AMD Radeon 840M vs AMD Ryzen Z1 GPU Comparison

Where Each One Wins

The AMD Radeon 840M and the AMD Ryzen Z1 GPU occupy different positions despite sharing the same underlying GPU core count. The Radeon 840M is a 15 W integrated graphics processor built on the Krackan Point chip, while the Ryzen Z1 GPU is a 30 W console-class part based on the Phoenix chip. The data shows that the Ryzen Z1 GPU holds the raw compute advantage, while the Radeon 840M counters with higher clock speeds and architectural refinements.

In terms of raw floating-point throughput, the Ryzen Z1 GPU delivers 2.560 TFLOPS of FP32 performance, which is substantially higher than the Radeon 840M's 1,484.8 GFLOPS (approximately 1.48 TFLOPS). This makes the Ryzen Z1 GPU the clear winner for compute-heavy workloads such as shader compilation, physics simulations, and any task that scales with raw ALU throughput. The FP16 numbers reinforce this split: the Ryzen Z1 GPU outputs 5.120 TFLOPS with a 2:1 FP16 ratio, while the Radeon 840M delivers 1,484.8 GFLOPS with a 1:1 ratio. The Ryzen Z1 GPU's FP16 advantage is more than 3.4 times, which matters for applications that use half-precision arithmetic.

The Radeon 840M, however, wins on clock speed. Its boost clock reaches 2900 MHz compared to the Ryzen Z1 GPU's 2500 MHz, a 400 MHz gap that partially offsets the Ryzen Z1 GPU's compute lead. The base clocks differ even more dramatically: 400 MHz for the Radeon 840M versus 1500 MHz for the Ryzen Z1 GPU. This suggests the Ryzen Z1 GPU maintains a higher floor under sustained load, while the Radeon 840M relies on aggressive boosting to reach its peak.

Pixel and texture throughput tell a nuanced story. The Radeon 840M achieves 23.20 GPixel/s and 46.40 GTexel/s, while the Ryzen Z1 GPU posts 20.00 GPixel/s and 40.00 GTexel/s. The Radeon 840M leads by approximately 16% in both metrics, a direct consequence of its higher boost clock. For rasterization-heavy scenes with simple geometry but high resolution, the Radeon 840M's pixel rate advantage could translate into smoother fill-rate-bound performance.

Memory configuration further separates the two. The Ryzen Z1 GPU has a dedicated 16 GB LPDDR5 pool on a 64-bit bus, yielding 51.20 GB/s of bandwidth. The Radeon 840M uses system-shared memory with bandwidth described as "System Dependent." This means the Ryzen Z1 GPU's memory bandwidth is fixed and guaranteed, while the Radeon 840M's performance depends entirely on the host system's memory configuration. In scenarios where system memory bandwidth is inadequate, the Radeon 840M could fall behind despite its clock advantage.

The Verdict

The data indicates that the Ryzen Z1 GPU is the stronger performer for sustained, compute-intensive workloads. Its 30 W TDP allows for higher sustained clocks, and its dedicated 16 GB LPDDR5 memory with 51.20 GB/s bandwidth removes the variable of system memory contention. The Ryzen Z1 GPU also has a massive FP32 lead (2.560 TFLOPS versus 1,484.8 GFLOPS), making it the appropriate choice for applications that repeatedly execute heavy math.

The Radeon 840M, on the other hand, is the more power-efficient option at 15 W, which is half the TDP of the Ryzen Z1 GPU. Its higher boost clock (2900 MHz) and superior pixel/texture rates (23.20 GPixel/s and 46.40 GTexel/s) suggest it can hold its own in fill-rate-limited scenarios, provided the system memory is fast enough. The Radeon 840M also uses the newer RDNA 3.5 architecture versus the Ryzen Z1 GPU's RDNA 3.0, which may bring architectural efficiency improvements not captured by raw clock and throughput numbers.

