AMD Radeon 840M vs AMD Ryzen Z2 Extreme GPU Comparison

# FAQ

Q: How do the AMD Radeon 840M and AMD Ryzen Z2 Extreme GPU compare in raw compute performance?

A: The Ryzen Z2 Extreme GPU delivers 5.530 TFLOPS of FP32 performance, which is roughly 3.7 times the 1,484.8 GFLOPS of the Radeon 840M. The gap is substantial across every computational metric.

Q: What is the memory configuration difference between the two?

A: The Radeon 840M uses system-shared memory with system-dependent bandwidth, while the Ryzen Z2 Extreme GPU has 16 GB of dedicated LPDDR5X on a 128-bit bus with 128.0 GB/s of bandwidth. The latter provides a fixed, high-bandwidth memory subsystem.

Q: Which chip has a higher boost clock?

A: The Radeon 840M boosts to 2900 MHz, while the Ryzen Z2 Extreme GPU boosts to 2700 MHz. The 840M has a 200 MHz higher boost clock, but the Z2 Extreme GPU compensates with far more execution resources.

Q: How do the transistor counts and die sizes compare?

A: The Ryzen Z2 Extreme GPU uses a Strix Point chip with 34,000 million transistors on a 233 mm² die, giving a density of 145.9M / mm². The Radeon 840M's Krackan Point chip has unknown transistor and die size figures, though both are fabricated on TSMC's 4 nm process.

Q: What were the recorded benchmark results for each GPU?

A: The Ryzen Z2 Extreme GPU scored 516 in the 3DMark Steel Nomad DX12 test, placing it in the 1st percentile of all GPUs. The Radeon 840M has no recorded benchmark entries in the database.

Q: Who are the closest rivals to the Ryzen Z2 Extreme GPU?

A: The nearest rivals include Intel HD Graphics P4000 (avg score 534, 3.4% higher), AMD Radeon HD 6870 (avg score 536, 3.7% higher), AMD Radeon HD 6750M (avg score 484, 6.6% lower), and AMD Radeon HD 6770M (avg score 569, 9.3% higher).

# The Verdict

The data presents an unambiguous performance hierarchy. The AMD Ryzen Z2 Extreme GPU holds every meaningful advantage in the recorded specifications: it has 4 times the shading units (1024 vs 256), 4 times the texture units (64 vs 16), 6 times the ROPs (48 vs 8), and 4 times the ray tracing cores (16 vs 4). Its FP32 throughput of 5.530 TFLOPS dwarfs the 840M's 1,484.8 GFLOPS.

The 840M counters only in clock speed, with a 2900 MHz boost versus 2700 MHz, and in power efficiency, with a 15 W TDP versus 28 W. It is also an integrated graphics solution with no power connectors, making it suited for ultra-low-power portable devices.

However, the Ryzen Z2 Extreme GPU's benchmark result tells a sobering story. Its 3DMark Steel Nomad DX12 score of 516 places it in the 1st percentile of all GPUs, barely competitive with decade-old discrete graphics cards. The Intel HD Graphics P4000 scores 3.4% higher, and the AMD Radeon HD 6870 scores 3.7% higher. Only the AMD Radeon HD 6750M sits below it, with a score 6.6% lower.

For users requiring the highest compute throughput within a mobile or console platform, the Ryzen Z2 Extreme GPU is the clear choice based on raw specifications. Its 16 GB of LPDDR5X memory and 128.0 GB/s bandwidth provide a dedicated memory pool that the 840M lacks entirely. The 840M, conversely, fits into the lowest power envelope and may be preferable for devices prioritizing battery life over graphics performance. The 840M has no benchmark data, so its real-world standing cannot be quantified, but its specification sheet indicates a much weaker part.

# Head-to-Head Benchmarks

No direct head-to-head benchmark entries exist in the database for these two GPUs. However, the Ryzen Z2 Extreme GPU has a single recorded benchmark result, while the Radeon 840M has none. This absence of comparative data means the only quantitative assessment comes from the Z2 Extreme GPU's 3DMark Steel Nomad DX12 score of 516.

This score positions the Z2 Extreme GPU at the 1st percentile among all GPUs, indicating that 99% of recorded GPUs outperform it. The nearest rivals confirm this placement. The Intel HD Graphics P4000, an integrated solution from a much older generation, outperforms it by 3.4%. The AMD Radeon HD 6870, a discrete graphics card from 2010, beats it by 3.7%. The AMD Radeon HD 6770M, a laptop GPU from the same era, leads by 9.3%.

Only the AMD Radeon HD 6750M trails the Z2 Extreme GPU, scoring 6.6% lower. This cluster of results suggests the Z2 Extreme GPU performs at the level of entry-level discrete GPUs from roughly a decade ago. Despite its modern RDNA 3.5 architecture, 1024 shading units, and 16 GB of LPDDR5X memory, the recorded performance does not break out of this low percentile band.

The Radeon 840M cannot be placed in this comparison due to missing benchmark data. Its specification sheet, with one quarter the shading units and one eighth the ROPs of the Z2 Extreme GPU, strongly implies it would land well below the Z2 Extreme GPU in any 3DMark test. The pixel rate of the 840M is 23.20 GPixel/s versus 129.6 GPixel/s for the Z2 Extreme GPU, a 5.6 times difference. The texture rate gap is similar: 46.40 GTexel/s versus 172.8 GTexel/s.

