Recently, Nvidia and MediaTek have been in talks to be secretly developing a new APU, and now it seems to bolster the gaming laptop market. The reports unfolded that Nvidia and MediaTek have entered into an alliance for their ‘rumored’ APU (integrating CPU and GPU chips) for gaming laptops. It is anticipated that the laptops will offer the same performance as an RTX 4070-equipped laptop, but with better energy efficiency. Sounds interesting, right? Let’s know the truth and explore what gaming laptop APU brings to the table.
What’s the Rumour?
The rumour reveals that Nvidia APU will use 65W power to compete against the 120W RTX 4070 gaming laptop, together with the Arm-based chip tipped to adopt a Blackwell-based GPU, which we have seen in the latest GPU Driver. There is nothing clear about what GPU it will be based on. However, the past claims show it would utilise a smaller GP10 Blackwell chip for laptops. However, there is a potential reason to trust that it could use the GB206 model, which is the same as RTX 5060 Ti or RTX 5060.
Reportedly, the gaming brand Alienware, by Dell, will be one of the leaders to introduce new gaming laptop APU. This suggests that we could see new Alienware notebooks that are thinner and deliver extended battery life, if the rumours about Nvidia and MediaTek’s chip are true. This new APU is anticipated to roll out in late 2025 or early 2026. This suggests that it would be announced during CES 2026. Firstly, the chip was tipped for a launch during Computex 2025. However, it now seems we will get a closer look later.
Furthermore, another rumour is coming from Moore’s Law is Dead, showing that Nvidia’s APU for gaming laptops would deliver a Thermal Design Power of 80W to 120W. However, it also indicates a late 2025 or early 2026 release.
What is an APU, And Why is Nvidia interested?
APU or Accelerated Processing Unit is a term derived by AMD to market a CPU and GPU integrated into the same chip. This may sound the same as System-on-Chip (SOC), and you are absolutely right if thinking so. However, an SOC includes more than mere CPU and GPU, but also controllers and networking.
Recently, it has been paired with a Neural Processing Unit for basic AI tasks. For example, Qualcomm’s Snapdragon X chips are found in Copilot+ PCs. The report that Nvidia is working on an APU did not come suddenly, but the rumors were already there since early 2023 when Reuters sowed the seeds about Nvidia and AMD collaborating on ARM processors for Windows PCs.
We already know that Nvidia is charging up the much-anticipated Nintendo Switch 2 with its Tegra T239 and APU using an ARM processor and Ampere GPU. Nvidia is also focusing on DGX Spark, which is an AI supercomputer tool backed by its GB10 Grace Blackwell chip. MSI unveiled a new EdgeXpert MS-C931 desktop PC on the basis of the GDGX Spark platform at Computex 2025. It is prominent that Team Green is set to leave its GPU comfort and enter a dominant role in the world of Windows laptops.
Rise of Thinner, Powerful Gaming Laptop APU
Nvidia and MediaTek’s ARM–based chip is all set to boost the ultraportable gaming laptops with great power efficiency. This probably means a greater battery life in gaming notebooks. We have already seen Qualcomm’s Snapdragon X Series, which provides a major battery and performance hike. Though Nvidia’s own chip supports its GPU tech, a gaming laptop APU may experience a major performance boost. It already seems to garner attention from the competition.
Nvidia RTX 4070 Performance
Overall, the latest RTX 4070 graphics card performs similarly to the last-gen RTX 3080. It is faster in some games but may fall in others. However, the margins are pretty slim, and the card appears to perform the same as the RTX 3080. When it comes to competition, the RTX 4070 presently has none–at least while talking about the new-gen GPUs in the market. AMD has not launched mid-range GPUs into its RX series yet. Hence, only GPUs that may compete with the 4070 presently are the last-gen RX 6000 series GPUs. These are based on the previous structure, and the efficiency is not as same when compared to the new-gen GPUs.
I am eager to see the competition between Nvidia and AMD for the newest generation after the launch of GPUs. However, presently, the RTX 4070 is the budget-friendly Nvidia GPU that is associated with the RTX 40 series.
Features of RTX 4070
This new graphics card from Nvidia is built upon the ‘Ada Lovelace’ architecture that utilizes TSMC’s 4N process. This manufacturing process is an improved version of the 5nm process. As it is the new architecture, there are larger improvements to the performance. In a nutshell, this means greater performance with lower power.
The initial user experience with RTX 4070 suggests that it performs like RTX 3080 but with dramatically low power consumption. This suggests that the new architecture brought a marginal improvement in performance.
DLSS 3 AI Frame Generation is NVIDIA’s innovative upscaling tool, which is designed for RTX 40 series graphics cards. By enabling DLSS 3, the framerate immediately turns on to improve the FPS irrespective of the setup being used. It is still artificial frames generated by AI, and the gaming experience may be a little disappointing sometimes. However, it is one of the best additions to this upgraded version of GPUs.
The Dual AV1 encoding on the RTX 4070 allows gamers to stream their game on channels like Twitch or YouTube at an improved fidelity in comparison to the x264 codec. AV1 support is gradually expanding in several places, and it will soon become an important feature.
Summary
Nvidia and MediaTek are jointly in rumours for launching an APU that integrates CPU, GPU, and NPU in a single chip that would make a dramatic shift in gaming laptops. However, we have to wait and see how the chip comes up and launch later this year or later.