For an integrated solution in a thin-and-light portable device where power draw is critical, the Radeon 840M's 15 W TDP and system-shared memory make it the logical fit. For a console-style device with a fixed memory allocation and higher power budget, the Ryzen Z1 GPU's 30 W TDP and dedicated 16 GB memory are clearly designed for that purpose. The Ryzen Z1 GPU also carries a launch MSRP of 599 USD, which contextualizes its console positioning.

Head-to-Head Benchmarks

The recorded data shows no direct benchmark results between these two parts, so the comparison relies on their published specifications. The most significant gap is in FP32 compute: the Ryzen Z1 GPU's 2.560 TFLOPS is 72.4% higher than the Radeon 840M's 1,484.8 GFLOPS. This is the single largest performance delta in the comparison.

In FP16, the difference widens further. The Ryzen Z1 GPU's 5.120 TFLOPS is more than three times the Radeon 840M's 1,484.8 GFLOPS. This indicates that the Ryzen Z1 GPU's architecture processes half-precision data at twice the rate of FP32, while the Radeon 840M processes both at the same rate. Applications that leverage FP16 for machine learning inference or certain graphics effects would see a disproportionate benefit from the Ryzen Z1 GPU.

The Radeon 840M counters in rasterization throughput. Its 23.20 GPixel/s pixel rate is 16% higher than the Ryzen Z1 GPU's 20.00 GPixel/s. Similarly, its 46.40 GTexel/s texture rate is 16% higher than the Ryzen Z1 GPU's 40.00 GTexel/s. These wins stem directly from the Radeon 840M's 2900 MHz boost clock versus 2500 MHz on the Ryzen Z1 GPU.

Clock speeds themselves present a mixed picture. The Radeon 840M's base clock of 400 MHz is extremely low, suggesting it idles or scales down aggressively to save power. The Ryzen Z1 GPU's base clock of 1500 MHz is 3.75 times higher, meaning it maintains a much higher minimum performance level. Under sustained load, the Ryzen Z1 GPU is likely to maintain a higher average clock, whereas the Radeon 840M may fluctuate more.

Memory bandwidth is another decisive factor. The Ryzen Z1 GPU's 51.20 GB/s is a fixed quantity, while the Radeon 840M's bandwidth is system-dependent and could be higher or lower depending on the host platform. For a 64-bit bus, 51.20 GB/s is a moderate figure, but it is guaranteed. The Radeon 840M's system-shared memory could theoretically exceed this with dual-channel high-speed DDR5, but that is not guaranteed by any data in the database.

FAQ

Q: Which GPU has higher raw compute performance?

A: The Ryzen Z1 GPU, with 2.560 TFLOPS FP32 versus the Radeon 840M's 1,484.8 GFLOPS. In FP16, the Ryzen Z1 GPU delivers 5.120 TFLOPS versus 1,484.8 GFLOPS.

Q: Does the Radeon 840M have any performance advantage?

A: Yes, the Radeon 840M has a higher boost clock (2900 MHz versus 2500 MHz) and higher pixel/texture rates (23.20 GPixel/s and 46.40 GTexel/s versus 20.00 GPixel/s and 40.00 GTexel/s).

Q: How do their power requirements differ?

A: The Radeon 840M has a 15 W TDP, while the Ryzen Z1 GPU has a 30 W TDP. The Ryzen Z1 GPU consumes twice the power.

Q: What memory configurations do they use?

A: The Radeon 840M uses system-shared memory with system-dependent bandwidth. The Ryzen Z1 GPU has 16 GB of LPDDR5 on a 64-bit bus with 51.20 GB/s bandwidth.

Q: Which architecture does each GPU use?

A: The Radeon 840M uses RDNA 3.5, while the Ryzen Z1 GPU uses RDNA 3.0. Both are built on TSMC's 4 nm process.

Q: What are their physical form factors?

