GPU Comparison

The Intel UHD Graphics 710 and NVIDIA GeForce 920MX are two end-of-life mobile-class graphics solutions that trade blows depending on the API used. In the head-to-head benchmarks, the NVIDIA GeForce 920MX takes the Geekbench OpenCL test with a score of 4068, beating the Intel UHD Graphics 710’s 3496 by a margin of 14.1%. However, the Intel UHD Graphics 710 strikes back decisively in Geekbench Vulkan, scoring 4088 against the 920MX’s 2987, a 36.9% advantage. This split result means the data shows a 1-1 tie in wins, with each GPU dominating in a different workload type.

Head-to-Head Benchmarks

The most significant victory for the NVIDIA GeForce 920MX is in OpenCL compute performance. With a score of 4068 versus the Intel UHD Graphics 710’s 3496, the 920MX is 572 points ahead, translating to a 14.1% lead. This is a substantial margin, suggesting that the discrete NVIDIA solution handles general-purpose compute tasks, often found in older game engines and productivity applications, with notably more efficiency. The Intel part’s score of 3496 places it slightly below its own average, indicating that OpenCL is not its strong suit.

Conversely, the Intel UHD Graphics 710 achieves a commanding win in the Vulkan graphics test. Its score of 4088 eclipses the 920MX’s 2987 by a massive 1101 points, a 36.9% difference. This is a stark reversal of the OpenCL result. Vulkan, a modern low-level API, clearly favors the Intel architecture’s design. The 920MX’s Vulkan score of 2987 is notably lower than its own OpenCL score, suggesting the older Maxwell architecture struggles with this newer API’s requirements.

Looking at the average benchmark scores across all tests, the Intel UHD Graphics 710 holds a slight edge. Its average of 3792 is higher than the 920MX’s 3528, a difference of 264 points. This places the Intel part in the 22nd percentile of all GPUs, while the NVIDIA part sits at the 21st percentile. In terms of nearest rivals, the Intel UHD Graphics 710 is essentially tied with the NVIDIA GeForce GTX 650, which has an average score of 3823 (a 0.8% difference), and the NVIDIA GeForce MX110, at 3834 (1.1% difference). The 920MX, meanwhile, is closest to the NVIDIA GeForce GT 545 (3594, 1.8% difference) and the NVIDIA RTX 5000 Mobile Ada Generation (3596, 1.9% difference), an interesting pairing of a low-end Maxwell part with a top-end Ada part in aggregate scoring.

Architecture Differences

The architectural divide between these two is a clash of generations. The Intel UHD Graphics 710 is built on Intel’s Alder Lake chip using the Generation 12.2 architecture and a 10 nm process node fabricated by Intel. In contrast, the NVIDIA GeForce 920MX uses the GM108S chip, based on the older Maxwell architecture, manufactured on TSMC’s 28 nm process. This process node difference is significant; the Intel part uses a much smaller manufacturing process, allowing for higher efficiency per transistor.

The core configurations differ sharply. The Intel UHD Graphics 710 has 128 shading units, 8 texture mapping units (TMUs), and 8 render output units (ROPs). The NVIDIA GeForce 920MX doubles the shading units to 256, triples the TMUs to 24, but keeps the same 8 ROPs. This gives the 920MX a theoretical texture fillrate of 23.83 GTexel/s, more than double the Intel part’s 10.40 GTexel/s. However, the Intel part has a higher pixel rate at 10.40 GPixel/s versus the 920MX’s 7.944 GPixel/s, despite having the same number of ROPs, due to its higher boost clock.

Clock speeds also tell a story of different design philosophies. The Intel UHD Graphics 710 has a base clock of 300 MHz and a boost clock of 1300 MHz, a wide range that allows for aggressive power saving at idle and higher performance under load. The NVIDIA 920MX operates at a much higher base clock of 965 MHz but only boosts to 993 MHz, a narrow range limited by its 16W TDP. The Intel part has a slightly lower TDP of 15W. The 920MX features 1,020 million transistors on a 77 mm² die, with a transistor density of 13.2M / mm², while the Intel part’s transistor count and die size are not listed.

Memory configurations are fundamentally different. The Intel UHD Graphics 710 uses System Shared memory, meaning its size, type, and bus width are all system-dependent, with bandwidth listed as "System Dependent." The NVIDIA 920MX has a dedicated 2 GB of DDR3 memory on a 64-bit bus, providing a fixed bandwidth of 14.40 GB/s. This is a critical distinction for gaming, as dedicated VRAM avoids competing with the CPU for system memory bandwidth. The 920MX also has a dedicated memory clock of 900 MHz (1800 Mbps effective), whereas the Intel part’s memory clock is tied to the system.

FAQ

Q: Which GPU has the higher average benchmark score?

A: The Intel UHD Graphics 710 has a higher average benchmark score of 3792, compared to the NVIDIA GeForce 920MX’s 3528. This places the Intel part in the 22nd percentile of all GPUs, one point higher than the 920MX’s 21st percentile.

Q: Is the NVIDIA GeForce 920MX faster in all APIs?

A: No. The data shows the 920MX wins in Geekbench OpenCL with a score of 4068 versus 3496 for the Intel part. However, the Intel UHD Graphics 710 wins decisively in Geekbench Vulkan, scoring 4088 against the 920MX’s 2987.

Q: What is the transistor count of the NVIDIA GeForce 920MX?

A: The NVIDIA GeForce 920MX has 1,020 million transistors on a 77 mm² die, manufactured on a 28 nm process by TSMC. The Intel UHD Graphics 710’s transistor count is not listed in the data.

