Report an Error

NVIDIA GeForce GTX 670MX

Graphics Processor
GK104
Cores
960
TMUs
80
ROPs
24
Memory Size
3 GB
Memory Type
GDDR5
Bus Width
192 bit
GPU Chip
GPU
The GeForce GTX 670MX was a mobile graphics chip by NVIDIA, launched on October 1st, 2012. Built on the 28 nm process, and based on the GK104 graphics processor, in its N13E-GR-A2 variant, the chip supports DirectX 12. Even though it supports DirectX 12, the feature level is only 11_0, which can be problematic with newer DirectX 12 titles. The GK104 graphics processor is an average sized chip with a die area of 294 mm² and 3,540 million transistors. Unlike the fully unlocked GeForce GTX 680, which uses the same GPU but has all 1536 shaders enabled, NVIDIA has disabled some shading units on the GeForce GTX 670MX to reach the product's target shader count. It features 960 shading units, 80 texture mapping units, and 24 ROPs. NVIDIA has paired 3,072 MB GDDR5 memory with the GeForce GTX 670MX, which are connected using a 192-bit memory interface. The GPU is operating at a frequency of 601 MHz, which can be boosted up to 601 MHz, memory is running at 700 MHz (2.8 Gbps effective).
Its power draw is rated at 75 W maximum. This device has no display connectivity, as it is not designed to have monitors connected to it. Rather it is intended for use in laptop/notebooks and will use the output of the host mobile device. GeForce GTX 670MX is connected to the rest of the system using a PCI-Express 3.0 x16 interface.

Graphics Processor

GPU Name
GK104
GPU Variant
N13E-GR-A2
Architecture
Kepler
Foundry
TSMC
Process Size
28 nm
Transistors
3,540 million
Density
12.0M / mm²
Die Size
294 mm²
Chip Package
BGA-1745

Mobile Graphics

Release Date
Oct 1st, 2012
Generation
GeForce 600M
Predecessor
GeForce 500M
Successor
GeForce 700M
Production
End-of-life
Bus Interface
PCIe 3.0 x16

Relative Performance

GeForce RTX 5090
3661%
GeForce RTX 4090
2792%
GeForce RTX 5080
2412%
GeForce RTX 4080 SUPER
2154%
GeForce RTX 4080
2136%
Radeon RX 7900 XTX
2134%
GeForce RTX 5070 Ti
2104%
Radeon RX 9070 XT
2010%
Radeon RX 7900 XT
1838%
GeForce RTX 3090 Ti
1804%
GeForce RTX 4070 Ti SUPER
1803%
Radeon RX 9070
1801%
GeForce RTX 4070 Ti
1650%
GeForce RTX 5070
1633%
GeForce RTX 3090
1609%
GeForce RTX 3080 Ti
1588%
GeForce RTX 4070 SUPER
1537%
Radeon RX 7900 GRE
1514%
Radeon RX 6950 XT
1489%
GeForce RTX 4070
1468%
Radeon RX 6900 XT
1454%
GeForce RTX 3080
1431%
Radeon RX 7800 XT
1426%
Radeon RX 6800 XT
1376%
GeForce RTX 5060 Ti 16 GB
1277%
GeForce RTX 5060 Ti 8 GB
1269%
Radeon RX 7700 XT
1251%
Radeon RX 9060 XT 16 GB
1212%
GeForce RTX 3070 Ti
1210%
Radeon RX 6800
1170%
Radeon RX 9060 XT 8 GB
1144%
GeForce RTX 3070
1132%
GeForce RTX 2080 Ti
1123%
GeForce RTX 4060 Ti 8 GB
1118%
GeForce RTX 5060
1100%
Radeon RX 6750 XT
1074%
GeForce RTX 3060 Ti
991%
Radeon RX 6700 XT
973%
GeForce RTX 2080 SUPER
904%
Radeon RX 7600 XT
897%
GeForce RTX 4060
894%
Arc B580
892%
GeForce RTX 5050
870%
GeForce RTX 2080
863%
Radeon RX 6650 XT
844%
Radeon RX 7600
843%
GeForce RTX 2070 SUPER
802%
GeForce GTX 1080 Ti
793%
Arc A770
793%
GeForce RTX 3060 12 GB
764%
Radeon RX 6600 XT
753%
Radeon VII
751%
Arc A750
736%
Radeon RX 5700 XT
719%
GeForce RTX 2070
709%
Radeon RX 6600
677%
Arc A580
658%
GeForce RTX 2060 SUPER
657%
Radeon RX Vega 64
629%
GeForce RTX 2060
609%
Radeon RX 5700
606%
GeForce GTX 1080
594%
GeForce GTX 1070 Ti
583%
Radeon RX 5600 XT
578%
Radeon RX Vega 56
549%
GeForce GTX 1070
529%
GeForce GTX 1660 SUPER
501%
GeForce GTX 1660 Ti
500%
GeForce GTX 980 Ti
474%
GeForce RTX 3050 8 GB
472%
Radeon R9 FURY X
466%
GeForce GTX 1660
461%
Radeon RX 590
435%
Radeon R9 FURY
424%
GeForce GTX 980
419%
GeForce GTX 1650 SUPER
415%
Radeon RX 6500 XT
410%
Radeon RX 5500 XT
406%
Radeon RX 580
403%
GeForce GTX 1060 6 GB
393%
Radeon R9 390X
387%
GeForce GTX 690
382%
Radeon RX 480
379%
Radeon HD 7990
377%
GeForce GTX 780 Ti
370%
GeForce GTX 970
361%
Radeon R9 290X
360%
Radeon R9 390
354%
Radeon RX 570
350%
Radeon R9 290
332%
Radeon RX 470
327%
GeForce GTX 1650
308%
Radeon RX 6400
302%
Arc A380
298%
GeForce GTX 780
297%
Radeon HD 7970 GHz Edition
269%
GeForce GTX 590
268%
Radeon R9 280X
265%
Radeon HD 6990
258%
GeForce GTX 1050 Ti
246%
GeForce GTX 680
241%
Radeon R9 380
238%
Radeon HD 7970
238%
GeForce GTX 770
237%
Radeon R9 285
236%
GeForce GTX 960
228%
GeForce GTX 670
223%
Radeon HD 5970
201%
GeForce GTX 1630
201%
Radeon HD 7950
197%
GeForce GTX 1050
197%
GeForce GTX 760
196%
GeForce GTX 580
195%
GeForce GTX 660 Ti
195%
Radeon R9 270X
192%
Radeon HD 7870 GHz Edition
188%
GeForce GTX 950
186%
Radeon RX 560
175%
GeForce GTX 570
172%
GeForce GTX 660
172%
Radeon RX 460
171%
Radeon R7 370
171%
Radeon HD 6970
165%
GeForce GTX 480
158%
Radeon HD 7850
155%
GeForce GTX 295
152%
Radeon HD 6950
147%
GeForce GTX 750 Ti
146%
Radeon HD 4870 X2
144%
GeForce GTX 560 Ti
143%
Radeon HD 5870
139%
Radeon HD 6870
130%
GeForce GTX 470
125%
GeForce GTX 650 Ti
120%
Radeon HD 5850
118%
GeForce GTX 285
110%
Radeon HD 6850
108%
GeForce GTX 280
101%
Radeon HD 7770 GHz Edition
101%
GeForce GTX 275
100%
GeForce GTX 670MX
100%
GeForce GT 1030
99%
GeForce GTX 460
98%
Radeon HD 4890
94%
Radeon HD 6790
93%
Radeon HD 5830
88%
GeForce GTX 650
83%
GeForce GTX 550 Ti
83%
Radeon HD 4870
82%
GeForce GTX 260
81%
Radeon HD 5770
79%
Radeon HD 7750
79%
Radeon HD 5750
69%
GeForce GTS 450
66%
GeForce GTS 250
65%
Radeon HD 4850
63%
Radeon HD 4770
60%
Radeon HD 4830
56%
Radeon HD 6670
54%
GeForce 9800 GT
52%
GeForce GT 640
50%
Radeon HD 5670
47%
GeForce 9600 GT
44%
GeForce GT 240
39%
GeForce GT 440
38%
Radeon HD 4670
36%
Radeon HD 5570
33%
GeForce GT 430
33%
GeForce GT 220
25%
GeForce GT 520
20%
Radeon HD 6450
19%
Radeon HD 5450
16%
Radeon HD 4550
13%
GeForce 9400 GT
10%
GeForce 210
8%
Based on TPU review data: "Performance Summary" at 1920x1080, 4K for RTX 3080 and faster.
Performance estimated based on architecture, shader count and clocks.

