NVIDIA GeForce GTX 465 vs NVIDIA GeForce GTX 950M Comparison

The NVIDIA GeForce GTX 465 and the NVIDIA GeForce GTX 950M occupy interesting positions in the database: a desktop Fermi card from the GeForce 400 generation and a mobile Maxwell part from the GeForce 900M generation, both now end-of-life. On the single recorded head-to-head benchmark, the newer GTX 950M wins, scoring 9745 in Geekbench OpenCL against 9294 for the GTX 465, a 4.6 percent gap. Yet the recorded data shows the two are far closer than their architectural distance might suggest, and each holds clear advantages in specific measurable categories that matter for different use cases.

Where Each One Wins

The benchmark ledger is one-sided: the GTX 950M takes the only recorded head-to-head test, so winsA stands at zero and winsB at one. That single OpenCL result, however, sits within a margin the database treats as effectively a tie between differently balanced designs, and the raw specification data explains why the split is not wider.

The GTX 465 wins on everything related to memory throughput and output bandwidth. Its GDDR5 subsystem delivers 102.7 GB/s across a 256-bit bus, against 28.80 GB/s on a 128-bit DDR3 configuration for the GTX 950M. That is a bandwidth advantage of roughly three and a half times, a gap that directly benefits texture-heavy workloads, high-resolution output, and any compute task that streams large datasets through memory. The desktop card also leads in raster output: 32 ROPs versus 16, giving it the fill-rate foundation for higher-resolution rendering, even though its absolute pixel rate is lower due to clock speeds.

The GTX 950M wins on raw compute and efficiency. Its FP32 throughput of 1,438.7 GFLOPS exceeds the GTX 465's 855.4 GFLOPS, it posts a higher pixel rate at 17.98 GPixel/s versus 13.38 GPixel/s, and a higher texture rate at 44.96 GTexel/s versus 26.75 GTexel/s. Add a TDP of 75 W against 200 W, no external power connectors against two 6-pin connectors, and the mobile part is the clear choice where power envelope or battery life matters. Its 4 GB of memory also doubles the 1024 MB on the GTX 465, which matters for modern workloads that exceed a 1 GB framebuffer.

Architecture Differences

The GTX 465 is built on the GF100 chip, Fermi architecture, GeForce 400 generation, manufactured at TSMC on a 40 nm process. It packs 3,100 million transistors onto a 529 mm² die, a transistor density of 5.9M per mm². The GTX 950M uses the GM107 chip, Maxwell architecture, GeForce 900M generation, also from TSMC but on 28 nm, with 1,870 million transistors on a 148 mm² die and a density of 12.6M per mm². The generational jump in process node is the single most consequential difference in this pairing: more than double the transistor density on a die less than a third the size, and it is the reason the newer chip delivers substantially more compute at well under half the power draw.

Shader configuration diverges sharply. The GTX 465 fields 352 shading units, 44 TMUs, and 32 ROPs. The GTX 950M counters with 640 shading units and 40 TMUs, but only 16 ROPs. Maxwell's approach of more, more efficient shading units is visible in the FP32 numbers despite the mobile part's memory disadvantage.

Clocking reflects the platform split. The GTX 465 has no base or boost clock recorded, while the GTX 950M runs a 993 MHz base and 1124 MHz boost. Memory clocks tell the opposite story: 802 MHz GDDR5 (3.2 Gbps effective) on the desktop card versus 900 MHz DDR3 (1800 Mbps effective) on the mobile one, with the bus width difference compounding into that 102.7 GB/s versus 28.80 GB/s bandwidth gap.

Platform features also differ. The GTX 465 connects over PCIe 2.0 x16 and offers fixed outputs, two DVI plus one mini-HDMI 1.3a, in a dual-slot card 241 mm long. The GTX 950M uses PCIe 3.0 x8 with display outputs described as portable device dependent, since it is an integrated mobile solution with no power connectors. Both support DirectX 12 with feature level 11_0 and OpenGL 4.6, but only the GTX 950M records Vulkan support, at version 1.4, which matters for titles and applications built on that API. Neither GPU has RT cores or tensor cores.

The GTX 465 carried a launch MSRP of 279 USD. The GTX 950M has no launch MSRP recorded in the database, consistent with mobile parts that shipped inside complete systems.

