AMD FirePro D500 vs NVIDIA GeForce GTX 690 Comparison

The GeForce GTX 690 and the FirePro D500 are two very different solutions from the same 28 nm TSMC generation, aimed at different worlds. The GTX 690 is a dual-GPU flagship from NVIDIA’s GeForce 600 line, while the D500 is a single-GPU workstation card from AMD’s FirePro Data Center series. Benchmark data shows the AMD FirePro D500 holds a 10% lead in the one shared test, the Geekbench Vulkan benchmark, scoring 18,533 against the GTX 690’s 16,675. However, the GTX 690 counters with a higher average benchmark score of 17,037 across two tests, and its architecture tells a story of raw compute throughput versus workstation stability.

Where Each One Wins

The data splits the two cards by workload type. The AMD FirePro D500 wins in the only direct head-to-head benchmark available: Geekbench Vulkan. It scores 18,533, which is 10% higher than the GTX 690’s 16,675. This suggests the D500 has an advantage in modern, low-level graphics APIs that the GTX 690 lacks. The D500 also holds a higher percentile rank among all GPUs at 62, compared to the GTX 690’s 60.

The NVIDIA GeForce GTX 690, despite losing the Vulkan test, posts a higher average benchmark score of 17,037 across its two recorded tests (Geekbench OpenCL and Vulkan). Its OpenCL score of 17,399 is its best result, indicating strength in compute workloads that use OpenCL. The GTX 690’s nearest rival data shows it trading blows with much newer cards: it is only 0.3% behind the AMD Radeon RX 7600 XT and 1% behind the NVIDIA GeForce RTX 3070 in average score. The D500, by comparison, sits 0.8% behind the AMD Radeon Pro 5700 XT in its average score. For gaming-oriented or general compute tasks, the GTX 690’s dual-GPU design appears to provide a broader performance umbrella, while the D500 is the more consistent performer in the Vulkan API specifically.

Architecture Differences

Both cards use a 28 nm process from TSMC, but their internal designs diverge sharply. The GTX 690 uses the GK104 chip, based on the Kepler architecture, with 3,540 million transistors on a 294 mm² die. The D500 uses the Tahiti chip, based on GCN 1.0, with 4,313 million transistors on a larger 352 mm² die. The transistor density is nearly identical (12.0M per mm² for the GTX 690 versus 12.3M per mm² for the D500), but the D500’s larger die allows for more raw silicon area.

Memory configurations also differ. The GTX 690 has 2 GB of GDDR5 on a 256-bit bus, delivering 192.3 GB/s of bandwidth. The D500 has 3 GB of GDDR5 on a 384-bit bus, delivering 243.8 GB/s. The D500’s wider bus and higher capacity give it a clear bandwidth advantage. The GTX 690 operates at higher clocks: a base of 915 MHz and boost of 1019 MHz, with memory at 6 Gbps effective. The D500’s memory runs at 5.1 Gbps effective, and no base or boost clocks are listed in the data.

Compute resources are a mixed bag. Both cards have 1,536 shading units and 32 ROPs. The GTX 690 has 128 TMUs, while the D500 has 96. The GTX 690’s higher clock speeds push its pixel rate to 32.61 GPixel/s and texture rate to 130.4 GTexel/s, versus the D500’s 23.20 GPixel/s and 69.60 GTexel/s. In FP32 compute, the GTX 690 leads with 3.130 TFLOPS versus the D500’s 2.227 TFLOPS. API support shows the D500 has a slight edge on DirectX (12 with 11_1 versus the GTX 690’s 12 with 11_0), but both support OpenGL 4.6 and Vulkan (1.2.170 for the D500, 1.2.175 for the GTX 690).

FAQ

Q: Which card has more memory bandwidth?

A: The AMD FirePro D500 has a significant lead, with 243.8 GB/s over a 384-bit bus, compared to the GTX 690’s 192.3 GB/s over a 256-bit bus.

Q: Is the GTX 690 faster in raw compute throughput?

A: Yes. The GTX 690’s FP32 performance is 3.130 TFLOPS, which is roughly 40% higher than the D500’s 2.227 TFLOPS, due to its higher clock speeds and additional texture units.

Q: Why does the FirePro D500 win the Vulkan benchmark if it has lower compute specs?

A: The D500 scores 18,533 in Geekbench Vulkan against the GTX 690’s 16,675, a 10% difference. This result likely reflects better driver optimization or architectural efficiency for Vulkan in the GCN 1.0 design, despite the GTX 690’s higher theoretical peak rates.

Q: Do both cards support the same modern APIs?

A: Both support OpenGL 4.6 and Vulkan 1.2.x. The D500 supports DirectX 12 (11_1), while the GTX 690 supports DirectX 12 (11_0), so the D500 has a marginally newer DirectX feature level.

Q: Which card has a higher transistor count?

