The GeForce 10 series is a series of graphics processing units developed by Nvidia, initially based on the Pascal microarchitecture announced in March 2014. This design series succeeded the GeForce 900 series, and is succeeded by the GeForce 16 series and GeForce 20 series using the Turing microarchitecture.

Quick Facts Release date, Manufactured by ...
GeForce 10 series
Thumb
GeForce GTX 1080 Ti released in 2017, the series' flagship unit
Release dateMay 27, 2016; 8 years ago
Manufactured byTSMC
Samsung
Designed byNvidia
Marketed byNvidia
CodenameGP10x
Architecture
ModelsGeForce GTX series
Transistors
  • 1.8B (GP108) 14 nm
  • 3.3B (GP107) 14 nm
  • 4.4B (GP106) 16 nm
  • 7.2B (GP104) 16 nm
  • 12B (GP102) 16 nm
Fabrication process
Cards
Entry-level
  • GeForce GT 1010
  • GeForce GT 1030
Mid-range
  • GeForce GTX 1050
  • GeForce GTX 1050 Ti
  • GeForce GTX 1060
High-end
  • GeForce GTX 1070
  • GeForce GTX 1070 Ti
  • GeForce GTX 1080
Enthusiast
  • GeForce GTX 1080 Ti
  • Nvidia Titan X (Pascal)
  • Nvidia Titan Xp
API support
DirectXDirect3D 12.0 (feature level 12_1)
Shader Model 6.7
OpenCLOpenCL 3.0[1][a]
OpenGLOpenGL 4.6[2]
Vulkan
History
PredecessorGeForce 900 series
Successor
Support status
Supported
Close

Architecture

The Pascal microarchitecture, named after Blaise Pascal, was announced in March 2014 as a successor to the Maxwell microarchitecture.[4] The first graphics cards from the series, the GeForce GTX 1080 and 1070, were announced on May 6, 2016, and were released several weeks later on May 27 and June 10, respectively. The architecture incorporates either 16 nm FinFET (TSMC) or 14 nm FinFET (Samsung) technologies. Initially, chips were only produced in TSMC's 16 nm process, but later chips were made with Samsung's newer 14 nm process (GP107, GP108).[5]

New Features in GP10x:

  • CUDA Compute Capability 6.0 (GP100 only), 6.1 (GP102, GP104, GP106, GP107, GP108)
  • DisplayPort 1.4 (No DSC)
  • HDMI 2.0b
  • Fourth generation Delta Color Compression
  • PureVideo Feature Set H hardware video decoding HEVC Main10 (10 bit), Main12 (12 bit) & VP9 hardware decoding (GM200 & GM204 did not support HEVC Main10/Main12 & VP9 hardware decoding)[6]
  • HDCP 2.2 support for 4K DRM protected content playback & streaming (Maxwell GM200 & GM204 lack HDCP 2.2 support, GM206 supports HDCP 2.2)[7]
  • NVENC HEVC Main10 10 bit hardware encoding (except GP108 which doesn't support NVENC[8])
  • GPU Boost 3.0
  • Simultaneous Multi-Projection
  • HB SLI Bridge Technology
  • New memory controller with GDDR5X & GDDR5 support (GP102, GP104, GP106)[9]
  • Dynamic load balancing scheduling system. This allows the scheduler to dynamically adjust the amount of the GPU assigned to multiple tasks, ensuring that the GPU remains saturated with work except when there is no more work that can safely be distributed. Nvidia therefore has safely enabled asynchronous compute in Pascal's driver.[10]
  • Instruction-level preemption. In graphics tasks, the driver restricts this to pixel-level preemption because pixel tasks typically finish quickly and the overhead costs of doing pixel-level preemption are much lower than performing instruction-level preemption. Compute tasks get either thread-level or instruction-level preemption. Instruction-level preemption is useful because compute tasks can take long times to finish and there are no guarantees on when a compute task finishes, so the driver enables the very expensive instruction-level preemption for these tasks.[11]
  • Triple buffering implemented in the driver level. Nvidia calls this "Fast Sync". This has the GPU maintain three frame buffers per monitor. This results in the GPU continuously rendering frames, and the most recently completely rendered frame is sent to a monitor each time it needs one. This removes the initial delay that double buffering with vsync causes and disallows tearing. The costs are that more memory is consumed for the buffers and that the GPU will consume power drawing frames that might be wasted because two or more frames could possibly be drawn between the time a monitor is sent a frame and the time the same monitor needs to be sent another frame. In this case, the latest frame is picked, causing frames drawn after the previously displayed frame but before the frame that is picked to be completely wasted.[12] This feature has been backported to Maxwell-based GPUs in driver version 372.70.[13]

