they will not. quantum computers as understood today will never run an LLM. read = destroy. and no cloning. wanna reload the entire model for each token? preparing quantum states is slow by any standard.
there is not yet a single approach to quantum computing that is provably scalable. so called experts may quibble, the uninformed (or financially aligned) will bloviate, bluster, and talking point us to death. but it's sadly just true. we're closer to useful fusion power than useful QC.
Bambu has proven time and again that they don't understand security. Unless, of course, it's theater and by design because real security would be inconvenient to state actors. Regardless, they gaslight and bludgeon those who wish to use the hardware they purchased in peace offline and away from prying eyes.
Having said all that, the hardware is very good. Software, not so much.
why not a proper Progress Web App so it can run on any device independent of app stores? it's not as though a social app needs deep OS integration. I'm sure Claude or Codex could vibe code that in an afternoon.
Frigate NVR + Amcrest cameras. 100% local, private, on-device AI object recognition and classification. Can use a Google Coral USB TPU to speed that up. Runs on hardware as modest as a Raspberry Pi.
I did not criticize your app. I offered that your blanket statement that "The browser is an extremely poor medium to deliver applications" does not comport with my experience. And it looks like I nailed it, too. It is the nature of your application. Had you said "the browser does not offer a real time API which I need for my application", there would have been nothing to say. This is obviously true. Even native desktop apps provide an inadequate environment for "hard RT". So I suspect that is also not a true requirement, either.
Makerspaces and education are two areas of focus. no SW install, fully loads in under a second. through the Onshape integration and ability to run on Chromebooks, it's made its way into high school and university STEM curriculum.
it's a combination of JS, WASM, and WebGPU. the JIT engines are so much faster than you would imagine, especially if you tune your code right. workers allow for parallel processing on all of your CPU cores. WebGPU, at least in Chrome, is kind of amazing.
second KiCad. just had my first board printed a few months back. its an esp32 stackable daughterboard. first time doing anything like that outside of breadboarding, and it worked great.
this does not track with my experience, so possibly it's the nature of your app or the way it's coded. frameworks like react are notoriously crap. stick to pure html5/css/js and it can be extremely fast and light.
this. webkit is intentionally hobbled and years behind the standards. browsers on iOS are forced to use webkit for ginned up security excuses/reasons so that no real browsers that implement full standards can complete with heavily taxed app store spyware.
there are desktop builds https://grid.space/downloads.html that run entirely offline. you can also use the center menu to "install" it as a progressive web app.
probably adaptive milling, which will be in an upcoming release. sharp path changes in harder metals can wear or break tools if you don't go slow, which has other issues.
no shared lineage. Cura and Kiri started around the same time (2011/2012), but as completely separate projects. Cura is a C++ desktop app and Kiri has always been 100% browser-based (no cloud, all computation in the browser sandbox). the licenses are different, too. Cura/Prusa/Orca are GPL based and Kiri is MIT.
how about making the player smaller as you get closer to the core? then each layer would be the same. there would be no seams where layers double/halve.