I genuinely thought we were past the nightmare of GPU shortages and scalper prices. The crypto crash felt like a breath of fresh air—gamers could finally buy graphics cards at reasonable prices. I was catastrophically wrong.
Nvidia and AMD GPU prices are about to climb significantly in 2026, and this time it’s not crypto miners to blame. It’s artificial intelligence, and there’s no end in sight.
Key takeaways:
- AMD raises GPU prices starting January 2026, Nvidia follows in February
- The RTX 5090 could reach $5,000—not a typo
- Memory costs now account for over 80% of GPU manufacturing expenses
- Both companies plan monthly price increases throughout 2026
The DRAM Crisis Behind Nvidia and AMD GPU Prices
The global semiconductor industry is experiencing what analysts call a “DRAM supercycle.” Memory demand from AI data centers has grown so aggressively that manufacturers like Samsung, SK Hynix, and Micron have redirected production away from consumer devices toward high-margin enterprise components.
Here’s a stat that floored me: producing one bit of HBM consumes approximately three times the wafer capacity required for standard DDR5. AI infrastructure is literally eating the memory supply that would otherwise go into our graphics cards.
The IDC warns the PC market could shrink by 9% in 2026 due to skyrocketing RAM pricing. This isn’t a typical cycle—it’s a fundamental reallocation of silicon capacity that could persist for years. As someone tracking industry trends closely, I’ve never seen disruption at this level.
RTX 5090 Price Increase: From $1,999 to $5,000?
According to Korean outlet Newsis, the flagship RTX 5090 could soar from its $1,999 MSRP to approximately $5,000 by late 2026. For context, that’s a functional used car, several months of rent, or roughly ten PlayStation 5 consoles.
Currently, the RTX 5090 already trades around $2,999 on Amazon—far above MSRP. It was the only current-gen GPU to increase in price over the past months. With 32GB of GDDR7 memory, the 5090 faces proportionally larger impacts as memory prices climb.
The RTX 5090 features 21,760 CUDA cores and a 575W TDP. Those specs make it a beast, but that 32GB VRAM allocation becomes increasingly expensive as GDDR7 prices surge. The $5,000 figure starts looking less like clickbait and more like reality.
AMD vs Nvidia GPU Prices: Timeline and Impact
Here’s the timeline: AMD kicks off price hikes in January 2026, with Nvidia following in February. Both companies reportedly plan to continue raising graphics card prices monthly throughout the year.
AMD has already confirmed adjustments for its Radeon RX 9000 series. AIB partners now spend $10 more per 8GB of VRAM. The RX 9070 XT moves from $599 to $619 MSRP; the RX 9070 goes from $549 to $569. These initial increases seem modest, but a larger wave is scheduled for January 2026.
For those exploring new gaming experiences, the current prices represent a relative low point before things get worse.
Current GPU Pricing Snapshot
Nvidia RTX 50 Series:
- RTX 5090: $1,999 MSRP, street price around $2,999
- RTX 5080: $999 MSRP, available near $1,000-$1,100
- RTX 5070 Ti: $749 MSRP, average around $820
- RTX 5070: $549 MSRP, available near MSRP
AMD RX 9000 Series:
- RX 9070 XT: $599 MSRP, finally available at MSRP
- RX 9070: $549 MSRP, available near MSRP
- RX 9060 XT: $349 MSRP, competitive budget option
Compared to six months ago, pricing has actually improved. The RX 9070 XT dropped from $837 in June to around $655. The RTX 5080’s average decreased by $155. That window of opportunity is closing fast.
Why AI Is Driving Up Graphics Card Prices
The AI boom has created a market dynamic where gamers fight for scraps while data centers feast. Industry estimates suggest AI will consume 20% of total DRAM production in 2026.
Memory manufacturers have made a rational business decision: prioritize high-margin enterprise sales. SK Hynix reported its HBM, DRAM, and NAND capacity is completely sold out through 2026. Samsung doubled the cost of 32GB DDR5 modules from $149 to $239 in months.