Nvidia has announced that the Pascal GP100 GPU will feature four High Bandwidth Memory stacks, allowing a total of 16 GB HBM2 on the highest-end models,[14] 16 nm technology,[5] Unified Memory and NVLink.[15]

Starting with Windows 10 version 2004, support has been added for using a hardware graphics scheduler to reduce latency and improve performance, which requires a driver level of WDDM 2.7.

Products

Founders Edition

Announcing the GeForce 10 series products, Nvidia introduced Founders Edition graphics card versions of the GTX 1060, 1070, 1070 Ti, 1080 and 1080 Ti. These are what were previously known as reference cards, i.e. which were designed and built by Nvidia and not by its authorized board partners. These cards were used as reference to measure performance of partner cards. The Founders Edition cards have a die cast machine-finished aluminum body with a single radial fan and a vapor chamber cooling (1070 Ti, 1080, 1080 Ti only[16]), an upgraded power supply and a new low profile backplate (1070, 1070 Ti, 1080, 1080 Ti only).[17] Nvidia also released a limited supply of Founders Edition cards for the GTX 1060 that were only available directly from Nvidia's website.[18] Founders Edition cards prices (with the exception of the GTX 1070 Ti and 1080 Ti) are greater than MSRP of partners cards; however, some partners' cards, incorporating a complex design, with liquid or hybrid cooling may cost more than Founders Edition.

Reintroduction of older cards

Due to production problems surrounding the RTX 30-series cards and a general shortage of graphics cards due to production issues caused by the ongoing COVID-19 pandemic, which led to a global shortage of semiconductor chips, and general demand for graphics cards increasing due to an increase in cryptocurrency mining, the GTX 1050 Ti, alongside the RTX 2060 and its Super counterpart,[19] was brought back into production in 2021.[20][21]

In addition, Nvidia quietly released the GeForce GT 1010 in January 2021.[22]

GeForce 10 (10xx) series for desktops

More information Model, Launch ...
Model Launch Code name(s) Fab (nm) Transistors (billion) Die size (mm2) Bus
interface
Core
config[c]
SM
count[d]
L2 cache
(KB)
Clock speeds[e] Fillrate[f][g] Memory[e] Processing power (GFLOPS)[h] TDP
(watts)
SLI HB
support[i]
Launch MSRP (USD)
Base
core
clock
(MHz)
Boost
core
clock
(MHz)
Memory
(MT/s)
Pixel
(GP/s)
Texture
(GT/s)
Size
(GB)
Bandwidth
(GB/s)
Type Bus
width
(bit)
Single precision
(boost)
Double precision
(boost)
Half precision
(boost)[24]
Standard Founders
Edition

