AMD Radeon R9 370X vs NVIDIA GeForce GTX 1080 Ti Comparison

AMD
RADEON

AMD Radeon R9 370X

CORE STATE Trinidad
VRAM 2 GB
CLOCK SPEED 1030 MHz
TDP 180 W
BUS WIDTH 256 bit
ARCHITECTURE GCN 1.0
nm
PROCESS 28 nm
LAUNCH DATE 2015
VS
NVIDIA
GEFORCE

GeForce GTX 1080 Ti

CORE STATE GP102
VRAM 11 GB
CLOCK SPEED 1582 MHz
TDP 250 W
BUS WIDTH 352 bit
ARCHITECTURE Pascal
nm
PROCESS 16 nm
LAUNCH DATE 2017

PERFORMANCE BENCHMARKS

geekbench_metal
12,676
30,624
geekbench_opencl
25,893
67,929
geekbench_vulkan
9,018
40,511
3dmark_3dmark_steel_nomad_dx12
N/A
2,231
passmark_directx_10
N/A
118
passmark_directx_11
N/A
151
passmark_directx_12
N/A
66
passmark_directx_9
N/A
231
passmark_g2d
N/A
939
passmark_g3d
N/A
18,600
passmark_gpu_compute
N/A
9,632

Analysis: AMD Radeon R9 370X vs NVIDIA GeForce GTX 1080 Ti

The AMD Radeon R9 370X and the NVIDIA GeForce GTX 1080 Ti represent two vastly different eras of GPU design, separated by nearly two years of architectural evolution. The data reveals a stark performance chasm, with the GTX 1080 Ti winning all three head-to-head benchmark comparisons by enormous margins, yet both cards occupy the same 58th percentile ranking among all GPUs, an intriguing statistical coincidence that masks their true performance difference. The R9 370X, built on AMD's aging GCN 1.0 architecture, delivers an average benchmark score of 15,862, while the GTX 1080 Ti, powered by NVIDIA's Pascal architecture, scores 15,548 on average, a difference of just 2% in the aggregate despite the individual test results telling a completely different story.

Architecture Differences

The architectural divide between these two GPUs is profound and explains nearly every performance discrepancy in the data. The R9 370X uses the Trinidad chip based on GCN 1.0, manufactured on a 28 nm process at TSMC, packing 2,800 million transistors into a 212 mm² die. The GTX 1080 Ti uses the GP102 chip based on Pascal, built on a more advanced 16 nm process, also at TSMC, but cramming 11,800 million transistors into a much larger 471 mm² die. The transistor density tells the story of manufacturing advancement: the R9 370X achieves 13.2M transistors per mm², while the GTX 1080 Ti reaches 25.1M per mm², nearly double the density.

Core configuration differences are equally stark. The R9 370X fields 1,280 shading units, 80 texture mapping units, and 32 ROPs, while the GTX 1080 Ti more than doubles that with 3,584 shading units, 224 TMUs, and 88 ROPs. Clock speeds further widen the gap, with the R9 370X running at a 980 MHz base and 1030 MHz boost, versus the GTX 1080 Ti's 1481 MHz base and 1582 MHz boost. Neither card features ray tracing or tensor cores, as both predate those technologies.

Memory subsystems reveal another generational leap. The R9 370X uses 2 GB of GDDR5 on a 256-bit bus, delivering 179.2 GB/s of bandwidth. The GTX 1080 Ti uses 11 GB of GDDR5X on a 352-bit bus, achieving 484.4 GB/s. The memory clock difference is notable: the R9 370X runs at 1400 MHz (5.6 Gbps effective), while the GTX 1080 Ti runs at 1376 MHz (11 Gbps effective), with the higher effective data rate enabled by the newer GDDR5X standard. The GTX 1080 Ti also supports DirectX 12 (12_1) and Vulkan 1.4, while the R9 370X is limited to DirectX 12 (11_1) and Vulkan 1.2.170, making the NVIDIA card more future-proof for modern API workloads.

