Select any two GPUs for comparison
VS

Gaming Performance Comparison

Recommended System Requirements
Game GeForce GTX 660 MSI Hawk 2GB Edition GeForce GTX 580
Red Dead Redemption 2 102% 98%
Call of Duty: Modern Warfare 102% 98%
Cyberpunk 2077 178% 172%
Fortnite: Chapter 2 7% 9%
Need For Speed Heat 102% 98%
The Outer Worlds 68% 65%
Microsoft Flight Simulator 178% 172%
Zombieland: Double Tap - Road Trip 89% 85%
Borderlands 3 114% 109%
FIFA 20 8% 6%

In terms of overall gaming performance, the graphical capabilities of the Nvidia GeForce GTX 580 are marginally better than the Nvidia GeForce GTX 660 MSI Hawk 2GB Edition.

The GTX 660 has a 313 MHz higher core clock speed and 16 more Texture Mapping Units than the GeForce GTX 580. This results in the GTX 660 providing 37.4 GTexel/s better texturing performance. This still holds weight but shader performance is generally more relevant, particularly since both of these GPUs support at least DirectX 10.

The GTX 660 has a 313 MHz higher core clock speed than the GeForce GTX 580, but the GeForce GTX 580 has 24 more Render Output Units than the GTX 660. As a result, the GeForce GTX 580 exhibits a 11.1 GPixel/s better Pixel Fill Rate than the GTX 660. However, both GPUs support DirectX 9 or above, and pixeling performance is only really relevant when comparing older cards.

The GTX 660 was released over a year more recently than the GeForce GTX 580, and so the GTX 660 is likely to have better driver support, meaning it will be more optimized for running the latest games when compared to the GeForce GTX 580.

Both GPUs exhibit very powerful performance, so it probably isn't worth upgrading from one to the other, as both are capable of running even the most demanding games at the highest settings.

The GTX 660 has 512 MB more video memory than the GeForce GTX 580, so is likely to be much better at displaying game textures at higher resolutions. However, the overall memory performance is about the same.

The GeForce GTX 580 has 48.2 GB/sec greater memory bandwidth than the GTX 660, which means that the memory performance of the GeForce GTX 580 is noticeably better than the GTX 660.

The GeForce GTX 660 MSI Hawk 2GB Edition has 960 Shader Processing Units and the GeForce GTX 580 has 512. However, the actual shader performance of the GTX 660 is 1104 and the actual shader performance of the GeForce GTX 580 is 791. The GTX 660 having 313 better shader performance is not particularly notable, as altogether the GeForce GTX 580 performs better when taking into account other relevant data.

The GTX 660 transistor size technology is 12 nm (nanometers) smaller than the GeForce GTX 580. This means that the GTX 660 is expected to run slightly cooler and achieve higher clock frequencies than the GeForce GTX 580. While they exhibit similar graphical performance, the GTX 660 should consume less power than the GeForce GTX 580.

The GeForce GTX 580 requires 244 Watts to run but there is no entry for the GTX 660. We would recommend a PSU with at least 600 Watts for the GeForce GTX 580, but we do not have a recommended PSU wattage for the GTX 660.

GPU Architecture

Core Speed1085 MHzvs772 MHz
Boost Clock1150 MHzvs-
ArchitectureKepler GK106-400-A1Fermi GF110-375-A1
OC Potential Poor vs Poor
Driver Support Good vs Poor
Release Date01 Oct 2012vs09 Nov 2010
GPU LinkGD LinkGD Link
Approved
Comparison

Resolution Performance

1366x768 - vs green tick
10
1600x900
9.2
vs green tick
9.3
1920x1080
7.3
green tick vs green tick
7.3
2560x1440
5.3
green tick vs green tick
5.3
3840x2160
3.6
green tick vs -

GPU Memory

Memory2048 MBvs1536 MB
Memory Speed1502 MHzvs1002 MHz
Memory Bus192 Bitvs384 Bit
Memory TypeGDDR5vsGDDR5
Memory Bandwidth144.2GB/secvs192.4GB/sec
L2 Cache 384 KB vs green tick 768 KB
Delta Color Compression no vs no
Memory Performance 0% green tick vs green tick 0%
Comparison

GPU Display

Shader Processing Units960vs512
Actual Shader Performance53%vs38%
Technology28nmvs40nm
Texture Mapping Units80vs64
Texture Rate86.8 GTexel/svs49.4 GTexel/s
Render Output Units24vs48
Pixel Rate26 GPixel/svs37.1 GPixel/s
Comparison

GPU Outputs

Max Digital Resolution (WxH)4096x2160vs2560x1600
VGA Connections0vs0
DVI Connections2vs2
HDMI Connections1vs1
DisplayPort Connections-vs0
Comparison

GPU Power Requirements

Max Power-244 Watts
Recommended PSU-600 Watts & 42 Amps

GPU Features

DirectX12.0vs12.0
Shader Model5.0vs5.0
Open GL4.5vs4.5
Open CL-vs-
Notebook GPUnono
SLI/Crossfireyesvsyes
Dedicatedyesvsyes
Comparison

GPU Supporting Hardware

Recommended ProcessorIntel Core i5-3470 3.2GHzvsIntel Core i7-880 Quad 3.06GHz
Recommended RAM8 GBvs8 GB
Maximum Recommended Gaming Resolution1920x1080vs1920x1080

Gaming Performance Value

Performance Value

GPU Mini Review

Mini ReviewGeForce GTX 660 MSI Hawk 2GB Edition a special edition of GeForce GTX 660.
The new features include the Twin Frozr III Thermal Design cooling solution which, according to MSI, generates 20% more airflow than traditional fan design.
In terms of overclocking, the base-clock was increased from 980MHz to 1085MHz and the boost clock can now reach 1150MHz, instead of 1033MHz. The memory clock was left untouched.
The overclocking is therefore extreme and the performance boost is around 8% when compared to the standard GeForce GTX 660.
Overview
GeForce GTX 580 is a High-End Graphics Card based on the Fermi Architecture.

Architecture
The Fermi Architecture was first manufactured with a 40nm technology and uses a technique known as Hot Clocking: The Shaders are clocked twice as fast as the Central Unit. While this leads to a reasonable performance boost, it causes enormous amounts of energy dissipation, leading, ultimately, to a significantly higher operating temperature.
Fermi is also the first GPU architecture with fully cached memory access which increases memory performance.

GPU
It equips a GPU Codenamed GF110-375-A1 which has 16 Stream Multiprocessors activated and thus offers 512 Shader Processing Units, 64 TMUs and 48 ROPs. The Central Unit is clocked at 772MHz.

Memory
The GPU accesses a 1.5GB frame buffer of fast GDDR5, through a 384-bit memory interface. The size of the frame buffer is adequate. The Memory Clock Operates at 1002Hz.

Power Consumption
With a rated board TDP of 244W, it requires at least a 600W PSU with one available 6-pin and one available 8-pin connectors.

Performance
GeForce GTX 580 competes with Radeon HD 6970 and proves to be around 10% faster.

System Suggestions
GeForce GTX 580 is best suited for resolutions up to and including 1920x1080. We recommend a Decent Processor and 8GB of RAM for Optimal Performance.
Recommended CPU
Possible GPU Upgrades
GPU Variants
-

Title

Body