AMD Radeon HD 8790M vs AMD Radeon RX 6400 Comparison

FAQ

Q: How much faster is the AMD Radeon RX 6400 in OpenCL compute benchmarks?

A: The RX 6400 scores 32,011 in Geekbench OpenCL, while the HD 8790M scores 5,017. That is a 538.1% advantage for the newer card.

Q: Does the Radeon HD 8790M beat the RX 6400 in any benchmark?

A: No. In the two shared head-to-head tests — Geekbench OpenCL and Geekbench Vulkan — the RX 6400 wins both, giving it a 2–0 record.

Q: What is the Vulkan performance gap between the two cards?

A: The RX 6400 scores 16,372 in Geekbench Vulkan versus 6,365 for the HD 8790M, a 157.2% difference.

Q: How do their overall GPU percentile rankings compare?

A: The RX 6400 sits in the 35th percentile of all GPUs, while the HD 8790M is in the 33rd percentile. Despite the huge benchmark gap, both are near the low end of the database.

Q: Which card has more memory, and what type?

A: The RX 6400 has 4 GB of GDDR6 on a 64-bit bus, while the HD 8790M has 2 GB of GDDR5 on a 128-bit bus.

Q: What is the manufacturing process difference?

A: The RX 6400 is built on a 6 nm process at TSMC, whereas the HD 8790M uses TSMC's 28 nm node.

Architecture Differences

The AMD Radeon RX 6400 and AMD Radeon HD 8790M represent two completely different eras of graphics architecture. The RX 6400 is based on the Navi 24 chip using RDNA 2.0 architecture, part of the Navi II (RX 6000) generation. The HD 8790M uses the Mars chip with GCN 1.0 architecture, from the Solar System (HD 8700M) generation. This architectural leap spans nearly a decade of design evolution.

The process technology tells a stark story. The RX 6400 is manufactured on a 6 nm process at TSMC, while the HD 8790M uses a 28 nm process, also at TSMC. Transistor counts reflect this: the RX 6400 packs 5,400 million transistors on a 107 mm² die, achieving a transistor density of 50.5 million per mm². The HD 8790M has just 950 million transistors on a 77 mm² die, with a density of 12.3 million per mm². The RX 6400 crams over five times the transistors into a smaller area than its predecessor.

Shader resources differ dramatically. The RX 6400 has 768 shading units, 48 texture mapping units, and 32 ROPs. The HD 8790M has 384 shading units, 24 TMUs, and only 8 ROPs — exactly half the shaders and TMUs, but a quarter of the ROPs. The RX 6400 also introduces 12 ray tracing cores, a feature entirely absent from the GCN 1.0-based HD 8790M.

Memory architecture diverges in approach. The RX 6400 uses 4 GB of GDDR6 on a 64-bit bus, yielding 128.0 GB/s of bandwidth. The HD 8790M uses 2 GB of GDDR5 on a 128-bit bus, but only achieves 64.00 GB/s. The newer card achieves double the bandwidth on half the bus width, thanks to much faster memory: 16 Gbps effective on the RX 6400 versus 4 Gbps effective on the HD 8790M.

Clock speeds reflect the efficiency gains. The RX 6400 runs at a base clock of 1923 MHz, a game clock of 2039 MHz, and a boost clock of 2321 MHz. The HD 8790M operates at 850 MHz base and 900 MHz boost — less than half the RX 6400's base frequency. The RX 6400's rated TDP is 53 W, while the HD 8790M's TDP is not listed in the data.

API support differs meaningfully. The RX 6400 supports DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. The HD 8790M supports DirectX 12 (11_1), OpenGL 4.6, and Vulkan 1.2.170. The RX 6400 also has a higher FP32 compute rating at 3.565 TFLOPS versus 691.2 GFLOPS for the HD 8790M, and supports FP16 at 7.130 TFLOPS (2:1), a feature the HD 8790M lacks entirely.

Head-to-Head Benchmarks

The benchmark data shows a decisive and one-sided contest. In the only two tests where both cards have recorded scores, the AMD Radeon RX 6400 wins outright, and the margins are enormous.

Starting with Geekbench OpenCL, the RX 6400 scores 32,011 against the HD 8790M's 5,017. That represents a 538.1% delta — the newer card is more than six times faster in raw compute throughput. This aligns with the architectural differences: the RX 6400 has double the shading units, significantly higher clocks, and a much newer design.

In Geekbench Vulkan, the gap narrows but remains overwhelming. The RX 6400 scores 16,372, while the HD 8790M manages 6,365. The delta is 157.2%. Vulkan is a lower-level API that can expose efficiency differences more directly, and the RDNA 2.0 architecture's advantage is still clear.

The wins tally is straightforward: RX 6400 takes 2 wins, HD 8790M takes 0. There is no benchmark in the shared dataset where the older card comes out ahead.

