RTX 5000 Ada Generation vs GeForce RTX 4070

#ad 
Buy on Amazon
VS

Aggregate performance score

We've compared GeForce RTX 4070 with RTX 5000 Ada Generation, including specs and performance data.

RTX 4070
2023, $599
12 GB GDDR6X, 200 Watt
64.05
RTX 5000 Ada Generation
2023
32 GB GDDR6, 250 Watt
74.40
+16.2%

RTX 5000 Ada Generation outperforms RTX 4070 by a moderate 16% based on our aggregate benchmark results.

Primary details

GPU architecture, market segment, value for money and other general parameters compared.

Place in the ranking3416
Place by popularity66not in top-100
Cost-effectiveness evaluation60.15no data
Power efficiency25.0123.20
ArchitectureAda Lovelace (2022−2024)Ada Lovelace (2022−2024)
GPU code nameAD104AD102
Market segmentDesktopWorkstation
Release date12 April 2023 (3 years ago)9 August 2023 (3 years ago)
Launch price (MSRP)$599 no data

Cost-effectiveness evaluation

The higher the ratio, the better. We use the manufacturer's recommended prices.

no data

Performance to price scatter graph

Currently popular graphics cards are shown for comparison.

Detailed specifications

General parameters such as number of shaders, GPU core base clock and boost clock speeds, manufacturing process, texturing and calculation speed. Real power consumption of some graphics cards can well exceed their nominal TDP, especially if overclocked.

Pipelines / CUDA cores588812800
Core clock speed1920 MHz1155 MHz
Boost clock speed2475 MHz2550 MHz
Number of transistors35,800 million76,300 million
Manufacturing process technology5 nm5 nm
Power consumption (TDP)200 Watt250 Watt
Texture fill rate455.41,020
Floating-point processing power29.15 TFLOPS65.28 TFLOPS
ROPs64176
TMUs184400
Tensor Cores184400
Ray Tracing Cores46100
L1 Cache5.8 MB12.5 MB
L2 Cache36 MB72 MB

Form factor & compatibility

Information on compatibility with other computer components. Useful when choosing a future computer configuration or upgrading an existing one. For desktop graphics cards it's interface and bus (motherboard compatibility), additional power connectors (power supply compatibility).

InterfacePCIe 4.0 x16PCIe 4.0 x16
Length240 mm267 mm
Width2-slot2-slot
Supplementary power connectors1x 16-pin1x 16-pin

VRAM capacity and type

Parameters of VRAM installed: its type, size, bus, clock and resulting bandwidth. Integrated GPUs have no dedicated video RAM and use a shared part of system RAM.

Memory typeGDDR6XGDDR6
Maximum RAM amount12 GB32 GB
Memory bus width192 Bit256 Bit
Memory clock speed1313 MHz2250 MHz
Memory bandwidth504.2 GB/s576.0 GB/s
Shared memory--
Resizable BAR++

Connectivity and outputs

This section shows the types and number of video connectors on each GPU. The data applies specifically to desktop reference models (for example, NVIDIA’s Founders Edition). OEM partners often modify both the number and types of ports. On notebook GPUs, video‐output options are determined by the laptop’s design rather than the graphics chip itself.

Display Connectors1x HDMI 2.1, 3x DisplayPort 1.4a4x DisplayPort 1.4a
HDMI+-

API and SDK support

List of supported 3D and general-purpose computing APIs, including their specific versions.

DirectX12 Ultimate (12_2)12 Ultimate (12_2)
Shader Model6.76.8
OpenGL4.64.6
OpenCL3.03.0
Vulkan1.31.3
CUDA8.98.9
DLSS++

Synthetic benchmarks

Non-gaming benchmark results comparison. The combined score is measured on a 0-100 point scale.


Combined synthetic benchmark score

This is our combined benchmark score.

RTX 4070 64.05
RTX 5000 Ada Generation 74.40
+16.2%

Passmark

This is the most ubiquitous GPU benchmark. It gives the graphics card a thorough evaluation under various types of load, providing four separate benchmarks for Direct3D versions 9, 10, 11 and 12 (the last being done in 4K resolution if possible), and few more tests engaging DirectCompute capabilities.

RTX 4070 26851
Samples: 21324
RTX 5000 Ada Generation 31189
+16.2%
Samples: 240

Gaming performance

Let's see how good the compared graphics cards are for gaming. Particular gaming benchmark results are measured in FPS.

