NVIDIA GeForce GTX 460 vs NVIDIA GeForce GTX 675M Comparison

Head-to-Head Benchmarks

The recorded benchmark data shows a single decisive comparison between these two GPUs. In the Geekbench OpenCL test, the NVIDIA GeForce GTX 460 scores 7,925 points, while the NVIDIA GeForce GTX 675M scores 6,946 points. This gives the GTX 460 a 14.1% advantage, a clear and meaningful margin in compute workloads. The GTX 460 wins the only head-to-head test recorded in the database, with a 1 to 0 win count.

Placing these scores in context, the GTX 460 sits at the 41st percentile among all GPUs in the database, while the GTX 675M sits at the 39th percentile. The gap between them is not enormous in percentile terms, but the raw score difference of 979 points is substantial. A 14.1% lead in OpenCL performance suggests the desktop-oriented GTX 460 has a real compute edge over the mobile GTX 675M, despite the latter being a newer product.

The nearest rivals for the GTX 460 reinforce its positioning. The NVIDIA Quadro K4100M scores 7,906, just 0.2% below the GTX 460, making it essentially a statistical tie. The GTX 460 is 1.4% ahead of both the NVIDIA Quadro P5000 and the NVIDIA GeForce GTX 880M, and 1.6% ahead of the NVIDIA GeForce GTX 650 Ti. These small deltas indicate the GTX 460 is clustered tightly with a group of mid-range GPUs, none of which clearly outclasses it in this workload.

For the GTX 675M, the nearest rival picture is different. The AMD Radeon R5 M240 scores 6,975, which is 0.4% above the GTX 675M, essentially a tie. The NVIDIA GeForce GTX 680M scores 7,023, 1.1% higher, and the NVIDIA T600 scores 7,035, 1.3% higher. On the other side, the AMD FirePro M5100 scores 6,830, which the GTX 675M beats by 1.7%. This shows the GTX 675M is also in a tight competitive cluster, but its cluster sits at a lower performance level than the GTX 460's cluster.

The key takeaway from the head-to-head data is straightforward: the GTX 460 outperforms the GTX 675M by a meaningful margin in the single recorded benchmark. The desktop card's 14.1% lead is not a marginal difference; it is large enough to affect real-world compute performance in OpenCL-accelerated applications.

Where Each One Wins

The data records only one benchmark test, so the use-case split is necessarily limited to what the measurements show. In OpenCL compute workloads, the GTX 460 is the clear winner. Its 7,925 score versus 6,946 for the GTX 675M gives it a 14.1% advantage, which translates directly to faster execution of general-purpose GPU compute tasks that rely on OpenCL.

The GTX 675M does not win any recorded benchmark. However, its profile suggests strengths outside the measured workload. It carries 2 GB of GDDR5 memory on a 256-bit bus, delivering 96.00 GB/s of bandwidth, compared to the GTX 460's 768 MB on a 192-bit bus at 86.40 GB/s. In scenarios where memory capacity matters more than raw compute throughput, such as holding larger textures or datasets in VRAM, the GTX 675M has a structural advantage. The database does not include a benchmark that isolates memory-bound performance, so this remains an inference from the specifications rather than a measured result.

The GTX 675M also has a much lower thermal design power at 100 W versus 160 W for the GTX 460. In thin-and-light or thermally constrained environments, the GTX 675M would be the more practical choice, as it generates less heat and requires no external power connectors. The GTX 460, by contrast, needs two 6-pin power connectors and a 450 W suggested power supply, making it unsuitable for compact systems.

For gaming, both cards support DirectX 12 (11_0) and OpenGL 4.6, so API-level compatibility is identical. Neither card has recorded Vulkan support in the database. The GTX 460's higher shading unit count at 336 versus 384 for the GTX 675M might suggest the mobile card has more raw shader throughput, but the measured OpenCL score contradicts that assumption, likely due to clock and memory subsystem differences.

In summary, the GTX 460 wins the only measured benchmark and is the better choice for OpenCL compute performance. The GTX 675M wins on memory capacity, bandwidth, and power efficiency, though none of those advantages are reflected in the recorded scores.

