AMD Radeon HD 6770 vs NVIDIA GeForce GT 735M Comparison

The AMD Radeon HD 6770 and the NVIDIA GeForce GT 735M are separated by two years of GPU design philosophy, yet their aggregate benchmark scores place them within 0.9% of each other. The AMD card, a 2011 desktop part built on TeraScale 2, edges out the 2013 mobile-focused NVIDIA chip in the sole Geekbench OpenCL test, scoring 3649 against 3616. This near-parity in raw compute output belies stark differences in power consumption, memory architecture, and feature support, making the choice between them highly dependent on the specific workload and platform constraints.

Architecture Differences

The foundational architectures of these two GPUs could hardly be more different. The AMD Radeon HD 6770 is built on the Juniper chip using the TeraScale 2 architecture, produced on a 40 nm process at TSMC. This older, larger design packs 1,040 million transistors into a 166 mm² die, resulting in a transistor density of 6.3M per mm². In contrast, the NVIDIA GeForce GT 735M uses the GK208 chip based on Kepler 2.0, fabricated on a more advanced 28 nm process. Despite having nearly the same transistor count at 1,020 million, the NVIDIA chip fits into a much smaller 87 mm² die, achieving a significantly higher density of 11.7M per mm².

The compute resources diverge dramatically. The AMD card fields 800 shading units, 40 texture mapping units (TMUs), and 16 raster output units (ROPs). The NVIDIA part, by comparison, is far more modest in raw unit counts, with 384 shading units, 32 TMUs, and only 8 ROPs. This disparity in execution resources explains the AMD card’s higher theoretical peak rates: it delivers 1,360.0 GFLOPS of FP32 compute, 34.00 GTexel/s of texture fill, and 13.60 GPixel/s of pixel throughput. The GT 735M counters with just 482.3 GFLOPS, 20.10 GTexel/s, and 5.024 GPixel/s. However, the NVIDIA card compensates with a base clock of 575 MHz and a boost clock of 628 MHz, while the AMD card’s clock speeds are not specified in the data.

Feature support also differs by generation. The AMD Radeon HD 6770 supports DirectX 11.2 (11_0) and OpenGL 4.4, but has no Vulkan support listed. The NVIDIA GeForce GT 735M supports DirectX 12 (11_0), OpenGL 4.6, and Vulkan 1.2.175, making it more forward-compatible with modern APIs despite its lower raw compute power. The memory subsystems are equally divergent: the AMD card uses 1024 MB of GDDR5 on a 128-bit bus, yielding 76.80 GB/s of bandwidth, while the NVIDIA card has 2 GB of DDR3 on a 64-bit bus, providing only 14.40 GB/s. This five-fold bandwidth advantage for AMD is critical for memory-intensive workloads.

Where Each One Wins

The benchmark data shows a single head-to-head result, with the AMD Radeon HD 6770 winning the Geekbench OpenCL test by 0.9%. This suggests that in raw compute throughput, the AMD card holds a slight edge. The AMD card’s much higher pixel rate (13.60 GPixel/s vs. 5.024 GPixel/s) and texture rate (34.00 GTexel/s vs. 20.10 GTexel/s) indicate it would be substantially faster in traditional rasterization workloads, such as older DirectX 11 games or OpenGL applications that rely on fill-rate-bound rendering.

The NVIDIA GeForce GT 735M, despite losing the compute benchmark, wins in efficiency and platform flexibility. Its 33 W TDP is less than a third of the AMD card’s 108 W, making it suitable for thin-and-light laptops where power draw and thermals are paramount. The GT 735M’s support for DirectX 12, Vulkan 1.2.175, and OpenGL 4.6 means it can run newer titles that leverage these APIs, even if at lower fidelity. The 2 GB memory capacity is double the AMD card’s 1 GB, which helps in texture-heavy scenarios where capacity trumps bandwidth, though the vastly lower bandwidth limits its effectiveness.

