Wait for the right chip—or wait for profits to stabilize

Valve is still deciding whether and when to make Steam Deck 2, waiting for what its designer Pierre-Loup Griffais calls the 'right chip.' This is the language of engineering patience, and it is almost certainly marketing. Nintendo spent five years saying the exact same thing about a Switch Pro, citing semiconductor constraints while a 139-million-unit install base made a $400 successor economically dangerous to the $300 original. Valve is now running that same clock.

The difference should matter but doesn't. Steam Deck has 3 million units in the wild. That is not a market saturation problem. That is not a situation where releasing a second-generation device threatens to cannibalize a thriving ecosystem. By Nintendo's own logic, Valve has room to innovate without economic risk. Yet the company waits anyway.

Current chips — Qualcomm's Snapdragon X, AMD's upcoming options — are more than adequate for a handheld refresh. They have been for two years. The engineering problem was solved before Steam Deck 2 was even announced as 'coming eventually.' What remains unsolved is not the chip. It is the decision to accept smaller margins on a device Valve knows will cost more to produce than the original, or to delay release until current-generation inventory burns through and profit per unit stops mattering less than total volume.

This is not hypocrisy. Companies protect margins. But calling it a semiconductor constraint is a useful half-truth, one that sounds like Valve is being patient rather than cautious. Nintendo's five-year hold taught the industry that timing a hardware refresh is partly engineering and partly mathematics. Valve learned the math part better than the engineering part.

What matters is recognizing the difference between patience and delay. When you wait because conditions genuinely aren't ready, you are building something. When you wait because the current thing still works, you are protecting something. The work of hope. The thing you actually do. Looks different in each case. One requires you to improve. The other just requires you to hold.