Architecture Differences

Both GPUs are built by NVIDIA and share the same 40 nm process node at TSMC. Both use 1,950 million transistors on a 332 mm² die, giving each a transistor density of 5.9 million per square millimeter. The foundry, transistor count, die size, and process node are identical between the two parts.

The architectural generation differs. The GTX 460 uses the GF104 chip based on the Fermi architecture, part of the GeForce 400 generation. The GTX 675M uses the GF114 chip based on the Fermi 2.0 architecture, part of the GeForce 600M generation. Fermi 2.0 is a refined iteration of the original Fermi design, though the database does not specify the exact improvements beyond the naming.

Despite the same transistor budget, the two chips allocate resources differently. The GTX 675M has 384 shading units, 64 texture mapping units, and 32 ROPs. The GTX 460 has 336 shading units, 56 TMUs, and 24 ROPs. The GTX 675M leads in every compute and rendering resource category, yet it loses in the recorded OpenCL benchmark. This suggests that clock speeds and memory configuration play a larger role in the measured performance than raw unit counts.

Memory configurations differ significantly. The GTX 460 has 768 MB of GDDR5 on a 192-bit bus, with memory clocked at 900 MHz, or 3.6 Gbps effective, yielding 86.40 GB/s of bandwidth. The GTX 675M has 2 GB of GDDR5 on a 256-bit bus, with memory clocked at 750 MHz, or 3 Gbps effective, yielding 96.00 GB/s of bandwidth. The GTX 675M has more capacity and higher bandwidth, despite its lower memory clock, because the wider bus compensates.

Pixel and texture rates reflect the resource differences. The GTX 675M achieves 9.920 GPixel/s and 39.68 GTexel/s, while the GTX 460 achieves 9.450 GPixel/s and 37.80 GTexel/s. The GTX 675M is ahead in both. Floating-point performance shows a similar pattern: the GTX 675M delivers 952.3 GFLOPS of FP32 compute, while the GTX 460 delivers 907.2 GFLOPS. Neither card has recorded FP16 or ray tracing cores.

Power and physical design diverge sharply. The GTX 460 has a 160 W TDP, is dual-slot, requires two 6-pin power connectors, and has a 450 W suggested power supply. It is a desktop card with a length of 210 mm, or 8.3 inches, and display outputs of 2x DVI and 1x mini-HDMI 1.3a. The GTX 675M has a 100 W TDP, uses an MXM module form factor, requires no power connectors, and has no suggested power supply. Its display outputs are listed as portable device dependent, reflecting its mobile nature.

Both cards support DirectX 12 (11_0) and OpenGL 4.6. Neither has Vulkan support recorded. The GTX 460 uses PCIe 2.0 x16, while the GTX 675M uses MXM-B (3.0). The release dates are 2010-07-11 for the GTX 460 and 2012-03-21 for the GTX 675M, making the GTX 675M the newer product by roughly a year and a half.

FAQ

Q: Which GPU is faster in the recorded benchmark?

A: The NVIDIA GeForce GTX 460 scores 7,925 in Geekbench OpenCL, while the NVIDIA GeForce GTX 675M scores 6,946. The GTX 460 leads by 14.1% and wins the only head-to-head test in the database.

Q: How do these GPUs compare to their nearest rivals?

A: The GTX 460 is within 0.2% of the NVIDIA Quadro K4100M, 1.4% behind the NVIDIA Quadro P5000 and NVIDIA GeForce GTX 880M, and 1.6% behind the NVIDIA GeForce GTX 650 Ti. The GTX 675M is 0.4% behind the AMD Radeon R5 M240, 1.1% behind the NVIDIA GeForce GTX 680M, 1.3% behind the NVIDIA T600, and 1.7% ahead of the AMD FirePro M5100.

Q: Which GPU has more memory bandwidth?

A: The GTX 675M has 96.00 GB/s of bandwidth, compared to 86.40 GB/s for the GTX 460. It also has 2 GB of memory versus 768 MB, on a 256-bit bus versus a 192-bit bus.

Q: What are the power requirements for each card?

A: The GTX 460 has a 160 W TDP, requires two 6-pin power connectors, and needs a 450 W suggested power supply. The GTX 675M has a 100 W TDP and requires no power connectors.

