AMD Radeon Pro 455 vs NVIDIA GeForce GTX 680 Comparison

The Verdict

The NVIDIA GeForce GTX 680 and AMD Radeon Pro 455 serve completely different roles despite their similar average benchmark scores. The GTX 680 takes two of three head-to-head tests decisively, while the Radeon Pro 455 dominates the remaining test by an even larger margin. The GTX 680 is the pick for OpenCL and Vulkan compute workloads, where it leads by 63.6% and 44.2% respectively. The Radeon Pro 455 is the pick for Metal-based applications, where it leads by 50.4%. The overall database average tells a close story: the GTX 680 sits at 14150 against the Pro 455's 12831, a gap of roughly 10% in favor of the older NVIDIA card. However, the percentile rankings are nearly even, with the GTX 680 at the 55th percentile of all GPUs and the Pro 455 at the 52nd. The GTX 680's nearest rivals include the NVIDIA Tesla K10 (0.9% behind) and the GeForce GTX 1070 Ti (0.9% ahead), placing it in a competitive mid-range cluster. The Pro 455 sits in a similar band, with the GeForce GTX 590 matching it exactly at 0% delta and the GTX 670 just 0.5% ahead. For raw compute throughput in OpenCL and Vulkan, the GTX 680 wins clearly. For Metal-centric workflows, the Pro 455 is the only rational choice. The GTX 680 is end-of-life, as is the Pro 455, so neither is a future-proof investment, but each remains relevant in its niche.

Architecture Differences

The GTX 680 uses the GK104 chip built on Kepler architecture, fabricated by TSMC on a 28 nm process. The Pro 455 uses the Baffin chip built on GCN 4.0 architecture, fabricated by GlobalFoundries on a 14 nm process. The transistor counts are similar, with the GTX 680 at 3,540 million and the Pro 455 at 3,000 million, but the die sizes diverge sharply: the GTX 680 measures 294 mm² while the Pro 455 is just 123 mm². This yields a transistor density of 12.0M per mm² for the GTX 680 versus 24.4M per mm² for the Pro 455, reflecting the newer manufacturing node. The GTX 680 carries 1536 shading units, 128 texture mapping units, and 32 ROPs. The Pro 455 is much leaner with 768 shading units, 48 TMUs, and 16 ROPs. Clock behavior differs too: the GTX 680 has a base clock of 1006 MHz and a boost clock of 1058 MHz, while the database records no base or boost clock for the Pro 455, only a memory clock of 1270 MHz with 5.1 Gbps effective. The GTX 680's memory runs at 1502 MHz with 6 Gbps effective. The GTX 680 supports DirectX 12 (11_0), OpenGL 4.6, and Vulkan 1.2.175. The Pro 455 supports DirectX 12 (12_0), OpenGL 4.6, and Vulkan 1.3, giving it a more recent API feature level. The GTX 680 is a dual-slot card requiring 2x 6-pin power connectors and a 450 W PSU, while the Pro 455 is an MXM module with no power connectors and a 35 W TDP. The GTX 680 draws 195 W. The bus interface also differs: PCIe 3.0 x16 for the GTX 680 versus PCIe 3.0 x8 for the Pro 455.

Head-to-Head Benchmarks

The Geekbench Metal test delivers the largest single delta in the comparison. The Radeon Pro 455 scores 15916 against the GTX 680's 7897, a 50.4% advantage for AMD. This is not a marginal edge; it is a near doubling of performance. The Pro 455's GCN 4.0 architecture clearly maps better to Metal workloads, which matters for macOS environments. The Geekbench OpenCL test flips the result completely. The GTX 680 scores 16906 against the Pro 455's 10336, a 63.6% lead for NVIDIA. This is the biggest win for the GTX 680 in the entire comparison. The Geekbench Vulkan test also goes to the GTX 680, with a score of 17646 against 12240, a 44.2% advantage. The GTX 680 wins two of three tests, but the Pro 455 wins the one test that matters most for Metal-specific software. The average benchmark score reinforces the GTX 680's overall edge: 14150 versus 12831. The GTX 680's nearest rival, the Tesla K10 at 14029, sits 0.9% behind, while the GTX 1070 Ti at 14277 sits 0.9% ahead. The Pro 455's nearest rival, the GTX 590 at 12830, matches it exactly at 0% delta, and the FirePro W5100 at 12847 trails by 0.1%. These clustering patterns show both cards operating in a tightly packed performance tier, yet the head-to-head results reveal that API-specific behavior separates them far more than average scores suggest.

FAQ

Q: Which GPU wins in Metal benchmarks?

A: The AMD Radeon Pro 455 wins decisively, scoring 15916 in Geekbench Metal against the GTX 680's 7897. That is a 50.4% advantage.

Q: Which GPU wins in OpenCL benchmarks?