Where Each One Wins

Based on the benchmark data, the GTX 1080 Ti wins every single comparison, leaving the R9 370X with zero victories across all tested workloads. The most lopsided result comes in the Geekbench Vulkan test, where the GTX 1080 Ti scores 40,511 against the R9 370X's 9,018, a staggering 77.7% advantage for NVIDIA. This suggests the Pascal architecture's Vulkan implementation is vastly more efficient than GCN 1.0's, a critical factor for modern games and compute workloads that leverage the low-level API.

The Geekbench OpenCL test shows the GTX 1080 Ti scoring 67,929 versus 25,893 for the R9 370X, a 61.9% lead. This compute-heavy workload highlights the raw throughput advantage of the GTX 1080 Ti's 11.34 TFLOPS FP32 performance, compared to the R9 370X's 2.637 TFLOPS. The Geekbench Metal test, while still a decisive NVIDIA win at 30,624 versus 12,676, shows the smallest margin at 58.6%, suggesting the R9 370X's GCN architecture holds up relatively better in Apple's Metal API than in other compute frameworks.

The GTX 1080 Ti also demonstrates superior pixel and texture throughput, with a pixel rate of 139.2 GPixel/s and texture rate of 354.4 GTexel/s, versus the R9 370X's 32.96 GPixel/s and 82.40 GTexel/s. These figures translate directly to real-world gaming performance at high resolutions and detail settings, where the NVIDIA card has clear headroom that the AMD card simply cannot match.

FAQ

Q: Why does the GTX 1080 Ti have such a massive lead in Vulkan performance?

A: The Geekbench Vulkan test shows the GTX 1080 Ti scoring 40,511 versus 9,018 for the R9 370X, a 77.7% difference. This likely stems from the GTX 1080 Ti supporting Vulkan 1.4, while the R9 370X is limited to Vulkan 1.2.170, combined with the Pascal architecture's more efficient command processing and driver overhead.

Q: Are these two cards actually comparable in overall performance?

A: Despite both having a 58th percentile ranking, their average benchmark scores are close (15,862 for the R9 370X vs 15,548 for the GTX 1080 Ti), but this is misleading. The head-to-head results show the GTX 1080 Ti winning by 58.6% to 77.7% in individual tests, indicating the R9 370X's average is likely buoyed by other benchmark scenarios not included in the head-to-head comparison.

Q: What explains the GTX 1080 Ti's superior memory bandwidth?

A: The GTX 1080 Ti uses 11 GB of GDDR5X memory on a 352-bit bus, achieving 484.4 GB/s, while the R9 370X uses 2 GB of GDDR5 on a 256-bit bus for 179.2 GB/s. The combination of a wider bus and faster effective memory speed (11 Gbps vs 5.6 Gbps) gives the NVIDIA card nearly 2.7 times the bandwidth.

Q: Which card has better compute performance for GPGPU workloads?

A: The GTX 1080 Ti dominates in compute, with 11.34 TFLOPS FP32 versus 2.637 TFLOPS for the R9 370X. The Geekbench OpenCL test confirms this, with the GTX 1080 Ti scoring 67,929 versus 25,893 for the AMD card, a 61.9% lead.

Q: Do both cards support the same modern APIs?

A: No. The GTX 1080 Ti supports DirectX 12 (12_1) and Vulkan 1.4, while the R9 370X only supports DirectX 12 (11_1) and Vulkan 1.2.170. Both support OpenGL 4.6, but the NVIDIA card's higher API versions provide better compatibility with newer games and applications.

Q: How do the power requirements compare?

A: The R9 370X has a 180 W TDP and requires a 450 W power supply with 2x 6-pin connectors, while the GTX 1080 Ti has a 250 W TDP and requires a 600 W power supply with 1x 6-pin plus 1x 8-pin connectors. The GTX 1080 Ti draws significantly more power but delivers proportionally much greater performance.