The AMD card wins on raw compute and fill-rate-bound tasks, while the NVIDIA card wins on API modernity, power efficiency, and memory capacity. For a desktop user with a 300 W power supply and a 6-pin connector available, the AMD card is the clear performance pick. For a laptop user prioritizing battery life and the ability to play modern API-based titles, the GT 735M is the only viable option.

Head-to-Head Benchmarks

The only benchmark recorded in the data is Geekbench OpenCL, where the AMD Radeon HD 6770 scores 3649 against the NVIDIA GeForce GT 735M’s 3616. This 33-point difference translates to a 0.9% delta in favor of the AMD card. While this is a narrow margin, the context from nearest rivals adds depth. The AMD card sits at the 22nd percentile of all GPUs, with an average benchmark score of 3649. Its closest competitor, the NVIDIA GeForce GTX 1050, scores 3629, just 0.6% lower. The GT 735M, at the 21st percentile, scores 3616, and is 0.9% behind the AMD card.

Interestingly, the AMD card is also 1.2% ahead of the NVIDIA GeForce 825M (which scores 3694) and 1.5% ahead of the NVIDIA RTX 5000 Mobile Ada Generation (which scores 3596). The GT 735M, meanwhile, is 0.3% behind the GTX 1050, 0.6% ahead of the RTX 5000 Mobile Ada Generation, and 0.6% ahead of the NVIDIA GeForce GT 545. These deltas suggest that both GPUs are clustered within a very tight performance band, where architectural differences translate into only marginal real-world compute differences in this specific OpenCL workload.

The single benchmark result indicates that the AMD card’s raw FP32 throughput (1,360.0 GFLOPS) and memory bandwidth (76.80 GB/s) do not translate into a decisive victory over the NVIDIA card’s lower specs (482.3 GFLOPS, 14.40 GB/s). This implies that the Geekbench OpenCL test may be more sensitive to factors like driver optimization, memory latency, or compute unit scheduling than to peak theoretical numbers. The NVIDIA card’s higher boost clock of 628 MHz and newer architecture may help it close the gap despite its smaller execution footprint.

Specification Differences

The two GPUs differ across nearly every major specification field. The AMD Radeon HD 6770 uses a 40 nm process, while the NVIDIA GeForce GT 735M uses 28 nm. The AMD chip has a die size of 166 mm², versus 87 mm² for NVIDIA. Transistor counts are nearly identical (1,040 million vs. 1,020 million), but AMD’s density is 6.3M per mm² compared to NVIDIA’s 11.7M per mm². Memory configurations diverge sharply: AMD offers 1024 MB of GDDR5 on a 128-bit bus with 76.80 GB/s bandwidth; NVIDIA offers 2 GB of DDR3 on a 64-bit bus with 14.40 GB/s bandwidth.

Compute units differ: AMD has 800 shading units, 40 TMUs, and 16 ROPs; NVIDIA has 384 shading units, 32 TMUs, and 8 ROPs. The AMD card’s pixel rate is 13.60 GPixel/s and texture rate is 34.00 GTexel/s, while NVIDIA’s are 5.024 GPixel/s and 20.10 GTexel/s. FP32 performance is 1,360.0 GFLOPS for AMD and 482.3 GFLOPS for NVIDIA. Power consumption is a major differentiator: AMD draws 108 W, while NVIDIA draws only 33 W. The AMD card requires a dual-slot cooler, one 6-pin power connector, and a 300 W suggested power supply; the NVIDIA card is an integrated graphics processor (IGP) with no power connectors and no suggested PSU.

The bus interface also differs: AMD uses PCIe 2.0 x16, while NVIDIA uses PCIe 3.0 x8. Display outputs reflect the desktop vs. mobile split: AMD has 2x DVI, 1x HDMI 1.3a, and 2x mini-DisplayPort 1.1; NVIDIA’s outputs are listed as “Portable Device Dependent.” API support differs as noted: AMD supports DirectX 11.2 and OpenGL 4.4, while NVIDIA supports DirectX 12, OpenGL 4.6, and Vulkan 1.2.175. The AMD card has a launch MSRP of 129 USD, while the NVIDIA card has no launch MSRP listed.