Average FPS across all PC games

Here are the average frames per second in a large set of popular games across different resolutions:

Full HD198220−230
+11.1%
1440p104120−130
+15.4%
4K6370−75
+11.1%

Cost per frame, $

1080p3.03no data
1440p5.76no data
4K9.51no data

FPS performance in popular games

Full HD
Low

Counter-Strike 2 300−350 350−400
+12.9%
Cyberpunk 2077 216 250−260
+15.7%
Hogwarts Legacy 188 210−220
+11.7%
Resident Evil 4 Remake 240 270−280
+12.5%

Full HD
Medium

Battlefield 5 170−180 200−210
+14.9%
Counter-Strike 2 300−350 350−400
+12.9%
Cyberpunk 2077 174 200−210
+14.9%
Far Cry 5 249 280−290
+12.4%
Fortnite 300−350 350−400
+15.9%
Forza Horizon 4 250−260 290−300
+13.3%
Hogwarts Legacy 158 180−190
+13.9%
PLAYERUNKNOWN'S BATTLEGROUNDS 170−180 200−210
+15.6%
Valorant 350−400 400−450
+5%

Full HD
High

Battlefield 5 170−180 200−210
+14.9%
Counter-Strike 2 300−350 350−400
+12.9%
Counter-Strike: Global Offensive 270−280 300−310
+7.9%
Cyberpunk 2077 143 160−170
+11.9%
Far Cry 5 234 270−280
+15.4%
Fortnite 300−350 350−400
+15.9%
Forza Horizon 4 250−260 290−300
+13.3%
Grand Theft Auto V 174 200−210
+14.9%
Hogwarts Legacy 131 150−160
+14.5%
Metro Exodus 170 190−200
+11.8%
PLAYERUNKNOWN'S BATTLEGROUNDS 170−180 200−210
+15.6%
The Witcher 3: Wild Hunt 351 400−450
+14%
Valorant 350−400 400−450
+5%

Full HD
Ultra

Battlefield 5 170−180 200−210
+14.9%
Cyberpunk 2077 128 140−150
+9.4%
Far Cry 5 218 250−260
+14.7%
Forza Horizon 4 250−260 290−300
+13.3%
Hogwarts Legacy 103 110−120
+6.8%
PLAYERUNKNOWN'S BATTLEGROUNDS 170−180 200−210
+15.6%
The Witcher 3: Wild Hunt 171 190−200
+11.1%
Valorant 350−400 400−450
+5%

Full HD
Epic

Fortnite 300−350 350−400
+15.9%

1440p
High

Counter-Strike 2 190−200 220−230
+11.1%
Counter-Strike: Global Offensive 500−550 550−600
+7.6%
Grand Theft Auto V 137 150−160
+9.5%
Metro Exodus 104 120−130
+15.4%
PLAYERUNKNOWN'S BATTLEGROUNDS 170−180 200−210
+14.3%
Valorant 450−500 500−550
+9.6%

1440p
Ultra

Battlefield 5 160−170 190−200
+15.9%
Cyberpunk 2077 81 90−95
+11.1%
Far Cry 5 173 200−210
+15.6%
Forza Horizon 4 220−230 250−260
+13.6%
Hogwarts Legacy 77 85−90
+10.4%
The Witcher 3: Wild Hunt 140 160−170
+14.3%

1440p
Epic

Fortnite 150−160 170−180
+12.6%

4K
High

Counter-Strike 2 85−90 100−105
+12.4%
Grand Theft Auto V 146 160−170
+9.6%
Hogwarts Legacy 54 60−65
+11.1%
Metro Exodus 65 75−80
+15.4%
The Witcher 3: Wild Hunt 117 130−140
+11.1%
Valorant 300−350 350−400
+10.1%

4K
Ultra

Battlefield 5 120−130 130−140
+8.3%
Counter-Strike 2 85−90 100−105
+12.4%
Cyberpunk 2077 36 40−45
+11.1%
Far Cry 5 93 100−105
+7.5%
Forza Horizon 4 170−180 190−200
+11.1%
Hogwarts Legacy 40 45−50
+12.5%
PLAYERUNKNOWN'S BATTLEGROUNDS 95−100 110−120
+14.6%

4K
Epic

Fortnite 75−80 90−95
+13.9%

This is how RTX 4070 and RTX 5000 Ada Generation compete in popular games:

  • RTX 5000 Ada Generation is 11% faster in 1080p
  • RTX 5000 Ada Generation is 15% faster in 1440p
  • RTX 5000 Ada Generation is 11% faster in 4K

Pros & cons summary


Performance score 64.05 74.40
Recency 12 April 2023 9 August 2023
Maximum RAM amount 12 GB 32 GB
Power consumption (TDP) 200 Watt 250 Watt

RTX 4070 has 20% lower power consumption.

RTX 5000 Ada Generation, on the other hand, has a 16% higher aggregate performance score, an age advantage of 3 months, and a 167% higher maximum VRAM amount.

The RTX 5000 Ada Generation is our recommended choice as it beats the GeForce RTX 4070 in performance tests.

Be aware that GeForce RTX 4070 is a desktop graphics card while RTX 5000 Ada Generation is a workstation one.

Other comparisons

We selected several comparisons of graphics cards with performance close to those reviewed, providing you with more options to consider.

Community ratings

Here you can see the user ratings of the compared graphics cards, as well as rate them yourself.


4.2 12938 votes

Rate GeForce RTX 4070 on a scale of 1 to 5:

  • 1
  • 2
  • 3
  • 4
  • 5
3.5 95 votes

Rate RTX 5000 Ada Generation on a scale of 1 to 5:

  • 1
  • 2
  • 3
  • 4
  • 5

Comments

Here you can give us your opinion about GeForce RTX 4070 or RTX 5000 Ada Generation, agree or disagree with our ratings, or report errors or inaccuracies on the site.