The GPU or graphics processing unit will probably be a part of Nvidia’s newest technology Blackwell-architecture AI processors and is predicted to be priced between $6,500 and $8,000, effectively under the $10,000-$12,000 the H20 bought for, in response to two of the sources.
The lower cost displays its weaker specs and less complicated manufacturing necessities.
It is going to be primarily based on Nvidia’s RTX Professional 6000D, a server-class graphics processor and can use standard GDDR7 memory as an alternative of extra superior excessive bandwidth reminiscence, the 2 sources stated.
They added it might not use Taiwan Semiconductor Manufacturing Co’s superior Chip-on-Wafer-on-Substrate (CoWoS) packaging expertise.
The brand new chip’s worth, specs and manufacturing timing haven’t beforehand been reported.
Uncover the tales of your curiosity
The three sources Reuters spoke to for this text declined to be recognized as they weren’t authorised to talk to media. An Nvidia spokesperson stated the corporate was nonetheless evaluating its “restricted” choices. “Till we choose a brand new product design and obtain approval from the U.S. authorities, we’re successfully foreclosed from China’s $50 billion information heart market.”
TSMC declined to remark.
Market share plunge
China stays an enormous marketplace for Nvidia, accounting for 13% of its gross sales prior to now monetary 12 months. It is the third time that Nvidia has needed to tailor a GPU for the world’s second-largest economic system after restrictions from U.S. authorities who’re eager to stymie Chinese language technological improvement.
After the U.S. successfully banned the H20 in April, Nvidia initially thought-about creating a downgraded model of the H20 for China, sources have stated, however that plan did not work out.
Nvidia CEO Jensen Huang stated final week the corporate’s older Hopper structure – which the H20 makes use of – can not accommodate additional modifications beneath present U.S. export restrictions.
Reuters was unable to find out the product’s remaining title.
Chinese language brokerage GF Securities stated in a word printed on Tuesday that the brand new GPU would doubtless be known as the 6000D or the B40, although it didn’t disclose pricing or cite sources for the knowledge.
Based on two of the sources, Nvidia can also be creating one other Blackwell-architecture chip for China that’s set to start manufacturing as early as September. Reuters was not instantly capable of verify specs of that variant.
Nvidia’s market share in China has plummeted from 95% earlier than 2022, when U.S. export curbs that impacted its merchandise started, to 50% at present, Huang informed reporters in Taipei this week. Its fundamental competitor is Huawei which produces the Ascend 910B chip.
Huang additionally warned that if U.S. export curbs proceed, extra Chinese language prospects will purchase Huawei’s chips.
The H20 ban compelled Nvidia to jot down off $5.5 billion in stock and Huang informed the Stratechery podcast on Monday that the corporate additionally needed to stroll away from $15 billion in gross sales.
The newest export restrictions launched new limits on GPU reminiscence bandwidth – an important metric measuring information transmission speeds between the principle processor and reminiscence chips. This functionality is especially necessary for AI workloads that require in depth information processing.
Funding financial institution Jefferies estimates that the brand new rules cap reminiscence bandwidth at 1.7-1.8 terabytes per second. That compares with the 4 terabytes per second that the H20 is able to.
GF Securities forecast the brand new GPU will obtain roughly 1.7 terabytes per second utilizing GDDR7 reminiscence expertise, simply throughout the export management limits.