Most of these comparisons are against general purpose processors. This article doesn't say ANYTHING about the hardware in question. Pit the quantum computers against systems that accelerate work using GPUs and other accelerators (FPGAs, for example). I'm not disputing quantum speedup exists, of course. But it depends on the quantum and conventional systems in question, and the algorithms (algorithms can be optimized for convential computing as well in many cases). Quantum computing has a long way to go, and these declarations of quantum supremacy are often light on facts and employ a lot of hand-waving.
More likely this tell that quantum Computer Are farter is spesific type of situations. The thing that was expected but now They did prove it.
In most normal situations old fachisoned computers still have a lead.
Aka They have provided first time a proof that quantum computers can be faster than normal in some situation maybe in the far far away Galaxy... Sorry time.
Before this it was suposed that quantum computers can be faster. So supose vs proof is the Main course of this news.
I think that probably the biggest quantum computer is the human brain. And quantum computers don't need to compete against the classical computational model, it will necessary we use them as a complement to achieve bigger paralellism: break down the problem and combine all the solutions.
Most of these comparisons are against general purpose processors.
It turns out that's all you need to do. According to Amdahl's Law, a N-wide parallel implementation of a classical computer can only ever be N times as fast as one of those computing elements. So, all you need to show is that a quantum computer can out-scale classical computers.
This article doesn't say ANYTHING about the hardware in question.
Anyway, did you check the paper the article is about? It's really not couched in specifics of particular modern computing systems, but rather trying to demonstrate a fundamental characteristic of quantum computers vs. classical ones.
Hint: Lucian always cites his source right at the top (as do most authors on this site). On the linked arxiv.org page, you'll find a PDF link to the paper.
And quantum computers don't need to compete against the classical computational model, it will necessary we use them as a complement to achieve bigger paralellism: break down the problem and combine all the solutions.
The question isn't moot. Any type of big computer is expensive, so it's important to know what type of computer is needed before buying/building it. And in this case, the cost is particularly large, since it includes the cost of many inventions and innovations to make quantum computing at scale both possible and practical.