In a few years when all these rack servers are outdated for AI workloads, there will be a glut in the market of cheap only-slightly-dated servers.

My JellyFin setup in 5 years is going to be fucking awesome.

▲

ivanjermakov 6 days ago | parent [-]

Same, after getting into homelabbing, "post-lease" became my favorite word.

▲

rbanffy 6 days ago | parent [-]

When this bubble bursts, we'll have a lot of number crunching capability being sold for peanuts.

▲

bee_rider 6 days ago | parent [-]

Using these tiny floats for classic engineering workloads—if I were starting gradschool right now, that’s definitely what I’d look at. Mixed precision numerical algorithms are already a topic, but with the hardware glut there’ll definitely be room to push things.

So it’s in a nice spot where there’s some scaffolding, but it isn’t totally done, and the hardware is probably going to make things possible.

	▲	rbanffy 5 days ago \| parent [-]
		> Using these tiny floats for classic engineering workloads I think that's the idea. Back in the stone age, I did that with Apple II floats (40-bits) because I was hitting quantization problems in a Mandelbrot explorer program I wrote. Wrapping my head around it in BASIC was hard. You can only go so far when your abstraction level is that low.