NVIDIA GeForce GTX 650 Ti Boost vs NVIDIA GeForce GTX 670M Comparison

FAQ

Q: How do the two GPUs compare in overall benchmark performance?

A: The NVIDIA GeForce GTX 650 Ti Boost has an average benchmark score of 8067, while the NVIDIA GeForce GTX 670M scores 6513. The desktop card sits in the 42nd percentile of all GPUs, whereas the mobile part lands in the 38th percentile.

Q: Which GPU wins the available head-to-head benchmark comparison?

A: The GTX 650 Ti Boost wins the sole head-to-head test, Geekbench OpenCL, with a score of 9302 versus 6513 for the GTX 670M. That represents a 42.8% advantage for the desktop card.

Q: What are the closest rivals to each GPU according to the database?

A: The GTX 650 Ti Boost sits between the NVIDIA GRID K2 (avg score 8080, 0.2% behind) and the NVIDIA GeForce GTX 650 Ti (avg score 8053, 0.2% ahead). The GTX 670M is closely matched with the Intel UHD Graphics P750 (avg score 6554, 0.6% behind) and the NVIDIA GeForce GT 555M (avg score 6493, 0.3% ahead).

Q: Do both GPUs support DirectX 12?

A: Yes, both support DirectX 12 (11_0) and OpenGL 4.6. However, the GTX 650 Ti Boost also supports Vulkan 1.2.175, while the GTX 670M has no Vulkan support listed.

Q: What memory configurations do the two cards use?

A: The GTX 650 Ti Boost has 2 GB of GDDR5 on a 192-bit bus with 144.2 GB/s bandwidth. The GTX 670M has 1536 MB of GDDR5, also on a 192-bit bus, but bandwidth drops to 72.00 GB/s.

Q: What are the power requirements for each GPU?

A: The GTX 650 Ti Boost has a TDP of 134 W and requires a 6-pin power connector with a suggested 300 W PSU. The GTX 670M has a TDP of 75 W, uses no power connectors, and has no suggested PSU rating since it is an MXM module.

Architecture Differences

The two NVIDIA parts come from different architectural generations. The GTX 650 Ti Boost uses the GK106 chip built on Kepler architecture, manufactured on a 28 nm process at TSMC. The GTX 670M, in contrast, uses the GF114 chip based on the older Fermi 2.0 architecture, fabricated on a 40 nm process, also at TSMC.

The transistor counts reflect the generational shift. The GK106 packs 2,540 million transistors into a 221 mm² die, giving a transistor density of 11.5 million per square millimeter. The GF114 contains 1,950 million transistors spread across a much larger 332 mm² die, resulting in a density of only 5.9 million per square millimeter. The newer process allows the Kepler part to fit more transistors in a smaller area.

The compute cores also differ significantly. The GTX 650 Ti Boost features 768 shading units, 64 texture mapping units, and 24 ROPs. The GTX 670M has 336 shading units, 56 TMUs, and the same 24 ROPs. That is more than double the shader count on the desktop card, and 8 additional texture units.

Clock behavior differs as well. The GTX 650 Ti Boost runs at a base clock of 980 MHz with a boost clock of 1032 MHz. The GTX 670M has no base or boost clock listed in the database, so its operating frequencies are not recorded. The memory clocks also diverge: the desktop card runs at 1502 MHz with 6 Gbps effective data rate, while the mobile card runs at 750 MHz with 3 Gbps effective.

The form factors are completely different. The GTX 650 Ti Boost is a dual-slot desktop card with a 241 mm length (9.5 inches) and a PCIe 3.0 x16 bus interface. The GTX 670M is an MXM module using the MXM-B (3.0) interface, designed for portable devices, and its display outputs are dependent on the host laptop.

Both cards share the same API foundation for DirectX (12, feature level 11_0) and OpenGL (4.6), and neither has ray tracing or tensor cores. The GTX 650 Ti Boost adds Vulkan support at version 1.2.175, but the GTX 670M lacks Vulkan entirely.

Head-to-Head Benchmarks

The database contains one recorded head-to-head benchmark between these two GPUs, and it is a decisive win for the desktop part. In Geekbench OpenCL, the GTX 650 Ti Boost scores 9302 against 6513 for the GTX 670M. The delta is 42.8% in favor of the Kepler card.

This result aligns with the raw specification gaps. The GTX 650 Ti Boost has 768 shading units versus 336, a 28 nm process versus 40 nm, nearly double the texture rate, and more than double the pixel rate. The FP32 throughput tells a similar story: 1.585 TFLOPS for the desktop card versus 803.7 GFLOPS for the mobile part.

The bandwidth difference is also stark. The GTX 650 Ti Boost delivers 144.2 GB/s of memory bandwidth, exactly double the 72.00 GB/s of the GTX 670M, even though both use a 192-bit bus. The higher memory clock on the desktop card (1502 MHz versus 750 MHz) explains this gap.

