Report an Error

NVIDIA GeForce GTX 645 OEM

Graphics Processor
GK106
Cores
576
TMUs
48
ROPs
16
Memory Size
1024 MB
Memory Type
GDDR5
Bus Width
128 bit
The GeForce GTX 645 OEM was a graphics card by NVIDIA, launched in April 2013. Built on the 28 nm process, and based on the GK106 graphics processor, in its GK106-200-A1 variant, the card supports DirectX 12.0. The GK106 graphics processor is an average sized chip with a die area of 221 mm² and 2,540 million transistors. Unlike the fully unlocked GeForce GTX 660, which uses the same GPU but has all 960 shaders enabled, NVIDIA has disabled some shading units on the GeForce GTX 645 OEM to reach the product's target shader count. It features 576 shading units, 48 texture mapping units and 16 ROPs. NVIDIA has placed 1,024 MB GDDR5 memory on the card, which are connected using a 128-bit memory interface. The GPU is operating at a frequency of 824 MHz, memory is running at 1000 MHz.
Being a single-slot card, the NVIDIA GeForce GTX 645 OEM does not require any additional power connector, its power draw is rated at 65 W maximum. Display outputs include: 1x DVI, 1x HDMI, 1x DisplayPort. GeForce GTX 645 OEM is connected to the rest of the system using a PCI-Express 3.0 x16 interface. The card measures 147 mm in length, and features a single-slot cooling solution.

Graphics Processor

GPU Name
GK106
GPU Variant
GK106-200-A1
Architecture
Kepler
Foundry
TSMC
Process Size
28 nm
Transistors
2,540 million
Die Size
221 mm²

Graphics Card

Release Date
Apr 22nd, 2013
Generation
GeForce 600
Production
End-of-life
Bus Interface
PCIe 3.0 x16

Relative Performance

12%
GeForce 210
15%
GeForce 9400 GT
18%
Radeon HD 4550
23%
Radeon HD 5450
28%
Radeon HD 6450
29%
GeForce GT 520
36%
GeForce GT 220
47%
GeForce GT 430
48%
Radeon HD 5570
52%
Radeon HD 4670
55%
GeForce GT 440
56%
GeForce GT 240
64%
GeForce 9600 GT
68%
Radeon HD 5670
73%
GeForce GT 640
75%
GeForce 9800 GT
79%
Radeon HD 6670
81%
Radeon HD 4830
86%
Radeon HD 4770
91%
Radeon HD 4850
93%
GeForce GTS 250
96%
GeForce GTS 450
99%
Radeon HD 5750
100%
GeForce GTX 645 OEM
113%
Radeon HD 7750
114%
Radeon HD 5770
116%
GeForce GTX 260
118%
Radeon HD 4870
118%
Radeon Vega 8
119%
GeForce GTX 550 Ti
120%
GeForce GTX 650
127%
Radeon HD 5830
134%
Radeon HD 6790
136%
Radeon HD 4890
141%
GeForce GTX 460
142%
Radeon RX Vega 11
144%
GeForce GT 1030
145%
GeForce GTX 275
145%
Radeon HD 7770 GHz Edition
146%
GeForce GTX 280
148%
GeForce GTX 465
156%
Radeon HD 6850
159%
GeForce GTX 285
170%
Radeon HD 5850
173%
GeForce GTX 650 Ti
173%
Radeon HD 7790
181%
Radeon RX 550
181%
GeForce GTX 470
188%
Radeon HD 6870
200%
Radeon HD 5870
206%
GeForce GTX 560 Ti
208%
Radeon HD 4870 X2
211%
GeForce GTX 750 Ti
212%
Radeon HD 6950
220%
GeForce GTX 295
222%
GeForce GTX 650 Ti Boost
224%
Radeon HD 7850
228%
GeForce GTX 480
239%
Radeon HD 6970
243%
Radeon R7 265
247%
Radeon R7 370
247%
Radeon RX 560
248%
GeForce GTX 660
249%
GeForce GTX 570
269%
GeForce GTX 950
271%
Radeon HD 7870 GHz Edition
277%
Radeon R9 270X
281%
GeForce GTX 660 Ti
282%
GeForce GTX 580
284%
GeForce GTX 1050
285%
Radeon HD 7950
290%
Radeon RX 460
291%
Radeon HD 5970
299%
GeForce GTX 760
322%
GeForce GTX 670
330%
GeForce GTX 960
338%
Radeon R9 380
340%
Radeon R9 285
344%
Radeon HD 7970
347%
GeForce GTX 680
355%
GeForce GTX 1050 Ti
365%
GeForce GTX 770
373%
Radeon HD 6990
382%
Radeon R9 280X
388%
GeForce GTX 590
388%
Radeon HD 7970 GHz Edition
430%
GeForce GTX 780
473%
Radeon RX 470
479%
Radeon R9 290
511%
Radeon R9 390
514%
Radeon R9 290X
521%
GeForce GTX 970
528%
GeForce GTX TITAN
532%
Radeon RX 570
535%
GeForce GTX 780 Ti
543%
Radeon R9 390X
545%
Radeon HD 7990
548%
Radeon RX 480
552%
GeForce GTX 690
582%
Radeon RX 580
594%
GeForce GTX 980
602%
GeForce GTX 1060 6 GB
602%
Radeon R9 FURY
605%
Radeon RX 590
662%
Radeon R9 295X2
673%
Radeon R9 FURY X
685%
GeForce GTX 980 Ti
736%
GeForce GTX TITAN X
772%
GeForce GTX 1660 Ti
776%
GeForce GTX 1070
810%
Radeon RX Vega 56
879%
GeForce GTX 1070 Ti
901%
Radeon RX Vega 64
901%
GeForce RTX 2060
924%
GeForce GTX 1080
1061%
Radeon VII
1061%
GeForce RTX 2070
1112%
TITAN X Pascal
1141%
GeForce GTX 1080 Ti
1232%
GeForce RTX 2080
1426%
GeForce RTX 2080 Ti
Based on TPU review data: "Performance Summary" at 1920x1080
Performance estimated based on architecture, shader count and clocks.

Clock Speeds

GPU Clock
824 MHz
Memory Clock
1000 MHz
4000 MHz effective

Memory

Memory Size
1024 MB
Memory Type
GDDR5
Memory Bus
128 bit
Bandwidth
64.00 GB/s

Render Config

Shading Units
576
TMUs
48
ROPs
16
SMX Count
3
L1 Cache
16 KB (per SMX)
L2 Cache
256 KB

Theoretical Performance

Pixel Rate
9.888 GPixel/s
Texture Rate
39.55 GTexel/s
FP32 (float) performance
949.2 GFLOPS
FP64 (double) performance
39.55 GFLOPS (1:24)

Board Design

Slot Width
Single-slot
Length
5.78 inches
147 mm
TDP
65 W
Outputs
1x DVI
1x HDMI
1x DisplayPort
Power Connectors
None
Board Number
P2010 SKU 20

Graphics Features

DirectX
12.0 (11_0)
OpenGL
4.6
OpenCL
1.2
Vulkan
1.1.103
CUDA
3.0
Shader Model
5.1

GK106 GPU Notes

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