AMD Radeon HD 8790M vs NVIDIA GeForce 940M Comparison

# NVIDIA GeForce 940M vs AMD Radeon HD 8790M

The NVIDIA GeForce 940M and AMD Radeon HD 8790M are two end-of-life mobile graphics solutions from different architectural generations, with the data revealing a split decision: NVIDIA wins the OpenCL compute test by 20%, while AMD counters with a 28.5% margin in Vulkan performance. Averaging both benchmark scores, the AMD Radeon HD 8790M edges ahead at 5691 versus 5284 for the GeForce 940M, placing it in the 33rd percentile of all GPUs compared to NVIDIA's 31st.

The Verdict

The benchmark results paint a nuanced picture. For users prioritizing raw compute throughput in OpenCL-based workloads, the NVIDIA GeForce 940M is the clear choice — its 6018 OpenCL score beats the AMD Radeon HD 8790M's 5017 by a full 20%. Conversely, if Vulkan API performance matters most, the AMD Radeon HD 8790M dominates with a 6365 Vulkan score versus NVIDIA's 4549, a 28.5% advantage that flips the overall average in AMD's favor.

Looking at the aggregate data, the AMD Radeon HD 8790M's average benchmark score of 5691 places it 0.4% behind the Intel Iris Pro Graphics P6300 and 1.6% ahead of the NVIDIA Quadro M500M. The NVIDIA GeForce 940M, by contrast, sits 0.4% behind the GeForce GTX 980M and 0.9% ahead of the GeForce GTX 760M. The AMD card's higher percentile ranking (33rd versus 31st) and superior average score suggest it offers marginally better all-around performance in the tested workloads.

However, the choice is not simply about averages. The GeForce 940M's architectural advantages — 512 shading units versus 384, 32 TMUs versus 24, and roughly 63% higher FP32 throughput — make it the more capable compute device in theory. The Radeon HD 8790M counters with GDDR5 memory and a 128-bit bus, delivering 64.00 GB/s of bandwidth compared to NVIDIA's 14.40 GB/s. For gaming or graphics workloads that lean on memory bandwidth, AMD's solution is preferable; for compute-heavy tasks, NVIDIA holds the edge. Buyers should select based on the specific API and workload mix they expect.

Architecture Differences

The two GPUs come from fundamentally different architectural schools. The NVIDIA GeForce 940M uses the GM107 chip built on the Maxwell architecture, fabricated on a 28 nm process at TSMC with 1,870 million transistors packed into a 148 mm² die, yielding a transistor density of 12.6M per mm². The AMD Radeon HD 8790M uses the Mars chip based on GCN 1.0 architecture, also on TSMC's 28 nm node, but with just 950 million transistors on a 77 mm² die, resulting in a slightly lower density of 12.3M per mm².

Core configurations differ substantially. The GeForce 940M fields 512 shading units, 32 texture mapping units, and 16 ROPs. The Radeon HD 8790M counters with 384 shading units, 24 TMUs, and only 8 ROPs. Clock speeds favor NVIDIA as well: the 940M runs at a 1020 MHz base and 1098 MHz boost, while the 8790M operates at 850 MHz base and 900 MHz boost. These differences manifest in the theoretical throughput numbers: NVIDIA achieves 1,124.4 GFLOPS FP32, 35.14 GTexel/s texture fill, and 17.57 GPixel/s pixel fill, versus AMD's 691.2 GFLOPS, 21.60 GTexel/s, and 7.200 GPixel/s respectively.

Memory architecture represents the most dramatic divergence. Both cards carry 2 GB, but the GeForce 940M uses DDR3 on a 64-bit bus, yielding just 14.40 GB/s bandwidth. The Radeon HD 8790M employs GDDR5 on a 128-bit bus, delivering 64.00 GB/s — more than quadruple the bandwidth. Memory clocks also differ: NVIDIA runs 900 MHz (1800 Mbps effective) while AMD runs 1000 MHz (4 Gbps effective). API support shows NVIDIA with Vulkan 1.4 and DirectX 12 (11_0), while AMD offers Vulkan 1.2.170 and DirectX 12 (11_1), a slightly higher DirectX feature level.

FAQ

Q: Which GPU has better overall average benchmark performance?

A: The AMD Radeon HD 8790M has a higher average benchmark score of 5691 compared to the NVIDIA GeForce 940M's 5284, a difference of roughly 7.7% in AMD's favor.

Q: How do the two GPUs compare in OpenCL performance?

A: The NVIDIA GeForce 940M scores 6018 in Geekbench OpenCL versus 5017 for the AMD Radeon HD 8790M, giving NVIDIA a 20% advantage in this compute workload.

Q: What about Vulkan performance?

A: The AMD Radeon HD 8790M scores 6365 in Geekbench Vulkan, while the NVIDIA GeForce 940M manages 4549 — AMD leads by 28.5% in this API.