The Verdict

The data presents an unambiguous picture: the NVIDIA GeForce GTX 1080 Ti is in a completely different performance class than the AMD Radeon R9 370X. The GTX 1080 Ti wins all three head-to-head benchmarks with margins ranging from 58.6% to 77.7%, making it the clear choice for anyone prioritizing raw performance in Vulkan, OpenCL, or Metal workloads. The R9 370X, despite its lower specifications, remains competitive only in the context of its much lower launch MSRP of 199 USD versus the GTX 1080 Ti's 699 USD, but the benchmark data shows no scenario where the AMD card wins or even comes close.

For gamers and professionals seeking maximum frame rates and compute throughput, the GTX 1080 Ti is the only logical choice from this pairing. Its 11 GB of memory, 484.4 GB/s bandwidth, and 11.34 TFLOPS FP32 performance provide substantial headroom for demanding applications. The R9 370X's 2 GB memory capacity and 179.2 GB/s bandwidth would likely bottleneck modern games and large datasets, making it suitable only for older or less demanding titles. The GTX 1080 Ti's higher API support (Vulkan 1.4, DirectX 12_1) also ensures better longevity for future software compatibility.

Head-to-Head Benchmarks

The three head-to-head benchmark results illustrate the performance gap with brutal clarity. In Geekbench Metal, the GTX 1080 Ti scores 30,624 against the R9 370X's 12,676, a delta of -58.6% for the AMD card. This is the closest contest between the two, yet it still represents the NVIDIA card delivering more than double the performance. The Metal API appears to be the R9 370X's relative strength, likely due to GCN's mature compute shader handling, but it is nowhere near enough to close the architectural gap.

The Geekbench OpenCL test widens the margin to -61.9%, with the GTX 1080 Ti scoring 67,929 versus 25,893 for the R9 370X. This result aligns with the raw specification differences: the GTX 1080 Ti's 3,584 shading units operating at higher clocks produce 11.34 TFLOPS of FP32 throughput, while the R9 370X's 1,280 shading units at lower clocks produce only 2.637 TFLOPS. The nearly 4.3x difference in theoretical compute power is somewhat tempered in the real-world benchmark, suggesting the R9 370X's GCN architecture has efficient compute scheduling despite lower peak numbers.

The most decisive victory comes in Geekbench Vulkan, where the GTX 1080 Ti scores 40,511 versus 9,018 for the R9 370X, a massive -77.7% delta. This benchmark highlights the Pascal architecture's superior low-level API implementation, likely benefiting from newer driver optimizations and hardware design. The R9 370X's older GCN 1.0 design, combined with its Vulkan 1.2.170 support versus the GTX 1080 Ti's Vulkan 1.4, creates a perfect storm of disadvantages for the AMD card in this test. For developers targeting Vulkan as their primary API, the GTX 1080 Ti is clearly the superior platform based on this data.

Specification Differences

The specification differences between these two cards are extensive and directly explain the benchmark results. The R9 370X uses a 28 nm process with 2,800 million transistors on a 212 mm² die, while the GTX 1080 Ti uses a 16 nm process with 11,800 million transistors on a 471 mm² die. The R9 370X has 1,280 shading units, 80 TMUs, and 32 ROPs, compared to the GTX 1080 Ti's 3,584 shading units, 224 TMUs, and 88 ROPs. Clock speeds differ significantly, with the R9 370X at 980 MHz base and 1030 MHz boost versus the GTX 1080 Ti's 1481 MHz base and 1582 MHz boost.

