pitches for the device was its use as a diving computer. Now Oceanic+, the app that makes that feature possible, launched exclusively for the Ultra, Apple announced today.
A lot of the features focus on either planning dives in advance or viewing dive reports after youâre done, but for those that you use underwater, the app utilizes haptics to send you alerts. The Watch Ultraâs very bright screen can help with legibility underwater, too.
Quantum memory devices can store data as quantum states instead of binary states, as classical computer memories do. While some existing quantum memory technologies have achieved highly promising results, several challenges will need to be overcome before they can be implemented on a large scale.
Researchers at the AWS Center for Quantum Networking and Harvard University have recently developed a promising quantum memory capable of error detection and with a lifetime or coherence time (i.e., the time for which a quantum memory can hold a superposition without collapsing) exceeding 2 seconds. This memory, presented in a paper in Science, could pave the way towards the creation of scalable quantum networks.
Quantum networks are systems that can distribute entangled quantum bits, or qubits, to users who are in different geographic locations. While passing through the networks, qubits are typically encoded as photons (i.e., single particles of light).
Researchers at the UPCâs Department of Electronic Engineering have developed a new type of magnetometer that can be integrated into microelectronic chips and that is fully compatible with the current integrated circuits. Of great interest for the miniaturization of electronic systems and sensors, the study has been recently published in Microsystems & Nanoengineering.
Microelectromechanical systems (MEMS) are electromechanical systems miniaturized to the maximum, so much so that they can be integrated into a chip. They are found in most of our day-to-day devices, such as computers, car braking systems and mobile phones. Integrating them into electronic systems has clear advantages in terms of size, cost, speed and energy efficiency. But developing them is expensive, and their performance is often compromised by incompatibilities with other electronic systems within a device.
MEMS can be used, among many others, to develop magnetometersâa device that measures magnetic field to provide direction during navigation, much like a compassâfor integration into smartphones and wearables or for use in the automotive industry. Therefore, one of the most promising lines of work are Lorentz force MEMS magnetometers.
The neuroscience of perception has recently been revolutionized with an integrative modeling approach in which computation, brain function, and behavior are linked across many datasets and many computational models. By revealing trends across models, this approach yields novel insights into cognitive and neural mechanisms in the target domain. We here present a systematic study taking this approach to higher-level cognition: human language processing, our speciesâ signature cognitive skill. We find that the most powerful âtransformerâ models predict nearly 100% of explainable variance in neural responses to sentences and generalize across different datasets and imaging modalities (functional MRI and electrocorticography). Modelsâ neural fits (âbrain scoreâ) and fits to behavioral responses are both strongly correlated with model accuracy on the next-word prediction task (but not other language tasks). Model architecture appears to substantially contribute to neural fit. These results provide computationally explicit evidence that predictive processing fundamentally shapes the language comprehension mechanisms in the human brain.
Physicists have discovered a new quantum state in a material with the chemical formula Mn3SiTe6. The new state forms due to long-theorized but never previously observed internal currents that flow in loops around the materialâs honeycomb-like structure. According to its discoverers, this new state could have applications for quantum sensors and memory storage devices for quantum computers.
Mn3SiTe6 is a ferrimagnet, meaning that its component atoms have opposing but unequal magnetic moments. It usually behaves like an insulator, but when physicists led by Gang Cao of the University of Colorado, Boulder, US, exposed it to a magnetic field applied along a certain direction, they found that it became dramatically more conducting â almost like it had morphed from being a rubber to a metal.
This effect, known as colossal magnetoresistance (CMR), is not itself new. Indeed, physicists have known about it since the 1950s, and it is now employed in computer disk drives and many other electronic devices, where it helps electric currents shuttle across along distinct trajectories in a controlled way.
Year 2011 face_with_colon_three
UC Berkeley scientists have developed a system to capture visual activity in human brains and reconstruct it as digital video clips. Eventually, this process will allow you to record and reconstruct your own dreams on a computer screen.