Yeah makes sense they knew what to call it... they got some of the best hardware engineers in the world ^_^
On a different note, I think it would be cool if you did some more engineer type articles if you are able to. More breakdowns of new microarchitectures and whatnot.
Inverse measures like Frames per Joule(And FPS) are usually avoided in engineering because it's really messy to try and use these inverse forms in calculations, they're mostly a marketing thing so that better things have higher numbers, if you wanted to for instance take an average of an array off FPS values to do it correctly you need to first inverse all the values then average the sum off the inverse's then inverse the result again, same with FpJ, so generally you'll use Frametimes for performance or Joules per Frame for energy efficiency up until the marketing material as these values are easier to throw around and more intuitive I'd say. This article explains the mentality in avoiding it well from when NVidia started using FpJ with Kepler marketing material http://fileadmin.cs.lth.se/graphics/research/papers/2012/perfperwhat/perfperwhat.pdf
Yeah makes sense they knew what to call it... they got some of the best hardware engineers in the world ^_^
On a different note, I think it would be cool if you did some more engineer type articles if you are able to. More breakdowns of new microarchitectures and whatnot.
My training is in Mechanical Engineering, not computer science. The problem is that I don't get access to a lot of the architecture info early.
Makes it difficult to do an architectural deep dive without being late to the party. TBH I love seeing how the architectures behind everything is shifting, but there are only a certain number of hours in the day and a lot of things that require my attention.