Gigabyte Aorus GeForce RTX 5090 Xtreme WaterForce specifications and in-depth review

Gigabyte Aorus GeForce RTX 5090 Xtreme WaterForce

Manufacturer: Gigabyte

The Gigabyte Aorus GeForce RTX 5090 Xtreme WaterForce is a high-end graphics card from Gigabyte's Aorus lineup, featuring an integrated water cooling solution and 115.5 TFLOPS of floating-point performance. Built on NVIDIA's Blackwell architecture using a 5nm semiconductor process, it carries 92,200 million transistors and operates with a base GPU clock of 2017 MHz that boosts up to 2655 MHz under load.

On the memory side, the card is equipped with 32GB of GDDR7 VRAM across a 512-bit bus, delivering a maximum bandwidth of 1790 GB/s at an effective speed of 28,000 MHz. It supports DirectX 12 Ultimate, OpenGL 4.6, ray tracing, and DLSS, while connectivity is handled through one HDMI 2.1b port and three DisplayPort outputs, allowing up to four simultaneous displays. The card connects via PCIe 5.0, includes Intel Resizable BAR support, and comes with a four-year warranty.

Pros
  • The 512-bit memory bus paired with 32GB of GDDR7 VRAM delivers a maximum bandwidth of 1790 GB/s, which is well-suited for memory-intensive workloads
  • Water cooling is built into the card itself, removing the need for a separate cooling solution while keeping thermals manageable for a 575W component
  • Support for up to four simultaneous displays through a combination of one HDMI 2.1b and three DisplayPort outputs offers flexible multi-monitor setups
  • ECC memory support adds data integrity protection, making the card more reliable for compute and professional workloads
  • The four-year warranty period is longer than the standard coverage typically offered for graphics cards
  • Ray tracing and DLSS support are both present, broadening compatibility with modern rendering workloads
Cons
  • A TDP of 575W places substantial demands on system power delivery and case airflow, requiring careful build planning
  • The absence of USB-C and DVI outputs limits connectivity options for users with older monitors or USB-C display setups
  • The card does not support XeSS (XMX), which restricts upscaling options to DLSS only
  • At 220mm wide, the physical footprint may present fitment challenges in smaller or mid-tower cases
Who is this for?

This card is well-matched for users running demanding compute and rendering workloads that can take full advantage of 115.5 TFLOPS of floating-point performance and 32GB of ECC-enabled GDDR7 VRAM. The integrated water cooling system makes it a practical fit for workstation builds where sustained load over long sessions is expected, while the four-display output support suits professionals who rely on multi-monitor configurations for creative, scientific, or technical work. PCIe 5.0 compatibility also makes it appropriate for users building on current-generation platforms who want to avoid bandwidth bottlenecks.

Who is this NOT for?

The card's 575W TDP makes it a poor fit for users with constrained power supplies or cases that lack adequate airflow, as the thermal and electrical demands are significant even with the built-in water cooling. It is also not well-suited for compact or small-form-factor builds, given its 220mm width, which may not physically accommodate the card. Users whose display setup relies on USB-C or DVI connections will also find this card lacking, as neither output type is available on this model.

Performance:

GPU clock speed 2017 MHz
GPU turbo 2655 MHz
pixel rate 467.3 GPixel/s
floating-point performance 115.5 TFLOPS
texture rate 1805 GTexels/s
GPU memory speed 1750 MHz
shading units 21760
texture mapping units (TMUs) 680
render output units (ROPs) 176
Has Double Precision Floating Point (DPFP)

The Performance section of this card centers on a base GPU clock of 2017 MHz with a turbo speed reaching 2655 MHz, backed by 115.5 TFLOPS of floating-point performance and support for Double Precision Floating Point (DPFP). The 21,760 shading units work alongside 680 texture mapping units and 176 render output units (ROPs), contributing to a texture rate of 1805 GTexels/s and a pixel rate of 467.3 GPixel/s. GPU memory operates at a speed of 1750 MHz, rounding out a specification set oriented toward sustained, compute-intensive workloads.