A: The AMD FirePro D500 has 4,313 million transistors on its Tahiti chip, while the NVIDIA GeForce GTX 690 has 3,540 million on its GK104 chip.

Q: What is the power draw difference?

A: The GTX 690 has a TDP of 300 W and requires a 700 W suggested PSU, while the D500 has a TDP of 274 W with a 600 W suggested PSU. The GTX 690 also needs two 8-pin power connectors, while the D500’s connectors are not listed.

Specification Differences

The two cards differ in nearly every core specification. The GTX 690 uses the GK104 chip with 3,540 million transistors on a 294 mm² die, while the D500 uses the Tahiti chip with 4,313 million transistors on a 352 mm² die. The GTX 690 has a base clock of 915 MHz and a boost of 1019 MHz; the D500 has no listed base or boost clocks. Memory speeds differ: the GTX 690 runs at 6 Gbps effective, the D500 at 5.1 Gbps.

Memory size and bus width favor the D500: 3 GB on a 384-bit bus versus 2 GB on a 256-bit bus. Bandwidth is 243.8 GB/s for the D500 versus 192.3 GB/s for the GTX 690. The GTX 690 has 128 TMUs versus 96 for the D500, but both have 1,536 shading units and 32 ROPs. Pixel rate is 32.61 GPixel/s for the GTX 690 versus 23.20 GPixel/s for the D500; texture rate is 130.4 GTexel/s versus 69.60 GTexel/s. FP32 compute is 3.130 TFLOPS versus 2.227 TFLOPS.

Power and display outputs differ as well. The GTX 690 draws 300 W with a 700 W suggested PSU, while the D500 draws 274 W with a 600 W suggested PSU. The GTX 690 has 3x DVI and 1x mini-DisplayPort 1.2 outputs; the D500 has 6x mini-DisplayPort 1.2 and 1x SDI output. The GTX 690 requires 2x 8-pin power connectors; the D500’s power connector requirements are not specified. Both are dual-slot cards with a 279 mm length. The GTX 690 has a launch MSRP of 999 USD; the D500 has no listed launch MSRP.

Head-to-Head Benchmarks

The only direct comparison is the Geekbench Vulkan test, where the AMD FirePro D500 wins decisively. The D500 scores 18,533, while the GTX 690 scores 16,675, giving the D500 a 10% advantage. This is a substantial gap in a modern API, and it aligns with the D500’s higher percentile rank of 62 versus the GTX 690’s 60.

However, the broader benchmark picture favors the GTX 690 in aggregate. The GTX 690’s average score of 17,037 comes from two tests: Geekbench OpenCL at 17,399 and Geekbench Vulkan at 16,675. The D500’s average of 18,533 comes from a single Vulkan run. When comparing to their respective nearest rivals, the GTX 690 sits within 1% of the NVIDIA GeForce RTX 3070 (deltaPct -1) and 0.3% of the AMD Radeon RX 7600 XT (deltaPct -0.3). The D500 sits 0.5% behind the AMD Radeon RX 560X (deltaPct -0.5) and 0.8% behind the AMD Radeon Pro 5700 XT (deltaPct -0.8). The GTX 690’s dual-GPU design appears to provide a higher peak compute rate, but the D500’s single-GPU implementation is more efficient in Vulkan. The GTX 690 wins on FP32 throughput (3.130 TFLOPS), pixel rate (32.61 GPixel/s), and texture rate (130.4 GTexel/s), but loses on memory bandwidth, which likely contributes to the Vulkan deficit.

The Verdict

The data points to two distinct purchasing decisions. The AMD FirePro D500 is the choice for workloads that prioritize Vulkan performance and memory bandwidth. Its 10% lead in the head-to-head Vulkan test, combined with 243.8 GB/s of bandwidth and 3 GB of VRAM, makes it the better option for modern graphics APIs and applications that are sensitive to memory capacity. Its higher percentile rank (62 versus 60) reinforces this. The D500 also runs cooler on paper, with a 274 W TDP versus the GTX 690’s 300 W, and offers six mini-DisplayPort outputs for multi-monitor setups.

The NVIDIA GeForce GTX 690 is the pick for raw compute throughput and legacy application support. Its FP32 rate of 3.130 TFLOPS is over 40% higher than the D500’s 2.227 TFLOPS, and its texture rate of 130.4 GTexel/s is nearly double the D500’s 69.60 GTexel/s. The GTX 690’s higher average benchmark score (17,037) also suggests it handles a wider range of tasks well, even if it loses to the D500 in Vulkan. It matches the D500 in shading units (1,536) and ROPs (32), but its additional TMUs (128 versus 96) and higher clocks give it a theoretical edge in fill-rate-bound scenes. For users running OpenCL workloads or older DirectX applications, the GTX 690 is the stronger card, despite its higher power draw and larger memory bus deficit. The choice is clear: the D500 for Vulkan and bandwidth, the GTX 690 for compute and fill rate.