Clock Speeds

Base Clock
601 MHz
Boost Clock
601 MHz
Memory Clock
700 MHz
2.8 Gbps effective

Memory

Memory Size
3 GB
Memory Type
GDDR5
Memory Bus
192 bit
Bandwidth
67.20 GB/s

Render Config

Shading Units
960
TMUs
80
ROPs
24
SMX Count
5
L1 Cache
16 KB (per SMX)
L2 Cache
384 KB

Theoretical Performance

Pixel Rate
12.02 GPixel/s
Texture Rate
48.08 GTexel/s
FP32 (float)
1,153.9 GFLOPS
FP64 (double)
48.08 GFLOPS (1:24)

Board Design

TDP
75 W
Outputs
Portable Device Dependent
Power Connectors
None

Graphics Features

DirectX
12 (11_0)
OpenGL
4.6
OpenCL
3.0
Vulkan
1.2.175
CUDA
3.0
Shader Model
6.5 (5.1)

GK104 GPU Notes

NVENC: 1st Gen
NVDEC: 1st Gen
PureVideo HD: VP5
VDPAU: Feature Set D

L1 Cache is configurable from 16 KB up to 48 KB per SMX

Latest Drivers:
Windows XP / Server 2003 x64:
Quadro Release R319 U2 (321.01)

Windows Vista:
GeForce Release 365.19
Quadro Release R346 U7 (348.40) / R352 BETA (352.86)

Windows 7 / 8 / 8.1 / 10 / 11 (x32 / x64):
GeForce Release 391.35 / 475.06
Quadro Release R390 U9 (392.37) / R470 U16 (474.82)
Data Center Release 427.11
Tesla Release 441.22

Windows 10 / 11 x64:
GeForce Release 475.06
Quadro Release R470 U16 (474.82)
Data Center Release 474.82

Devices based on this design (2)

Name GPU Clock Boost Clock Memory Clock Other Changes
601 MHz 601 MHz 700 MHz
615 MHz 615 MHz 700 MHz
Aug 1st, 2025 07:54 CDT change timezone

New Forum Posts

Popular Reviews

TPU on YouTube

Controversial News Posts