Q: Do both cards support the same APIs?

A: Yes. Both support DirectX 12 (11_0) and OpenGL 4.6. Neither has Vulkan support recorded in the database.

Q: Which card has more shading units?

A: The GTX 675M has 384 shading units, while the GTX 460 has 336. The GTX 675M also has more TMUs (64 versus 56) and more ROPs (32 versus 24), yet it still loses the recorded benchmark.

Specification Differences

The two GPUs differ in the following recorded specifications:

• Chip: GF104 for the GTX 460, GF114 for the GTX 675M
• Architecture: Fermi for the GTX 460, Fermi 2.0 for the GTX 675M
• Generation: GeForce 400 for the GTX 460, GeForce 600M for the GTX 675M
• Memory clock: 900 MHz (3.6 Gbps effective) for the GTX 460, 750 MHz (3 Gbps effective) for the GTX 675M
• Memory size: 768 MB for the GTX 460, 2 GB for the GTX 675M
• Memory bus width: 192 bit for the GTX 460, 256 bit for the GTX 675M
• Memory bandwidth: 86.40 GB/s for the GTX 460, 96.00 GB/s for the GTX 675M
• Shading units: 336 for the GTX 460, 384 for the GTX 675M
• TMUs: 56 for the GTX 460, 64 for the GTX 675M
• ROPs: 24 for the GTX 460, 32 for the GTX 675M
• Pixel rate: 9.450 GPixel/s for the GTX 460, 9.920 GPixel/s for the GTX 675M
• Texture rate: 37.80 GTexel/s for the GTX 460, 39.68 GTexel/s for the GTX 675M
• FP32: 907.2 GFLOPS for the GTX 460, 952.3 GFLOPS for the GTX 675M
• TDP: 160 W for the GTX 460, 100 W for the GTX 675M
• Slot width: Dual-slot for the GTX 460, MXM Module for the GTX 675M
• Power connectors: 2x 6-pin for the GTX 460, None for the GTX 675M
• Suggested PSU: 450 W for the GTX 460, not specified for the GTX 675M
• Bus interface: PCIe 2.0 x16 for the GTX 460, MXM-B (3.0) for the GTX 675M
• Display outputs: 2x DVI, 1x mini-HDMI 1.3a for the GTX 460, Portable Device Dependent for the GTX 675M
• Dimensions: 210 mm (8.3 inches) length for the GTX 460, not specified for the GTX 675M
• Release date: 2010-07-11 for the GTX 460, 2012-03-21 for the GTX 675M
• Predecessor: GeForce 200 for the GTX 460, GeForce 500M for the GTX 675M
• Successor: GeForce 500 for the GTX 460, GeForce 700M for the GTX 675M
• Launch MSRP: 199 USD for the GTX 460, not specified for the GTX 675M

The two cards share the same process node, foundry, transistor count, die size, transistor density, manufacturer, DirectX support, OpenGL support, and production status.

The Verdict

The NVIDIA GeForce GTX 460 is the performance winner in the recorded data. Its 7,925 OpenCL score beats the GTX 675M's 6,946 by 14.1%, placing it at the 41st percentile compared to the 39th. For users whose priority is OpenCL compute throughput, the GTX 460 is the clear choice, despite being the older product and having fewer shading units, TMUs, and ROPs.

The NVIDIA GeForce GTX 675M, however, should be chosen by users who need its physical and memory advantages. It consumes only 100 W versus 160 W, requires no power connectors, and fits an MXM module form factor, making it suitable for laptop or compact systems. Its 2 GB memory and 96.00 GB/s bandwidth also give it a capacity edge over the GTX 460's 768 MB and 86.40 GB/s. In memory-heavy workloads that are not captured by the recorded OpenCL test, the GTX 675M could plausibly perform better, but the database provides no measured evidence of this.

Users building a desktop system with adequate power delivery and space should select the GTX 460, as it delivers the higher benchmark score and superior compute performance. Users constrained by power, thermals, or form factor should select the GTX 675M, accepting a 14.1% compute deficit in exchange for a 60 W lower TDP and double the memory capacity. The data does not support a single universal winner; the right choice depends entirely on the usage context.