Commenting here as well for visibility, the author of this article mangled the math to produce this clickbait headline. It’s very cool tech with a lot of potential, but it unfortunately did not show 1700x the efficiency of a google TPU.

The paper claims that a simulated scaled-up 8-bit version of this tech (180nm CNT transistor TPUs) could theoretically reach 1TOPS/W. That is less than the efficiency the author specifies for the google TPU (4TOPS/2W = 2TOPS/W)

Then they go on to speculate that a lower process node will probably improve that efficiency greatly (very likely true, but no figures listed in the public preview of the paper, even simulations)

The author of the article assumed (wrongly) that the actual chip they made could do 1TOPS (it’s only 3000 transistors and can only do 2-bit math), and that it consumed 295 microwatts to do so, for an efficiency of 3389TOPS/W. (roughly 1700x the 2TOPS/W of the google chip) That’s of course ludicrous.

deleted by creator

on a desktop it might not be significant but I tried using flatpak apps on a device with very limited root emmc storage (16 GB) and ran out of space really fast. Its really common to see a couple multi-hundred-megabyte library downloads for each new app IME.

I like them for some stuff but there are glaring issues that I don’t like. I’ve posted about it before, poor integration of apps/not getting the right permissions is a big problem, the people packaging them don’t often do as good of a job as someone like a distro maintainer.

But admittedly my experience using it probably isn’t representative (pop os through their shop and arch on a mobile device). Neither were amazing, but not having to compile shit myself or install with an untrusted shell script was nice for some apps. Without some significant improvements it’s not a good replacement for a distro’s package repos but it might be a good way to broaden the available applications without having to maintain 10x more packages.