AMD Radeon RX 9070 对 NVIDIA GeForce RTX 5060 Ti (16GB). 规格

显卡的基本技术数据比较 AMD Radeon RX 9070 和 NVIDIA GeForce RTX 5060 Ti (16GB), 芯片，信息处理单元.

• AI Accelerators
  112 -
• 建筑学
  RDNA 4 Blackwell
• 基于
  AMD Radeon RX 9070 NVIDIA GeForce RTX 5060 Ti
• Compute Units
  56 -
• Execution Units
  56 -
• Ray Accelerators
  56 -
• 渲染输出单位
  128 -
• 着色器
  3584 4608
• CUDA Cores
  - 4608

Video memory capacity, type, bus width and bandwidth describe different properties. These values alone do not determine graphics performance.

• Infinity Cache
  64 MB -
• Memory Bandwidth
  640 GB/s 448 GB/s
• Memory Bus
  256 bit 128 bit
• Memory Data Rate
  20 Gbps -
• 内存大小
  16 GB 16 GB
• 存储器类型
  GDDR6 GDDR7

Published GPU clock frequencies describe different operating modes. Clock speed alone does not determine graphics performance.

• 最大频率
  2520 MHz 2570 MHz
• Game Clock
  2070 MHz -
• 基准频率
  - 2410 MHz

Graphics card power limits and recommended system power supply describe different specifications.

• System Power Supply
  650 W (System power requirement) 600 W (System power requirement)
• TDP
  220 W 180 W
• Typical Board Power
  220 W -
• Maximum GPU Temperature
  - 87 °C

连通性和连接.

• DisplayPort Version
  2.1a -
• HDMI Version
  2.1b -
• PCIe Interface
  PCIe 5.0 x16 5.0
• Power Connectors
  2x8-Pin -

内置对视频和图像压缩标准的支持.

• AV1
  Decode / Encode Decode / Encode

GPU的接口和发布日期的比较 NVIDIA GeForce RTX 5060 Ti (16GB) 和 AMD Radeon RX 9070.

• Availability window (manufacturer)
  2025-03-06 -
• Date evidence scope
  Manufacturer-announced OEM/board availability window; exact first shipment is not established by this source. -
• Official source
  https://www.amd.com/en/products/graphics/desktops/radeon/9000-series/amd-radeon-rx-9070.html https://www.nvidia.com/en-us/geforce/graphics-cards/50-series/rtx-5060-family/
• 发布日期
  2025-03-06 -
• Release date source
  https://ir.amd.com/news-events/press-releases/detail/1238/amd-unveils-next-generation-amd-rdna-4-architecture-with-the-launch-of-amd-radeon-rx-9000-series-graphics-cards -
• Verification scope
  Explicit manufacturer fields refreshed; other original historical fields retained, not all reverified. Partial composite conflicts are preserved. -
• Verified on
  2026-09-30 2026-09-30
• Specification source
  - https://www.nvidia.com/en-us/geforce/graphics-cards/compare/

• FP32 (theoretical)
  36.1 TFLOPS -