Recommended byte size of a single vertex

Started by
11 comments, last by 21st Century Moose 7 years ago

Well not anymore, Moore's law has technically been dead for around 2+ years now. And HBM is a huge boost in memory bandwidth!

[citation needed] :D

http://www.telegraph.co.uk/technology/2017/01/05/ces-2017-moores-law-not-dead-says-intel-boss/
Technically Moore's law is about how many transistors you can fit onto a surface, and Intel is still keeping up the pace in 2017. In practice the number of transistors you can fit onto a surface does roughly corellate with compute performance.
In recent years single-core perf has plateaued, but we can still just keep adding more and more cores to keep the performance charts growing.
On a timescale where bandwidth doubles every decade, sudden leaps like HBM don't really affect that long term trend -- or if they do, you can't be sure until a decade's time when you look back over the data :)

Nah, they missed the ship date for 14nm by a few months. Technically it was within "2 years" of their 22nm node, but technically Moore's law states "Every 18-24 months" not years. Of course an Intel PR announcement is going to tell you differently.

Advertisement

Nah, they missed the ship date for 14nm by a few months. Technically it was within "2 years" of their 22nm node, but technically Moore's law states "Every 18-24 months" not years. Of course an Intel PR announcement is going to tell you differently.

22nm to 14nm was a ~2.5x improvement in component area, not a 2x improvement though, so Moore's law states that this particular step is expected to take longer than 18-24 months :P

14nm to 10nm is a ~2x improvement, so should've been out last year instead of coming out this year. Still one outlying data point doesn't change the long term trend, yet.

Nah, they missed the ship date for 14nm by a few months. Technically it was within "2 years" of their 22nm node, but technically Moore's law states "Every 18-24 months" not years. Of course an Intel PR announcement is going to tell you differently.


We're talking about GPUs though.

Direct3D has need of instancing, but we do not. We have plenty of glVertexAttrib calls.

This topic is closed to new replies.

Advertisement