AMD Radeon HD 6770 vs NVIDIA GeForce 930M Comparison

FAQ

Q: How do the NVIDIA GeForce 930M and AMD Radeon 6770 perform based on benchmark scores?

A: The recorded data shows the GeForce 930M scores 5046 points in the Geekbench OpenCL benchmark, while the Radeon HD 6770 scores 3649. This gives the 930M a 38.3% lead in that test. The 930M holds an average benchmark score of 4388, while the HD 6770 sits at 3649.

Q: Which GPU ranks higher in the grand scheme of all GPUs?

A. The GeForce 930M sits at the 26th percentile of all GPUs, whereas the Radeon HD 6770 ranks at the 22nd percentile. The higher percentile for the 930M indicates it outperforms a greater share of the GPU population.

Q: What is the closest rival to each of these cards?

A: The GeForce 930M’s nearest rival is the NVIDIA GeForce GT 645M, with an average score of 4411, which is 0.5% lower. The Radeon HD 6770’s nearest rival is the NVIDIA GeForce 825M, with an average score of 3694, which is 1.2% higher.

Q: Does the AMD Radeon HD 6770 support Vulkan?

A: No. The Radeon HD 6770 lists no Vulkan support in its API specifications, while the GeForce 930M supports Vulkan 1.4. The HD 6770 supports DirectX 11.2 (11_0) and OpenGL 4.4, while the 930M supports DirectX 12 (11_0), OpenGL 4.6, and Vulkan 1.4.

Q: What is the difference in memory capacity and memory bandwidth?

A: The GeForce 930M comes with 2 GB of DDR3 memory across a 64-bit bus, yielding 12.80 GB/s of bandwidth. The Radeon HD 6770 uses 1024 MB GDDR5 on a 128-bit bus, delivering 76.80 GB/s of bandwidth. The HD 6770 has a much higher memory transfer rate at 4.8 Gbps effective versus 1600 Mbps effective for the 930M.

Q: What is the transistor count and die size for these two chips?

A: The GeForce 930M has 1,020 million transistors on a 77 mm² die with a density of 13.2M/mm². The Radeon HD 6770 has 1,040 million transistors on a 166 mm² die, giving a density of 6.3M/mm².

Architecture Differences

The two GPUs are built on fundamentally different architectures. NVIDIA’s GeForce 930M uses the Maxwell architecture with chip GM108S, while AMD’s Radeon HD 6770 is built on TeraScale 2 with the Juniper chip. The process nodes differ significantly: the 930M uses TSMC’s 28 nm process node, whereas the HD 6770 uses TSMC’s 40 nm process node. Both chips come from TSMC, but the difference in process size explains part of the divergence in die area, 77 mm² versus 166 mm². The transistor counts are nearly identical at 1,020 million versus 1,040 million, but the 930M packs them far more densely at 13.2M/mm² versus 6.3M/mm².

The AMD part belongs to the Northern Islands generation and the specific family line of HD 6700, while the NVIDIA card is from the GeForce 900M series. The GeForce 930M uses a PCIe 3.0 x8 bus interface, while the Radeon HD 6770 uses PCIe 2.0 x16. In terms of layout, the 930M is an IGP with no power connectors and no slot width, while the HD 6770 is a dual-slot card that is 198 mm (7.8 inches long and requires a single 6-pin power connector along with a 300 W PSU.

The API feature set marks a clear generation gap as well. The 930M supports DirectX 12 (11_0), OpenGL 4.6, and Vulkan 1.4, whereas the HD 6770 supports only DirectX 11.2 (11_0) and OpenGL 4.4 with no Vulkan support. The shading units, TMU and ROP counts also differ widely, with the 930M having 384 shading units, 24 TMUs, and 8 ROPs, while the HD 6770 has 800 shading units, 40 TMUs, 40 ROPs and 16 ROPs. AMD’s output configuration includes 2x DVI 1x HDMI 2x mini-DisplayPort 1.1.

Both cards are end-of-life products. The GeForce 930M was released in 2015 on 2015-03-12T17:00:00. The Radeon HD 6770 was released earlier on 2011-01-20T17:00:00. TDP differences are stark: the 930M draws 33 W and the HD 6770 draws 108 W. In pixel rate, the HD 6770 also holds ahead at 13.60 GPixel/s versus 4.60 GPixel/s for the 930M, 34.00 GTexel/s, and 1,360.0 GFLOPS FP32 performance versus 421.6 GFLOPS, 13.18 GTexel/s. The 930M uses 1600 Mbps effective memory speed, 4.8 Gbps effective for the HD 6770. The GeForce 930M is the successor to GeForce 800M series and predecessor to GeForce 10 Mobile, while the HD 6770 is preceded by Evergreen and followed by Southern Islands.

Head-to-Head Benchmarks

In the only head-to-head benchmark recorded, the GeForce 930M wins the Geekbench OpenCL test with a score of 5046 against the HD 6770’s 3649. That is a 38.3% advantage for the NVIDIA chip. The 930M also holds a higher average benchmark score of 4388 versus the HD 6770’s 3649, which reflects a broader performance spread across multiple workloads.

Looking at the nearest rivals, the 930M outperforms the GeForce GT 645M by 0.7% in average score (4388 versus 4411), 1.2% ahead of the Intel Iris Pro Graphics 5200 and 2.2% ahead of the AMD FirePro W2100. The 930M edges out the RTX 4070 GDDR6 in average score by 1.2% (4388 versus 4335).

The HD 6770 meanwhile trails its nearest rival in average score by 0.6%, losing to the GTX 1050 (3629 versus 3649 by 0.6%); it outperforms the GT 735M by 0.9% (3649 versus 3616), 1.2% ahead of the GeForce 825M (3649 versus 3694) and 1.5% ahead of the RTX 5000 Mobile Ada Generation (3649 versus 3596 1.5%). The GeForce 930M wins 1 head-to-head benchmark and 0 wins for the HD 6770. The 930M sits at percentile 26 versus 22 in the overall GPU ranking.

