Researchers have found a way to perform certain quantum operations more than 1,000 times faster, cutting thousands of repeated control cycles down to just one. The advance could reduce errors and bring reliable, fault-tolerant quantum computers closer to reality.
Call back when quantum computers can solve real world problems, not just quantum computer toy benchmarks.
When I hear this sort of thing all I’m hearing is a lack of imagination.
Programming quantum computers is fun as hell.
Whenever I hear another quantum computer “record”, all I’m hearing overbearing delusion.
People made the same comments about AI a decade ago when image generators produced holy horrors full of eyes. That, and quantum computing, and many other emerging technologies, progress on an S-shaped curve. Very slow for a long time, then a series of massive breakthroughs, then settles until the next cycle. This particular discovery is more focused around reducing errors than it is increasing speed.
What an unfair comparison. A quantum computer can do an asymptotically fast fast Fourier transform. That has a lot of value to a lot of people. Unlike the current state of generative ai.
The same statements were made about cars, airplanes, the internet, cell phones, medicine, etc. AI is just the latest “big thing” to have gotten this treatment.
No, I’m more apt description is cryptocurrency. Something that was hyped as the Advent of something important and turned out to be just a circle jerk for a bunch of already rich people.
In technology something actually has to demonstrate value and utility. We can’t just describe it for no reason.
Yes, about AI they said it a decade ago. About quantum computers they are saying it for way longer, nearly as long as fusion reactors, and I believe we are decades closer to a working fusion reactor than a usable quantum computer.
It’s almost like S-curves apply differently to different areas of research…
I’m going to go out of my way to guess that you have no idea how they actually work.
I can count bits and know how much complexity you can build into them, so I can see that the interesting problems where they promise that quantum computers could actually do something are far beyond what they can handle.
They can handle a few hundred to a thousand bits, and getting more into it seems to get more and more complicated instead of easier. I would not be surprised if quantum computing will hit a wall where adding just another bit would eat more energy than all the AI burners together.
Lol. Not even remotely the case.
Yeah! Let me know when it’s in an apple watch otherwise who cares
Reminder that most Quantum computing claims are smoke and mirrors, if not outright lies.
So did Q-Day just move up again?
so they can shovel AI slop 1000x faster
Nope, these are very different computers, they can’t even run Doom.
probably cant even email either.
Tbf, quantum computing researchers aren’t planning to ever displace silicon for everyday tasks - it requires too much infrastructure just to cool the chips down to the required operating temperatures and there’s only so much optimization to be had in that tech.
The goal of quantum computing is to assist in completely different tasks like cryptography and physics simulations
The quantum unit still run alongside of conventional hardware.
Dafuq? They will definitely be used in AI model training at least