A: The NVIDIA GeForce GTX 680 wins by a wide margin, scoring 16906 against the Pro 455's 10336. The delta is 63.6% in NVIDIA's favor.

Q: How do the two GPUs compare in Vulkan performance?

A: The GTX 680 leads with a score of 17646 versus 12240 for the Pro 455, a 44.2% advantage.

Q: What are the average benchmark scores for each GPU?

A: The GTX 680 has an average benchmark score of 14150, while the Pro 455 has an average benchmark score of 12831.

Q: How do these GPUs rank against all other GPUs in the database?

A: The GTX 680 sits at the 55th percentile of all GPUs, while the Pro 455 sits at the 52nd percentile.

Q: What are the power consumption figures for each card?

A: The GTX 680 has a TDP of 195 W, while the Pro 455 has a TDP of 35 W. The GTX 680 requires a 450 W PSU and dual 6-pin connectors, while the Pro 455 is an MXM module with no power connectors.

Where Each One Wins

The GTX 680 wins in OpenCL and Vulkan. Its OpenCL score of 16906 is 63.6% higher than the Pro 455's 10336. Its Vulkan score of 17646 is 44.2% higher than the Pro 455's 12240. This makes the GTX 680 the stronger choice for general-purpose compute workloads, cross-platform GPU tasks, and Vulkan-based applications. The GTX 680 also holds a higher average benchmark score (14150 versus 12831) and a higher percentile ranking (55th versus 52nd). Its nearest rivals include the Tesla K10 and the GTX 1070 Ti, placing it in a performance tier where it competes with much newer hardware despite its 2012 release. The Pro 455 wins in Metal. Its score of 15916 is 50.4% higher than the GTX 680's 7897. This makes it the clear choice for macOS applications and any workload that leverages Apple's Metal API. The Pro 455 also offers a much lower power envelope at 35 W, making it suitable for portable or compact systems where the GTX 680's 195 W requirement would be impractical. The Pro 455's nearest rivals include the GTX 590 and the FirePro W5100, and its 0% delta with the GTX 590 shows it trading blows with a dual-GPU flagship from an earlier era. The use-case split is clean: compute-heavy, cross-platform workloads favor the GTX 680; Metal-centric, low-power environments favor the Pro 455.

Specification Differences

The two GPUs differ across nearly every major specification category. The process nodes are different: 28 nm for the GTX 680 versus 14 nm for the Pro 455. The foundries differ as well, with TSMC producing the GTX 680 and GlobalFoundries producing the Pro 455. Transistor counts are close but not equal: 3,540 million for the GTX 680 versus 3,000 million for the Pro 455. Die sizes diverge substantially: 294 mm² versus 123 mm², leading to transistor densities of 12.0M per mm² and 24.4M per mm² respectively. The GTX 680 has a base clock of 1006 MHz and a boost clock of 1058 MHz; the Pro 455 has no recorded base or boost clock. Memory clocks also differ: 1502 MHz with 6 Gbps effective for the GTX 680 versus 1270 MHz with 5.1 Gbps effective for the Pro 455. Both cards have 2 GB of GDDR5 memory, but the bus widths differ at 256 bit versus 128 bit. This produces a large bandwidth gap: 192.3 GB/s for the GTX 680 versus 81.28 GB/s for the Pro 455. The shading unit counts are 1536 versus 768, TMU counts are 128 versus 48, and ROP counts are 32 versus 16. Pixel rates are 33.86 GPixel/s for the GTX 680 versus 13.68 GPixel/s for the Pro 455. Texture rates are 135.4 GTexel/s versus 41.04 GTexel/s. FP32 performance is 3.250 TFLOPS for the GTX 680 versus 1,313.3 GFLOPS for the Pro 455. The Pro 455 has FP16 performance of 1,313.3 GFLOPS with a 1:1 ratio, while the GTX 680 has no recorded FP16 figure. TDPs are 195 W versus 35 W. The GTX 680 is a dual-slot card with 2x 6-pin power connectors and a 450 W suggested PSU; the Pro 455 is an MXM module with no power connectors and no suggested PSU. The GTX 680 uses PCIe 3.0 x16 while the Pro 455 uses PCIe 3.0 x8. Display outputs are 2x DVI, 1x HDMI 1.4a, and 1x DisplayPort 1.2 for the GTX 680, while the Pro 455's outputs are portable device dependent. API support differs in DirectX and Vulkan: the GTX 680 supports DirectX 12 (11_0) and Vulkan 1.2.175, while the Pro 455 supports DirectX 12 (12_0) and Vulkan 1.3. Both support OpenGL 4.6. The GTX 680 measures 256 mm in length, 111 mm in height, and 38 mm in width; the Pro 455 has no recorded dimensions. The GTX 680 was released on 2012-03-21 with a launch MSRP of 499 USD, while the Pro 455 was released on 2016-10-29 with no launch MSRP recorded. Both are end-of-life products.