ATI Radeon X1900 XT

The ATI Radeon X1900 XT is a 90 nm graphics card from AMD, built on the R580 chip with the Ultra-Threaded SE architecture and released on January 23, 2006, at a launch MSRP of 299 USD. It sits at the 50th percentile of all GPUs in the benchmark database, pairing 256 MB of GDDR3 memory on a 256-bit bus with 16 texture mapping units and 16 raster operation units to produce a pixel rate of 10.00 GPixel/s and a texture rate of 10.00 GTexel/s. This dual-slot, 242 mm card carries a 100 W TDP and requires a single 1x 6-pin power connector.

Benchmark Performance

The database records no averaged benchmark score for the X1900 XT; the average benchmark score field is 0, and no nearest-rival entries are associated with this part. The only positional metric available is the 50th percentile standing across all GPUs, which places this card at the exact median of the tracked field. Without rival scores or deltaPct values, direct percentage comparisons are impossible, so the performance picture must be assembled from the fixed-function throughput figures in the specification.

The pixel rate of 10.00 GPixel/s and texture rate of 10.00 GTexel/s are identical, indicating a balanced design between the 16 ROPs and 16 TMUs. The memory clock of 725 MHz, operating at 1450 Mbps effective, drives the 256-bit bus to deliver 46.40 GB/s of bandwidth. These raw numbers define the card's ceiling: every frame must be shaded, textured, and rasterized within these rates. The 50th percentile ranking suggests that, among the GPUs tracked in this database, the X1900 XT lands squarely in the middle of the performance distribution. That median position is consistent with a card whose transistor count of 384 million, spread across a 352 mm² die at a density of 1.1M transistors per mm², reflects a mid-generation design rather than a flagship or entry-level part. The absence of benchmark scores likely reflects the card's end-of-life production status and the age of the platform; nevertheless, the 10.00 GPixel/s fill rate and 10.00 GTexel/s texture rate are the figures that governed its real-world behavior in the DirectX 9.0c era.

Ray Tracing and Feature Set

The X1900 XT has no dedicated ray tracing cores and no tensor cores; both fields are null in the specification. This is an immediate indicator that the card predates hardware-accelerated ray tracing by a wide margin. The feature set is defined entirely by the API support listed in the database: DirectX 9.0c with shader model 9_3, OpenGL 2.1 in full and OpenGL 3.0 in partial form, and no Vulkan support.

The DirectX 9.0c (9_3) target means the card was built for the pixel shader and vertex shader workloads of the mid-2000s, not for compute-heavy or ray-traced rendering. The partial OpenGL 3.0 support is notable: applications that require the full OpenGL 3.0 feature set will not run correctly, while OpenGL 2.1 applications are fully covered. The Ultra-Threaded SE architecture in the R580 chip is the underlying mechanism that schedules the 16 TMUs and 16 ROPs. With no tensor cores, any machine-learning or AI-accelerated features are absent. With no RT cores, any form of hardware ray tracing is absent. The card's feature set is therefore entirely rasterization-based, relying on the 10.00 GPixel/s pixel rate and 10.00 GTexel/s texture rate to deliver frames. The display outputs — 2x DVI and 1x S-Video — complete the feature set, providing dual digital outputs and one analog video output.

How It Compares

The benchmark database lists no nearest rivals for the X1900 XT, so there are no rival names, scores, or deltaPct values to cite. The card's position must therefore be inferred from its generation and percentile. The predecessor is the Radeon R400 PCIe series, and the successor is the Radeon R600 series. The X1900 XT sits between these two families as part of the Radeon R500 PCIe (X1900) generation.

The 50th percentile ranking across all GPUs places it at the median of the entire tracked field, which is a useful anchor even without direct rivals. A card at the 50th percentile is neither a performance outlier nor a laggard; it represents the typical experience of the GPU population in the database. The R580 chip, manufactured on TSMC's 90 nm process, uses 384 million transistors on a 352 mm² die. That die size and transistor count are mid-range figures for the era. Compared to the R400 generation it succeeded, the X1900 XT brings the Ultra-Threaded SE architecture and the higher memory clock of 725 MHz. Compared to the R600 generation that followed, it lacks the architectural advances of the successor, but the database does not quantify those differences. What can be stated is the card's balanced specification: 16 TMUs and 16 ROPs both operating at rates of 10.00 GTexel/s and 10.00 GPixel/s respectively. This balance suggests that the X1900 XT was designed to avoid bottlenecks between texturing and rasterization, a trait that would have served it well in the DirectX 9.0c workloads of its time.

Power and Cooling

The X1900 XT carries a thermal design power of 100 W. The suggested power supply rating is 300 W, and the card requires a single 1x 6-pin PCIe power connector. These three figures define the electrical requirements: a 300 W PSU with one 6-pin connector is the minimum configuration.

