Intel spent months promising that Arrow Lake would reset the desktop PC.
The Core Ultra 200S lineup arrived with new architecture, a new naming scheme, and a new socket, the kind of generational break that usually justifies an upgrade.
Reviewers who tested the flagship Core Ultra 9 285K found something stranger than a flop.
In several popular games, it lost to the cheaper chip it was meant to replace.
This is not a niche complaint from overclocking forums.
Gaming is the single biggest reason ordinary Americans spend $1,500 or more on a desktop.
If the new part can't beat the old part at the thing buyers care about most, the launch story collapses into a simple question: why would anyone pay more for less?
Intel's explanation is that reviewers didn't test properly.
The company pointed to missed software updates and insisted that with the right Windows patches and power settings, the picture improves.
Some outlets retested and saw modest gains.
When a company's defense is that the benchmarks were done wrong, buyers hear something else entirely: we shipped a product that needs excuses.
The reviews landed during a rough stretch for Intel.
The company has been cutting jobs, its stock has taken a beating, and its foundry business is burning cash while it tries to catch up to competitors it once dominated.
Arrow Lake was supposed to be the proof that the turnaround was real.
Instead, it became another data point for people already skeptical.
There's a quieter cost, too, and it lands on regular households.
A family replacing a five-year-old PC now faces a mess of names that mean nothing to them.
Core Ultra 9 is not obviously better than Core i9.
The socket is new, so the motherboard has to be replaced.
What used to be a straightforward trip to Best Buy has turned into a research project, and the payoff at the end is a machine that may not even feel faster.
Meanwhile, AMD keeps selling chips that fit into existing boards and win the benchmarks people actually run.
When one company asks you to rebuild your whole system for a sideways step, and the other lets you keep your motherboard for a meaningful jump, the decision makes itself.
For heavy productivity work, video editing, and multitasking, it holds its own, and it runs cooler and draws less power than the previous generation.
If you do professional rendering all day, there's a case.
But "it's efficient" is a strange pitch to a gamer who just wants more frames.
Intel has spent years telling customers to wait for the next generation.
When the flagship underperforms its own predecessor, the next promise sounds hollow, and the brand that once meant "just buy it" now means "check the reviews first." Intel will recover or it won't, but the lesson for the rest of us is already here.
A familiar name on a box is no longer a guarantee of anything, and the burden of figuring out what's actually inside has been quietly handed to the people least equipped to carry it.
Our take: this is what happens when marketing outruns engineering.
Final Thoughts
If a company can't beat its own last product, it shouldn't ask you to pay for the privilege of finding out.