Me: Points out the obvious fact that a monospace terminal renderer requires nothing but a full-screen quad with a pixel shader that fetches glyphs from a coverage atlas. Microsoft:pic.twitter.com/75Kb5fvebL
You can add location information to your Tweets, such as your city or precise location, from the web and via third-party applications. You always have the option to delete your Tweet location history. Learn more
I mean you certainly _could_ make an entire Unicode atlas but... seems like it would waste a lot of memory?
143859 unicode characters, you could probably store each in a 16x16 grayscale sdf and have it look decent in most places, = ~37mb texture. About on par with a single 4k RGBA texture
In the case of a terminal it would be required to handle unicode, so you would have to be intelligent about it of course. So I suspect the benchmark should be cat'ing a file with continously random unicode chars and the accepted subset of control characters.
I would disagree. While that's a fine benchmark to _also_ optimize for, that is never actually what people are doing. You don't sacrifice your 99.9% case for a .1% case that never happens in practice. So benchmarks should focus on the common case.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.