A: The Radeon 840M is an IGP with no slot width specified and no power connectors. The Ryzen Z1 GPU measures 280 mm by 111 mm by 21 mm and also has no power connectors.

Architecture Differences

The Radeon 840M is built on the Krackan Point chip using RDNA 3.5 architecture, while the Ryzen Z1 GPU uses the Phoenix chip with RDNA 3.0. Both are fabricated on TSMC's 4 nm process, so the manufacturing node is identical. The Ryzen Z1 GPU has known silicon details: 25,390 million transistors on a 178 mm² die, giving a transistor density of 142.6M per mm². The Radeon 840M's transistor count and die size are listed as unknown.

Both GPUs share the same core configuration: 256 shading units, 16 texture mapping units, 8 raster operations units, and 4 ray tracing cores. Neither has tensor cores listed. This means the architectural differences come down to the RDNA version and clock behavior rather than core counts.

The RDNA 3.5 architecture in the Radeon 840M represents a newer iteration of AMD's graphics design. The database lists its generation as "Navi III IGP (Strix Point Mobile)," indicating a mobile-focused integrated design. The Ryzen Z1 GPU's generation is listed as "Console GPU (AMD)," positioning it as a component for handheld or console-class devices.

The FP16 execution ratio differs significantly. The Radeon 840M processes FP16 at a 1:1 ratio with FP32, meaning it does not gain a throughput advantage from half-precision math. The Ryzen Z1 GPU processes FP16 at a 2:1 ratio, doubling its throughput when using half-precision data. This architectural choice suggests the Ryzen Z1 GPU is better suited for workloads that can exploit FP16, such as certain machine learning operations or graphics effects.

The Ryzen Z1 GPU's dimensions (280 mm by 111 mm by 21 mm) indicate a physical card or module form factor, whereas the Radeon 840M is an IGP with no dimensions listed. The Ryzen Z1 GPU also has no display outputs, while the Radeon 840M's display outputs are listed as "Portable Device Dependent." This suggests the Ryzen Z1 GPU is designed to output through a host system rather than directly driving displays.

Specification Differences

The two GPUs differ across several specification fields. The Radeon 840M has a base clock of 400 MHz and a boost clock of 2900 MHz, while the Ryzen Z1 GPU has a base clock of 1500 MHz and a boost clock of 2500 MHz. The memory clock differs as well: the Radeon 840M uses system-shared memory, while the Ryzen Z1 GPU has a memory clock of 800 MHz with 6.4 Gbps effective speed.

Memory capacity and type are major differentiators. The Radeon 840M has system-shared memory with system-shared type and bus width. The Ryzen Z1 GPU has 16 GB of LPDDR5 on a 64-bit bus with 51.20 GB/s bandwidth. The Radeon 840M's bandwidth is listed as "System Dependent," while the Ryzen Z1 GPU's bandwidth is fixed.

The TDP differs by a factor of two: 15 W for the Radeon 840M versus 30 W for the Ryzen Z1 GPU. The Radeon 840M is classified as an IGP with no slot width, while the Ryzen Z1 GPU has physical dimensions of 280 mm by 111 mm by 21 mm. Both have no power connectors.

The bus interface differs: the Radeon 840M uses PCIe 4.0 x8, while the Ryzen Z1 GPU has no bus interface listed. Display outputs also differ: the Radeon 840M has "Portable Device Dependent" outputs, while the Ryzen Z1 GPU has "No outputs."

The release dates are distinct: the Radeon 840M was released on 2025-02-28, while the Ryzen Z1 GPU was released on 2023-09-17. The Radeon 840M lists its predecessor as "Navi II IGP," while the Ryzen Z1 GPU has no predecessor listed. The Ryzen Z1 GPU has a launch MSRP of 599 USD, while the Radeon 840M has no launch MSRP recorded.

Both GPUs support DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. Both have a percentile rank of 50 against all GPUs in the database, and both have an average benchmark score of 0. The production status for both is "Active."