Nvidia reportedly drops 24GB RTX 5080 plan as GDDR7 costs rise

By Tom Whitfield

MSI plans November distributor price increases of up to about $1,200 on Nvidia’s GeForce RTX 5080, while Nvidia reportedly abandoned a 24GB model for China. Both developments put pressure on the same 16GB product that Nvidia is now reported to be keeping.

A Wednesday report by Notebookcheck and VideoCardz cited a Board Channels post, relayed by Japan’s Gazlog, saying Nvidia would drop the 24GB RTX 5080 and keep selling the 16GB model. The post said Nvidia had informed add-in board partners and supplied no new release window, and Nvidia has not confirmed the decision or commented.

Notebookcheck describes the cancellation as applying to China, where the card had been expected in January 2027. No replacement release date has been reported, leaving the 16GB model as the reported China plan.

What changed in the RTX 5080 line?

The late-October version of the plan pointed in the opposite direction. The same forum had said Nvidia was pulling the 24GB card forward to December or January while “winding down the 16GB model.”

Within weeks, the reported decision reversed that sequence and preserved the 16GB card. The commercial result is simple: the 24GB option has no launch date, while the 16GB card remains in the reported China plan.

Why does memory cost matter?

The 24GB design retained the RTX 5080’s 256-bit memory bus. It changed the memory configuration by replacing eight 2GB GDDR7 modules with eight 3GB modules, which the Board Channels report said cost roughly three times as much. The reported explanation for dropping the card is memory cost.

That creates the commercial mechanism behind the reported reversal. Nvidia could keep the same bus and product family, but higher-cost memory components would raise the bill for each board. The constraint therefore sits in the cost of the modules, not in a change to bus width.

What will distributors pay?

Wccftech reported on Nov. 10 about a supply-chain document from the account ChannelGate. Its title reads: “MSI graphics cards will implement a distributor price adjustment policy in November.”

The document listed two RTX price ranges, both denominated in yuan:

• GeForce RTX 5080: increases of 1,300 to 8,000 yuan, equivalent to about $194 to $1,200.
• GeForce RTX 5070 Ti: increases of 800 to 4,000 yuan.

These figures describe distributor adjustments, not confirmed retail prices, and MSI has not commented on the document. The report says graphics-card makers are passing memory costs on, making the RTX 5080 increase part of a broader pricing move rather than an isolated change.

The immediate effect is a higher distributor cost for both listed RTX models, while the retail effect remains open. What buyers ultimately pay will depend on how much of those increases reaches retail shelves.

How does AI demand reach consumer graphics cards?

The report attributes the pressure to the same AI demand that has lifted memory makers’ margins and reached Apple and PC makers. Memory suppliers are also trying to reduce the amount of DRAM used per system.

AMD said this week that its MEXT software lets servers run on half the DRAM. That approach addresses component demand directly, whereas the graphics-card response relies on higher-priced modules and distributor price increases.

The next test is whether MSI’s November increases reach retail shelves and whether the reported GDDR7 cost problem persists. Until then, the 16GB RTX 5080 remains the reported China plan, and the 24GB version has no confirmed release date.