AMD Radeon RX 570X vs NVIDIA GeForce GTX 590 Comparison

FAQ

Q: Which GPU has the higher average benchmark score?

A: The AMD Radeon RX 570X has an average benchmark score of 13,871, while the NVIDIA GeForce GTX 590 scores 12,830. The RX 570X also holds a higher percentile rank at 55 versus the GTX 590's 52.

Q: What is the biggest performance gap between the two cards?

A: In the Geekbench OpenCL test, the RX 570X scores 38,939 against the GTX 590's 12,830. That represents a delta of -67.1% from the perspective of the GTX 590, meaning the RX 570X is more than three times faster in this specific compute workload.

Q: How do the memory configurations differ?

A: The GTX 590 has 1,536 MB of GDDR5 on a 384-bit bus with 164.0 GB/s bandwidth, while the RX 570X has 8 GB of GDDR5 on a 256-bit bus with 224.0 GB/s bandwidth. Despite the narrower bus, the RX 570X delivers higher total memory bandwidth.

Q: Which card has more shading units?

A: The RX 570X has 2,048 shading units, which is four times the 512 shading units found on the GTX 590. The RX 570X also has 128 texture mapping units versus 64, though the GTX 590 has 48 ROPs compared to the RX 570X's 32.

Q: What are the power requirements for each card?

A: The GTX 590 has a TDP of 365 W with a suggested PSU of 750 W and requires two 8-pin power connectors. The RX 570X has a TDP of 150 W with a suggested PSU of 450 W and requires a single 6-pin connector.

Q: Do both cards support the same API levels?

A: Both support DirectX 12 and OpenGL 4.6. However, the RX 570X supports DirectX 12 (12_0) and Vulkan 1.3, while the GTX 590 supports DirectX 12 (11_0) and has no listed Vulkan support.

Architecture Differences

The two GPUs come from fundamentally different architectural eras. The NVIDIA GeForce GTX 590 is built on the Fermi 2.0 architecture using the GF110 chip, fabricated on a 40 nm process at TSMC. The AMD Radeon RX 570X uses the GCN 4.0 architecture with the Polaris 20 chip, manufactured on a 14 nm process at GlobalFoundries. This process node difference is stark: 40 nm versus 14 nm.

Transistor counts and die sizes tell the story of the architectural generational gap. The GTX 590 packs 3,000 million transistors on a 520 mm² die, giving a transistor density of 5.8 million per mm². The RX 570X contains 5,700 million transistors on a much smaller 232 mm² die, yielding a density of 24.6 million per mm². The RX 570X achieves nearly double the transistor count on less than half the silicon area.

Compute resources differ dramatically. The GTX 590 has 512 shading units, 64 TMUs, and 48 ROPs. The RX 570X has 2,048 shading units and 128 TMUs but only 32 ROPs. The RX 570X's shading unit count is four times higher, which directly feeds its raw FLOP advantage. The GTX 590's higher ROP count suggests a different balance between compute and pixel processing.

Memory architecture also diverges. The GTX 590 uses 1,536 MB of GDDR5 on a 384-bit bus, while the RX 570X uses 8 GB on a 256-bit bus. The RX 570X's memory runs at 1,750 MHz (7 Gbps effective) versus the GTX 590's 854 MHz (3.4 Gbps effective). This results in 224.0 GB/s for the RX 570X versus 164.0 GB/s for the GTX 590.

The RX 570X also supports PCIe 3.0 x16, while the GTX 590 uses PCIe 2.0 x16. Display output capabilities differ as well: the GTX 590 offers 3x DVI and 1x mini-DisplayPort, while the RX 570X provides 1x DVI, 1x HDMI 2.0b, and 3x DisplayPort 1.4a. The RX 570X lists Vulkan 1.3 support; the GTX 590 has no Vulkan entry.

Where Each One Wins

The benchmark data shows a single head-to-head result, and it is decisive. The AMD Radeon RX 570X wins the only shared test (Geekbench OpenCL) by a massive margin. In that test, the RX 570X scores 38,939 versus the GTX 590's 12,830.

The RX 570X also holds the overall average benchmark advantage. Its average score of 13,871 includes results from three tests: Geekbench OpenCL (38,939), Passmark G2D (750), and Passmark G3D (1,923). The GTX 590's average of 12,830 comes from its single Geekbench OpenCL result.

However, the GTX 590 is not without its own strengths. Its nearest rivals include the AMD Radeon Pro 455 (scoring 12,831) and the NVIDIA GeForce GTX 670 (scoring 12,773). The GTX 590 sits within 0.4% of the GTX 670 and is essentially tied with the Radeon Pro 455. This placement suggests the GTX 590 remains competitive within its immediate performance class, even though it trails the RX 570X.

The RX 570X's nearest rivals show a different picture. It sits near the NVIDIA RTX A2000 Mobile (13,821) and the AMD Radeon 660M (13,812), while being 0.9% ahead of the AMD Radeon RX 7900 XT (13,745). The RX 570X's average score of 13,871 places it slightly above these competitors, indicating it performs near the upper boundary of its comparison group.

For compute-heavy workloads measured by OpenCL, the RX 570X is clearly the superior choice. The GTX 590, despite its age and architectural limitations, still benchmarks close to several modern mid-range GPUs in its percentile tier. The data does not show any test where the GTX 590 wins; the wins tally stands at 0 for the GTX 590 and 1 for the RX 570X.

Specification Differences

The two cards differ across nearly every specification category. Process node: the GTX 590 uses 40 nm, the RX 570X uses 14 nm. Transistor count: 3,000 million versus 5,700 million. Die size: 520 mm² versus 232 mm². Transistor density: 5.8M per mm² versus 24.6M per mm².

