Things have changed a bit over the last 20 years. The vast majority of takeoffs are now twin jets, which as has been pointed out are most often limited by the 2.4% climb gradient engine out scenario in the second segment. But in almost all cases we have such a surplus of thrust that we see pretty healthy initial climb rates once we finally get airborne.
The noise per aircraft has obviously reduced over time. But overall number of takeoffs has also increased dramatically. The question is could we do better? Ironically I can find no evidence that reduced thrust takeoffs save fuel, nor can I find any conclusive proof that they are noise optimized at least at low weights with low acceleration heights. What they definitely do is make engines last longer as the recent post about the TUI 737 engine record shows. It depends who the engine belongs to as to who benefits from that long term.