Nvidia confirms GeForce RTX GPU supply is “constrained”
Nvidia confirms that “memory supply is constrained”
In a short statement to HardwareLuxx, Nvidia has confirmed that its ability to produce GeForce RTX GPUs is “constrained” by “memory supply”. This seemingly confirms rumours from last month that claimed that Nvidia could not procure enough GDDR7 memory to meet demand.
In response to allegations that Nvidia has effectively discontinued memory-rich GPUs like its RTX 5070 Ti and RTX 5060 Ti 16GB, Nvidia has stated that “we continue to ship all GeForce SKUs”. However, it is worth noting that Nvidia only needs to ship small quantities of affected GPU models for this statement to be correct.
Nvidia also states that it is working with its partners to “maximise memory availability”. That said, memory manufacturers are currently operating at capacity. This makes it hard for them to increase their supply of GDDR7 memory modules.
Question (HLuxx): What changes are occurring in the GeForce ecosystem given the current memory market?
Response (Nvidia): Demand for GeForce RTX GPUs is strong, and memory supply is constrained. We continue to ship all GeForce SKUs and are working closely with our suppliers to maximize memory availability.
Nvidia’s statement confirms what we already suspected. Not enough GDDR7 memory is being manufactured to meet demand. This is reportedly why Nvidia has postponed or cancelled its planned RTX 50 SUPER series of GPUs. Why would Nvidia produce GPUs with more memory when their GPU manufacturing is memory-constrained? If these reports are true, Nvidia’s RTX 50 SUPER series GPUs are a victim of the global (AI-induced) memory shortage.
You can join the discussion about Nvidia’s statement on RTX 50-series GPU availability on the OC3D Forums.