Memory configuration shows the GTX 1080 Ti's clear advantage: 11 GB of GDDR5X on a 352-bit bus with 484.4 GB/s bandwidth, versus 2 GB of GDDR5 on a 256-bit bus with 179.2 GB/s. The pixel rate is 32.96 GPixel/s for the R9 370X and 139.2 GPixel/s for the GTX 1080 Ti, while texture rates are 82.40 GTexel/s and 354.4 GTexel/s respectively. FP32 performance is 2.637 TFLOPS for the AMD card and 11.34 TFLOPS for the NVIDIA card. Power consumption rises from 180 W TDP with a 450 W suggested PSU for the R9 370X to 250 W TDP with a 600 W suggested PSU for the GTX 1080 Ti. The R9 370X uses 2x 6-pin power connectors, while the GTX 1080 Ti uses 1x 6-pin plus 1x 8-pin. Display outputs differ as well, with the R9 370X offering 2x DVI, 1x HDMI 1.4a, and 1x DisplayPort 1.2, while the GTX 1080 Ti offers 1x HDMI 2.0 and 3x DisplayPort 1.4a. Physical dimensions show the R9 370X at 221 mm length and 111 mm height, while the GTX 1080 Ti is longer at 267 mm with 112 mm height and 40 mm width. Both cards are dual-slot designs with PCIe 3.0 x16 interfaces, but the NVIDIA card's superior everything else makes the comparison one-sided.

DETAILED SPECIFICATIONS

SPECIFICATION
R9 370X
GTX 1080 Ti
Core Specs
Shading Units
1,280
3,584 +180.0%
Shaders
1,280
3,584 +180.0%
TMUs
80
224 +180.0%
ROPs
32
88 +175.0%
Compute Units
24
—
SM Count
—
28
Clocks
Base Clock
980 MHz
1481 MHz
Boost Clock
1030 MHz
1582 MHz
Memory Clock
1400 MHz 5.6 Gbps effective
1376 MHz 11 Gbps effective
Memory
Memory Size
2 GB
11 GB
VRAM (MB)
2,048
11,264 +450.0%
Memory Type
GDDR5
GDDR5X
Memory Bus
256 bit
352 bit
Bandwidth
179.2 GB/s
484.4 GB/s
Cache
L1 Cache
16 KB (per CU)
48 KB (per SM)
L2 Cache
512 KB
2.75 MB
Performance
Pixel Rate
32.96 GPixel/s
139.2 GPixel/s
Texture Rate
82.40 GTexel/s
354.4 GTexel/s
FP32 (TFLOPS)
2.637 TFLOPS
11.34 TFLOPS
FP64 (TFLOPS)
164.8 GFLOPS (1:16)
354.4 GFLOPS (1:32)
FP16 (TFLOPS)
—
177.2 GFLOPS (1:64)
Power
TDP
180 W
250 W
TDP (W)
180
250 +38.9%
Suggested PSU
450 W
600 W
Power Connectors
2x 6-pin
1x 6-pin + 1x 8-pin
Architecture
Architecture
GCN 1.0
Pascal
GPU Name
Trinidad
GP102
Generation
Pirate Islands (R9 300)
GeForce 10
Process Size
28 nm
16 nm
Transistors
2,800 million
11,800 million
Die Size
212 mm²
471 mm²
Foundry
TSMC
TSMC
Density
13.2M / mm²
25.1M / mm²
API Support
DirectX
12 (11_1)
12 (12_1)
OpenGL
4.6
4.6
Vulkan
1.2.170
1.4
OpenCL
2.1 (1.2)
3.0
CUDA
—
6.1
Shader Model
6.5 (5.1)
6.8
Physical
Slot Width
Dual-slot
Dual-slot
Length
221 mm 8.7 inches
267 mm 10.5 inches
Height
111 mm 4.4 inches
112 mm 4.4 inches
Outputs
2x DVI1x HDMI 1.4a1x DisplayPort 1.2
1x HDMI 2.03x DisplayPort 1.4a
Bus Interface
PCIe 3.0 x16
PCIe 3.0 x16
Other
Launch Price
199 USD
699 USD
Production
End-of-life
End-of-life
Predecessor
Volcanic Islands
GeForce 900
Successor
Arctic Islands
GeForce 20
View Radeon R9 370X Details View GeForce GTX 1080 Ti Details