Select any two GPUs for comparison
VS

Gaming Performance Comparison

Recommended System Requirements
Game GeForce GTX 650 Gigabyte OC 1GB Edition Radeon HD 4870
Cyberpunk 2077 372% 371%
Assassins Creed: Valhalla 332% 331%
Call of Duty: Black Ops Cold War 318% 317%
Watch Dogs Legion 356% 354%
FIFA 21 138% 138%
Microsoft Flight Simulator 318% 317%
Godfall 667% 665%
Grand Theft Auto VI 565% 563%
World of Warcraft: Shadowlands 577% 575%
Genshin Impact 372% 371%

In terms of overall gaming performance, the graphical capabilities of the AMD Radeon HD 4870 are marginally better than the Nvidia GeForce GTX 650 Gigabyte OC 1GB Edition.

The GTX 650 has a 360 MHz higher core clock speed but 8 fewer Texture Mapping Units than the Radeon HD 4870. The lower TMU count doesn't matter, though, as altogether the GTX 650 manages to provide 5.5 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 650 has a 360 MHz higher core clock speed than the Radeon HD 4870 and the same number of Render Output Units. This results in the GTX 650 providing 5.8 GPixel/s better pixeling performance. However, both GPUs support DirectX 9 or above, and pixeling performance is only really relevant when comparing older cards.

The GTX 650 was released over three years more recently than the Radeon HD 4870, and so the GTX 650 is likely to have far better driver support, meaning it will be much more optimized and ultimately superior to the Radeon HD 4870 when running the latest games.

The GTX 650 has 512 MB more video memory than the Radeon HD 4870, so is likely to be much better at displaying game textures at higher resolutions. However, the overall memory performance is about the same.

The Radeon HD 4870 has 35.2 GB/sec greater memory bandwidth than the GTX 650, which means that the memory performance of the Radeon HD 4870 is noticeably better than the GTX 650.

The GeForce GTX 650 Gigabyte OC 1GB Edition has 384 Shader Processing Units and the Radeon HD 4870 has 800. However, the actual shader performance of the GTX 650 is 426 and the actual shader performance of the Radeon HD 4870 is 390. The GTX 650 having 36 better shader performance is not particularly notable, as altogether the Radeon HD 4870 performs better when taking into account other relevant data.

The GTX 650 transistor size technology is 27 nm (nanometers) smaller than the Radeon HD 4870. This means that the GTX 650 is expected to run slightly cooler and achieve higher clock frequencies than the Radeon HD 4870. While they exhibit similar graphical performance, the GTX 650 should consume less power than the Radeon HD 4870.

The Radeon HD 4870 requires 150 Watts to run but there is no entry for the GTX 650. We would recommend a PSU with at least 400 Watts for the GTX 650 and a PSU with at least 500 Watts for the Radeon HD 4870. The Radeon HD 4870 has been recommended a PSU with 100 Watts more than the GTX 650. The difference is significant enough that the Radeon HD 4870 may have an adverse affect on your yearly electricity bills in comparison to the GTX 650.

Game FPS Benchmarks On Ultra

GPU Architecture

Core Speed1110 MHzvs750 MHz
Boost Clock-vs-
ArchitectureKepler GK107-450-A2RV770 PRO
OC Potential Poor vs Fair
Driver Support Good vs -
Release Date01 Sep 2012vs01 Jun 2008
GPU LinkGD LinkGD Link
Approved
Comparison

Resolution Performance

1366x768 - vs green tick
8.5
1600x900
5.9
vs green tick
6.4
1920x1080
4.9
green tick vs green tick
4.9
2560x1440
3.1
green tick vs
2.8
3840x2160
2.1
green tick vs
1.8

GPU Memory

Memory1024 MBvs512 MB
Memory Speed1250 MHzvs900 MHz
Memory Bus128 Bitvs256 Bit
Memory TypeGDDR5vsGDDR5
Memory Bandwidth80GB/secvs115.2GB/sec
L2 Cache 256 KB green tick vs -
Delta Color Compression no vs no
Memory Performance 0% green tick vs green tick 0%
Comparison

GPU Display

Shader Processing Units384vs800
Actual Shader Performance21%vs19%
Technology28nmvs55nm
Texture Mapping Units32vs40
Texture Rate35.5 GTexel/svs30 GTexel/s
Render Output Units16vs16
Pixel Rate17.8 GPixel/svs12 GPixel/s
Comparison

GPU Outputs

Max Digital Resolution (WxH)4096x2160vs2560x1600
VGA Connections1vs0
DVI Connections2vs2
HDMI Connections1vs0
DisplayPort Connections-vs-
Comparison

GPU Power Requirements

Max Power-150 Watts
Recommended PSU400 Watts & 33 Ampsvs500 Watts & 32 Amps

GPU Features

DirectX12.0vs10.1
Shader Model5.0vs4.1
Open GL4.5vs3.2
Open CL-vs-
Notebook GPUnono
SLI/Crossfirenovsyes
Dedicatedyesvsyes
Comparison

GPU Supporting Hardware

Recommended ProcessorIntel Core i3-3220 3.3GHzvsIntel Core 2 Duo E8600 3.33GHz
Recommended RAM8 GBvs6 GB
Maximum Recommended Gaming Resolution1600x900vs1600x900

Gaming Performance Value

Performance Value

GPU Mini Review

Mini ReviewGeForce GTX 650 Gigabyte OC 1GB Edition is a special edition of the fast-middle-class GeForce GTX 650 1GB.
This special edition comes overclocked out of the box to 1110MHz and features a custom-designed 10 cm cooler.
Benchmarks indicate a 4% performance boost when compared to the reference card and so this GPU is suitable for medium to high settings on most demanding games but not with extreme resolutions.
Radeon HD 4870 is a high-end GPU based on the 55nm, third unified shader architecture, R700.
It's based on the RV770 PRO Core and offers 800 Shader Processing Units, 40 TMUs and 16 ROPs a 256-bit memory interface of fast GDDR5. The central unit runs at 750MHz and the memory clock runs at up to 900MHz.
Compared to Radeon HD 4890, it offers slightly less bandwidth (thanks to the higher operating memory clock), the central unit is lower clocked and features twice as less video memory. However, it only consumes up to 150W (40W less). The performance difference between is not very significant and so Radeon HD 4870 might be the best choice, over 4890 since it can be easily overclocked.
Most of today's modern games should run fluently at high settings but with modest resolutions. DirectX 11 games aren't supported.
Recommended CPU
Possible GPU Upgrades
-
GPU Variants
-