NVIDIA’s rumored GeForce RTX 50 Super refresh is reportedly on hold as the cost of the 3GB GDDR7 memory chips needed for its higher-VRAM designs rises sharply. TrendForce, citing VideoCardz and Tom’s Hardware, says at least one add-in-board partner has received GPUs but was told not to proceed with a launch while memory pricing remains unsettled.
The key claim is the price gap: sources cited by VideoCardz put a 3GB GDDR7 chip at roughly $60–$70, versus about $20 for a conventional 2GB part. NVIDIA has not announced an RTX 50 Super lineup, confirmed those prices, or set a release date, so the entire situation remains a supply-chain rumor rather than a product delay in the formal sense.
The reported refresh would use 3GB GDDR7 devices to increase VRAM capacity without changing the width of the graphics card’s memory bus. A 256-bit card populated with eight 3GB chips could carry 24GB of memory, for example, rather than 16GB with eight 2GB chips. A 192-bit design with six chips could reach 18GB instead of 12GB.
That is a meaningful upgrade for modern games with large texture sets, GPU-accelerated creative workloads, and local AI use. It is also a costly way to get there if the reported module pricing is accurate. A card using eight of the 3GB chips would carry several hundred dollars more in memory cost than an otherwise comparable design using 2GB parts.
Tom’s Hardware reports that NVIDIA would face an unattractive choice: lift retail pricing above planned MSRPs, accept lower margins across its board partners, or wait for the GDDR7 market to cool. Board makers generally have little room to absorb a large bill-of-materials increase, especially when they must also account for cooling, power delivery, logistics, and retailer margins.
For Windows PC builders, the immediate implication is simple: do not delay a needed upgrade based on expectation of an imminent RTX 50 Super release. The current RTX 50-series cards remain the only announced consumer products, and any Super launch could arrive later than expected or at prices that blunt the value of its additional memory.
The key claim is the price gap: sources cited by VideoCardz put a 3GB GDDR7 chip at roughly $60–$70, versus about $20 for a conventional 2GB part. NVIDIA has not announced an RTX 50 Super lineup, confirmed those prices, or set a release date, so the entire situation remains a supply-chain rumor rather than a product delay in the formal sense.
Why the memory matters
The reported refresh would use 3GB GDDR7 devices to increase VRAM capacity without changing the width of the graphics card’s memory bus. A 256-bit card populated with eight 3GB chips could carry 24GB of memory, for example, rather than 16GB with eight 2GB chips. A 192-bit design with six chips could reach 18GB instead of 12GB.That is a meaningful upgrade for modern games with large texture sets, GPU-accelerated creative workloads, and local AI use. It is also a costly way to get there if the reported module pricing is accurate. A card using eight of the 3GB chips would carry several hundred dollars more in memory cost than an otherwise comparable design using 2GB parts.
Tom’s Hardware reports that NVIDIA would face an unattractive choice: lift retail pricing above planned MSRPs, accept lower margins across its board partners, or wait for the GDDR7 market to cool. Board makers generally have little room to absorb a large bill-of-materials increase, especially when they must also account for cooling, power delivery, logistics, and retailer margins.
A refresh without a schedule
The rumored products include RTX 5080 Super, RTX 5070 Ti Super, and RTX 5070 Super models, with higher VRAM their principal differentiator. Earlier reporting from VideoCardz suggested the refresh might not arrive until CES 2027 at the earliest, but there is no public roadmap from NVIDIA to corroborate that timing.For Windows PC builders, the immediate implication is simple: do not delay a needed upgrade based on expectation of an imminent RTX 50 Super release. The current RTX 50-series cards remain the only announced consumer products, and any Super launch could arrive later than expected or at prices that blunt the value of its additional memory.
References
- Primary source: TrendForce
Published: 2026-07-20T14:44:27+00:00
[News] NVIDIA’s RTX 50 Super Refresh Faces Launch Uncertainty as 3GB GDDR7 Reportedly Costs 3× Than 2GB
Memory is becoming a critical bottleneck in the AI era, and even NVIDIA is not immune to the pressure. Citing VideoCardz, Tom’s Hardware reports that ...www.trendforce.com - Related coverage: tomshardware.com
Nvidia RTX 50 Super GPUs are reportedly ready, but stuck in limbo due to excessive GDDR7 pricing — 3GB GDDR7 module costs triple the price of 2GB | Tom's Hardware
These GPUs will probably be more expensive than ever.www.tomshardware.com - Related coverage: techradar.com
Nvidia RTX 5000 Super GPUs rumored to be ready — but they're on hold, and the reason why makes me nervous about pricing | TechRadar
These Super refreshes are still coming, by all accounts, but they could make quite the dent in your bank balance.www.techradar.com - Related coverage: videocardz.com
- Related coverage: tomshw.it
- Related coverage: tweaktown.com
NVIDIA's RTX 50 Super GPUs have reached board partners, but launch is on hold over 3GB GDDR7 pricing
The 3GB GDDR7 chips the Super series depends on cost three times more than standard modules, adding hundreds of dollars to each card's bill of materials.
www.tweaktown.com
