NVIDIA’s next-gen “AI PC” SOC is rumored to incorporate Arm-based Blackhawk CPUs, Blackwell RTX GPUs & LPDDR6 memory.
NVIDIA’s Next-Gen SOC Could Shake Things Up In The AI PC Segment If These Rumored Specs Are Correct, Blackhawk Arm CPU Cores, Blackwell RTX GPU & LPDDR6 Memory All Onboard The Same Package
Recently, we have seen several rumors and reports of NVIDIA potentially planning an entry into the client PC space with its own SOC. There are reports that these SOCs will also open the door for new consoles, primarily handhelds in the beginning. Just yesterday, CEOs of both NVIDIA & Dell teased this venture & it looks like we have a new rumor that might give us an early hint at what these SOCs would look like.
https://x.com/XpeaGPU/status/1793469879334687122
The details come from @XpeaGPU who we quoted within the previous article that talked about a potential handheld and console SOC being designed by NVIDIA based on the same architecture. It looks like NVIDIA is looking to incorporate several next-generation IPs on a single package which is looking like the way to go when it comes to modern-day PC designs. We have seen with Apple and now with Intel’s Lunar Lake which offers multiple IPs on the same die with on-package memory and it makes for a very efficient and powerful solution in a small footprint.
https://x.com/Kepler_L2/status/1793523940968820899
For its next-generation SOC, NVIDIA is allegedly going to leverage a 3nm process node, reportedly either Intel or TSMC, and an advanced packaging solution to bring Arm-based Cortex X5 BlackHawk CPU cores & the Blackwell RTX GPU cores together. It’s plausible that NVIDIA would use the BlackHawk CPU cores since they are reportedly partnering with MediaTek to create these SOCs.
MediaTek is reported to be using the same Arm architecture for its upcoming Dimensity 9400 chip and is said to tackle the likes of Apple A17 Pro and Qualcomm’s Oryon architecture.
The use of Blackwell RTX GPUs isn’t that far-fetched either since MediaTek previously confirmed in a press release that they were going to use the latest RTX & AI GPU IP within its Dimensity SOCs so maybe we will be getting a modified version of that designed and optimized further by NVIDIA. The NVIDIA “AI PC” SOCs are also stated to feature LPDDR6 memory as an on-package design. LPDDR6 is going to feature 10,667 MT/s speeds from scratch and would go as high as 14,400 MT/s during its life cycle as revealed recently.
Another hint at this development comes directly from NVIDIA’s CEO, Jensen Huang, who states that the recent introduction of Copilot+ PCs by Microsoft has paved the way for opportunities within the client segment. Jensen says that Copilot+ PCs support ARM beautifully and Qualcomm Snapdragon X CPUs are a testament to that so that gives NVIDIA the motivation to try and do something even better within this segment.
Today, as you saw at the build yesterday, which I thought was really excellent, Satya announced the next-generation PCs, Copilot+ PC, which runs fantastically on NVIDIA’s RTX GPUs that are shipping in laptops. But it also supports ARM beautifully. And so it opens up opportunities for system innovation even for PCs.
Jensen Huang – NVIDIA CEO (Q1 2024 Earnings Call)
Once again, its too early to say what the final design and specifications for NVIDIA’s “AI PC” SOC for client PCs would look like but the company has industry-proven experience of incorporating high-end GPU architectures alongside datacenter ARM CPUs (Grace) so we look forward to see NVIDIA bring the same combo on the client side.