GeForce GT
1010 (DDR4)[j][25][26][27]
Jun 7, 2022 GP108-200-A1 14 1.8 74 PCIe 3.0
×4
256:16:16 2 256 1151 1379 2100 18.42 18.42 2 16.8 DDR4 64 589.3
(706.1)
24.56
(29.42)
20 No  ?
GeForce GT
1010[j][22][28]
Jan 13, 2021 GP108-200-A1 1228 1468 5000 23.49 23.49 40.1 GDDR5 629
(752)
26.2
(31.3)
30 $70[29]
GeForce GT
1030 (DDR4)[j][30][31]
Mar 12, 2018 GP108-310-A1 384:24:16 3 512 1151 1379 2100 18.41 27.6 16.8 DDR4 883
(1059)
27
(33)
13
(16)
20 $80[32]
GeForce GT
1030[j][30][33]
May 17, 2017 GP108-300-A1 1227 1468 6000 19.6 29.4 48 GDDR5 942
(1127)
29
(35)
15
(18)
30
GeForce GTX
1050 (2GB)[34][35]
Oct 25, 2016 GP107-300-A1 3.3 132 PCIe 3.0
×16
640:40:32 5 1024 1354 1455 7000 43.3 54.2 112 128 1733
(1862)
54
(58)
27
(29)
75 $109
GeForce GTX
1050 (3GB)[36]
May 21, 2018 GP107-301-A1 768:48:24 6 768 1392 1518 33.4 66.8 3 84 96 2138
(2332)
66
(72)
33
(36)
 ?
GeForce GTX
1050 Ti[34][37][38]
Oct 25, 2016 GP107-400-A1 768:48:32 1024 1290 1392 41.3 61.9 4 112 128 1981
(2138)
62
(67)
31
(33)
$139
GeForce GTX
1060 (3GB)[39][40]
Aug 18, 2016 GP106-300-A1 16 4.4 200 1152:72:48 9 1536 1506 1708 8000 72.3 108.4 3 192 192 3470
(3935)
108
(123)
54
(61)
120 $199
GeForce GTX
1060 (5GB)[41][42]
Dec 26, 2017
(Only available in China)
GP106-350-K3-A1 1280:80:40 10 1280 8000 60.2 120.5 5 160 160 3855
(4372)
120
(137)
60
(68)
OEM
GeForce GTX
1060[39][43][44]
Jul 19, 2016 GP106-400-A1
GP106-410-A1
1280:80:48 1536 8000
9000
72.3 6 192
216
192 $249 $299
GeForce GTX
1060 (GDDR5X)[45]
Oct 18, 2018 GP104-150-KA-A1 7.2 314 8000 192 GDDR5X
GeForce GTX
1070[46][47]
Jun 10, 2016 GP104-200-A1 1920:120:64 15 2048 1683 96.4[k][48] 180.7 8 256 GDDR5 256 5783
(6463)
181
(202)
90
(101)
150 2-way SLI HB[49]
or 2/3/4-way SLI[50]
$379 $449
GeForce GTX
1070 Ti[51]
Nov 2, 2017 GP104-300-A1 2432:152:64 19 1607 102.8 244.3 7816
(8186)
244
(256)
122
(128)
180 $449
GeForce GTX
1080[23][52][53]
May 27, 2016 GP104-400-A1
GP104-410-A1
2560:160:64 20 1733 10000
11000
257.1 320
352
GDDR5X 8228
(8873)
257
(277)
128
(139)
$599 $699
GeForce GTX
1080 Ti[54]
Mar 10, 2017 GP102-350-K1-A1 12 471 3584:224:88 28 2816 1480 1582 11000 130.2 331.5 11 484 352 10609
(11340)
332
(354)
166
(177)
250 $699
Nvidia
Titan X[55][56]
Aug 2, 2016 GP102-400-A1 3584:224:96 3072 1417 1531 10000 136 317.4 12 480 384 10157
(10974)
317
(343)
159
(171)
$1200
Nvidia
Titan Xp[57][58]
Apr 6, 2017 GP102-450-A1 3840:240:96 30 1405 1582 11410 135 337.2 547.7 10790
(12150)
337
(380)
169
(190)
Model Launch Code name(s) Fab (nm) Transistors (billion) Die size (mm2) Bus
interface
Core
config[c]
SM
count[d]
L2 cache
(KB)
Clock speeds[e] Fillrate[f][g] Memory Processing power (GFLOPS)[h] TDP
(watts)
SLI HB
support[i]
Launch MSRP (USD)
Base
core
clock
(MHz)
Boost
core
clock
(MHz)
Memory
(MT/s)
Pixel
(GP/s)
Texture
(GT/s)
Size
(GB)
Bandwidth
(GB/s)
Type Bus
width
(bit)
Single precision
(boost)
Double precision
(boost)
Half precision
(boost)[59]
Standard Founders
Edition
Close
  1. In OpenCL 3.0, OpenCL 1.2 functionality has become a mandatory baseline, while all OpenCL 2.x and OpenCL 3.0 features were made optional.
  2. The Nvidia Titan Xp & the Founders Edition GTX 1080 Ti does not have a dual link DVI port, but a DisplayPort to single link DVI adapter is included in the box.
  3. The number of streaming multiprocessors on the GPU.
  4. GTX 1060 and GTX 1080 cards shipped after April 2017 feature increased memory speeds, thus increasing memory bandwidth.
  5. Pixel fillrate is calculated as the lowest of three numbers: number of ROPs multiplied by the base core clock speed, number of rasterizers multiplied by the number of fragments they can generate per rasterizer multiplied by the base core clock speed, and the number of streaming multiprocessors multiplied by the number of fragments per clock that they can output multiplied by the base clock rate.
  6. Texture fillrate is calculated as the number of TMUs multiplied by the base core clock speed.
  7. For calculating the processing power, see the Performance subsection of the Pascal architecture article.
  8. SLI HB only supports a maximum of 2-way SLI using SLI HB bridges, however if using traditional SLI bridges it can support a maximum of 4-way SLI but the performance is mostly improved in synthetic benchmarks only.
  9. Lacks hardware video encoder
  10. The GTX 1070 has one of the four GPCs disabled in the die. Losing one of the Raster Engines only allows for the use of 48 ROPs per cycle.