Framework announced another memory price hike with more coming. Japanese PC shops halted desktop orders until 2026. Dell plans 15-20% PC price increases. When even preparing for a cozy gaming session requires financial planning, you know the market has gone sideways.
Should You Buy a GPU Now or Wait?
Multiple industry insiders, including representatives from PowerColor and MAINGEAR, explicitly advise not to wait. MAINGEAR’s CEO stated the situation will deteriorate further in 2026.
The RX 9070 XT at $599 offers exceptional 1440p and 4K performance, often matching the RTX 5070 Ti in rasterization while costing $100-$150 less. AMD’s RDNA 4 brought significant ray tracing improvements, and FSR 4 now genuinely rivals DLSS.
The RTX 5070 Ti at $749-$820 remains solid if you prioritize DLSS and ray tracing. For enthusiasts exploring detailed game guides, these mid-range options still deliver impressive visuals.
Budget alternatives include the Intel Arc B580 at $249-$279 for 1080p gaming, or the RX 9060 XT at $349 with its generous 16GB VRAM—double what Nvidia offers at that price point.
How Long Will the GPU Price Crisis Last?
Based on industry analysis, the memory shortage should persist through at least 2027, with some analysts predicting pressure through 2028. Building new semiconductor facilities takes years and billions of dollars.
A Sapphire representative predicted DRAM prices will begin stabilizing in 6-8 months, but cautioned “it may not be the prices we want.” TrendForce analyst Avril Wu expects DRAM prices to rise another 40% in the coming quarter with no 2026 relief.
Final Thoughts on Nvidia and AMD GPU Prices
PC gaming isn’t dying, and graphics cards won’t become exclusively for billionaires. But we’re entering a period requiring more careful planning than in years.
My honest advice: if you’ve been shopping for a GPU and have the budget, pull the trigger now. Current Nvidia and AMD GPU prices represent a relative low point before 2026 increases. The RX 9070 XT at $599 and RTX 5070 at $549 offer value that may not exist in three months.
If your current card handles games adequately, you might ride it out—but be prepared for “waiting for better prices” to mean waiting until 2028. The “PC Master Race” used to be about performance superiority. In 2026, it might just be about who has the deepest pockets.
Frequently Asked Questions
Which GPU is better, Nvidia or AMD?
It depends on priorities. Nvidia excels in ray tracing and DLSS, ideal for cutting-edge gaming and content creation. AMD offers better price-to-performance and more VRAM at similar prices. For value, AMD’s RX 9070 XT is excellent; for absolute performance, Nvidia’s RTX 5090 leads. The performance gap has narrowed significantly with RDNA 4.
Can you have an AMD and Nvidia GPU?
Technically yes, with sufficient PCIe slots and power supply capacity. However, for gaming, running both offers no practical advantage—games only utilize one GPU at a time. This setup is primarily useful for specific professional workloads.
Do Nvidia GPUs work well with AMD?
Nvidia GPUs work perfectly with AMD processors. The Ryzen 9800X3D paired with an RTX 5090 is currently one of the best gaming combinations available. There are no compatibility issues between the manufacturers.
Why is the dual GPU dead?
Modern single GPUs are powerful enough for most tasks, developers rarely optimize for multi-GPU setups, and technologies like DLSS and FSR make extracting more performance from one card more practical than running multiple GPUs together.
What is the disadvantage of AMD?
AMD’s main disadvantages include historically weaker ray tracing (though RDNA 4 closed much of this gap), less mature AI software support since CUDA dominates, and no ultra-high-end option competing with the RTX 5090.
Do gamers prefer NVIDIA or AMD?
Steam surveys show Nvidia holds 75-80% market share due to ray tracing, DLSS, and consistent driver support. However, AMD’s share is growing among budget-conscious gamers who value the RX 9070 series’ performance-per-dollar.