GPU Comparison

The AMD Radeon HD 6770 and the NVIDIA GeForce 920M occupy adjacent tiers in the database, separated by just over 2% in their average benchmark scores. The data shows a narrow victory for the NVIDIA part, which posts a 3725 score in Geekbench OpenCL compared to the AMD’s 3649, a 2% margin. However, the overall picture is more nuanced than a single test result, as the two cards differ fundamentally in architecture, memory configuration, and power envelope, making their respective strengths and weaknesses highly context-dependent.

Head-to-Head Benchmarks

The only direct head-to-head benchmark available in the data is Geekbench OpenCL, and the result is close. The NVIDIA GeForce 920M scores 3725, while the AMD Radeon HD 6770 scores 3649, giving the NVIDIA part a 2% lead. This is a slim margin, and both scores fall within a tight cluster of comparable GPUs. For reference, the AMD card’s nearest rivals include the NVIDIA GeForce GTX 1050 (3629, 0.6% behind) and the NVIDIA GeForce 825M (3694, 1.2% ahead), whereas the 920M’s nearest rivals include the Intel HD Graphics 530 (3332, 1.4% behind) and the NVIDIA GeForce GT 640 (3210, 2.4% ahead). The fact that the 920M beats the 6770 by only 2% in OpenCL, while sitting in a similar percentile band (20th vs 22nd percentile overall), suggests that raw compute performance is nearly equivalent.

That said, the 920M’s average benchmark score of 3287 is notably lower than its OpenCL score of 3725, because its Vulkan score of 2849 drags the average down. The 6770 has no Vulkan score listed, so its average is identical to its OpenCL score (3649). This discrepancy means that in pure compute workloads, the 920M edges ahead, but when looking at a broader average across multiple test types, the 6770 actually holds a 10% advantage (3649 vs 3287). The head-to-head table only counts one win for the 920M, and that is in OpenCL; the 6770’s lack of a second benchmark prevents a fuller comparison.

Looking at the rivals, the 6770’s position is interesting. It is 0.6% ahead of the GTX 1050 and 0.9% ahead of the GT 735M, but 1.2% behind the 825M. The 920M, meanwhile, is 2.4% ahead of the GT 640 and 3% behind the Intel HD Graphics P4600. Neither card is a performance leader; both sit in the lower quartile of all GPUs (22nd and 20th percentile, respectively). In a direct comparison, the OpenCL delta of 2% is within run-to-run variance for many workloads, so the practical difference is minimal, though the data technically favors the 920M.

FAQ

Q: Which GPU has the higher Geekbench OpenCL score?

A: The NVIDIA GeForce 920M scores 3725, which is 2% higher than the AMD Radeon HD 6770’s 3649. This is the only head-to-head benchmark available, and it favors the NVIDIA card.

Q: How do the two compare in average benchmark score?

A: The AMD Radeon HD 6770 has an average benchmark score of 3649, while the NVIDIA GeForce 920M averages 3287. This gives the AMD card a 10% advantage, driven by the 920M’s lower Vulkan score of 2849.

Q: What is the memory configuration of each card?

A: The AMD Radeon HD 6770 uses 1024 MB of GDDR5 memory on a 128-bit bus, yielding 76.80 GB/s of bandwidth. The NVIDIA GeForce 920M uses 2 GB of DDR3 memory on a 64-bit bus, providing only 14.40 GB/s of bandwidth.

Q: Which card has a higher transistor density?

A: The NVIDIA GeForce 920M has a transistor density of 11.7M / mm², which is higher than the AMD Radeon HD 6770’s 6.3M / mm². This is due to the 920M’s smaller die size (87 mm²) despite similar transistor counts (1,020 million vs 1,040 million).

Q: What are the power requirements for each card?

A: The AMD Radeon HD 6770 has a TDP of 108 W and requires a 300 W suggested PSU plus a 1x 6-pin power connector. The NVIDIA GeForce 920M has a TDP of 33 W, uses no power connectors, and has no suggested PSU listed, reflecting its integrated mobile design.

Q: Which card supports the Vulkan API?

A: The NVIDIA GeForce 920M lists Vulkan support at version 1.2.175, along with DirectX 12 (11_0) and OpenGL 4.6. The AMD Radeon HD 6770 lists no Vulkan support, only DirectX 11.2 (11_0) and OpenGL 4.4.

