NVIDIA GeForce GTX 560 vs NVIDIA GeForce GTX 650 Ti Boost Comparison

NVIDIA’s GeForce GTX 560 and GeForce GTX 650 Ti Boost represent two distinct moments in the company’s desktop GPU timeline. The GTX 560 arrives from the GeForce 500 generation, built on the Fermi 2.0 architecture with a 40 nm process. The GTX 650 Ti Boost comes from the GeForce 600 generation, using the Kepler architecture on a 28 nm node. The database includes a single direct head-to-head benchmark between these two cards, but the full specification sheets reveal deep differences in how each GPU approaches memory, compute, and power delivery. This analysis walks through the measured data, architectural shifts, and practical implications for both cards.

Head-to-Head Benchmarks

The recorded head-to-head result comes from the Geekbench OpenCL test. In that workload, the GTX 650 Ti Boost scores 9302, while the GTX 560 scores 9058. The difference works out to about 2.6 percent in favor of the newer card. That is a narrow margin, but it is consistent with the broader benchmark profiles for both GPUs. The GTX 650 Ti Boost’s OpenCL score places it just ahead of the GTX 560, yet the GTX 560 still holds its own in this compute-oriented test.

Looking at the average benchmark scores, the GTX 560 posts 9058, which is identical to its single OpenCL result. The GTX 650 Ti Boost, however, has a lower average of 8067 because the database includes additional test types. Its OpenCL score of 9302 is strong, but the card also records 4858 in Geekbench Metal and 10041 in Geekbench Vulkan. The Vulkan result is notably the highest single score for either card in the available data. The Metal score is much lower, likely reflecting the workload characteristics of that API.

The percentile rankings place the GTX 560 at the 45th percentile among all GPUs, while the GTX 650 Ti Boost sits at the 42nd percentile. Despite the newer card winning the direct OpenCL comparison, the GTX 560 ranks slightly higher overall in the database’s distribution. This suggests that the GTX 560’s consistent performance across the measured workload gives it an edge in percentile placement, even if the raw OpenCL score favors the GTX 650 Ti Boost.

The nearest rivals for the GTX 560 include the NVIDIA TITAN V CEO Edition with an average score of 9037, which is only 0.2 percent behind the GTX 560. The GeForce GTX 660 scores 9022, within 0.4 percent. The AMD Radeon 550X trails by 1.6 percent with 8918, while the AMD Radeon 890M leads the GTX 560 by 1.7 percent with a score of 9210. These close margins show that the GTX 560 lands in a crowded performance neighborhood where small differences separate cards.

For the GTX 650 Ti Boost, the nearest rivals cluster even tighter. The NVIDIA GRID K2 scores 8080, just 0.2 percent ahead. The GeForce GTX 650 Ti scores 8053, within 0.2 percent behind. The GeForce GTX 880M is 0.3 percent behind at 8040, and the Quadro P5000 sits at 8039, also 0.3 percent behind. The GTX 650 Ti Boost’s average score of 8067 places it in the middle of this group, with rivals on both sides separated by fractions of a percent.

The data shows a clear pattern: the GTX 650 Ti Boost wins the only direct comparison, but the margin is small. The GTX 560 compensates with a higher percentile rank and a tighter spread of rival scores. In practical terms, neither card dominates the other by a wide margin in the measured workload.

Architecture Differences

The GTX 560 uses the GF114 chip, part of the Fermi 2.0 architecture. It is built on a 40 nm process at TSMC. The die measures 332 mm² and contains 1,950 million transistors. That works out to a transistor density of 5.9 million per mm². The GTX 650 Ti Boost uses the GK106 chip, based on the Kepler architecture. This chip is manufactured on a 28 nm process, also at TSMC. The die is smaller at 221 mm², but it packs more transistors: 2,540 million. The density jumps to 11.5 million per mm². The smaller, denser Kepler die represents a significant manufacturing advance over the older Fermi design.

