Dark Mode
PassMark & Geekbench GPU Data
Product Reviews & Feedback
Tags and Keywords
Trusted By
"No reviews yet"
Free
About
This dataset compiles benchmark results data for Graphics Processing Units (GPUs), allowing for straightforward performance comparisons between various graphics cards. It serves as an invaluable resource for evaluating GPU performance across standardised tests, particularly useful for individuals looking to purchase or build a new PC. The data includes results from PassMark PerformanceTest benchmarks, updated daily, and Geekbench 5 scores, averaged from user uploads. The dataset captures the substantial parallel computation capacity of GPUs, supporting various programming interfaces like DirectCompute and OpenCL, with the latest data available as of 8th May 2022.
Columns
The dataset includes the following columns:
- gpuName: The model name of the Graphics Processing Unit.
- G3Dmark: The Graphics 3D mark score, indicating 3D performance.
- G2Dmark: The Graphics 2D mark score, indicating 2D performance.
- price: The last recorded market price in USD ($). Please note that a significant portion of values for this column are missing.
- gpuValue: A calculated metric representing the G3D Mark divided by the price, offering a value-for-money indicator. Many values in this column are missing.
- TDP: The maximum Thermal Design Power in watts (W), indicating power consumption. A substantial number of values are missing.
- powerPerformance: A calculated metric representing the G3D Mark divided by the TDP, indicating performance per watt. Many values in this column are missing.
- testDate: The date of the first benchmark submission for a given GPU.
- category: The category of the GPU, such as 'Unknown' or 'Desktop'.
Distribution
This dataset is provided as a CSV file (GPU_benchmarks_v7.csv), approximately 117.83 kB in size. It contains 9 columns. The dataset includes 2317 unique GPU models and corresponding benchmark entries. While many columns have complete data, some, such as
price, gpuValue, TDP, and powerPerformance, have a notable percentage of missing values.Usage
This dataset is ideal for:
- Comparing the performance of different graphics cards for informed purchasing decisions.
- Assessing GPU value for money by analysing the
gpuValuemetric. - Evaluating power efficiency using the
powerPerformancemetric. - Supporting PC building and upgrade planning by identifying suitable GPUs.
- Analysing trends in GPU performance over time.
- Researching GPU capabilities for parallel computing using APIs like OpenCL and DirectCompute.
Coverage
The dataset's time range spans from 2009 up to 8th May 2022. It includes benchmark results for a wide array of GPU models. Data is sourced from user-uploaded results to global benchmarking platforms. GPU categories are identified, including 'Unknown', 'Desktop', and 'Other', providing insights into various types of GPUs.
License
CC0: Public Domain
Who Can Use It
This dataset is intended for:
- PC enthusiasts and gamers looking to optimise their system builds.
- Hardware reviewers and tech journalists for performance comparisons and analyses.
- Researchers and data scientists interested in hardware trends and performance modelling.
- Consumers seeking to make informed decisions when buying new graphics cards.
- Developers working with GPU-accelerated computing.
Dataset Name Suggestions
- GPU Benchmark Results (2009-2022)
- Graphics Card Performance Database
- PC GPU Benchmarks and Specs
- PassMark & Geekbench GPU Data
- GPU Performance Comparison
Attributes
Original Data Source: PassMark & Geekbench GPU Data
QUALITY