Memory:

effective memory speed 28000 MHz
maximum memory bandwidth 1790 GB/s
VRAM 32GB
GDDR version GDDR7
memory bus width 512-bit
Supports ECC memory

The card features 32GB of GDDR7 VRAM running across a 512-bit memory bus, with an effective memory speed of 28,000 MHz that translates to a maximum memory bandwidth of 1790 GB/s. ECC memory support is also included, adding a layer of data integrity protection for workloads where memory accuracy is a priority.

Features:

DirectX version DirectX 12 Ultimate
OpenGL version 4.6
OpenCL version 3
Supports multi-display technology
supports ray tracing
Supports 3D
supports DLSS
has XeSS (XMX)
AMD SAM / Intel Resizable BAR Intel Resizable BAR
has LHR
has RGB lighting
supported displays 4

On the software and feature side, the card supports DirectX 12 Ultimate, OpenGL 4.6, and OpenCL 3, covering a broad range of graphics and compute workloads. Ray tracing and DLSS are both supported, while XeSS (XMX) is not included. The card also supports stereoscopic 3D, multi-display technology with up to four simultaneous outputs, and Intel Resizable BAR for improved CPU-to-GPU data access. LHR is not present, and RGB lighting is built in. It does not include XeSS (XMX) functionality.

Ports:

has an HDMI output
HDMI ports 1
HDMI version HDMI 2.1b
DisplayPort outputs 3
USB-C ports 0
DVI outputs 0
mini DisplayPort outputs 0

The card's output configuration consists of one HDMI 2.1b port and three DisplayPort outputs, giving a total of four available display connections. There are no USB-C, DVI, or mini DisplayPort outputs on this model.

General info:

GPU architecture Blackwell
Thermal Design Power (TDP) 575W
PCI Express (PCIe) version 5
semiconductor size 5 nm
number of transistors 92200 million
warranty period 4 years
Has air-water cooling
width 220 mm
height 130 mm

This card is based on the Blackwell GPU architecture, fabricated on a 5nm semiconductor process and packing 92,200 million transistors onto the die. It connects via PCIe 5.0 and carries a Thermal Design Power (TDP) of 575W, with an integrated air-water cooling solution to manage thermals. The physical dimensions measure 220mm in width and 130mm in height, and the card ships with a four-year warranty.

Final Verdict

The Gigabyte Aorus GeForce RTX 5090 Xtreme WaterForce is a card built for users who place serious demands on both compute throughput and memory capacity, with its 32GB of GDDR7 VRAM across a 512-bit bus delivering 1790 GB/s of bandwidth standing out as a defining characteristic of the overall package. The integrated water cooling, four-year warranty, and PCIe 5.0 support reflect a product designed for sustained, professional-grade use rather than casual deployment. That said, the 575W TDP and physical dimensions mean it demands a well-planned system environment to operate as intended. For users whose workloads align with what this card offers, the Aorus RTX 5090 Xtreme WaterForce represents a thoroughly specified option within the high-end graphics card category.

Popular Comparisons

Gigabyte Aorus GeForce RTX 5090 Xtreme WaterForce
Gigabyte Aorus GeForce RTX 5090 Xtreme WaterForce
VS
Asus ROG Astral LC GeForce RTX 5090
Asus ROG Astral LC GeForce RTX 5090
Gigabyte Aorus GeForce RTX 5090 Xtreme WaterForce
Gigabyte Aorus GeForce RTX 5090 Xtreme WaterForce
VS
Asus ROG Astral LC GeForce RTX 5090 OC Edition
Asus ROG Astral LC GeForce RTX 5090 OC Edition
Gigabyte Aorus GeForce RTX 5090 Xtreme WaterForce
Gigabyte Aorus GeForce RTX 5090 Xtreme WaterForce
VS
Gigabyte GeForce RTX 5080 Aero OC SFF
Gigabyte GeForce RTX 5080 Aero OC SFF
Gigabyte Aorus GeForce RTX 5090 Xtreme WaterForce
Gigabyte Aorus GeForce RTX 5090 Xtreme WaterForce
VS
MSI GeForce RTX 5090 Suprim SOC
MSI GeForce RTX 5090 Suprim SOC