Context from the nearest rivals helps frame these scores. The RX 6400's average benchmark score is 6,001, which places it near the NVIDIA GeForce GTX 770M (6,000, 0% delta), the NVIDIA RTX PRO 6000 Blackwell Server (5,996, 0.1% delta), the AMD FirePro W4100 (5,987, 0.2% delta), and the NVIDIA Quadro K4000M (5,986, 0.3% delta). The HD 8790M's average is 5,691, sitting near the Intel Iris Pro Graphics P6300 (5,712, -0.4% delta), the NVIDIA GeForce GTX 670MX (5,721, -0.5% delta), the NVIDIA GeForce GTX 550 Ti (5,731, -0.7% delta), and the NVIDIA Quadro M500M (5,604, 1.6% delta). The RX 6400's average is about 5.4% higher than the HD 8790M's, despite the individual benchmark deltas being far larger.

Specification Differences

The following fields differ between the two cards:

• Chip: Navi 24 (RX 6400) versus Mars (HD 8790M)
• Architecture: RDNA 2.0 versus GCN 1.0
• Generation: Navi II (RX 6000) versus Solar System (HD 8700M)
• Process Node: 6 nm versus 28 nm
• Transistors: 5,400 million versus 950 million
• Die Size: 107 mm² versus 77 mm²
• Transistor Density: 50.5M / mm² versus 12.3M / mm²
• Base Clock: 1923 MHz versus 850 MHz
• Boost Clock: 2321 MHz versus 900 MHz
• Game Clock: 2039 MHz (RX 6400 only; HD 8790M has none)
• Memory Clock: 2000 MHz / 16 Gbps effective versus 1000 MHz / 4 Gbps effective
• Memory Size: 4 GB versus 2 GB
• Memory Type: GDDR6 versus GDDR5
• Bus Width: 64 bit versus 128 bit
• Bandwidth: 128.0 GB/s versus 64.00 GB/s
• Shading Units: 768 versus 384
• TMUs: 48 versus 24
• ROPs: 32 versus 8
• RT Cores: 12 versus none
• Pixel Rate: 74.27 GPixel/s versus 7.200 GPixel/s
• Texture Rate: 111.4 GTexel/s versus 21.60 GTexel/s
• FP32: 3.565 TFLOPS versus 691.2 GFLOPS
• FP16: 7.130 TFLOPS (2:1) versus none
• TDP: 53 W versus not listed
• Slot Width: Single-slot versus MXM Module
• Bus Interface: PCIe 4.0 x4 versus MXM-A (3.0)
• Display Outputs: 1x HDMI 2.1, 1x DisplayPort 1.4a versus Portable Device Dependent
• DirectX Support: 12 Ultimate (12_2) versus 12 (11_1)
• Vulkan Support: 1.4 versus 1.2.170
• Release Date: 2022-01-18 versus 2013-03-31
• Predecessor: Navi versus London
• Successor: Navi III versus Gem System
• Launch MSRP: 159 USD (RX 6400 only; HD 8790M has none listed)

The Verdict

The data presents a clear hierarchy. The AMD Radeon RX 6400 is categorically superior to the AMD Radeon HD 8790M in every measurable way — compute performance, memory bandwidth, feature support, and efficiency.

For users choosing between these two, the RX 6400 is the only rational pick for any modern workload. Its 538.1% advantage in OpenCL and 157.2% advantage in Vulkan make the comparison effectively a foregone conclusion. The HD 8790M's 384 shaders, 8 ROPs, and 2 GB of GDDR5 memory are relics of the GCN 1.0 era, and its 28 nm process puts it at a fundamental disadvantage in power efficiency and clock headroom.

The RX 6400's 4 GB of GDDR6 memory at 128.0 GB/s, combined with 12 ray tracing cores and DirectX 12 Ultimate support, means it can handle contemporary gaming and compute tasks that would be impossible or unplayable on the HD 8790M. The older card's DirectX 12 (11_1) support and lack of FP16 capabilities further limit its usefulness.

However, the percentile rankings show both cards are near the low end of the GPU spectrum. The RX 6400 sits in the 35th percentile, and the HD 8790M in the 33rd. Neither is a high-performance part by modern standards; the RX 6400's nearest rivals include the NVIDIA GeForce GTX 770M and NVIDIA Quadro K4000M, all within 0.3% of its average score. The HD 8790M's rivals — Intel Iris Pro Graphics P6300, NVIDIA GeForce GTX 670MX, and NVIDIA GeForce GTX 550 Ti — are similarly low-end.

The HD 8790M's only potential advantage is its MXM-A (3.0) form factor, which could make it a drop-in replacement in certain portable systems. But the RX 6400's single-slot design, PCIe 4.0 x4 interface, and standard display outputs (1x HDMI 2.1, 1x DisplayPort 1.4a) make it far more versatile for desktop use.

In short: the RX 6400 wins decisively on every benchmark, every architectural metric, and every feature comparison. The HD 8790M belongs to a bygone generation, and the data offers no reason to choose it over the RX 6400 for any task.