NVIDIA GeForce GTX 650 Ti Boost vs NVIDIA GeForce GTX 950M Comparison

FAQ

Q: Which GPU has the higher average benchmark score?

A: The NVIDIA GeForce GTX 950M posts an average benchmark score of 8135, while the NVIDIA GeForce GTX 650 Ti Boost scores 8067. The 950M leads by approximately 0.8% in aggregate performance.

Q: How do the two cards compare in OpenCL performance?

A: The GTX 950M wins the Geekbench OpenCL test with a score of 9745 against the GTX 650 Ti Boost's 9302. That is a 4.8% advantage for the mobile Maxwell part.

Q: Which GPU wins in Vulkan performance?

A: The GTX 650 Ti Boost dominates the Geekbench Vulkan test, scoring 10041 versus the GTX 950M's 6525. The desktop Kepler card leads by 35% in this specific API workload.

Q: What are the memory specifications of each card?

A: The GTX 950M has 4 GB of DDR3 memory on a 128-bit bus with 28.80 GB/s bandwidth. The GTX 650 Ti Boost has 2 GB of GDDR5 memory on a 192-bit bus with 144.2 GB/s bandwidth.

Q: How do the transistor counts and die sizes differ?

A: The GTX 650 Ti Boost uses 2,540 million transistors on a 221 mm² die, while the GTX 950M uses 1,870 million transistors on a 148 mm² die. Both are built on TSMC's 28 nm process.

Q: What is the TDP difference between the two cards?

A: The GTX 950M has a 75 W TDP and is an IGP (integrated graphics processor) with no power connectors. The GTX 650 Ti Boost has a 134 W TDP, is a dual-slot card, and requires a single 6-pin power connector with a 300 W suggested PSU.

Where Each One Wins

The GTX 950M claims victory in the OpenCL compute workload, which often reflects general-purpose GPU compute tasks and certain productivity applications. Its 4.8% edge over the GTX 650 Ti Boost in this test suggests that the Maxwell architecture's efficiency translates into competitive raw compute throughput despite having fewer shading units (640 vs 768).

The GTX 650 Ti Boost takes a commanding lead in Vulkan, a low-overhead graphics API that tends to favor higher memory bandwidth and raw fill rates. The 35% margin is substantial and indicates that for Vulkan-based games and applications, the desktop Kepler card is clearly the stronger performer.

Beyond benchmark wins, the GTX 950M offers 4 GB of memory versus 2 GB, which matters for texture-heavy workloads or higher-resolution assets. The GTX 650 Ti Boost counters with GDDR5 memory and 144.2 GB/s bandwidth, a fivefold advantage over the 950M's DDR3 at 28.80 GB/s, which directly impacts memory-bound scenarios.

The GTX 650 Ti Boost also wins on texture rate (66.05 GTexel/s vs 44.96 GTexel/s) and has more TMUs (64 vs 40) and ROPs (24 vs 16). The GTX 950M edges out on pixel rate (17.98 GPixel/s vs 16.51 GPixel/s) despite fewer ROPs, thanks to its higher boost clock relative to its architecture.

Architecture Differences

The GTX 950M is built on NVIDIA's Maxwell architecture using the GM107 chip. The GTX 650 Ti Boost uses the older Kepler architecture with the GK106 chip. This generational gap explains several performance characteristics.

Maxwell was designed for improved power efficiency per clock, which is evident in the 950M's 75 W TDP versus the 650 Ti Boost's 134 W. The 950M achieves this with 1,870 million transistors on a 148 mm² die, while the 650 Ti Boost packs 2,540 million transistors on a 221 mm² die. Transistor density is slightly higher on the 950M at 12.6M per mm² versus 11.5M per mm² on the 650 Ti Boost.

Both GPUs support DirectX 12 (11_0) and OpenGL 4.6. The Vulkan API support differs: the 950M supports Vulkan 1.4, while the 650 Ti Boost supports Vulkan 1.2.175. This could affect long-term compatibility with newer Vulkan applications, though the 650 Ti Boost's superior Vulkan benchmark score suggests its hardware is better suited to the API's workloads.

The memory subsystems are fundamentally different. The 950M uses DDR3 with a 128-bit bus, while the 650 Ti Boost uses GDDR5 with a 192-bit bus. This explains the massive bandwidth gap (28.80 GB/s vs 144.2 GB/s) and likely contributes to the 650 Ti Boost's Vulkan advantage.

Specification Differences

The GTX 950M and GTX 650 Ti Boost differ across nearly every major specification category. Clock speeds are close: the 950M has a 993 MHz base and 1124 MHz boost, while the 650 Ti Boost has a 980 MHz base and 1032 MHz boost. The 950M's memory runs at 900 MHz (1800 Mbps effective), whereas the 650 Ti Boost's memory runs at 1502 MHz (6 Gbps effective).

