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

The NVIDIA GeForce GTX 650 Ti Boost and the NVIDIA GeForce GTX 650 Ti are two closely related Kepler-based graphics cards from the GeForce 600 generation, but the benchmark data reveals a clear performance hierarchy between them. In the available head-to-head tests, the Boost variant wins both comparisons, with its advantage ranging from 18.1% to 22%. The average benchmark scores place the two cards nearly level, with the Boost model at 8067 and the standard model at 8053, a difference of just 0.2%, but the specific workload tests show a more pronounced separation. The data indicates that while these cards share a common architectural foundation, the Boost model is engineered for higher throughput, and that difference manifests consistently across compute and graphics API benchmarks.

Head-to-Head Benchmarks

The head-to-head results are unambiguous: the GTX 650 Ti Boost outperforms the GTX 650 Ti in both recorded benchmark tests. In Geekbench OpenCL, the Boost card scores 9302 against 7877 for the standard card, yielding a delta of 18.1% in favor of the Boost. This is a substantial margin in a compute-oriented workload, suggesting that the Boost model’s higher memory bandwidth and pixel throughput translate directly into faster data processing. The gap widens further in Geekbench Vulkan, where the Boost card reaches 10041 while the standard card manages 8229, a 22% advantage. Vulkan is a low-level graphics API that often rewards raw hardware throughput, so this larger delta indicates that the Boost card’s additional ROPs and faster clock speeds provide a meaningful edge in draw-call-heavy scenarios.

Looking at the broader context, the average benchmark scores tell a slightly different story. The GTX 650 Ti Boost posts an average of 8067, while the GTX 650 Ti posts 8053, a delta of just 0.2%. This near-tie is reflected in the nearest rival data: the Boost card’s closest rival is the NVIDIA GRID K2 at 8080 (a -0.2% difference), followed by the GTX 650 Ti itself at 8053 (a 0.2% difference). The standard GTX 650 Ti’s nearest rival is the GTX 880M at 8040 (0.2% ahead), with the Boost card at 8067 (-0.2%). This suggests that in a mixed workload average, the two cards are effectively interchangeable, but the head-to-head tests expose where the Boost model’s specifications make a decisive difference. The 18.1% and 22% wins are not marginal; they indicate that the Boost card is significantly faster in compute and modern API graphics, even if older or less demanding workloads may not show the same spread.

Architecture Differences

Both cards are built on the Kepler architecture using a 28 nm process at TSMC, and they share the same transistor count of 2,540 million on a 221 mm² die, yielding a transistor density of 11.5M / mm². The core chips, however, differ: the GTX 650 Ti Boost uses the GK106, while the GTX 650 Ti uses the GK106S. This chip designation is not merely cosmetic, as the Boost model enables a higher configuration of key execution resources. Both cards have 768 shading units and 64 texture mapping units (TMUs), but the Render Output Units (ROPs) differ: the Boost card has 24 ROPs, while the standard card has only 16. This 50% increase in ROPs directly explains the Boost card’s higher pixel rate of 16.51 GPixel/s versus 14.85 GPixel/s.

The memory subsystem is another major differentiator. The GTX 650 Ti Boost features 2 GB of GDDR5 memory on a 192-bit bus, resulting in a bandwidth of 144.2 GB/s. The GTX 650 Ti, in contrast, has 1024 MB of GDDR5 on a 128-bit bus, yielding only 86.40 GB/s. This is a substantial gap in memory bandwidth, and it is the likely driver of the Boost card’s superior Vulkan performance. The memory clock also differs: the Boost card runs at 1502 MHz (6 Gbps effective), while the standard card runs at 1350 MHz (5.4 Gbps effective). The Boost card also has a higher base clock of 980 MHz with a boost clock of 1032 MHz, whereas the standard card’s base and boost clocks are not listed in the data. The FP32 compute throughput reflects these clock and ROP differences: the Boost card delivers 1.585 TFLOPS, while the standard card delivers 1,425.4 GFLOPS.

The physical and power characteristics also diverge. The GTX 650 Ti Boost has a TDP of 134 W and is a dual-slot card, while the GTX 650 Ti has a TDP of 110 W and is a single-slot card. Both require a single 6-pin power connector and a 300 W suggested PSU, and both use a PCIe 3.0 x16 bus interface. The Boost card is longer at 241 mm (9.5 inches), compared to the standard card’s 145 mm (5.7 inches). Display outputs also differ: the Boost card offers 2x DVI, 1x HDMI 1.4a, and 1x DisplayPort 1.2, while the standard card provides 2x DVI and 1x mini-HDMI 1.4a, with no DisplayPort. Both cards support DirectX 12 (11_0), OpenGL 4.6, and Vulkan 1.2.175.

Where Each One Wins

The GTX 650 Ti Boost wins in every benchmark category where both cards were tested, so the analysis of “where each one wins” is more about the magnitude and nature of the victories. In compute-heavy workloads, as represented by Geekbench OpenCL, the Boost card’s 18.1% lead is significant. This is likely attributable to the combination of higher memory bandwidth (144.2 GB/s versus 86.40 GB/s) and higher FP32 throughput (1.585 TFLOPS versus 1,425.4 GFLOPS). For users running OpenCL-accelerated applications, such as video encoding or physics simulations, the Boost card offers a clear performance advantage that justifies its higher TDP and larger physical footprint.

