NVIDIA GeForce GTX 950 Review: Affordable Maxwell For The Masses

NVIDIA GeForce GTX 950 Specifications & Features

	NVIDIA GeForce GTX 950
Graphics Processing Clusters	2
Streaming Multiprocessors	6
CUDA Cores (single precision)	768
Texture Units	48
ROP Units	32
Base Clock	1024MHz
Boost Clock	1188MHz
Memory Clock (Data rate)	3300MHz (Effective Speed - ~6.6Gbps)
L2 Cache Size	1024KB
Total Video Memory	2048 MB GDDR5
Memory Interface	128-Bit
Total Memory Bandwidth	105.6 GB/s
Texture Filtering Rate (Bilinear)	49.2 GigaTexels/sec
Fabrication Process	28 nm
Transistor Count	2.94 Billion
Connectors	3 x Display Port 1.2 1 x Dual-Link DVI 1 x HDMI 2.0
Form Factor	Dual Slot
Power Connectors	One 6-Pin
Recommended Power Supply	350 Watts
Thermal Design Power (TDP)	90 Watts
Thermal Threshold	95°C
Price	$159 MSRP - Find Them At Amazon

---

The GM206 GPU at the heart of the GeForce GTX 950 is the same chip found on the GTX 960, with a couple of blocks disabled. It features roughly 2.94 billion transistors and has die size of 227mm². And like its predecessors, it’s built on TSMC’s 28nm process node.

Like the bigger GM204 powering the GeForce GTX 980, the GM206 is based on NVIDIA's "Maxwell 2.0" microarchitecture. Maxwell is designed to boost efficiency through better GPU utilization, and ultimately improve performance-per-watt and per die area versus older GPUs. Maxwell’s Streaming Multiprocessors, or SMs, are somewhat different than the older designs used in Kepler. With Maxwell, NVIDIA has made improvements to the control logic partitions for better workload balancing, and it also has finer-grained clock-gating and better compiler-based scheduling. Maxwell can also issue more instructions per clock cycle, all of which allow the Maxwell SM (also called an SMM in some NVIDIA docs) to exceed Kepler’s SMX in terms of efficiency. NVIDIA is claiming that Maxwell’s new SM architecture can deliver 40% more performance per CUDA core on shader-limited workloads than Kepler, with up to double the performance per watt, despite using the same 28nm manufacturing process.


A full-featured GM206 GPU contains two GPCs, eight Maxwell Streaming Multiprocessors (SM), and two 64-bit memory controller partitions (128-bit total)—it is essentially half of a GM204. Each SM is partitioned into four separate processing blocks, each with its own instruction buffer, scheduler and 32 CUDA cores. But the version of the chip used on the GTX 950 has a couple of SMs disabled.

NVIDIA’s reference specifications for the GeForce GTX 950 call for a base clock of 1024MHz and a Boost clock of 1188MHz. The GPU is packing 768 CUDA cores, 48 texture units, and 32 ROPs.

The 2GB of video memory on GeForce GTX 950 cards is clocked at a 6.6GHz (effective GDDR5 data rate) and the memory links to the GPU via a 128-bit interface. At those clocks, the GeForce GTX 950 offers up a peak textured fillrate of 49.2 GTexels/s and 105.6 GB/s of memory bandwidth. Those numbers might seem low in light of some previous gen GeForce cards and current Radeon cards with wider memory interfaces, but the memory compression technology in Maxwell affords up to an additional 30% of effective bandwidth, so the max effective data rate is somewhat higher with many workloads.

Though the GM206 is similar to the GM204 in many ways, NVIDIA did make one addition. The GM206 has support for H.265 (HEVC) encoding and decoding. The GTX 980’s NVENC video engine offers native support for H.265 encode only, no decode. Support for other features like NVIDIA’s Voxel Global Illumination (VXGI), Multi-Frame sampled AA (MFAA), and Dynamic Super Resolution (DSR) are all present on the GeForce GTX 950, as is support DirectX 12. Since we’ve already talked about all of these features in detail in our GeForce GTX 980 launch coverage, we won’t rehash them again here. If you’d like to learn more about them, we’d suggest taking a look at this article. Now, let's look at the cards...