Shading units favor the 650 Ti Boost with 768 versus 640. TMUs are 64 vs 40, and ROPs are 24 vs 16, both favoring the desktop card. The 650 Ti Boost also leads in FP32 performance at 1.585 TFLOPS versus 1,438.7 GFLOPS for the 950M.

Memory capacity favors the 950M at 4 GB versus 2 GB, but memory type and bus width heavily favor the 650 Ti Boost. The 650 Ti Boost uses a PCIe 3.0 x16 interface, while the 950M uses PCIe 3.0 x8. The 650 Ti Boost has a 241 mm length and dual-slot design with 2x DVI, 1x HDMI 1.4a, and 1x DisplayPort 1.2 outputs; the 950M has portable-device-dependent outputs.

The 650 Ti Boost carries a launch MSRP of 169 USD. The 950M has no listed launch MSRP.

Head-to-Head Benchmarks

The two GPUs split their benchmark matchups exactly one win apiece. The GTX 950M wins Geekbench OpenCL with a score of 9745 against 9302 for the GTX 650 Ti Boost, a 4.8% margin. This is a moderate but consistent lead in a compute-oriented workload.

The GTX 650 Ti Boost wins Geekbench Vulkan decisively with 10041 against 6525, a 35% advantage. This is the largest gap between the two cards in any measured test and represents a significant performance cliff for the 950M in Vulkan-based applications.

Looking at the nearest rivals for context, the GTX 950M's 8135 average score places it just above the AMD Radeon R9 M360 (8129, +0.1%), the NVIDIA GeForce GTX 980 (8167, -0.4%), the NVIDIA GeForce 945M (8099, +0.4%), and the NVIDIA GRID K2 (8080, +0.7%). The GTX 650 Ti Boost's 8067 average sits near the NVIDIA GRID K2 (8080, -0.2%), the NVIDIA GeForce GTX 650 Ti (8053, +0.2%), the NVIDIA GeForce GTX 880M (8040, +0.3%), and the NVIDIA Quadro P5000 (8039, +0.3%).

Both cards land at the 42nd percentile among all GPUs, indicating they occupy a similar overall performance tier despite their architectural differences. The benchmark data shows two cards with nearly identical aggregate performance but very different workload-specific strengths.

The Verdict

The data paints a nuanced picture. The GTX 950M and GTX 650 Ti Boost achieve nearly identical average benchmark scores (8135 vs 8067, a 0.8% difference) and share the same 42nd percentile ranking among all GPUs. Neither card can claim overall superiority based on aggregate metrics.

For users prioritizing OpenCL compute performance, the GTX 950M is the clear choice with a 4.8% lead. Its 4 GB memory capacity also provides advantages for texture-heavy workloads, and its 75 W TDP with IGP form factor makes it suitable for portable systems where power and space are constrained.

For users prioritizing Vulkan graphics performance, the GTX 650 Ti Boost is unequivocally superior with a 35% margin. Its GDDR5 memory and 144.2 GB/s bandwidth provide a massive advantage in memory-bound scenarios, and its higher texture rate (66.05 GTexel/s) and FP32 throughput (1.585 TFLOPS) support demanding graphical workloads. The 650 Ti Boost also offers a longer list of display outputs for multi-monitor setups.

Architecture differences reinforce the performance split. The Maxwell-based 950M is more transistor-dense (12.6M / mm² vs 11.5M / mm²) and power-efficient, but the Kepler-based 650 Ti Boost has more raw resources: more shading units, TMUs, ROPs, and higher bandwidth. The 650 Ti Boost's older architecture nevertheless supports a newer Vulkan revision in its driver stack, though its benchmark lead in Vulkan likely stems from hardware capabilities rather than software version.

The 650 Ti Boost's higher TDP (134 W vs 75 W) and dual-slot form factor make it unsuitable for thin-and-light laptops, but its desktop-oriented design with a 6-pin connector and 300 W suggested PSU indicates it was built for performance without power constraints. The 950M's IGP design with no power connectors reflects its mobile-first positioning.

In practical terms, the choice depends on the intended use case. The GTX 950M suits compute-focused workloads and systems requiring low power draw. The GTX 650 Ti Boost suits Vulkan-based gaming and applications where memory bandwidth is critical. The 35% Vulkan gap is the single largest performance differential in the data and should weigh heavily for any user targeting that API. Conversely, the 950M's OpenCL lead, while smaller, still represents a measurable advantage for compute tasks. Users should match the card to their dominant workload rather than relying on aggregate scores, which obscure these significant per-API differences.