Clock speeds: the GTX 590 has no listed base or boost clock, while the RX 570X runs at 1,168 MHz base and 1,244 MHz boost. Memory clock: 854 MHz (3.4 Gbps effective) on the GTX 590 versus 1,750 MHz (7 Gbps effective) on the RX 570X.

Memory: 1,536 MB versus 8 GB. Bus width: 384-bit versus 256-bit. Bandwidth: 164.0 GB/s versus 224.0 GB/s. Shading units: 512 versus 2,048. TMUs: 64 versus 128. ROPs: 48 versus 32.

Pixel rate: 19.46 GPixel/s on the GTX 590 versus 39.81 GPixel/s on the RX 570X. Texture rate: 38.91 GTexel/s versus 159.2 GTexel/s. FP32 performance: 1,244.2 GFLOPS versus 5.095 TFLOPS. The RX 570X also lists FP16 at 5.095 TFLOPS (1:1), a feature absent from the GTX 590.

Power and cooling: the GTX 590 has a 365 W TDP with 2x 8-pin connectors and a 750 W suggested PSU. The RX 570X has a 150 W TDP with 1x 6-pin connector and a 450 W suggested PSU. Both are dual-slot cards.

Physical dimensions: the GTX 590 measures 279 mm (11 inches) in length, 111 mm (4.4 inches) in height, and 40 mm (1.6 inches) in width. The RX 570X is 241 mm (9.5 inches) long with no listed height or width.

Bus interface: PCIe 2.0 x16 versus PCIe 3.0 x16. Display outputs: 3x DVI and 1x mini-DisplayPort versus 1x DVI, 1x HDMI 2.0b, and 3x DisplayPort 1.4a. API support: DirectX 12 (11_0) versus DirectX 12 (12_0), with Vulkan 1.3 only on the RX 570X. Release dates: March 23, 2011 for the GTX 590 versus April 10, 2018 for the RX 570X. The GTX 590 has a launch MSRP of 699 USD; the RX 570X has no listed launch MSRP.

Head-to-Head Benchmarks

The head-to-head comparison table contains one entry: Geekbench OpenCL. The AMD Radeon RX 570X scores 38,939, while the NVIDIA GeForce GTX 590 scores 12,830. The delta percentage is -67.1%, indicating the GTX 590 trails by more than two-thirds in this workload.

This single result dominates the comparison. The RX 570X's score is roughly three times that of the GTX 590. In absolute terms, the gap is 26,109 points. That is not a marginal difference; it represents a completely different performance tier in OpenCL compute.

Looking at broader context, the GTX 590's nearest rival is the AMD Radeon Pro 455 at 12,831, a 0% delta. The AMD FirePro W5100 scores 12,847 (-0.1%), and the AMD Radeon 740M scores 12,870 (-0.3%). The NVIDIA GeForce GTX 670 is slightly behind at 12,773 (+0.4% for the GTX 590). This places the GTX 590 within a tight cluster of GPUs all scoring near 12,800 in average benchmark performance.

The RX 570X's nearest rivals occupy a higher band. The NVIDIA RTX A2000 Mobile scores 13,821 (+0.4% for the RX 570X), and the AMD Radeon 660M scores 13,812 (+0.4%). The AMD Radeon RX 7900 XT is at 13,745 (+0.9% for the RX 570X), while the NVIDIA Tesla K10 sits above at 14,029 (-1.1%). The RX 570X's average of 13,871 puts it at the top of this group except for the Tesla K10.

The percentile ranks reinforce the separation: the GTX 590 sits at the 52nd percentile of all GPUs, while the RX 570X sits at the 55th. Though only three percentile points apart, the average score gap is 1,041 points. The RX 570X also benefits from having three benchmark results contributing to its average, including a Passmark G3D score of 1,923 and a Passmark G2D score of 750.

The Verdict

The data is unambiguous. The AMD Radeon RX 570X outperforms the NVIDIA GeForce GTX 590 in every measured category that matters for compute performance. The Geekbench OpenCL score of 38,939 versus 12,830 is a 67.1% advantage. The average benchmark score of 13,871 versus 12,830 further confirms the RX 570X's superiority.

Buyers looking for modern API support should choose the RX 570X. It offers DirectX 12 (12_0) and Vulkan 1.3, while the GTX 590 is limited to DirectX 12 (11_0) with no Vulkan support. The RX 570X also has substantially more memory (8 GB versus 1,536 MB), higher bandwidth (224.0 GB/s versus 164.0 GB/s), and a much higher pixel rate (39.81 GPixel/s versus 19.46 GPixel/s).

Power efficiency strongly favors the RX 570X. Its 150 W TDP is less than half the GTX 590's 365 W, and it requires a 450 W PSU versus 750 W. The single 6-pin connector versus dual 8-pin connectors further simplifies installation.

The GTX 590's only comparative advantages are its higher ROP count (48 versus 32) and its wider memory bus (384-bit versus 256-bit). Neither compensates for the massive compute deficit. Its nearest rivals include the GTX 670 and Radeon Pro 455, both scoring near 12,800, which suggests the GTX 590 is firmly anchored in an older performance tier.

For any workload measured by OpenCL, the RX 570X is the clear recommendation. For users prioritizing modern connectivity, lower power draw, and higher memory capacity, the RX 570X again wins. The GTX 590, released in 2011 with a 699 USD launch MSRP, remains competitive only within its own aging peer group. The RX 570X, released in 2018, delivers roughly triple the compute performance with a fraction of the power requirement. The verdict from the data: choose the AMD Radeon RX 570X.