It's your assumption the framerate will take a hit at higher graphical settings, but you don't actually have a means of proving that, do you? The presence of the lower graphical settings was a bug in itself, or a settings mistake if you want to be more partial. That's what one of the DiRT 5 devs told John on Twitter when the results were brought to their attention.
Whether the framerate takes a hit at higher graphical settings depends on optimizations more than anything, even the PS5 settings are not very high in relation to the higher PC settings for that very same game.
You're better than ignoring commonly expressed devkit tardiness/optimization rumors around MS that have been persistent for about a year (and therefore newness to 3P devs WRT the GDK API tools) in lieu of pegging the repeatable performance results in certain games down to inferior hardware design on MS's part.
You're better than that.
I don't care about the semantics of repeatable vs. random, because that's not at the heart of my discussion. My main point was pretty clear in the first post I made in this thread, it's only a page back.
But to recap, and just brushing up on what you specify: just because you can ascertain a set of conditions that cause a bug to manifest itself, doesn't change the nature of the problem as a bug. Which is what the flame thing in Valhalla on Series X comes off as. That or a poorly optimized block of code, but the two probably go hand-in-hand.
At least you are willing to acknowledge it's something that can be patched; a lot of other people in that thread (similar to what they're doing in the DiRT 5 thread) pretend like patches suddenly don't exist and no patches can be done; according to them the results must 100% be down to hardware deficiencies or poor engineering.
Maybe you can try explaining to them how bug patches work.
Even if it was Codemaster's choice that doesn't tell us much of anything regarding the hardware itself. They probably wanted a 120 FPS mode there (heck, it could've been at MS's request to have it ready) and took the path of least resistance WRT optimizations to get the game in what they considered a presentable state to provide the 120 FPS.
Considering that even though they might have had some GDK devkits for a decent bit and find them suitable for what they are doing, that doesn't mean they have mastered the API features and tools of the devkits. It also doesn't mean they're up on the latest GDK revisions, or have the full suite of updated revisions.
Also stop trying to turn this into an attack on DF; what's done is done. They didn't know the framerate issue might've been a checkpoint load bug, someone else did and let them know. Now they know and got further clarification on a few other things regarding the game. None of these analysis channels are God or omniscient; all of them make a couple of oversights here and there, that's the nature of being human.
It's a good thing then that several of them seem to share findings and details with each other, it's almost like they're a community who just enjoy doing what they do and don't have a problem with sharing results among peers if it means more accurate results can be had to the benefit of everyone. Now if only some gamers could try taking that type of mentality when it comes to multiple platform holders
.....