I'm not saying that incremental improvements are the sole strategy. I'm saying that before you design a 5th gen system, the tech[0] you want to use should be p

I'm not saying that incremental improvements are the sole strategy. I'm saying that before you design a 5th gen system, the tech[0] you want to use should be proven as an upgrade to a 4th gen system. I'm saying that running out of headroom is a normal part of the lifecycle, and trying to cram in the latest everything into a newly designed system in an attempt to put off the inevitable is futile and counterproductive. I'm also saying that you may not run out of headroom as soon as you think. [0] Note that I'm aware that I am drawing a distinction here between 'tech' and 'design' that isn't really so clearcut.