# Specification Differences

The two GPUs differ across nearly every measurable specification. The Radeon 840M uses the Krackan Point chip, while the Ryzen Z2 Extreme GPU uses the Strix Point chip. Both employ RDNA 3.5 architecture and TSMC's 4 nm process.

Clock speeds differ in both directions. The 840M has a 400 MHz base clock and 2900 MHz boost, while the Z2 Extreme GPU has an 800 MHz base and 2700 MHz boost. The Z2 Extreme GPU starts higher but peaks lower.

Memory configurations are fundamentally different. The 840M uses system-shared memory with system-dependent bandwidth and bus width. The Z2 Extreme GPU has 16 GB of LPDDR5X on a 128-bit bus with 128.0 GB/s bandwidth. The memory clock is listed as 1000 MHz with 8 Gbps effective for the Z2 Extreme GPU, while the 840M's memory clock is listed as system-shared.

Execution resources heavily favor the Z2 Extreme GPU: 1024 shading units versus 256, 64 TMUs versus 16, 48 ROPs versus 8, and 16 ray tracing cores versus 4. Neither GPU has tensor cores listed.

Rates follow the resource counts. The Z2 Extreme GPU achieves 129.6 GPixel/s and 172.8 GTexel/s, versus 23.20 GPixel/s and 46.40 GTexel/s for the 840M. FP32 and FP16 throughput are both listed at 5.530 TFLOPS for the Z2 Extreme GPU and 1,484.8 GFLOPS for the 840M, with both running 1:1 FP16 to FP32 ratios.

Power draw favors the 840M at 15 W TDP versus 28 W. Both have no power connectors. The 840M is an IGP with PCIe 4.0 x8 interface and portable-device-dependent display outputs. The Z2 Extreme GPU has no listed bus interface and one USB Type-C display output. Both support DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4.

# Architecture Differences

Both GPUs share the RDNA 3.5 architecture, but they belong to different generations and device classes. The Radeon 840M is classified under the Navi III IGP generation for Strix Point Mobile, while the Ryzen Z2 Extreme GPU is classified as a Console GPU under AMD.

The process node is identical: TSMC 4 nm. However, the chip implementations diverge sharply. The Z2 Extreme GPU's Strix Point chip measures 233 mm² and contains 34,000 million transistors, yielding a density of 145.9M / mm². The 840M's Krackan Point chip has unknown transistor count and die size, though the resource counts indicate a much smaller implementation.

The RDNA 3.5 architecture provides both GPUs with unified ray tracing cores, though the Z2 Extreme GPU has 16 versus 4 on the 840M. Neither has dedicated tensor or AI accelerator cores listed. Both support the same API set, including DirectX 12 Ultimate with the 12_2 feature level, OpenGL 4.6, and Vulkan 1.4.

The Z2 Extreme GPU's memory architecture uses dedicated LPDDR5X with a fixed 128-bit bus, while the 840M relies on shared system memory. This architectural difference affects not just bandwidth but also latency characteristics, though the database does not record latency figures.

The release dates differ, with the 840M appearing on 2025-02-28 and the Z2 Extreme GPU on 2025-07-08. Both remain in active production status. The 840M lists its predecessor as Navi II IGP, while the Z2 Extreme GPU has no predecessor listed.

# Where Each One Wins

The Ryzen Z2 Extreme GPU wins decisively in every computational category. Its FP32 throughput of 5.530 TFLOPS is 3.7 times higher than the 840M's 1,484.8 GFLOPS. Its pixel rate of 129.6 GPixel/s is 5.6 times higher. Its texture rate of 172.8 GTexel/s is 3.7 times higher. The Z2 Extreme GPU also provides 16 GB of dedicated memory with 128.0 GB/s bandwidth, eliminating the system-dependent variability of the 840M's shared memory.

The Z2 Extreme GPU is the appropriate choice for workloads requiring sustained graphics compute: gaming, 3D rendering, or GPU-accelerated content creation within a 28 W envelope. Its 16 ray tracing cores enable hardware-accelerated ray tracing, though the benchmark data suggests real-world performance remains modest. The 3DMark Steel Nomad score of 516 indicates it can run modern titles only at low settings and resolutions.

The Radeon 840M wins in power efficiency, with a 15 W TDP versus 28 W, and in boost clock, at 2900 MHz versus 2700 MHz. It is also a pure IGP with no power connectors and a PCIe 4.0 x8 interface, making it simpler to integrate into thin-and-light devices. For fanless or passively cooled designs, the 840M's lower power draw is the deciding factor.

The 840M's system-shared memory means its bandwidth scales with the host system's memory configuration. In a device with fast LPDDR5X or DDR5, the 840M could perform better than its fixed specifications suggest, but the database records no benchmark scores to confirm this. The Z2 Extreme GPU's fixed 128.0 GB/s bandwidth is predictable and does not depend on the host platform.

For users prioritizing battery life and thermal headroom in an ultraportable device, the 840M is the logical selection. For users needing maximum graphics throughput in a handheld console or compact gaming device, the Z2 Extreme GPU's superior resources and dedicated memory make it the only viable option between the two. The benchmark percentile, however, cautions that even the stronger part delivers performance comparable to decade-old discrete GPUs.