In graphics-intensive scenarios, particularly those leveraging the Vulkan API, the Boost card’s 22% lead is even more pronounced. This is where the additional ROPs (24 versus 16) and higher pixel rate (16.51 GPixel/s versus 14.85 GPixel/s) come into play. Vulkan’s low overhead design tends to expose raw rasterization and memory bandwidth limits, and the Boost card’s 192-bit memory bus provides a substantial advantage over the 128-bit bus on the standard card. For gaming at higher resolutions or with anti-aliasing enabled, the Boost card would likely maintain higher frame rates, though the data does not include direct gaming benchmarks.

The GTX 650 Ti, meanwhile, does not win any head-to-head tests, but it does have a distinct advantage in form factor and power efficiency. With a TDP of 110 W versus 134 W, and a single-slot design versus dual-slot, the standard card is easier to integrate into compact systems. Its 145 mm length (5.7 inches) is significantly shorter than the Boost card’s 241 mm (9.5 inches), making it suitable for smaller cases. While the standard card’s average benchmark score is only 0.2% lower than the Boost card’s, it achieves this with less power draw and a smaller footprint. For users who prioritize physical compatibility and thermal headroom over peak performance, the standard card remains a viable option.

Specification Differences

The following specifications differ between the two cards, based solely on the provided data:

• Chip: GK106 (Boost) versus GK106S (standard)
• Base clock: 980 MHz (Boost) versus not listed (standard)
• Boost clock: 1032 MHz (Boost) versus not listed (standard)
• Memory clock: 1502 MHz / 6 Gbps effective (Boost) versus 1350 MHz / 5.4 Gbps effective (standard)
• Memory size: 2 GB (Boost) versus 1024 MB (standard)
• Memory bus width: 192 bit (Boost) versus 128 bit (standard)
• Memory bandwidth: 144.2 GB/s (Boost) versus 86.40 GB/s (standard)
• ROPs: 24 (Boost) versus 16 (standard)
• Pixel rate: 16.51 GPixel/s (Boost) versus 14.85 GPixel/s (standard)
• Texture rate: 66.05 GTexel/s (Boost) versus 59.39 GTexel/s (standard)
• FP32 compute: 1.585 TFLOPS (Boost) versus 1,425.4 GFLOPS (standard)
• TDP: 134 W (Boost) versus 110 W (standard)
• Slot width: Dual-slot (Boost) versus Single-slot (standard)
• Display outputs: 2x DVI, 1x HDMI 1.4a, 1x DisplayPort 1.2 (Boost) versus 2x DVI, 1x mini-HDMI 1.4a (standard)
• Length: 241 mm / 9.5 inches (Boost) versus 145 mm / 5.7 inches (standard)
• Release date: 2013-03-25 (Boost) versus 2012-10-08 (standard)
• Launch MSRP: 169 USD (Boost) versus 149 USD (standard)

Shared specifications include 768 shading units, 64 TMUs, 28 nm process, 2,540 million transistors, 221 mm² die size, 11.5M / mm² transistor density, PCIe 3.0 x16 interface, 1x 6-pin power connector, 300 W suggested PSU, DirectX 12 (11_0), OpenGL 4.6, Vulkan 1.2.175, and end-of-life production status. The benchmarks also differ: the Boost card has Geekbench Metal, OpenCL, and Vulkan scores, while the standard card only has OpenCL and Vulkan scores.

FAQ

Q: How much faster is the GTX 650 Ti Boost than the GTX 650 Ti in OpenCL?

A: In Geekbench OpenCL, the GTX 650 Ti Boost scores 9302, while the GTX 650 Ti scores 7877, giving the Boost card an 18.1% advantage.

Q: What is the performance gap in Vulkan between the two cards?

A: The GTX 650 Ti Boost scores 10041 in Geekbench Vulkan, compared to 8229 for the GTX 650 Ti, a difference of 22%.

Q: Do the two cards have the same number of shading units and TMUs?

A: Yes, both cards have 768 shading units and 64 texture mapping units, but the Boost card has 24 ROPs versus 16 on the standard card.

Q: How does memory bandwidth compare between the two models?

A: The GTX 650 Ti Boost has a memory bandwidth of 144.2 GB/s due to its 192-bit bus and 2 GB of GDDR5, while the GTX 650 Ti has 86.40 GB/s from a 128-bit bus and 1024 MB of GDDR5.

Q: Which card has a lower TDP and smaller physical footprint?

A: The GTX 650 Ti has a TDP of 110 W and is a single-slot card measuring 145 mm (5.7 inches) in length, whereas the Boost model has a 134 W TDP, is dual-slot, and measures 241 mm (9.5 inches).

Q: What is the average benchmark score difference between the two cards?

A: The GTX 650 Ti Boost has an average benchmark score of 8067, while the GTX 650 Ti has 8053, a difference of 0.2% in favor of the Boost card.