Valve closes Steam Machine waitlist, admits it can't price what it builds
Valve has shut the 2TB queue with more people in line than units to build, saying memory and storage costs have hit 'unprecedented levels' and it cannot yet name a price or a production number until January.

Valve has stopped building Steam Machine waitlists after telling buyers it does not yet know how many units it can make or at what price, with the 2TB model 'especially constrained' and now listed as out-of-stock. The decision turns a component-cost squeeze driven by AI data center demand into a visible retail shortage, and it hands cloud gaming platforms an argument that expensive local hardware is no longer the easy default.
Valve has stopped taking names for the Steam Machine. In a new post on Steam, the company said it cannot say how many units it will be able to build, or what they will cost, and confirmed the corollary: the waitlist is closing rather than growing. For the 512GB model, Valve says it has fulfilled all current reservations with no waitlist remaining, and units can be bought directly while supplies last. Once that stock runs dry, the listing flips to out-of-stock rather than seeding another queue.
The 2TB picture is blunter. Valve says it has finished sending purchase invites to all remaining 2TB reservation holders, but nobody still on the waitlist is guaranteed a unit, and the line is being closed anyway because there are more people on the waitlist than units available. Position in line is retained, but the 2TB model will now show as out-of-stock in the store, and anyone who decides to take a 512GB unit instead has to leave the 2TB queue to do it. Valve's own explanation is unusually candid for a hardware launch: "We are continuing to build more Steam Machine units, but memory and storage costs continue to increase at unprecedented levels; as a result we don't yet know the number we are able to make, and at what price we'll be able to offer them," the company wrote, adding that "the 2TB model in particular will be especially constrained."
Read that sentence again, because it inverts how a product launch normally works. Demand is not the problem. Valve has more willing buyers than it can serve, which is the dream scenario every hardware founder writes into a pitch deck. The problem is the bill of materials. Memory and storage pricing has moved so far, so fast, that Valve would have to quote a price it might not be able to honor by the time the units ship. Faced with that, the company chose a shorter queue over a longer promise.
The macro backdrop is the AI buildout. The RAM and storage drought is projected to continue well beyond January amid data center fever, as hyperscalers absorb memory and storage capacity to feed training and inference workloads. Historically, consumer hardware makers buy components on contracts months ahead and price devices against those locked-in costs. When spot pricing outruns contracts, the pricing decision becomes a bet on a commodity market rather than a judgment about what customers will pay. Valve says it expects to have "more information" in January. More information is not more units.
There is a second act forming around the price tag. Amazon's games boss has said he hopes the Steam Machine and consoles "pushing well over $1,000" will drive more people toward cloud gaming services like Prime's Luna. That is a hope, not a forecast, but it is a revealing one. It frames high local hardware prices as a customer acquisition engine for streaming platforms, where the cost of the silicon lives on someone else's balance sheet. Every month Valve cannot name a number is a month that argument gets easier to make.
For operators, the more interesting decision is the waitlist itself. Queues are a demand-signaling tool: they create scarcity, they capture intent, and they generate free press. The moment you cannot guarantee supply, that same queue flips into a liability, because it is a public, timestamped record of promises you may not keep. Valve's move, closing 2TB reservations rather than stacking them deeper, is the least bad option. It caps the disappointing email count at whatever is already in the system.
For context on how unusual this is, hardware launches typically ship with a price locked well before orders open, even when volumes are uncertain. What Valve has done is separate the two variables in public: the product exists, the customers exist, and the price is a live unknown. That is a preview of what a sustained memory crunch does to the rest of the consumer hardware market, from handhelds to laptops to anything with a DRAM chip and a storage module inside it. Component costs are no longer a footnote in the earnings call. They are the product decision.
The strategic stakes land hardest on smaller players. A large platform owner can absorb margin compression, delay a launch, or eat a bad quarter. A hardware startup with pre-orders open and components unbought cannot. The practical read-across is that reservation mechanics and pricing need to be stress-tested against input costs, not just demand curves, and that the flex built into a supply agreement matters more than the headline unit forecast. Watch three things from here: what Valve says in January, whether the 512GB stock quietly runs dry without a replacement, and whether other device makers start using the same constrained, unpriced language. If they do, this stops being a story about one console and becomes a pricing regime.
This story's Key Insights and Take-aways are locked.
Create a free account to unlock Executive Actions for one credit.
Register to UnlockAlways free for Executives Club members. Join the Club
More in Technology
Anthropic mathematician claims Claude helped crack the Riemann hypothesis
If the proof holds, AI has toppled math's most famous open problem; if it does not, it is a case study in announcing breakthroughs before anyone verifies them.
AI's $4 Trillion Wipeout Erases Nasdaq's Gains for the Year
A brutal week took roughly $4 trillion off the AI sector and left the Nasdaq flat for the year, and the damage ran the full length of the supply chain.
Anthropic mathematician says AI proved the Riemann hypothesis. Verification will take time.
A single unverified claim from inside Anthropic now sits against more than 160 years of open mathematics, and the gap between asserting a proof and surviving peer review is the entire story.


