| Release date | March 12, 2014; 11 years ago (March 12, 2014)[1] |
|---|---|
| Codename | GF117 GK104 GM10x |
| Architecture | Fermi Kepler Maxwell |
| Models | GeForce series
|
| Transistors | 585M (GF117)
|
| Fabrication process | TSMC28 nm |
| Cards | |
| Entry-level | GeForce 800M GeForce 820M GeForce 825M GeForce 830M GeForce 840M GeForce 845M |
| Mid-range | GeForce GTX 850M GeForce GTX 860M |
| High-end | GeForce GTX 870M GeForce GTX 880M |
| API support | |
| DirectX | Direct3D 12.0 (feature level 11_0)[3][4]Shader Model 6.7 (Maxwell),Shader Model 6.5 (Kepler) orShader Model 5.1 (Fermi) |
| OpenCL | OpenCL 3.0[a] |
| OpenGL | OpenGL 4.6[2] |
| Vulkan | Vulkan 1.2 (Kepler) Vulkan 1.3 (Maxwell) SPIR-V |
| History | |
| Predecessor | GeForce 600 series |
| Variant | GeForce 700 series |
| Successor | GeForce 900 series |
| Support status | |
| Fermi andKepler unsupported. Maxwell limited support. | |
TheGeForce 800M series is a family ofgraphics processing units byNvidia for laptop PCs.[3] It consists of rebrands of mobile versions of theGeForce 700 series[3] and some newer chips that are lower end compared to the rebrands.
The GeForce 800 series name was originally planned to be used for both desktop and mobile chips based on theMaxwell microarchitecture (GM-codenamed chips), named after the Scottish theoretical physicistJames Clerk Maxwell, which was previously introduced into theGeForce 700 series in the GTX 750 and GTX 750 Ti, released on February 18, 2014.[5] However, because mobile GPUs under the GeForce 800M series had already been released using theKepler architecture, Nvidia decided to rename its GeForce 800 series desktop GPUs as theGeForce 900 series.[3]
TheMaxwell microarchitecture, the successor toKepler microarchitecture, was the first Nvidia architecture to feature an integratedARM CPU of its own.[6] This enabled Maxwell GPUs to be more independent from the main CPU according to Nvidia's CEOJen-Hsun Huang.[7] Nvidia expects three major things from the Maxwell architecture: improved graphics capabilities, simplified programming, and better energy-efficiency compared to theGeForce 700 series andGeForce 600 series.[8]
First generation Maxwell GM107/GM108 provides few consumer-facing additional features; Nvidia instead focused on power efficiency. Nvidia's video encoder, NVENC, is 1.5 to 2 times faster than on Kepler-based GPUs meaning it can encode video at 6 to 8 times playback speed.[5] Nvidia also claims an 8 to 10 times performance increase inPureVideo Feature Set E video decoding due to the video decoder cache paired with increases in memory efficiency. However,HEVC is not supported for full hardware decoding, relying on a mix of hardware and software decoding.[5] When decoding video, a new low power state "GC5" is used on Maxwell GPUs to conserve power.[5]
Nvidia increased the amount of L2 cache on GM107 to 2 MB, up from 256 KB on GK107, reducing the memory bandwidth needed. Accordingly, Nvidia cut the memory bus to 128 bit on GM107 from 192 bit on GK106, further saving power.[5] Nvidia also changed the streaming multiprocessor design from that of Kepler (SMX), naming it SMM. The layout of SMM units is partitioned so that each of the fourwarp schedulers controls isolated FP32CUDA cores, load/store units and special function units, unlike Kepler, where the warp schedulers share the resources. Texture units and FP64 CUDA cores are still shared.[5] SMM allows for a finer-grain allocation of resources than SMX, saving power when the workload isn't optimal for shared resources. Nvidia claims a 128 CUDA core SMM has 90% of the performance of a 192 CUDA core SMX.[5]
GM107/GM108 supportsCUDA Compute Capability 5.0 compared to 3.5 on GK110/GK208 GPUs and 3.0 on GK10x GPUs. Dynamic Parallelism and HyperQ, two features in GK110/GK208 GPUs, are also supported across the entire Maxwell product line.
Maxwell provides native shared memory atomic operations for 32-bit integers and native shared memory 32-bit and 64-bit compare-and-swap (CAS), which can be used to implement other atomic functions.
