I thought fragments of graphene were found in graphite, so shall we start to ban all pencils while we're at it?

Whatever happened to redundant/multiple video recordings, like you know, how we back up data? Failing that, a stronger signal wouldn't hurt surely?

I often wonder what a programming language would look like if CPU and memory were infinitely fast and big. What kind of constructs and algorithms would we never have to worry about any more?

Sorting? Scrap Quicksort - BogoSort to the rescue! Garbage collection? No worries - the waste disposal area is essentially a black hole! Bugs with threading driving you to despair? Never fear, say goodbye to race conditions forever! What does 'optimization' even mean - just brute force everything!

Elegant wouldn't even begin to describe the result. Perhaps it would look a little Basic-esque even... [ducks]

The good news though is that as memory increases, it becomes practical to do just as I said. My own OpalCalc program (shown in my sig) does that, restricting the number of undo/redos to say 500 (though I'm sure 5000 would be fine too). The class is less than around 25-40 lines of code. As (text) files inside OpalCalc tend to be on the small/tiny side, no harm is done, and it makes the chances for programmer error much smaller.

I'm guessing it's easy to code such a system in functional programming ...

If you had infinite memory/CPU (or at least a small enough file), you could implement multiple undo/redo simply by storing the whole string in the file for every character change, easing code writing immensely.

Interesting. Would you say (again assuming an infinitely fast CPU) that for maximum ease of programming, that it would be best to write code in an entirely functional style, or use a mix of functional and traditional imperative programming?

Seriously though, without AI, I was just wondering if the benefits of not using functional programming for low level functions had anything to do with speed, or if it was an ease of code writing or some other issue.

With an infinitely fast CPU, would it make sense to use functional style even for low-level control flow?

Apart from the ones that go extinct.

22 thousand *per year*.

How something is supposed to 'look' is something that can change within a generation.

If on the other hand you believe in universal aesthetics (like I do), and must dip into the mere look of the presented food, then there's no reason we can't have a BETTER looking food put onto our plate than what stone-age meat has to offer.

If this takes off, and I don't see why not (unless it's significantly more expensive), I almost worry that the existing range of quorn-like products will die off. There's a potentially infinite range of tastes and textures out there, and by eating just real meat, we're forcing ourselves to a tiny sub-portion of possible flavours.

It's about time we moved away from dark-age style slaughterhouses to a tasty meat substitute. Bring it on.

Finally. Something that gets hardware and software devs to care about latency. For too long, software, mobile phones, games, monitors and tons of other gadgets have often exhibited latency above 50-100ms. I'm hoping the tech will trickle down from VR headets to other devices so we can end the madness once and for all.

10ms would be better by the way as even 16ms can be perceptible by many.

All sound in the end is air pressure whether that's from an electronic speaker or a 10 trillion year old violin. The speaker is in itself a kind of instrument, except it's the most versatile instrument ever created because it can play not just any sound you have ever heard, but any sound you can POSSIBLY conceive of mathematically. I think that's an incredible concept.

Anyway, his point is that you are not hearing the 30kHz and a 34kHz pitches directly. So that still means the human ear can't hear above roughly 20kHz no matter what anyone claims. However, you're free to go for a double blind trial yourself and see if you're not wrong.

Excuse my ignorance, but wouldn't a 20kHz sine wave sound less harsh or loud compared to a 20kHz square wave? Has that ever been tested?