The GTX 670M does not win any recorded benchmarks in the head-to-head comparison. Its closest competitors in the database, the Intel UHD Graphics P750 at 6554 and the GeForce GT 555M at 6493, are all within 0.6% of its score, but the GTX 650 Ti Boost sits far above that cluster.

When placed in context of the broader database, the GTX 650 Ti Boost's average score of 8067 places it near the NVIDIA GRID K2 (8080) and the GTX 880M (8040), all within a fraction of a percent. The GTX 670M at 6513 is a tier below, closer to integrated graphics solutions than to the desktop Kepler card.

Specification Differences

The following fields differ between the two GPUs:

• Chip: GK106 (GTX 650 Ti Boost) versus GF114 (GTX 670M)
• Architecture: Kepler versus Fermi 2.0
• Generation: GeForce 600 versus GeForce 600M
• Process Node: 28 nm versus 40 nm
• Transistors: 2,540 million versus 1,950 million
• Die Size: 221 mm² versus 332 mm²
• Transistor Density: 11.5M / mm² versus 5.9M / mm²
• Base Clock: 980 MHz versus not listed
• Boost Clock: 1032 MHz versus not listed
• Memory Clock: 1502 MHz (6 Gbps effective) versus 750 MHz (3 Gbps effective)
• Memory Size: 2 GB versus 1536 MB
• Memory Bandwidth: 144.2 GB/s versus 72.00 GB/s
• Shading Units: 768 versus 336
• TMUs: 64 versus 56
• Pixel Rate: 16.51 GPixel/s versus 8.372 GPixel/s
• Texture Rate: 66.05 GTexel/s versus 33.49 GTexel/s
• FP32: 1.585 TFLOPS versus 803.7 GFLOPS
• TDP: 134 W versus 75 W
• Slot Width: Dual-slot versus MXM Module
• Power Connectors: 1x 6-pin versus None
• Suggested PSU: 300 W versus not listed
• Bus Interface: PCIe 3.0 x16 versus MXM-B (3.0)
• Display Outputs: 2x DVI, 1x HDMI 1.4a, 1x DisplayPort 1.2 versus Portable Device Dependent
• Vulkan Support: 1.2.175 versus not listed
• Dimensions: 241 mm length versus not listed
• Release Date: 2013-03-25 versus 2012-03-21
• Predecessor: GeForce 500 versus GeForce 500M
• Successor: GeForce 700 versus GeForce 700M
• Launch MSRP: 169 USD versus not listed

Fields that remain identical: manufacturer (NVIDIA), foundry (TSMC), ROPs (24), memory type (GDDR5), memory bus width (192 bit), DirectX support (12, 11_0), OpenGL support (4.6), and production status (End-of-life).

Where Each One Wins

The GTX 650 Ti Boost wins every recorded benchmark comparison against the GTX 670M. The 42.8% OpenCL lead, combined with more than double the shader count, double the memory bandwidth, and nearly double the FP32 throughput, makes it the clear choice for any task that leverages raw compute performance.

For desktop users, the GTX 650 Ti Boost offers a dual-slot card with a 6-pin power connector, a 134 W TDP, and a suggested 300 W PSU. It has four display outputs (2x DVI, HDMI 1.4a, DisplayPort 1.2), making it suitable for multi-monitor setups. Its Vulkan support gives it access to modern API workloads that the GTX 670M cannot run.

The GTX 670M, by contrast, has one clear advantage: power efficiency. Its 75 W TDP is nearly half that of the desktop card, and it requires no external power connectors. For laptop implementations, this lower power draw is essential. The MXM-B form factor allows it to fit into portable systems where a dual-slot PCIe card is physically impossible.

The GTX 670M also has a small footprint in terms of memory capacity: 1536 MB versus 2 GB. While the desktop card wins on capacity, the mobile part's 192-bit bus still provides reasonable bandwidth for its era, though at half the rate of the GTX 650 Ti Boost.

In terms of database positioning, the GTX 650 Ti Boost sits in the 42nd percentile of all GPUs, slightly above its nearest rival, the GRID K2, and just ahead of the GTX 650 Ti and GTX 880M by 0.2% to 0.3%. The GTX 670M, at the 38th percentile, is surrounded by mobile and integrated parts: the GT 555M, Quadro M5000M, Vega 10 Mobile, and UHD Graphics P750 are all within 0.6% of its score.

For users prioritizing compute performance, desktop compatibility, and modern API support, the GTX 650 Ti Boost is the unambiguous winner. For users constrained by a laptop chassis, thermal limits, or power budgets, the GTX 670M remains the only viable option of the two, but the benchmark data shows it trails the desktop part by a substantial margin in every measured test.