RTX 5090 Becomes Ground Zero for a New Wave of Price Inflation
The flagship GeForce RTX 5090 is rapidly turning into a case study in how component shortages inflate graphics card costs. Launched at USD 1999 (approx. RM9200), the GPU was already priced at the upper edge of the consumer market and has since drifted far above that level in real-world sales. Reports now suggest that the card, along with the RTX 5090D V2 variant, is facing fresh cost pressure tied directly to its memory configuration. Featuring 32GB of GDDR7—twice the capacity of the RTX 5080—the RTX 5090 is uniquely exposed to any spike in next‑generation VRAM pricing. While official MSRPs remain unchanged on paper, distributors and retailers are no longer buying these cards at those levels, setting the stage for yet another round of price hikes that users will ultimately shoulder.
GDDR7 Shortage: How Memory Costs Cascade Through the Supply Chain
At the heart of the RTX 5090 price increase is a tightening supply of GDDR7 memory. According to reports shared via Board Channels, NVIDIA has informed partners that higher GDDR7 procurement costs necessitated a price adjustment. On 13 May, the company reportedly raised what it charges board partners for RTX 5090 and RTX 5090D V2 chips and memory by approximately USD 300 (approx. RM1380). This cost is not absorbed by NVIDIA; instead, it is passed directly to add‑in card (AIC) vendors, who must then rework their own pricing to maintain margins. Because the RTX 5090 uses more GDDR7 than any other RTX 50‑series card, it feels the impact first and most severely. These moves underline how a GPU memory shortage at the component level can rapidly translate into higher end‑user prices, even without any official MSRP change.
Board Partners Squeezed as Another USD 300 Hit Lands
NVIDIA’s latest move reportedly adds another USD 300 (approx. RM1380) on top of already elevated costs for board partners building RTX 5090 models. For AIC vendors, this is not a minor tweak but a substantial hit to each card’s bill of materials. Since these companies operate on relatively tight margins, they have limited options: strip features, reduce quality, or raise prices. Most will choose the latter, pushing retail tags further away from the original launch price of USD 1999 (approx. RM9200). One retailer example already shows the “most affordable” RTX 5090 listed at £3299.99, a massive premium over the official £1799 MSRP. While retailers often get blamed by consumers, the underlying reality is that they are buying stock at inflated wholesale prices, leaving little room to maintain earlier, more consumer‑friendly price points.
Flagship GPU Prices Creep Toward the USD 5000 Frontier
With successive cost increases, the RTX 5090 is drifting into ultra‑luxury territory. Reports indicate the card, once hovering near USD 3000 (approx. RM13,800) even before the current “RAMpocalypse,” is now frequently seen selling around USD 4000 (approx. RM18,400) or more. Following the latest USD 300 (approx. RM1380) hike, analysts expect starting prices to move beyond USD 4000 (approx. RM18,400), with some custom models potentially touching the USD 4500–USD 5000 (approx. RM20,700–RM23,000) range. This escalation dramatically erodes FPS‑per‑dollar value for enthusiasts, especially when performance gains over previous generations no longer scale with these cost increases. While only the RTX 5090 and RTX 5090D V2 are confirmed to be affected so far, persistent GDDR7 constraints raise the risk that other RTX 50‑series cards could face similar adjustments if memory markets remain tight.