The card is a dual-slot design, meaning it occupies two expansion slots in a chassis. Its physical length is 242 mm, or 9.5 inches, which imposes a clearance requirement on the case. The 90 nm process from TSMC, with 384 million transistors on a 352 mm² die, generates the 100 W thermal load that the dual-slot cooler must dissipate. The dual-slot form factor allows for a larger heatsink and fan assembly than a single-slot design, which is a reasonable thermal solution for the era. The 1x 6-pin connector is a single supplemental power input, in addition to the power drawn from the PCIe 1.0 x16 slot. The 300 W suggested PSU is the total system power supply rating, not the card's draw; the card itself is rated at 100 W TDP. Users building a system around this card would need a PSU rated at 300 W or higher, with one 6-pin cable available. The dual-slot cooler and 242 mm length are the physical constraints to check against a case's interior dimensions.

FAQ

Q: What is the memory size and type on the X1900 XT?

A: The card has 256 MB of GDDR3 memory on a 256-bit bus, with a memory clock of 725 MHz (1450 Mbps effective) and a bandwidth of 46.40 GB/s.

Q: What power supply is recommended?

A: The suggested PSU rating is 300 W, and the card requires one 1x 6-pin power connector. The TDP is 100 W.

Q: What DirectX version does it support?

A: It supports DirectX 9.0c with shader model 9_3. OpenGL is supported at 2.1 (full) and 3.0 (partial). Vulkan is not listed.

Q: When was it released and at what price?

A: It was released on January 23, 2006, with a launch MSRP of 299 USD.

Q: What is the process node and die size?

A: It is manufactured on a 90 nm process at TSMC, with 384 million transistors on a 352 mm² die, yielding a density of 1.1M transistors per mm².

Q: What are the display outputs?

A: The card offers 2x DVI and 1x S-Video outputs.

Memory Subsystem

The memory subsystem of the X1900 XT is built around 256 MB of GDDR3 on a 256-bit bus. The memory clock is 725 MHz, which translates to 1450 Mbps effective due to double data rate signaling. This combination yields a bandwidth of 46.40 GB/s. The 256-bit bus width is a critical factor: it provides the data path width needed to feed the 16 TMUs and 16 ROPs.

The 46.40 GB/s bandwidth is the rate at which texture data, frame buffers, and geometry data can move between the GPU and memory. For high-resolution rendering, the 256 MB capacity is the primary constraint. A larger frame buffer allows more data to be held on-card, reducing the need to re-fetch data from system memory. The 256 MB size means that at higher resolutions and with higher texture detail settings, the card may run out of local memory, forcing it to spill to system memory over the PCIe 1.0 x16 bus. The 256-bit bus, however, is a wide interface that was a hallmark of performance-oriented cards of this generation. The 46.40 GB/s bandwidth is the product of the 256-bit bus and the 1450 Mbps effective memory clock. The memory type, GDDR3, was the standard for high-end cards of the mid-2000s, offering a balance of speed and power consumption. The 10.00 GPixel/s pixel rate and 10.00 GTexel/s texture rate are both served by this memory bandwidth; if the bandwidth were insufficient, these rates would be unattainable. In practice, the 46.40 GB/s figure is the ceiling for memory-intensive operations, and the 256 MB capacity, while modest by later standards, was appropriate for the DirectX 9.0c workloads of 2006.

Who Should Consider It

The X1900 XT is a card for users whose expectations align with its 50th percentile standing in the database. With an average benchmark score of 0 recorded, there is no performance metric to cite beyond the percentile, so recommendations must be grounded in the specification. The card's 256 MB of GDDR3 memory and 46.40 GB/s bandwidth, combined with a 10.00 GPixel/s pixel rate and 10.00 GTexel/s texture rate, define its capability envelope.

Users targeting the DirectX 9.0c (9_3) API are the natural audience; the card supports this fully. For OpenGL, the full 2.1 support and partial 3.0 support mean that OpenGL 2.1 applications are the safe target. The 100 W TDP and 300 W suggested PSU make it a relatively modest power consumer, suitable for systems with a 300 W or higher supply. The dual-slot cooler and 242 mm length require a case with adequate clearance. The 2x DVI and 1x S-Video outputs support dual digital monitors and one analog display.

Given the 50th percentile ranking, the card is appropriate for mid-range expectations — not for the highest settings of its era, but for standard settings at the resolutions common to 2006. The 256 MB frame buffer is the limiting factor for high-resolution texture-heavy workloads; users who prioritize large texture sets or very high resolutions would need more memory, but the database does not record a higher-memory variant. The card's end-of-life production status means it is no longer manufactured, so consideration is limited to the used market. For users building a retro DirectX 9.0c system, the X1900 XT's 16 TMUs, 16 ROPs, and 46.40 GB/s bandwidth provide a balanced foundation. For anyone expecting modern features like ray tracing, tensor cores, or Vulkan, this card is not the answer — those features are absent from its specification.