Qualcomm Snapdragon X Plus X1P-42-100 vs Intel Core i7-1265U

Qualcomm Snapdragon X Plus X1P-42-100

The Snapdragon X Plus 8-core (X1P-42-100) is a relatively affordable ARM architecture processor for use in Windows laptops that was unveiled in Sep 2024. This Qualcomm SoC features 8 Oryon CPU cores running at up to 3.4 GHz, along with the 1.7 TFLOPS X1-45 iGPU and the 45 TOPS Hexagon NPU. The super-fast LPDDR5x-8448 memory controller, USB 4.0 support, TB 4 support and PCIe 4 support are all onboard as well.

Architecture and Features

The Oryon cores (2 clusters of essentially identical cores; 8 threads) are mostly based on Nuvia IP and they reportedly make use of the ARM v8.7 microarchitecture. Much like modern AMD and Intel processors, the Snapdragon chip is compatible with USB 4 and thus with Thunderbolt 4 however it does not appear to support eGPUs as of September 2024.

The X1P-42-100 is based on the smaller die codenamed Purwa, unlike most X Elite and X Plus processors. It is believed to have at least 8 PCIe 4 and 4 PCIe 3 lanes for connecting various kinds of devices. NVMe SSDs are supported with a throughput of up to 7.9 GB/s; furthermore, most laptops built around the chip are expected to have 16 GB of LPDDR5x-8448 RAM. There is also a 45 TOPS NPU present for accelerating AI workloads.

Performance

A lot depends on the power targets of a specific system, just like it is with AMD, Intel and Nvidia products. That being said, the average X1P-42-100 competes with older chips such as the Core i5-1245U when executing multi-threaded x86 code. With apps compiled specifically with Windows on ARM in mind, a Core i7-1360P-like performance level is to be expected which is not bad at all for what is supposed to be a budget CPU.

The 3.4 GHz clock speed is only achievable in single-thread workloads. When under multi-threaded load, the CPU cores will run at no more than 3.2 GHz.

Graphics

The X1-45 used here delivers up to 1.7 TFLOPS of performance. Unlike the much faster 3.8 TFLOPS and 4.6 TFLOPS X1-85 iGPUs, this little guy here has much fewer unified shaders and runs at lower clock speeds, too. Games put it a little behind the GeForce MX350; this kind of performance is sufficient for older games and sub-900p resolutions only.

AVC, HEVC and AV1 video codecs can be both hardware-decoded and hardware-encoded whereas with VP9, only decoding is possible. The highest monitor resolution supported is UHD 2160p.

Power consumption

Expect to see anything between 15 W and 30 W under long-term workloads depending on the system and the power profile chosen. The number includes RAM.

The SoC is built with TSMC's N4P process for better-than-average power efficiency, as of H2 2024.

Intel Core i7-1265U

The Intel Core i7-1265U is a high-end mobile CPU for thin and light laptops based on the Alder Lake architecture. It was announced in early 2022 and offers 2  performance cores (P-cores, Golden Cove architecture) and 8 efficient cores (E-cores, Gracemont architecture). The P-cores support Hyper-Threading leading to 12 supported threads when combined with the E-cores. The clock rate ranges from 1.8 to 4.8 GHz on the performance cluster and 1.3 to 3.6 GHz on the efficient cluster. The performance of the E-cores should be similar to old Skylake cores (compare to the Core i7-6700HQ). All cores can use up to 12 MB L3 cache. The 1265U offers "Enterprise" vPro management features. Compared to the P-series (28W) or H-series (45W), the 15W TDP may limit the sustained performance.

Performance

Thanks to the 10 cores, the multi-threaded performance should be clearly faster than the older quad-core Tiger Lake CPUs with 15 Watt (like the Intel Core i7-1160G7). The sustained performance will be limited by the TDP setting and thermals and therefore only short term loads. Thanks to the architectural improvements of the big Golden Cove cores, the single thread performance should be better than similar clocked Tiger Lake CPUs.

Features

The integrated memory controller supports various memory types up to DDR5-4800, DDR4-3200, LPDDR5-5200, and LPDDR4x-4267. The Thread Director (in hardware) can support the operating system to decide which thread to use on the performance or efficient cores for the best performance. For AI tasks, the CPU also integrates GNA 3.0 and DL Boost (via AVX2). Quick Sync in version 8 is the same as in the Rocket Lake CPUs and supports MPEG-2, AVC, VC-1 decode, JPEG, VP8 decode, VP9, HEVC, and AV1 decode in hardware. The CPU only supports PCIe 4.0 (x8 for a GPU and two x4 for SSDs).

The integrated graphics adapter is based on the Xe-architecture and offers all 96 EUs (Execution Units) operating at up to 1.25 GHz.

The CPU is rated at 15 W base power (TDP, PL1) and 55 W max. Turbo power (PL2) and the whole SoC is manufactured on a 10nm process at Intel, which is known as Intel 7.