GeForce 10 (10xx) series for notebooks

Thumb
GTX 1080 Ti die (GP-102-350-K1-A1)

The biggest highlight to this line of notebook GPUs is the implementation of configured specifications close to (for the GTX 1060–1080) and exceeding (for the GTX 1050/1050 Ti) that of their desktop counterparts, as opposed to having "cut-down" specifications in previous generations. As a result, the "M" suffix is completely removed from the model's naming schemes, denoting these notebook GPUs to possess similar performance to those made for desktop PCs, including the ability to overclock their core frequencies by the user, something not possible with previous generations of notebook GPUs. This was made possible by having lower Thermal Design Power (TDP) ratings as compared to their desktop equivalents, making these desktop-level GPUs thermally feasible to be implemented into OEM notebook chassis with improved thermal dissipation designs, and, as such, are only available through the OEMs. In addition, the entire line of GTX Notebook GPUs also are available in lower-TDP and quieter variations called the "Max-Q Design", specifically made for ultra-thin gaming systems in conjunction with OEM Partners that incorporate enhanced heat dissipation mechanisms with lower operating noise volumes, which are also made available as an additional more powerful option to existing gaming notebooks as well, which was launched on 27 June 2017.

In addition, the GT series line of Notebook GPUs is no longer introduced starting from this generation, replaced by the MX series of Notebook GPUs. Only the MX150 is based on Pascal's GP108 die used on the GT1030 for Desktops, with higher clock frequencies compared to its Desktop counterpart, while the other chips in the MX series were re-branded versions of the previous generation GPUs (MX130 is a re-branded GT940MX GPU while MX110 is a re-branded GT920MX GPU).[citation needed]

  • Supported APIs are: Direct3D 12 (feature level 12_1 or 11_0 on MX110 and MX130), OpenGL 4.6, OpenCL 3.0 and Vulkan 1.3
  • Only GTX 1070 and GTX 1080 have SLI support.
More information Model, Launch ...
Model Launch Code
name
(s)
Fab (nm) Transistors (billion) Die size (mm2) Bus
interface
Core
config[a]
SM
Count[b]
L2
cache

(MB)
Clock speeds Fillrate[c][d] Memory Processing power (GFLOPS)[e] TDP
(watts)
Base
core
clock
(MHz)
Boost
core
clock
(MHz)
Memory
(MT/s)
Pixel
(GP/s)
Texture
(GT/s)
Size
(GB)
Bandwidth
(GB/s)
Type Bus
width
(bit)
Single
precision

(Boost)
Double
precision

(Boost)
Half
precision

(Boost)