Maxwell supports DirectX 12.[9]
Nvidia has announced that the company skipped the GeForce 800 series for desktop graphics cards, most likely because the GTX 800M series consists of high-end Kepler and low-end Maxwell based components. Instead, Nvidia had announced that the newly renamed GeForce GTX 980 and GTX 970 will be introduced formally on September 19, 2014.[3]
The GeForce 800M series is designed for notebooks. The processing power is obtained by multiplying shader clock speed, the number of cores and how many instructions the cores are capable of performing per cycle. Note that all GK104 based GPUs are using the olderKepler Architecture and the 820M uses GF117 cores based on theFermi Architecture.
| Model | Launch | Code name(s) | Fab (nm) | Bus interface | Core config1 | Clock speed | Fillrate | Memory | API support (version) | Processing Power2 (GFLOPS) | TDP (watts) | |||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Core (MHz) | Shader (MHz) | Memory (MT/s) | Pixel (GP/s) | Texture (GT/s) | Size (MB) | Bandwidth (GB/s) | Type | Bus width (bit) | DirectX | OpenGL | ||||||||
| GeForce 800M[b][10] | March 17, 2014 | GF117 | TSMC 28 | PCIe 2.0 ×8 | 48:8:8 | 738 | 1476 | 2000 | 5.9 | 5.9 | 2048 | 14.4 | DDR3 | 64 | 12.0 (11_0) | 4.6 | 141.7 | 15 |
| GeForce 820M[b][11] | November 27, 2013 | PCIe 2.0 ×16 | 96:16:4 | 719–954 | 1438–1908 | 2.9–3.8 | 11.5–15.3 | 16 | 276.1–366.3 | 15 | ||||||||
| GeForce 830M[c][12] | March 12, 2014 | GM108 | PCIe 3.0 ×16 | 256:16:8 (2 SMM) | 1082 | — | 8.2 | 16.5 | 14.4 | 554 | 30 | |||||||
| GeForce 840M[c][13] | 384:24:8 (3 SMM) | 1029 | 8.2 | 24.7 | 16 | 790.3 | 30 | |||||||||||
| GeForce GTX 850M[14] | GM107 | 640:40:16 (5 SMM) | 936+Boost 876+Boost | 5000 | 14.0 | 35.0 | 2048 4096 | 32 80 | DDR3 GDDR5 | 128 | 1198.1 | 40 | ||||||
| GeForce GTX 860M[15] | GM107 GK104 | 640:40:16 (5 SMM) 1152:96:16 (6 SMX) | 1029+Boost 797+Boost | 16.5 12.8 | 41.2 76.5 | 80 | GDDR5 | 1317.1 1836.3 | 40–45 75 | |||||||||
| GeForce GTX 870M[16] | GK104 | 1344:112:24 (7 SMX) | 941+Boost | 22.6 | 105.4 | 3072 6144 | 120 | 192 | 2529.4 | 100 | ||||||||
| GeForce GTX 880M[17] | 1536:128:32 (8 SMX) | 954+Boost | 30.5 | 122.1 | 4096 8192 | 160 | 256 | 2930.7 | 105 | |||||||||
The next generation of mobile cards was the900M series, which continued usage of the Maxwell archicture. Nvidia followed the Maxwell architecture with thePascal architecture in 2016 in theGeForce 10 series.
"Driver 368.81 is the last driver to support Windows XP/Windows XP 64-bit".
In March 2018, Nvidia discontinued support for 32-bit operating systems with the release of driver version 391.35.[18]
Notebook GPUs based on the Kepler architecture moved to legacy support in April 2019 and stopped receiving critical security updates in April 2020.[19][20] The Nvidia GeForce GTX 830M, 840M and 850M from the 8xxM GPU family were unaffected by this change.
Nvidia announced that after Release 470 drivers, it would transition driver support for the Windows 7 and Windows 8.1 operating systems to legacy status and continue to provide critical security updates for these operating systems through September 2024.[21]
In January 2025, Nvidia has announced plans to wind down developer support for the Maxwell architecture, which includes the Maxwell based models of the GTX 800M-Series. While CUDA support will be deprecated in a future release, driver updates are expected to continue for a limited time.[22]