GeForce GT 645M 

Released October, 2012
  • 709 MHz
  • 2 GB DDR3
6.3 Out of 10

It is based on the Fermi GF116, which features 144 CUDA cores, 1.5 GB or 3 GB GDDR3 192-bit memory interface, 720 MHz core clock speed, and 891 MHz memory clock speed.
by Doug-Crowthers (Apr, 2012)
We're not sure exactly how this is done, but it's essentially a GPU version of the Turbo Boost and Turbo Core technologies we've seen from Intel and AMD CPUs for some time.
by Dustin-Sklavos (Mar, 2012)

GPUBoss Review Our evaluation of the GeForce GT 645M among Laptop GPUs under 35W TDP

Gaming

GeForce GT 645M
5.5
Battlefield 3, Battlefield 4, Bioshock Infinite and 21 more

Graphics

GeForce GT 645M
7.2
T-Rex, Manhattan and Cloud Gate Factor

Computing

GeForce GT 645M
6.6
Face Detection, Ocean Surface Simulation and 3 more

Performance per Watt

GeForce GT 645M
5.9
Battlefield 3, Battlefield 4, Bioshock Infinite and 30 more

Value

GeForce GT 645M
n.d.
Battlefield 3, Battlefield 4, Bioshock Infinite and 30 more

Noise and Power

GeForce GT 645M
5.2
TDP, Idle Power Consumption, Load Power Consumption and 2 more

6.3

Overall Score

GeForce GT 645M
6.3

Benchmarks Real world tests of the GeForce GT 645M

Bitcoin mining Data courtesy CompuBench

GeForce GT 645M
19.12 mHash/s
GeForce GTX 1060
575.19 mHash/s
GeForce GTX 1050 Ti
328.03 mHash/s

Face detection Data courtesy CompuBench

GeForce GT 645M
8.49 mPixels/s
GeForce GTX 1060
159.16 mPixels/s
GeForce GTX 1050 Ti
93.18 mPixels/s

T-Rex (GFXBench 3.0) Data courtesy CompuBench

GeForce GT 645M
1,859.4
GeForce GTX 1060
3,359.74

Manhattan (GFXBench 3.0) Data courtesy CompuBench

GeForce GT 645M
1,745.92
GeForce GTX 1060
3,718.56

Fire Strike Factor Data courtesy FutureMark

GeForce GT 645M
14.86

Sky Diver Factor Data courtesy FutureMark

GeForce GT 645M
118.03

Bioshock Infinite

GeForce GT 645M
14.3

Crysis 3

GeForce GT 645M
7

In The News From around the web

20
Jul

Report: NVIDIA GeForce GTX 1070M and 1060M Specs Leaked

by Sebastian Peak |
www.pcper.com
The report comes via BenchLife, with the screenshot of GPU-Z showing the higher CUDA core count (though VideoCardz mentions the TMU count should be 128).
19
Mar

NVIDIA Announces Quadro M6000 & Quadro VCA (2015)

by Ryan Smith |
www.anandtech.com
However the company is also pushing the use of Physically Based Rendering (PBR), a compute-intensive rendering solution that uses far more accurate rendering algorithms to accurately model the physical characteristic of a material, in essence properly capturing how light will interact with that material and reflect off of it rather than using a rough approximation.
However as FP64 compute is not widely used in graphics, this is not something NVIDIA believes will be an issue.
So let’s talk about the GPUs themselves, NVIDIA’s first tier chips would be MXM design which means they would be interchangeable, upgradable and even offer SLI support on some notebook designs.
Next up, we have the entry level models which will feature the onboard design which mean that these GPUs will be permanently soldered to the mainboard hence no upgrading can be possible.
AMD’s Kaveri APUs would feature the latest core architecture along with enhanced GCN graphics on a single die package.
This means that higher frequency DDR3 memory just might come in handy with Kaveri APUs to unleash the full potential of the graphics core.

Specifications Full list of technical specs

gpu

GPU brand Nvidia
GPU name GK107
Market Laptop
Clock speed 709 MHz
Is dual GPU No
Reference card None

raw performance

Shading units 384
Texture mapping units 32
Render output processors 16
Pixel rate 5.67 GPixel/s
Texture rate 22.69 GTexel/s
Floating-point performance 544.5 GFLOPS

memory

Memory clock speed 900 MHz
Effective memory clock speed 1,800 MHz
Memory bus 128 bit
Memory 2,048 MB
Memory type DDR3
Memory bandwidth 28.8 GB/s

noise and power

TDP 32W
Report a correction

Comments

comments powered by Disqus