It is a fairly common "mistake" to compare numbers of totally different systems, in this case "closed" systems (consoles) with "open" systems (PC). And it is also a mistake to use TFs as the unit of measurement. These "numbers", for the user, can give an orientation (sometimes wrong), but that orientation, if it can be had, should only serve to compare systems with the same or very similar characteristics. In no case should it be used to compare the results between PC and Consoles.
That said, knowledgeable or not, I think a mental effort of sheer logic should be made. Since the answer is really always there in front of us and many times the light of the numbers blinds our minds (error).
Let's see ... (I'll use PS5 for the example, as it's the machine that supposedly disappointed you. And I'll use rounded numbers to symplify the example).
RTX 3090 ► 30TF
RTX 2080Ti ► 14TF
GTX 1080Ti ► 11TF
PS5 ► 10 TF
PS4 ► 2 TF
Initially, as I said before, comparing this data is a mistake, but "assuming" that we could do it, we would have as a result that a 2013 console with 2TF has achieved better "visual" results during its useful life than a 14TF card. No matter what the numbers say, what matters is what is achieved in practice. And really, in practice, it shouldn't matter what DF says either, what should matter is
visual perception.
I understand the need of the user to have the latest technology at any cost (I also have that need), and therefore I also understand the need of those same users to have to justify their purchase above all (I also had in its time that need). But it is self-deception.
All that extra power in PC is there because it is necessary to alleviate the "lack" of optimization of video games with respect to a specific hardware, being forced to have to be compatible with a great variety of practically "random" Hardware. With this you are very limited to using what exists and with little or no research margin. So much of that power is totally wasted on a practical level. And although it could be done (squeeze hardware on PC) the reality is that unfortunately PC manufacturers change technology (or hardware) before the previous hardware could have been used. And therefore developers cannot get to squeeze that hardware (and for many other reasons).
But obviously it is logical, it is a different business. They live exclusively (practically) on the hardware sold, not on the games sold in relation to that Hardware, and for this they must create a mental need in the buyer to want to change Hard every year practically.
So what do you do with that power? Well, you use that power to get better resolution and a better frame rate per second. But the visual quality, the graphic detail, is not in line with what is expected of a 14TF machine. You will never squeeze that power technically speaking. You are going to "waste" it in other aspects that are important for some and not for others (such as resolution or the rate of frames per second). Still, in those cases, the only really important thing is that the frame rate per second is stable, and the resolution is adequate depending on your distance from the monitor or television. But of course, there are invented monitors that have to be sold, although a person who is 30cm from the monitor does not really need an 8K monitor (and if you think so, then I would say that what you really need are glasses).
The same happens with the 3090. Professionals will take advantage of it for VFX, since rendering engines like Redshift (say one GPU) use about 95-98% of the power of the set of GPUs constantly, but as development of video games do not expect there to be a direct relationship between what you see on the screen and the power sold (that does not mean that what you see on the screen is not going to impress you). Surely you can play Fornite with RT (for example) and 8k resolutions at 120fps or 360Hz if you have the money to buy a monitor that supports it and upgrade your tower to avoid bottlenecks. But instead PS4 users are already enjoying TLOU2 (on a 2TF machine from 2013). Users of a 2080Ti do not.
What is "wrong" here (and is the essence of the problem) is the table of equivalences that exists in the minds of users (because of the manufacturers) with respect to what one expects to see in relation to the supposed power acquired .
What leads us to the usual absurd and repetitive discussion online (and outside of it) and is that, definitely, comparing TFs between different platforms, or evaluating a product by its TFs is a mistake, a very serious mistake.
I hope it was possible to understand what he was trying to say. It is not about knowing what is better or what is worse, it is about knowing what can be used and in what way. But I would stop trying to compare such different worlds, because you are really wasting valuable time that you could use to play games or create life (which is much more rewarding).
If there is an error due to mental lapses or a bad translation, feel free to let me know. I will feel free to answer or ignore you (just kidding

).