Intel Arc A370M vs Nvidia RTX 500 Ada Generation Laptop GPU

Intel Arc A370M

The Intel Arc A370M is a dedicated mobile entry level graphics card for laptops based on Xe HPG microarchitecture. It uses the ACM-G11 chip and offers all 8 Xe-cores (128 ALUs), 8 ray tracing units and a 64-bit memory bus for 4 GB GDDR6 graphics memory. The chip integrates 7.2 billion transistors and the die measures 157 mm2.

The performance is in the entry level of dedicated graphics cards for laptops in 2022. AMD did publish some gaming benchmarks with medium settings in 5 games that showed that the Radeon RX 6500M is clearly faster. Intel only compared the A370M (with a faster CPU) to the slower Iris iGPU in the i7-1280P (see below). Thanks to the additional 2 cores and higher clock speeds, the entry level Arc A350M should be clearly left behind.

The chip also integrates two media engines for VP9, AVC, HEVC, and AV1 8k en- and decoding. The 4 display pipes support up to 4x 4k120 HDR via DMI 2.0B or DisplayPort 2.0 10G. 

The A370M is produced in 6nm at TSMC (N6 process) and supports dynamic power share (using Deep Link) with 12th gen Intel CPUs (Alder Lake).

Nvidia RTX 500 Ada Generation Laptop GPU

The Nvidia RTX 500 Ada Generation, not to be confused with the A500, P500 and the T500, is a lower-end professional graphics card for use in laptops that sports 2,048 CUDA cores and a paltry 4 GB of GDDR6 VRAM. We believe this graphics card to be a heavily cut-down GeForce RTX 4050 Laptop; therefore, both should employ the Ada Lovelace AD107 chip built with TSMC's 5 nm process. The RTX 500 was launched in February 2024. The Nvidia-recommended TGP range for this graphics card is moderately wide at 35 W to 60 W [the second figure includes the Dynamic Boost, it seems] leading to noticeable performance differences between different systems powered by what is supposed to be the same graphics card.

Quadro series graphics cards ship with much different BIOS and drivers than GeForce cards and are targeted at professional users rather than gamers. Commercial product design, large-scale calculations, simulation, data mining, 24 x 7 operation, certified drivers - if any of this sounds familiar, then a Quadro card will make you happy.

Architecture and Features

Ada Lovelace brings a range of improvements over older graphics cards utilizing the outgoing Ampere architecture. It's not just a better manufacturing process and a higher number of CUDA cores that we have here; under-the-hood refinements are plentiful, including an immensely larger L2 cache, an optimized ray tracing routine (a different way to determine what is transparent and what isn't is used), and other changes. Naturally, these graphics cards can both encode and decode some of the most widely used video codecs, AVC, HEVC and AV1 included; they also support a host of proprietary Nvidia technologies, including Optimus and DLSS 3, and they can certainly be used for various AI applications.

The RTX 500 Ada features 16 RT cores of the 3rd generation, 64 Tensor cores of the 4th generation and 2,048 CUDA cores. Increase those numbers by 25%, and you get the RTX 1000 Ada - as long as we pay no attention to clock speed differences, of course. Unlike costlier Ada Generation professional laptop graphics cards, the RTX 500 comes with just 4 GB of non-ECC VRAM; the lack of error correction makes this card less suitable for super-important tasks and round-the-clock operation. The VRAM is just 64-bit wide, delivering an anemic bandwidth of ~128 GB/s.

The RTX 500 Ada Generation makes use of the PCI-Express 4 protocol, just like Ampere-based cards did. 8K SUHD monitors are supported, however, DP 1.4a video outputs may prove to be a bottleneck down the line.

Performance

At 50 W (35 W + 15 W Dynamic Boost), the graphics card can handle most 2023 and 2024 games like Baldur's Gate 3 at 1080p on high graphics settings. With a Geekbench 6.2 OpenCL GPU score of 61,500 points and a Blender v3.3 Classroom CUDA score of 71 seconds, it's clear the Ada is so much faster than any integrated GPUs on the market including the 890M.

Power consumption

With the latest Nvidia graphics cards, laptop makers are free to set the TGP according to their needs within a fairly wide range. With the RTX 500 Ada, we have the lowest value recommended sitting at just at 35 W while the highest value is 60 W [this most likely includes Dynamic Boost]. Real-world performance of the slowest RTX 500 Ada will probably be around 40% lower than that of the fastest one.

Last but not the least, the improved 5 nm process (TSMC 4N) the AD107 chip is built with makes for decent energy efficiency, as of early 2024.