Head-to-Head Benchmarks

The only direct comparison in the database is Geekbench OpenCL. The GTX 950M scores 9745; the GTX 465 scores 9294. The delta is 4.6 percent in favor of the mobile Maxwell chip. Interpreted against the percentile fields, the picture gets nuanced: the GTX 465 sits at the 46th percentile against all GPUs in the database, while the GTX 950M sits at the 42nd. The older desktop card actually ranks slightly higher on that distribution, because the percentile calculation uses each card's average benchmark score: 9294 for the GTX 465 versus 8135 for the GTX 950M, the latter dragged down by a second recorded test, Geekbench Vulkan, where the mobile part scores 6525.

The rival clustering reinforces how tight this matchup is. The GTX 465's nearest rivals include the NVIDIA GeForce GTX 850M at 9302 (just 0.1 percent apart), the AMD Radeon R7 M380 at 9313 (0.2 percent apart), the NVIDIA GeForce GTX 960 at 9273 (0.2 percent apart), and the AMD Radeon Vega 8 at 9221 (0.8 percent apart). The GTX 950M's neighborhood includes the AMD Radeon R9 M360 at 8129 (0.1 percent), the NVIDIA GeForce GTX 980 at 8167 (0.4 percent), the NVIDIA GeForce 945M at 8099 (0.4 percent), and the NVIDIA GRID K2 at 8080 (0.7 percent). Both cards sit in dense clusters, meaning neither separates itself from peers; this is a mid-pack pairing on compute.

Where the specifications diverge sharply, the wins are unambiguous. The GTX 950M wins FP32 compute by a wide absolute margin, 1,438.7 GFLOPS against 855.4 GFLOPS. It wins pixel rate, 17.98 versus 13.38 GPixel/s, and texture rate, 44.96 versus 26.75 GTexel/s. The GTX 465 wins memory bandwidth, 102.7 versus 28.80 GB/s, ROP count, 32 versus 16, bus width, 256-bit versus 128-bit, and framebuffer capacity is the GTX 950M's, 4 GB versus 1024 MB. Power draw splits them most dramatically: 200 W with two 6-pin connectors and a suggested 550 W PSU for the GTX 465, versus 75 W with no connectors for the GTX 950M.

The Verdict

The recorded data supports a clear division. Pick the GTX 950M if compute throughput, power efficiency, or memory capacity is the priority: higher FP32, higher pixel and texture rates, 4 GB of memory, Vulkan 1.4 support, and a 75 W envelope with no external power. Its single benchmark win and its newer API support make it the more practical chip for sustained use inside a constrained platform.

Pick the GTX 465 if memory bandwidth and raster output dominate the workload. Its 102.7 GB/s of GDDR5 bandwidth is more than triple the mobile part's, and its 32 ROPs double the GTX 950M's, advantages that show up in bandwidth-bound rendering and compute tasks that fit the 1024 MB framebuffer. Note the platform cost: a dual-slot, 241 mm card drawing 200 W with a 550 W suggested PSU.

Both are end-of-life parts sitting in the lower-middle of the database distribution, the GTX 465 at the 46th percentile and the GTX 950M at the 42nd. Neither is a performance leader by current data. For a mobile or power-limited scenario, the GTX 950M is the only sensible option of the two. For a desktop with a capable PSU where bandwidth matters more than efficiency, the GTX 465 remains the stronger fit on paper, though the single head-to-head benchmark favors its younger rival by 4.6 percent.

FAQ

Q: Which GPU wins the head-to-head benchmark? A: The GTX 950M. It scores 9745 in Geekbench OpenCL against 9294 for the GTX 465, a 4.6 percent margin.

Q: How much more memory bandwidth does the GTX 465 have? A: 102.7 GB/s versus 28.80 GB/s for the GTX 950M, thanks to GDDR5 on a 256-bit bus against DDR3 on a 128-bit bus.

Q: Which GPU is more power efficient? A: The GTX 950M, with a 75 W TDP and no power connectors. The GTX 465 has a 200 W TDP, requires two 6-pin connectors, and carries a suggested 550 W PSU.

Q: Which one has more video memory? A: The GTX 950M, with 4 GB. The GTX 465 has 1024 MB.

Q: Which GPU has higher raw compute throughput? A: The GTX 950M, at 1,438.7 GFLOPS FP32 versus 855.4 GFLOPS for the GTX 465.

Q: Where does each GPU rank against all GPUs in the database? A: The GTX 465 sits at the 46th percentile; the GTX 950M sits at the 42nd. Both cluster tightly with their nearest rivals, within roughly one percent in most cases.