Q: How does the memory bandwidth compare?

A: The NVIDIA GeForce 920MX has a fixed memory bandwidth of 14.40 GB/s from its dedicated 2 GB DDR3 memory on a 64-bit bus. The Intel UHD Graphics 710 uses system-shared memory, so its bandwidth is listed as "System Dependent" and cannot be directly compared.

Q: Which GPU has a higher boost clock?

A: The Intel UHD Graphics 710 has a higher boost clock of 1300 MHz, compared to the NVIDIA GeForce 920MX’s boost clock of 993 MHz. The 920MX does have a higher base clock at 965 MHz versus the Intel part’s 300 MHz.

Q: What are the direct API version supports?

A: Both GPUs support DirectX 12, OpenGL 4.6, and Vulkan 1.4. However, the Intel UHD Graphics 710 supports DirectX 12 (12_1), while the NVIDIA GeForce 920MX supports DirectX 12 (11_0), indicating a higher feature level for the Intel part.

The Verdict

Based strictly on the data, the choice depends entirely on your workload. For applications that leverage OpenCL, the NVIDIA GeForce 920MX is the clear winner, offering a 14.1% performance advantage that will be noticeable in compute-heavy tasks. Its dedicated 2 GB of VRAM also provides a fixed memory pool that is beneficial for storing textures and frame buffers without relying on system memory.

However, for modern gaming or applications that utilize the Vulkan API, the Intel UHD Graphics 710 is the superior choice. Its 36.9% lead in Vulkan is a massive gap, indicating that it is significantly more future-proof for newer software titles. Its higher boost clock of 1300 MHz and better DirectX feature level (12_1) also point to a more capable architecture for contemporary rendering techniques.

In terms of overall average performance, the Intel UHD Graphics 710 edges out the 920MX by 7.5%, placing it slightly higher in the global GPU percentile rankings. The Intel part also has the advantage of a more modern 10 nm process node, which should lead to better power efficiency per clock. The final verdict is that the Intel UHD Graphics 710 is the better all-rounder for modern workloads, while the NVIDIA GeForce 920MX is a specialized tool for OpenCL compute and legacy applications.

Specification Differences

The following specifications differ between the two GPUs:

• Manufacturer: Intel vs NVIDIA
• Chip: Alder Lake vs GM108S
• Architecture: Generation 12.2 vs Maxwell
• Generation: HD Graphics (Alder Lake) vs GeForce 900M
• Process Node: 10 nm vs 28 nm
• Foundry: Intel vs TSMC
• Transistors: Not listed vs 1,020 million
• Die Size: Not listed vs 77 mm²
• Transistor Density: Not listed vs 13.2M / mm²
• Base Clock: 300 MHz vs 965 MHz
• Boost Clock: 1300 MHz vs 993 MHz
• Memory Clock: System Shared vs 900 MHz (1800 Mbps effective)
• Memory Size: System Shared vs 2 GB
• Memory Type: System Shared vs DDR3
• Memory Bus Width: System Shared vs 64 bit
• Memory Bandwidth: System Dependent vs 14.40 GB/s
• Shading Units: 128 vs 256
• TMUs: 8 vs 24
• Pixel Rate: 10.40 GPixel/s vs 7.944 GPixel/s
• Texture Rate: 10.40 GTexel/s vs 23.83 GTexel/s
• FP32 Performance: 332.8 GFLOPS vs 508.4 GFLOPS
• FP16 Performance: 665.6 GFLOPS (2:1) vs Not listed
• TDP: 15 W vs 16 W
• Slot Width: IGP vs MXM Module
• Power Connectors: Not listed vs None
• Bus Interface: Ring Bus vs PCIe 3.0 x8
• Display Outputs: Motherboard Dependent vs Portable Device Dependent
• Release Date: 2022-01-03 vs 2016-03-24
• Predecessor: Not listed vs GeForce 800M
• Successor: Not listed vs GeForce 10 Mobile
• DirectX Support: 12 (12_1) vs 12 (11_0)

Where Each One Wins

Intel UHD Graphics 710 Wins:

• Vulkan API workloads: A 36.9% performance lead (4088 vs 2987) makes this the go-to for Vulkan-based games and applications.
• Overall average performance: A higher average benchmark score of 3792 vs 3528.
• Pixel fillrate: A higher pixel rate of 10.40 GPixel/s vs 7.944 GPixel/s, beneficial for fillrate-limited scenarios.
• FP16 compute: Offers 665.6 GFLOPS of FP16 performance, while the 920MX has no listed FP16 capability.
• Modern process technology: Built on a 10 nm node versus 28 nm, suggesting better power efficiency.
• Higher DirectX feature level: Supports DirectX 12_1, while the 920MX only supports 12_0.

NVIDIA GeForce 920MX Wins:

• OpenCL compute: A 14.1% lead in Geekbench OpenCL (4068 vs 3496), making it better for OpenCL-accelerated tasks.
• Texture fillrate: A significantly higher texture rate of 23.83 GTexel/s vs 10.40 GTexel/s, which helps with texture-heavy rendering.
• FP32 compute: Higher raw FP32 performance at 508.4 GFLOPS vs 332.8 GFLOPS.
• Dedicated memory: Has a fixed 2 GB of DDR3 VRAM with 14.40 GB/s bandwidth, ensuring consistent memory performance.
• Shading units: Double the shading units (256 vs 128) for more parallel processing.
• Higher base clock: A much higher base clock of 965 MHz vs 300 MHz, which can help in lightly-threaded or low-load scenarios.