Nvidia is considering an aggressive move to address the shortage of high-end high-bandwidth memory chips: using lower video memory configurations than originally planned on the next-generation graphics processor Rubin Ultra. According to three people familiar with the testing process, over the past few weeks, Nvidia has tested at least three versions of Rubin Ultra video cards, some of which have lower memory than the specifications initially announced by the company. The three people familiar with the matter said that part of the reason Nvidia is considering the low video memory version is that it may not be able to obtain enough high-end memory to meet the supply requirements of the original design. Reduced video memory will have an impact on chip performance, but Nvidia can make up for it in other ways. However, if an artificial intelligence company uses this reduced video memory chip to run a large AI model, it will need to use more chips than the original one. Nvidia's considerations reflect that under the AI boom, it is already difficult for the chip industry's production capacity to meet the skyrocketing demand of data centers. As a result, there was a serious shortage of memory chips, the core component of GPUs, which boosted the price of the product. Rising costs have triggered a ripple effect across the technology industry: companies are forced to spend more than their budgets, and companies such as Apple have raised the price of hardware terminals as a result.

Nvidia is considering an aggressive move to address the shortage of high-end high-bandwidth memory chips: using lower video memory configurations than originally planned on the next-generation graphics processor Rubin Ultra. According to three people familiar with the testing process, over the past few weeks, Nvidia has tested at least three versions of Rubin Ultra video cards, some of which have lower memory than the specifications initially announced by the company. The three people familiar with the matter said that part of the reason Nvidia is considering the low video memory version is that it may not be able to obtain enough high-end memory to meet the supply requirements of the original design. Reduced video memory will have an impact on chip performance, but Nvidia can make up for it in other ways. However, if an artificial intelligence company uses this reduced video memory chip to run a large AI model, it will need to use more chips than the original one. Nvidia's considerations reflect that under the AI boom, it is already difficult for the chip industry's production capacity to meet the skyrocketing demand of data centers. As a result, there was a serious shortage of memory chips, the core component of GPUs, which boosted the price of the product. Rising costs have triggered a ripple effect across the technology industry: companies are forced to spend more than their budgets, and companies such as Apple have raised the price of hardware terminals as a result.