The GeForce 930M’s performance per watt is not directly recorded, but the combination of a 33 W TDP and 421.6 GFLOPS FP32 output versus the HD 6770’s 108 W and 1,360.0 GFLOPS tells part of the story. The 6770 has 1,360.0 GFLOPS FP32 performance, 34.00 GTexel/s texture fill rate, and 13.60 GPixel/s pixel rate, all higher than the 930M. Still, the 930M’s higher benchmark score suggests that in the OpenCL workload, the NVIDIA architecture is more efficient at delivering results.

The Verdict

The data shows two very different products. The GeForce 930M is a 33 W IGP with a modern Maxwell architecture, PCIe 3.0 x8, DirectX 12 (11_0), Vulkan 1.4, and 2 GB DDR3 on a 64-bit bus. It has a lower pixel rate (4.392 GPixel/s) and lower texture rate (13.18 GTexel/s), but its benchmark result is much higher than the HD 6770, scoring 5046 versus 3649 in OpenCL, a 38.3% lead. This makes it the better option for any workload that is captured by the Geekbench OpenCL test.

The Radeon HD 6770 is a 108 W dual-slot desktop card with a 128-bit GDDR5 bus, 76.80 GB/s bandwidth, 800 shading units, 40 TMUs, and 16 ROPs. It has a higher pixel rate (13.60 GPixel/s) and higher texture rate (34.00 GTexel/s), and it offers 1,360.0 GFLOPS of FP32 performance, which is more than three times the 930M’s 421.6 GFLOPS. However, its OpenCL score is lower at 3649. The 6770 also requires a 300 W PSU and a 6-pin connector, while the 930M uses no power connectors.

Who should pick which? For a laptop or compact system where power draw and slot space are limited, the GeForce 930M is the sensible choice: it is an IGP with no power connectors, a 33 W TDP, and it wins the benchmark. For a desktop system with a PCIe 2.0 x16 slot and a 300 W PSU, the Radeon HD 6770 offers a much higher raw fill rate and FP32 compute, but the recorded benchmark favors the 930M. The data indicates that the 930M is the stronger performer in the OpenCL test, despite the HD 6770’s higher theoretical throughput. The 930M also supports a newer API set, including Vulkan 1.4, which the HD 6770 lacks.

The HD 6770’s lower percentile (22nd versus 26th) and its lower average score (3649 versus 4388) further cement the 930M as the better overall performer according to the database. While the HD 6770 has a higher memory bandwidth and a wider memory bus, the 930M’s benchmark result points to better real-world performance in the measured workload. The HD 6770 does have a larger memory footprint in terms of speed (4.8 Gbps effective versus 1600 Mbps effective), and its GDDR5 memory type is superior to the 930M’s DDR3, but the 930M’s overall score remains higher. The 930M is the winner in the head-to-head comparison, with 1 win and 0 losses.

Specification Differences

The two GPUs differ in the following fields:

• Process Node: 28 nm (930M) vs 40 nm (HD 6770)
• Die Size: 77 mm² (930M) vs 166 mm² (HD 6770)
• Transistor Density: 13.2M/mm² (930M) vs 6.3M/mm² (HD 6770)
• Shading Units: 384 (930M) vs 800 (HD 6770)
• TMUs: 24 (930M) vs 40 (HD 6770)
• ROPs: 8 (930M) vs 16 (HD 6770)
• Pixel Rate: 4.392 GPixel/s (930M) vs 13.60 GPixel/s (HD 6770)
• Texture Rate: 13.18 GTexel/s (930M) vs 34.00 GTexel/s (HD 6770)
• FP32: 421.6 GFLOPS (930M) vs 1,360.0 GFLOPS (HD 6770)
• TDP: 33 W (930M) vs 108 W (HD 6770)
• Slot Width: IGP (930M) vs Dual-slot (HD 6770)
• Power Connectors: None (930M) vs 1x 6-pin (HD 6770)
• Suggested PSU: None (930M) vs 300 W (HD 6770)
• Bus Interface: PCIe 3.0 x8 (930M) vs PCIe 2.0 x16 (HD 6770)
• Display Outputs: Portable Device Dependent (930M) vs 2x DVI 1x HDMI 1.3a 2x mini-DisplayPort 1.1 (HD 6770)
• Memory Size: 2 GB (930M) vs 1024 MB (HD 6770)
• Memory Type: DDR3 (930M) vs GDDR5 (HD 6770)
• Memory Bus: 64 bit (930M) vs 128 bit (HD 6770)
• Memory Bandwidth: 12.80 GB/s (930M) vs 76.80 GB/s (HD 6770)
• Memory Clock: 800 MHz (930M) vs 1200 MHz (HD 6770)
• DirectX: 12 (11_0) (930M) vs 11.2 (11_0) (HD 6770)
• OpenGL: 4.6 (930M) vs 4.4 (HD 6770)
• Vulkan: 1.4 (930M) vs None (HD 6770)
• Release Date: 2015-03-12T17:00:00.000Z (930M) vs 2011-01-20T17:00:00.000Z (HD 6770)
• Predecessor: GeForce 800M (930M) vs Evergreen (HD 6770)
• Successor: GeForce 10 Mobile (930M) vs Southern Islands (HD 6770)

The GeForce 930M uses the GM108S chip, while the HD 6770 uses the Juniper chip. The 930M is part of the GeForce 900M generation, and the HD 6770 is part of the Northern Islands (HD 6700) generation. The 930M has no launch MSRP recorded, while the HD 6770 has a launch MSRP of 129 USD.