Report an Error

NVIDIA GRID K2

Graphics Processor
GK104 x2
Cores
1536 x2
TMUs
128 x2
ROPs
32 x2
Memory Size
4 GB x2
Memory Type
GDDR5
Bus Width
256 bit x2
The GRID K2 was an enthusiast-class professional graphics card by NVIDIA, launched in May 2013. Built on the 28 nm process, and based on the GK104 graphics processor, in its GK104-895-A2 variant, the card supports DirectX 12. The GK104 graphics processor is an average sized chip with a die area of 294 mm² and 3,540 million transistors. GRID K2 combines two graphics processors to increase performance. It features 1536 shading units, 128 texture mapping units, and 32 ROPs, per GPU. NVIDIA has paired 8 GB GDDR5 memory with the GRID K2, which are connected using a 256-bit memory interface per GPU (each GPU manages 4,096 MB). The GPU is operating at a frequency of 745 MHz, memory is running at 1250 MHz (5 Gbps effective).
Being a dual-slot card, the NVIDIA GRID K2 draws power from 1x 8-pin power connector, with power draw rated at 225 W maximum. This device has no display connectivity, as it is not designed to have monitors connected to it. GRID K2 is connected to the rest of the system using a PCI-Express 3.0 x16 interface. The card measures 267 mm in length, and features a dual-slot cooling solution. Its price at launch was 5199 US Dollars.

Graphics Processor

GPU Name
GK104
GPU Variant
GK104-895-A2
Architecture
Kepler
Foundry
TSMC
Process Size
28 nm
Transistors
3,540 million
Die Size
294 mm²

Graphics Card

Release Date
May 11th, 2013
Generation
GRID K2
Production
End-of-life
Launch Price
5,199 USD
Bus Interface
PCIe 3.0 x16

Relative Performance

3%
GeForce 210
4%
GeForce 9400 GT
5%
Radeon HD 4550
6%
Radeon HD 5450
7%
Radeon HD 6450
7%
GeForce GT 520
9%
GeForce GT 220
12%
GeForce GT 430
12%
Radeon HD 5570
13%
Radeon HD 4670
14%
GeForce GT 440
14%
GeForce GT 240
16%
GeForce 9600 GT
17%
Radeon HD 5670
18%
GeForce GT 640
19%
GeForce 9800 GT
19%
Radeon HD 6670
20%
Radeon HD 4830
21%
Radeon HD 4770
23%
Radeon HD 4850
23%
GeForce GTS 250
24%
GeForce GTS 450
25%
Radeon HD 5750
28%
Radeon HD 7750
28%
Radeon HD 5770
29%
GeForce GTX 260
29%
Radeon HD 4870
29%
Radeon Vega 8
30%
GeForce GTX 550 Ti
30%
GeForce GTX 650
32%
Radeon HD 5830
33%
Radeon HD 6790
34%
Radeon HD 4890
35%
GeForce GTX 460
35%
Radeon RX Vega 11
36%
GeForce GT 1030
36%
GeForce GTX 275
36%
Radeon HD 7770 GHz Edition
36%
GeForce GTX 280
37%
GeForce GTX 465
39%
Radeon HD 6850
39%
GeForce GTX 285
42%
Radeon HD 5850
43%
GeForce GTX 650 Ti
43%
Radeon HD 7790
45%
Radeon RX 550
45%
GeForce GTX 470
46%
Radeon HD 6870
50%
Radeon HD 5870
51%
GeForce GTX 560 Ti
51%
Radeon HD 4870 X2
52%
GeForce GTX 750 Ti
53%
Radeon HD 6950
54%
GeForce GTX 295
55%
GeForce GTX 650 Ti Boost
55%
Radeon HD 7850
56%
GeForce GTX 480
59%
Radeon HD 6970
60%
Radeon R7 265
61%
Radeon RX 460
61%
Radeon R7 370
61%
GeForce GTX 660
62%
GeForce GTX 570
63%
Radeon RX 560
67%
GeForce GTX 950
67%
Radeon HD 7870 GHz Edition
69%
Radeon R9 270X
70%
GeForce GTX 660 Ti
70%
GeForce GTX 580
70%
GeForce GTX 1050
71%
Radeon HD 7950
72%
Radeon HD 5970
74%
GeForce GTX 760
80%
GeForce GTX 670
82%
GeForce GTX 960
84%
Radeon R9 380
84%
Radeon R9 285
85%
Radeon HD 7970
86%
GeForce GTX 680
88%
GeForce GTX 1050 Ti
90%
GeForce GTX 770
92%
Radeon HD 6990
95%
Radeon R9 280X
96%
GeForce GTX 590
96%
Radeon HD 7970 GHz Edition
100%
GRID K2
106%
GeForce GTX 780
110%
GeForce GTX 1650
117%
Radeon RX 470
119%
Radeon R9 290
125%
Radeon RX 570
127%
Radeon R9 390
127%
Radeon R9 290X
129%
GeForce GTX 970
131%
GeForce GTX TITAN
133%
GeForce GTX 780 Ti
134%
Radeon R9 390X
135%
Radeon HD 7990
136%
Radeon RX 480
137%
GeForce GTX 690
140%
GeForce GTX 1060 6 GB
142%
Radeon RX 5500 OEM
144%
Radeon RX 580
145%
Radeon RX 5500 XT
147%
GeForce GTX 980
148%
GeForce GTX 1650 SUPER
149%
Radeon R9 FURY
156%
Radeon RX 590
164%
Radeon R9 295X2
165%
GeForce GTX 1660
167%
Radeon R9 FURY X
170%
GeForce GTX 980 Ti
174%
GeForce GTX TITAN X
185%
GeForce GTX 1660 SUPER
189%
GeForce GTX 1660 Ti
189%
GeForce GTX 1070
201%
Radeon RX Vega 56
214%
GeForce GTX 1070 Ti
220%
Radeon RX Vega 64
223%
GeForce RTX 2060
223%
GeForce GTX 1080
232%
Radeon RX 5700
250%
GeForce RTX 2060 SUPER
259%
GeForce RTX 2070
266%
Radeon RX 5700 XT
269%
Radeon VII
276%
TITAN X Pascal
284%
GeForce GTX 1080 Ti
287%
GeForce RTX 2070 SUPER
305%
GeForce RTX 2080
324%
GeForce RTX 2080 SUPER
357%
GeForce RTX 2080 Ti
471%
GeForce RTX 3080
519%
GeForce RTX 3090
Based on TPU review data: "Performance Summary" at 1920x1080, 4K for 2080 Ti and faster.
Performance estimated based on architecture, shader count and clocks.

Clock Speeds

GPU Clock
745 MHz
Memory Clock
1250 MHz
5 Gbps effective

Memory

Memory Size
4 GB
Memory Type
GDDR5
Memory Bus
256 bit
Bandwidth
160.0 GB/s

Render Config

Shading Units
1536
TMUs
128
ROPs
32
SMX Count
8
L1 Cache
16 KB (per SMX)
L2 Cache
512 KB

Theoretical Performance

Pixel Rate
23.84 GPixel/s
Texture Rate
95.36 GTexel/s
FP32 (float) performance
2.289 TFLOPS
FP64 (double) performance
95.36 GFLOPS (1:24)

Board Design

Slot Width
Dual-slot
Length
267 mm
10.5 inches
TDP
225 W
Suggested PSU
550 W
Outputs
No outputs
Power Connectors
1x 8-pin
Board Number
P2055

Graphics Features

DirectX
12 (11_0)
OpenGL
4.6
OpenCL
1.2
Vulkan
1.1
CUDA
3.0
Shader Model
5.1

GK104 GPU Notes

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