Memory configurations differ substantially. The GTX 560 comes with 1024 MB of GDDR5 on a 256 bit bus. Its memory clock is 1000 MHz with 4 Gbps effective data rate, yielding 128.0 GB/s of bandwidth. The GTX 650 Ti Boost doubles the capacity to 2 GB of GDDR5, but narrows the bus to 192 bit. Its memory runs at 1502 MHz with 6 Gbps effective, pushing bandwidth to 144.2 GB/s. The newer card has more capacity and higher bandwidth despite the narrower bus.

Shading resources favor the GTX 650 Ti Boost. The GTX 560 has 336 shading units, 56 texture mapping units, and 32 ROPs. The GTX 650 Ti Boost has 768 shading units, 64 texture mapping units, and 24 ROPs. The Kepler card more than doubles the shading units, but loses ROPs. Pixel rate for the GTX 560 is 11.34 GPixel/s, while the GTX 650 Ti Boost reaches 16.51 GPixel/s. Texture rate is 45.36 GTexel/s for the GTX 560 and 66.05 GTexel/s for the GTX 650 Ti Boost. The newer card leads in both fill rates.

Compute throughput also shifts. The GTX 560 delivers 1,088.6 GFLOPS of FP32 performance. The GTX 650 Ti Boost reaches 1.585 TFLOPS, which is 1,585 GFLOPS. That is a substantial increase in raw floating point capability. Neither card lists FP16 performance in the database.

Power requirements move in the opposite direction. The GTX 560 has a TDP of 150 W and requires two 6 pin power connectors. The suggested power supply is 450 W. The GTX 650 Ti Boost draws less power at 134 W, needs only one 6 pin connector, and suggests a 300 W power supply. The newer card delivers higher compute and memory performance while consuming less power and requiring less from the power supply.

Interface support also differs. The GTX 560 uses PCIe 2.0 x16, while the GTX 650 Ti Boost uses PCIe 3.0 x16. Display outputs are distinct as well. The GTX 560 offers two DVI ports and one mini HDMI 1.3a. The GTX 650 Ti Boost offers two DVI ports, one HDMI 1.4a, and one DisplayPort 1.2. API support is similar for DirectX and OpenGL, with both listing DirectX 12 (11_0) and OpenGL 4.6. The GTX 650 Ti Boost additionally lists Vulkan 1.2.175 support, while the GTX 560 has no Vulkan entry in the database.

Physical dimensions differ slightly. The GTX 560 measures 210 mm, or 8.3 inches, in length. The GTX 650 Ti Boost is longer at 241 mm, or 9.5 inches. Both cards are dual slot designs.

FAQ

Q: Which card has the higher OpenCL benchmark score?

A: The GTX 650 Ti Boost scores 9302 in Geekbench OpenCL, while the GTX 560 scores 9058. The difference is about 2.6 percent in favor of the GTX 650 Ti Boost.

Q: How do the average benchmark scores compare?

A: The GTX 560 has an average benchmark score of 9058. The GTX 650 Ti Boost has an average of 8067, which includes additional test types beyond OpenCL.

Q: Which card has more memory?

A: The GTX 650 Ti Boost has 2 GB of GDDR5, double the 1024 MB found on the GTX 560.

Q: What is the power draw difference?

A: The GTX 560 has a TDP of 150 W and needs two 6 pin connectors. The GTX 650 Ti Boost has a TDP of 134 W and needs only one 6 pin connector.

Q: Which card supports Vulkan?

A: The GTX 650 Ti Boost lists Vulkan 1.2.175 support. The GTX 560 has no Vulkan entry in the database.

Q: How do the percentile rankings compare?

A: The GTX 560 sits at the 45th percentile among all GPUs. The GTX 650 Ti Boost sits at the 42nd percentile.

Specification Differences

The two cards differ across nearly every major specification category.

The manufacturing process moves from 40 nm on the GTX 560 to 28 nm on the GTX 650 Ti Boost. Transistor count rises from 1,950 million to 2,540 million. Die size shrinks from 332 mm² to 221 mm². Transistor density increases from 5.9 million per mm² to 11.5 million per mm².