GeForce
MX110[f][60][61]
Nov 17, 2017 GM108
(N16V-GMR1)
28  ?  ? PCIe 3.0
×4
384:24:8 3 1.0 965 993 1800
(DDR3)
5000
(GDDR5)
7.944 23.83 2 14.4
(DDR3)
40.1
(GDDR5)
DDR3
GDDR5
64 741.1
(762.6)
23.16
(23.83)
30
GeForce
MX130[f][62][63]
GM108
(N16S-GTR)
1122 1242 9.936 29.81 861.7
(953.9)
26.93
(29.81)
GeForce
MX150[f][64][65][66]
May 17, 2017 GP108
(N17S-LG)
14 1.8 74 384:24:16 0.5 937 1038 5000 14.99 22.49 2
4
40.1 GDDR5 719.6
(797.2)
22.49
(24.91)
11.24
(12.45)
10
GP108
(N17S-G1)
1468 1532 6000 23.49 35.23 48 1127
(1177)
35.23
(36.77)
17.62
(18.38)
25
GeForce GTX
1050 Max-Q
(Notebook)[67][68]
Jan 3, 2018 GP107
(N17P-G0)
3.3 132 PCIe 3.0
×16
640:40:16 5 1.0 9991189 11391328 7000 19.02 47.56 112 128 12781521
(14571699)
39.9647.56
(45.5653.12)
19.9823.78
(22.7826.56)
34-40
GeForce GTX
1050
(Notebook)[67][68]
Jan 3, 2017 1354 1493 21.66 54.16 1733
(1911)
54.16
(59.72)
27.08
(29.86)
53
GeForce GTX
1050 Ti Max-Q
(Notebook)[67][69]
Jan 3, 2018 GP107
(N17P-G1)
768:48:32 6 11511290 12901417 41.28 61.92 4 17671981
(19812176)
55.2461.92
(61.9268.02)
27.6230.96
(30.9634.01)
40-46
GeForce GTX
1050 Ti
(Notebook)[67][69]
Jan 3, 2017 1493 1620 47.78 71.66 2293
(2488)
71.66
(77.76)
35.83
(38.88)
64
GeForce GTX
1060 Max-Q
(Notebook)[67][70]
Jun 27, 2017 GP106
(N17E-G1)
16 4.4 200 1280:80:48 10 1.5 10631265 13411480 8000 60.72 101.2 3
6
192 192 27213238
(34323788)
85.04101.2
(107.3118.4)
42.5250.60
(53.6459.20)
60-70
GeForce GTX
1060
(Notebook)[67][70]
Aug 16, 2016 1404 1670 67.39 112.3 3594
(4275)
112.3
(133.6)
56.16
(66.80)
80
GeForce GTX
1070 Max-Q
(Notebook)[67][71]
Jun 27, 2017 GP104
(N17E-G2)
7.2 314 2048:128:64 16 2.0 11011215 12651379 77.76 155.5 8 256 256 45094977
(51815648)
140.9155.5
(161.9176.5)
70.4677.76
(80.9688.26)
80-90
GeForce GTX
1070
(Notebook)[67][71]
Aug 16, 2016 1442 1645 92.29 184.6 5906
(6738)
184.6
(210.6)
92.29
(105.3)
115
GeForce GTX
1080 Max-Q
(Notebook)[67][72]
Jun 27, 2017 GP104
(N17E-G3)
2560:160:64 20 11011290 12781458 10000 82.56 206.4 320 GDDR5X 56376605
(65437465)
176.2206.4
(204.5233.3)
88.08103.2
(102.2116.6)
90-110
GeForce GTX
1080
(Notebook)[67][72]
Aug 16, 2016 1556 1733 99.58 249.0 7967
(8873)
249.0
(277.3)
124.5
(138.6)
150
Close
  1. The number of streaming multiprocessors on the GPU.
  2. Pixel fillrate is calculated as the lowest of three numbers: number of ROPs multiplied by the base core clock speed, number of rasterizers multiplied by the number of fragments they can generate per rasterizer multiplied by the base core clock speed, and the number of streaming multiprocessors multiplied by the number of fragments per clock that they can output multiplied by the base clock rate.
  3. Texture fillrate is calculated as the number of TMUs multiplied by the base core clock speed.
  4. For calculating the processing power, see the Performance subsection of the Pascal architecture article.
  5. Lacks hardware video encoder and decoder

Discontinued support

Nvidia stopped releasing 32-bit drivers for 32-bit operating systems after driver 391.35 in March 2018.[73]

Nvidia announced that after release of the 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.[74] The GeForce 10 series is the last Nvidia GPU generation to support Windows 7/8.x or any 32-bit operating system; beginning with the Turing architecture, newer Nvidia GPUs now require a 64-bit operating system.

See also

References

Wikiwand in your browser!

Seamless Wikipedia browsing. On steroids.

Every time you click a link to Wikipedia, Wiktionary or Wikiquote in your browser's search results, it will show the modern Wikiwand interface.

Wikiwand extension is a five stars, simple, with minimum permission required to keep your browsing private, safe and transparent.