I can remember in around 2003, AMD's Athlon desktop processors were superior to Intel's Pentium 4 "Netburst" processors. Then Intel, under Sunil Shenoy's team designed the "Core" processors and Intel took a substantial lead that lasted for decades until Zen. Intel will do it again, for sure. It will be interesting what Intel comes up with and if it's powerful like Core was the consumer will win.
I agree with you there. But one of the things that gave Intel such a key advantage was it's advanced process nodes. AMD was stuck with GloFo for such a long time and they couldn't get the raw speeds with GloFo's technology. . If they could have improving Phenom, they with a better node, they might have been in a better place. But once again the bean counters took over and aimed for raw speed with deep pipelines, and cheap R&D with circuits being automated designed instead of hand optimizing sections as they should have.
It was the same with RTG. Todays landscape is a lot more competitive on the node size front. It will be difficult for Intel to truly establish the lead and maintain core count total performance. Rocket Lake demonstrates this. Alder lake could be a misfire as well unless it's mobile power consumption intel is worried about. And even that may be a bad bet. The #1 consumer of energy is GPU and Displays now. And whenever power savings advancements were made in the past, Laptop Mfg's made the battery Wh rating smaller. (cuts down on size and cost) And that battery loss negates any power saving benefit.
Low powered cores are good for smart phones and tablets where you are constantly unplugged. My laptop is plugged in 99% of the time. I don't care about power efficiency when I'm waiting on compiles.
I think this process node parity however is a good thing. Monopolies are never good for consumers.