The Verdict

From the data, the NVIDIA GeForce 920M wins the single head-to-head benchmark (OpenCL) by 2%, but the AMD Radeon HD 6770 wins on average benchmark score by 10% (3649 vs 3287). The 6770 also holds a massive advantage in memory bandwidth (76.80 GB/s vs 14.40 GB/s), which is a 5.3x difference, and in pixel rate (13.60 GPixel/s vs 7.632 GPixel/s). However, the 920M is far more power-efficient, with a TDP of 33 W versus 108 W, and it supports newer APIs including Vulkan and DirectX 12. For users who prioritize raw compute consistency and memory throughput, the 6770 is the better choice based on its higher average score and bandwidth. For users who need a low-power, modern-API-capable solution, the 920M justifies its slim OpenCL win. The 6770’s higher percentile ranking (22nd vs 20th) reinforces that it is the stronger overall performer in this pairing.

Specification Differences

The two cards differ on nearly every measured specification. The AMD Radeon HD 6770 uses a 40 nm process, while the NVIDIA GeForce 920M uses a 28 nm process, both from TSMC. The 6770 has 1,040 million transistors on a 166 mm² die, whereas the 920M has 1,020 million transistors on an 87 mm² die. Memory size favors the 920M (2 GB vs 1 GB), but the 6770 uses GDDR5 on a 128-bit bus (76.80 GB/s), while the 920M uses DDR3 on a 64-bit bus (14.40 GB/s). Shading units also differ: the 6770 has 800, versus 384 for the 920M, and the 6770 has 40 TMUs and 16 ROPs, compared to 32 TMUs and 8 ROPs on the 920M. The 6770’s pixel rate is 13.60 GPixel/s and texture rate is 34.00 GTexel/s, versus 7.632 GPixel/s and 30.53 GTexel/s for the 920M. FP32 performance is 1,360.0 GFLOPS for the 6770 and 732.7 GFLOPS for the 920M. The 6770 has a TDP of 108 W, a dual-slot form factor, and a 1x 6-pin connector, while the 920M is an IGP with 33 W TDP and no connectors. Bus interface differs: PCIe 2.0 x16 for the 6770, PCIe 3.0 x8 for the 920M. Display outputs are also distinct: the 6770 offers 2x DVI, 1x HDMI 1.3a, and 2x mini-DisplayPort 1.1, whereas the 920M is listed as portable device dependent.

Architecture Differences

The architectural split is stark. The AMD Radeon HD 6770 is built on TeraScale 2, part of the Northern Islands generation (HD 6700), and uses the Juniper chip. It is a 40 nm part with a transistor density of 6.3M / mm². The NVIDIA GeForce 920M is a Kepler 2.0 architecture, from the GeForce 900M generation, using the GK208B chip on a 28 nm process, with a higher transistor density of 11.7M / mm². The 6770’s predecessor is Evergreen and its successor is Southern Islands; the 920M’s predecessor is GeForce 800M and its successor is GeForce 10 Mobile. In terms of feature set, the 920M supports Vulkan 1.2.175 and DirectX 12 (11_0), while the 6770 only reaches DirectX 11.2 (11_0) and OpenGL 4.4, versus the 920M’s OpenGL 4.6. The 6770 has no listed Vulkan support. The 920M also has a boost clock of 954 MHz (same as its base clock), while the 6770 lists no base or boost clock, only a memory clock of 1200 MHz (4.8 Gbps effective). The 920M’s memory clock is 900 MHz (1800 Mbps effective). These differences reflect two very different design goals: the 6770 is a discrete desktop card from 2011, while the 920M is a mobile integrated part from 2015.

Where Each One Wins

The AMD Radeon HD 6770 wins decisively in memory bandwidth, with 76.80 GB/s versus 14.40 GB/s, a 5.3x advantage that matters for texture-heavy workloads and high-resolution framebuffers. It also wins on raw shader throughput (1,360.0 GFLOPS vs 732.7 GFLOPS), pixel fill rate (13.60 GPixel/s vs 7.632 GPixel/s), and texture fill rate (34.00 GTexel/s vs 30.53 GTexel/s). Its average benchmark score is 10% higher, and it ranks in the 22nd percentile versus the 920M’s 20th. The NVIDIA GeForce 920M wins on power efficiency, with a 33 W TDP versus 108 W, making it suitable for thin-and-light portable implementations. It also wins on API modernity, supporting Vulkan 1.2.175 and DirectX 12 (11_0), which the 6770 cannot claim. In the single OpenCL head-to-head, the 920M takes the win by 2%. For users with a desktop system and a power budget, the 6770 is the stronger card; for mobile users needing low power draw and newer API support, the 920M is the logical pick. The data does not support a universal winner, but it does show that the 6770’s compute advantage is more consistent across its available benchmarks.