FAQ

Q: Which GPU has higher raw compute performance?

A: The AMD Radeon HD 6770, with 1,360.0 GFLOPS of FP32 performance, compared to 482.3 GFLOPS for the NVIDIA GeForce GT 735M. This is reflected in the Geekbench OpenCL score, where AMD wins 3649 to 3616.

Q: Why does the NVIDIA card have lower power consumption?

A: The NVIDIA GeForce GT 735M has a 33 W TDP, while the AMD Radeon HD 6770 draws 108 W. This is due to the NVIDIA card’s smaller 87 mm² die on a 28 nm process and its reduced shading unit count (384 vs. 800), which lowers both performance and power draw.

Q: Can the NVIDIA card support modern APIs that the AMD card cannot?

A: Yes. The NVIDIA GeForce GT 735M supports DirectX 12, OpenGL 4.6, and Vulkan 1.2.175. The AMD Radeon HD 6770 only supports DirectX 11.2 and OpenGL 4.4, with no Vulkan support listed.

Q: Which card has more memory bandwidth?

A: The AMD Radeon HD 6770, with 76.80 GB/s from its 128-bit GDDR5 interface. The NVIDIA GeForce GT 735M has only 14.40 GB/s from a 64-bit DDR3 bus, despite having double the capacity (2 GB vs. 1 GB).

Q: Are these GPUs close in overall benchmark performance?

A: Yes. The AMD Radeon HD 6770 scores 3649 in Geekbench OpenCL, and the NVIDIA GeForce GT 735M scores 3616, a 0.9% difference. The AMD card ranks at the 22nd percentile of all GPUs, while the NVIDIA card ranks at the 21st.

Q: What are the physical form factor differences?

A: The AMD Radeon HD 6770 is a dual-slot desktop card measuring 198 mm (7.8 inches) in length, requiring a 6-pin power connector and a 300 W power supply. The NVIDIA GeForce GT 735M is an integrated graphics processor (IGP) with no power connectors, designed for portable devices.

The Verdict

The data points to a clear split based on use case. The AMD Radeon HD 6770 is the superior choice for anyone building or upgrading a desktop system where power draw is not a primary concern. It wins the only benchmark recorded (3649 vs. 3616), offers 2.8 times the FP32 compute (1,360.0 GFLOPS vs. 482.3 GFLOPS), and provides 5.3 times the memory bandwidth (76.80 GB/s vs. 14.40 GB/s). Its higher pixel and texture rates (13.60 GPixel/s and 34.00 GTexel/s vs. 5.024 GPixel/s and 20.10 GTexel/s) make it better suited for fill-rate-bound rendering, which is typical of the DirectX 11-era games it was designed for. The 129 USD launch MSRP also indicates it was positioned as an affordable mid-range part.

The NVIDIA GeForce GT 735M, conversely, is the only option for portable devices. Its 33 W TDP is a fraction of the AMD card’s 108 W, and its IGP form factor means it requires no power connectors or dedicated cooling slots. While it loses the compute benchmark by a slim 0.9%, its support for DirectX 12, OpenGL 4.6, and Vulkan 1.2.175 gives it access to newer rendering paths that the AMD card cannot use. The 2 GB memory capacity, though hampered by low bandwidth, is double the AMD card’s 1 GB, which can help in scenarios where texture storage exceeds 1 GB.

Neither card is a dominant winner. The AMD card wins on raw performance metrics and bandwidth, but it is an end-of-life desktop part with higher power demands and older API support. The NVIDIA card wins on efficiency, API modernity, and memory capacity, but its compute performance is measurably lower. Users with a desktop platform and a 300 W power supply should choose the AMD Radeon HD 6770 for its superior fill rates and compute throughput. Users needing a mobile solution with no external power should choose the NVIDIA GeForce GT 735M, accepting a 0.9% benchmark deficit in exchange for a dramatically lower power envelope and broader software compatibility.