Why chiplets? Can anyone explain to me why I'd want chiplets as opposed to a processor embedded in a Field Programmable Gate Array? Seems like development would be infinitely cheaper and the possibilities far more dynamic? Chiplets seem slow and expensive to develop. Once built they'd be completely static and probably necessarily replaced by the subsequent version in 3-5 years.
Sample size 1, but I've got a kid. I've got an ipad. I put the two together in with supervision, and what I've got now is a 4 year old that reads the walls in the bathroom stall while she's pooping.
I don't know if this is "better" or what you had in mind as "gains", but it's certainly entertaining for me!
In my youth, I was familiar with the facility that came before NIF, built in the 70's the goal at the time was to use a smaller target to demonstrate that the foundational principles that would underpin the success of NIF would work. As far as I know, they never succeeded. NIF was built anyway because this type of device is well suited to allowing access to certain types or demonstrations of physics that are otherwise unaccessible and important in a specific field of research not directly related to the flourishing of our race.
The article is old news before it was written. The article mentions the previous 'success' (yield was higher than previous experiments), and that was over a year ago now. They haven't been able to reproduce the previous experiment even knowing as precisely as they can what they perceive to be the preconditions necessary for an effective reaction. It also seems that this article was written about a single experiment. They will not be able to intentionally repeat the experiment. The manner in which they're exploring the pareto front is like groping in the dark to find a light switch that has an unknown texture and conformation. It's a classic monte-carlo simulation but they have one iteration every several weeks or months, and they cannot even possibly identify all the controlling parameters, nor do they have the necessary throughput or bandwidth to succeed in their pursuit without windfall.
The low hanging fruit providing the basic harmonics of the solution were discovered well before I was even introduced to this technology (in the 70's and 80's. Coincidentally around the moment of the genesis of many of our modern treaties on weapons testing).
You are overly optimistic, a 40-60% increase in nearly nothing is still nearly nothing. The PR campaign around this event is I think more significant in its political convenience, and in white washing the purpose of the facility. There are significant discoveries that still need to be made to even make the reactions consistent, and they will not come conveniently or quickly. Once the reactions are better understood and the mechanisms can be manipulated with intent the distance between the science and a practical industry / commercial product will require even more hurdles that stretch the imagination to be overcome. For instance I cannot conceive of a practical mechanism for actually utilizing any fraction of the massive amount of energy released in a fraction of a second in a chaotic murder of wavelengths and particles. The most practical way we've yet discovered for converting neutrons to electricity is through boiling water. Grossly inefficient in other contexts, I'm not sure that has even marginal utility in this scope.
I for one am 100% sure I barely know what I'm talking about. My disclaimer is that I'm not a physics guy, and high energy density physics was only a hobby of mine at one brief point in my life. Through perspicacity and access to papers and people, this is my honest mental model of the whole thing. You're welcome to your perspective, but although you seem well informed you sound very inexperienced.
My father worked for what was basically a bootstrapped startup at the bleeding edge of computing in the late 80's. Eventually his company, through a series of exchanges was part of Nortel Networks.
He'll tell you a different story. When you mention the "espionage / sabotage" he doesn't laugh or shake his head, it actually just outright infuriates him. Such a naive narrative only angers him because it seems to absolve the people who managed the company of outright incompetence and corruption, which is how it should be remembered.
For example: one story I vividly recall because I could never fully fathom it involved a specific female executive who was traveling for business - first class overseas. She never made her intended engagement because she was immediately arrested upon disembarking the plane. She had got drunk and (although married) decided to perform overt sexual acts with the gentleman sitting adjacent to her. On the plane. In her seat.[1]
At some point the culture of the executive at Nortel, for whatever reason, became completely incompetent and outright immoral. Rather than Huawei underhandedly perpetrating the perfect crime, it was simply the people at the head of the organization that solicited the crime to pursue their own benefit above all else.