Q: Why does the AMD card have such higher memory bandwidth?

A: The Radeon HD 8790M uses GDDR5 memory on a 128-bit bus, achieving 64.00 GB/s. The GeForce 940M uses DDR3 on a 64-bit bus, which limits it to 14.40 GB/s.

Q: Which GPU has more shading units?

A: The NVIDIA GeForce 940M has 512 shading units, compared to 384 on the AMD Radeon HD 8790M, a 33% advantage for NVIDIA.

Q: Are these GPUs still in production?

A: No, both are listed as end-of-life products. The NVIDIA GeForce 940M was released in March 2015, while the AMD Radeon HD 8790M came earlier, in March 2013.

Specification Differences

The two mobile GPUs differ across nearly every specification category. The GeForce 940M uses the GM107 chip with Maxwell architecture, while the Radeon HD 8790M employs the Mars chip with GCN 1.0. Transistor counts diverge significantly: NVIDIA integrates 1,870 million transistors on a 148 mm² die, versus AMD's 950 million on 77 mm² — a 920 million transistor and 71 mm² difference. Transistor density is nearly identical at 12.6M vs 12.3M per mm².

Clock speeds favor NVIDIA on both base and boost: 1020 MHz versus 850 MHz base, and 1098 MHz versus 900 MHz boost. Memory specifications strongly favor AMD: the Radeon HD 8790M uses GDDR5 with a 128-bit bus and 64.00 GB/s bandwidth, while the GeForce 940M uses DDR3 with a 64-bit bus and only 14.40 GB/s. Memory clock rates also differ, with AMD at 1000 MHz (4 Gbps effective) versus NVIDIA's 900 MHz (1800 Mbps effective).

Core counts show NVIDIA's advantage in shading units (512 vs 384), TMUs (32 vs 24), and ROPs (16 vs 8). Theoretical rates reflect these differences: NVIDIA achieves 17.57 GPixel/s pixel rate and 35.14 GTexel/s texture rate versus AMD's 7.200 GPixel/s and 21.60 GTexel/s. FP32 compute is 1,124.4 GFLOPS for NVIDIA versus 691.2 GFLOPS for AMD. The GeForce 940M specifies a 75 W TDP, while the Radeon HD 8790M has no TDP listed. Both use MXM modules, but NVIDIA uses MXM-B (3.0) while AMD uses MXM-A (3.0). API support differs in DirectX (12 (11_0) for NVIDIA, 12 (11_1) for AMD) and Vulkan (1.4 versus 1.2.170). Release dates also differ: NVIDIA in March 2015, AMD in March 2013.

Head-to-Head Benchmarks

Two benchmark tests define the head-to-head comparison, and each GPU takes one victory. In Geekbench OpenCL, the NVIDIA GeForce 940M scores 6018 against the AMD Radeon HD 8790M's 5017, a 20% delta in NVIDIA's favor. This result aligns with the underlying hardware: NVIDIA's 512 shading units, higher clock speeds (1098 MHz boost versus 900 MHz), and 1,124.4 GFLOPS of FP32 throughput provide substantial compute headroom over AMD's 384 units and 691.2 GFLOPS. The OpenCL test appears to reward raw shader throughput and fill rates, where NVIDIA holds commanding leads in both texture rate (35.14 GTexel/s vs 21.60 GTexel/s) and pixel rate (17.57 GPixel/s vs 7.200 GPixel/s).

In Geekbench Vulkan, the tables turn dramatically. The AMD Radeon HD 8790M scores 6365 versus NVIDIA's 4549, a 28.5% margin for AMD. This is where the memory subsystem becomes decisive. AMD's GDDR5 memory on a 128-bit bus delivers 64.00 GB/s — over four times NVIDIA's 14.40 GB/s from DDR3 on a 64-bit bus. Vulkan workloads often stress memory bandwidth heavily, and the Radeon HD 8790M's superior memory architecture plus its higher DirectX 12 feature level (11_1 versus 11_0) likely contribute to this result. The GeForce 940M's lower Vulkan score also reflects its older Maxwell architecture's less mature driver support for this API generation.

The split results yield one win each in the head-to-head tally. Averaging the two scores, AMD's 5691 average (computed from its two benchmark results) versus NVIDIA's 5284 gives AMD the aggregate edge. The Radeon HD 8790M also ranks higher in the overall GPU percentile distribution at 33rd versus 31st. However, the nature of the workloads matters more than the aggregate: users running OpenCL compute tasks will see NVIDIA perform 20% better, while those on Vulkan-based applications will experience AMD's 28.5% superiority. The GeForce 940M's nearest rival, the GeForce GTX 980M, scores 5308 (0.4% higher), while the Radeon HD 8790M's closest competitor, the Intel Iris Pro Graphics P6300, scores 5712 (0.4% higher). Both cards sit near the middle of the mobile GPU performance spectrum, with AMD holding a slight overall edge.