America’s new supercomputer beats China’s fastest machine to take title of world’s most powerful

America’s new supercomputer beats China’s fastest machine to take title of world’s most powerful

Summit is a stepping stone toward a world of exascale computing.

The winner: The US Department of Energy’s Oak Ridge National Laboratory in Tennessee has taken the wraps off Summit, which boasts peak computing power of 200 petaflops, or 200 million billion calculations a second. That makes it a million times faster than your typical laptop.

The loser: China. Summit is 60 percent faster than the previous supercomputing leader, the Sunway TaihuLight based in the Chinese city of Wuxi. Consolation prize: China still boasted way more entries than the US in a list of the fastest 500 supercomputers published last year.

AI smarts: Summit is the first supercomputer designed from the ground up to handle machine learning, neural networks, and other AI applications. Its many thousands of AI-optimized chips from Nvidia and IBM can handle demanding tasks, such as crunching through mountains of reports and medical images to help unearth hidden causes of disease.

(https://www.technologyreview.com/the-download/611385/americas-new-supercomputer-beats-chinas-fastest-machine-to-take-the-worlds-most/ more at link...)

These are very hyped and sure, they're powerful and fast. But they're simply scaling up the CPU and GPU architecture as it improves, which is verrrryyyy slowly. They're still stuck at the electron level and conflate electricity with ions, and they're just starting at the 7 nanometer level in production currently.

The crystal "cell" size of Silicon is .566 nanometers, allegedly, so they are nearing the limit that Silicon-on-Insulator tech can max out at in that regard as well. 14x smaller and they're at yet another bottleneck, versus the thousands of times smaller the photon is than the proton if they ever developed photon-transistor tech.

Jared wrote: if they ever developed photon-transistor tech.

Do you think this is coming down the road? I see this maybe in a crystal form? Patterns in surface materials affects the charge field directional output so directly (static electricity)....something I think they are aware of but can't design for currently.

I think it's coming, but it really has to be coming from people like us. Not me, I'm not an engineer - but that's part of what sparks me and spurs me on to learn and work on this charge theory. The promise of new technologies, if we can make them happen. Perhaps a cheaper or free energy source which we have already uncovered, but must learn to harness.

I have four main computers and they ALL love to overheat, especially in the summer when I'm rendering. If we think about it, all that heat is just wasted energy - wasted, unused photons being fed to the system to push a mere few electrons around. A proper, efficient charge device would perhaps make use of ALL that energy?

This is also how we know "quantum computers" are bullshit. They're just as hot and stinky as our current Silicon-on-Insulator systems are, and not even as fast. That's not the right direction. And we know that "qubits" aren't real either, so they're barking up the wrong tree.