Are you aware that Ryzen is practically brand new architecture wise, and is on a cutting edge 7nm process?
Intel on the other hand, are on a 5 year old architecture (Skylake) and a 5.5 year old process (14nm). The fact that Intel still wins in some areas (gaming) by a small margin, mean it will absolutely demolish Ryzen once Intel releases first Rocket Lake (new architecture backported to 14nm) and then Alder Lake on LGA1700 shortly after (new architecture & new process).
That's what I call interesting, not current Ryzen struggling to beat Intel in gaming, or Zen's 3 perhaps matching Intel in gaming.
Zen2 is the third generation of Zen.The first two generations of AMD CPUs had lower IPC,and Intel couldn't even after 5 years managed to improve the design. The company is in a mess,dumping CEO after CEO,and is being held back by internal problems in the company. You forgot 14NM had tons of problems too - Broadwell got massively delayed due to this.Skylake at the start clocked poorly. Intel had to rejig 14NM to make it clock higher.
Their 10NM node was a disaster and 7NM is repeatedly delayed.
Now you are saying in 1~2 years suddenly Intel will move to a perfect 7NM node,and make a Core2 level improvement. If after 5~6 years they couldn't actually push forward,then what makes you think suddenly any of their marketing slides have any merit??
If Intel cannot fix its own node problems,that is not my concern,especially if AMD decided to make a more intelligent design that can actually be ported quicker to new nodes. That is actually a design feature.
AMD is held back by inter-CCX latency,which is why Intel mesh bus CPUs loose in very Intel friendly games such as Fallout 4 as its higher latency. Those older engines were developed on Intel ring bus CPUs.
They hit the same latency problems and Rocket Lake is limited to 8 cores. Once Intel moves off ring bus,they will have problems too,as the engines don't work as well on mesh bus CPUs,and Intel will need to make changes if they hope to move past 10~12 cores.
Intel will also need to move away from monolithic cores,as yields will be the other problem too. So once they have chiplets,they will have to work around the latencies due to that. AMD would have 2~3 generations of working around the problems with chiplet based designs by then.
The consoles are using Zen2 based CPUs,so all the dev kits will be using AMD CPUs. So longterm lots of the new engines will be running on AMD Zen based dev kits.
With Ryzen getting more share,devs are slowly bothering to actually optimise for it. With consoles,every major AAA game will be developed on AMD CPUs.
To put in context the P4 beat the A64 at launch in a number of games,but it didn't last for long.
Intel 7NM is some time away. Rocket Lake is still on 14nm and is limited to 8 cores. When Intel moves to 7NM,AMD will be on 5NM/3NM and Zen4.
TSMC will deliver as they have so many customers,that they can dump cash on it. Intel did a stupid thing and never opened up their fabs properly. It would take TSMC screwing up too for Intel to gain some traction.
Intel will hit the same problems as AMD did. The newer nodes don't clock as high. The same happened to Broadwell and the 1st releases of Skylake,they clocked poorly on 14NM,and 32NM/22NM could clock higher.
The problem is also you ignore the performance of Ice Lake,which despite being on 10NM,and having similar IPC to Zen2(maybe a tad better) isn't really doing much demolishing unless you run SuperPi all the time. It is also competing against a depreciated Zen2 design using less L3 cache,and having double the cores in the same TDP,so it can't boost as high.