
Reblogged by cstanhope@social.coop ("Your friendly 'net denizen"):
EdS@mastodon.sdf.org ("Ed S") wrote:
Bit of recent #computerhistory : Matt Pharr tells
The story of ispc
in 12 parts. Very interesting - single person not in the compiler team brings up a new tool, gets great results, is allowed to take it open source. Lots of insight into vector processing and Intel culturehttps://pharr.org/matt/blog/2018/04/30/ispc-all
Inside, a bonus link to a history of Larrabee aka Knights chips. By Tom Forsyth
Why Didn't Larrabee Fail?
https://tomforsyth1000.github.io/blog.wiki.html#%5B%5BWhy%20didn%27t%20Larrabee%20fail%3F%5D%5D