Memory capacity doubles from 1024 MB to 2 GB. Bus width narrows from 256 bit to 192 bit. Memory clock changes from 1000 MHz to 1502 MHz, and effective data rate rises from 4 Gbps to 6 Gbps. Bandwidth increases from 128.0 GB/s to 144.2 GB/s.

Shading units rise from 336 to 768. Texture mapping units rise from 56 to 64. ROPs drop from 32 to 24. Pixel rate increases from 11.34 GPixel/s to 16.51 GPixel/s. Texture rate increases from 45.36 GTexel/s to 66.05 GTexel/s. FP32 performance rises from 1,088.6 GFLOPS to 1.585 TFLOPS.

Power figures improve on the GTX 650 Ti Boost. TDP drops from 150 W to 134 W. Power connectors go from two 6 pin to one 6 pin. Suggested power supply drops from 450 W to 300 W.

Bus interface upgrades from PCIe 2.0 x16 to PCIe 3.0 x16. Display outputs change from two DVI and one mini HDMI 1.3a to two DVI, one HDMI 1.4a, and one DisplayPort 1.2. Vulkan support appears only on the GTX 650 Ti Boost with version 1.2.175.

Physical length grows from 210 mm to 241 mm. Both cards remain dual slot designs.

Where Each One Wins

The GTX 650 Ti Boost wins the direct OpenCL head-to-head with a score of 9302 versus 9058. It also leads in shading units, texture mapping units, texture rate, pixel rate, FP32 compute, memory capacity, and memory bandwidth. The newer card supports Vulkan, uses a more efficient 28 nm process, draws less power, and requires a smaller power supply. For users prioritizing compute throughput, higher resolution textures, or modern API compatibility, the GTX 650 Ti Boost has the clear advantage in the recorded data.

The GTX 560 counters with a higher percentile ranking at 45 versus 42. Its average benchmark score of 9058 exceeds the GTX 650 Ti Boost’s average of 8067. The GTX 560 also has more ROPs at 32 against 24, a wider 256 bit memory bus, and a shorter physical length. The older card’s nearest rival group shows tighter margins around the 9058 average, with rivals within 0.2 percent on one side and 1.7 percent on the other. For scenarios where the database’s overall distribution matters more than a single workload result, the GTX 560 holds its ground.

The GTX 650 Ti Boost’s Vulkan score of 10041 stands as the highest single benchmark score available for either card. None of the GTX 560’s recorded tests exceed 9058. The Metal score of 4858 on the GTX 650 Ti Boost is the lowest recorded result, dragging down its average, but users running Vulkan workloads would see the strongest performance from the Kepler card.

The Verdict

The data paints a nuanced picture. The GTX 650 Ti Boost is the faster card in raw compute terms, winning the only direct comparison by about 2.6 percent and posting a higher Vulkan score of 10041. It also brings efficiency advantages: lower power draw at 134 W, a single 6 pin connector, and a 300 W suggested power supply. Users who prioritize these measured performance should favor the GTX 650 Ti Boost without hesitation.

The GTX 560, however, still commands respect in the database rankings. Its 45th percentile and average score of 9058 place it above the GTX 650 Ti Boost’s 42nd percentile and 8067 average. The GTX 560’s higher ROP count of 32 and wider 256 bit bus suggest it remains competitive in scenarios where the database weights different workload types differently.

Neither card suits modern high end gaming. The GTX 650 Ti Boost’s Vulkan support makes it the more versatile choice for users running a broader set of APIs. The GTX 560 lacks Vulkan entirely, limiting it to DirectX and OpenGL workloads. For those specific use cases, the GTX 650 Ti Boost is the clearer pick. The GTX 560 retains relevance for users who rely on OpenCL-backed applications, where the margin of 2.6 percent is small enough that real world differences may be imperceptible. The verdict is straightforward for most users: choose the GTX 650 Ti Boost for peak measured performance, and choose the GTX 560 only when the database’s percentile context matters more than raw workload performance.