GPU Comparison

Intel
GPU

Intel Arc B580

CORE STATE BMG-G21
VRAM 12 GB
CLOCK SPEED 2670 MHz
TDP 190 W
BUS WIDTH 192 bit
ARCHITECTURE Xe2-HPG
nm
PROCESS 5 nm
LAUNCH DATE 2024
VS
NVIDIA
GEFORCE

GeForce GTX 1630

CORE STATE TU117
VRAM 4 GB
CLOCK SPEED 1785 MHz
TDP 75 W
BUS WIDTH 64 bit
ARCHITECTURE Turing
nm
PROCESS 12 nm
LAUNCH DATE 2022

PERFORMANCE BENCHMARKS

3dmark_3dmark_steel_nomad_dx12
3,068
N/A
geekbench_opencl
92,821
24,858
geekbench_vulkan
109,672
23,695
passmark_directx_10
76
N/A
passmark_directx_11
128
N/A
passmark_directx_12
76
N/A
passmark_directx_9
183
N/A
passmark_g2d
709
N/A
passmark_g3d
15,748
N/A
passmark_gpu_compute
7,729
N/A

Analysis: Intel Arc B580 vs NVIDIA GeForce GTX 1630

NVIDIA GeForce GTX 1630 vs Intel Arc B580: A comparison of two graphics cards that occupy entirely different performance tiers. The GTX 1630 is an end-of-life Turing-era entry card, while the Arc B580 is an active Battlemage-generation mainstream card with a launch MSRP of 249 USD. Benchmark data shows the Intel card winning all shared tests by overwhelming margins.

The Verdict

The data presents a clear hierarchy. For any user deciding between these two cards, the Intel Arc B580 is the overwhelming choice based on raw performance. In the two benchmarks both cards share, Geekbench OpenCL and Vulkan, the B580 leads by 73.2% and 78.4%, respectively. There is no single metric in the FACT PACK where the GTX 1630 comes out ahead. The GTX 1630 sits at the 70th percentile of all GPUs, while the B580 sits at the 68th; their average benchmark scores are 24,277 and 23,021, respectively. This means the GTX 1630 actually has a slightly higher overall percentile rank, but that rank is based on a smaller benchmark set and does not reflect the head-to-head results where the B580 dominates.

Choose the GTX 1630 only if constraints like power delivery, physical size, or legacy interface compatibility are absolute dealbreakers. It draws a 75 W TDP, requires no power connectors, is single-slot, and measures 145 mm in length. It also uses a PCIe 3.0 x16 interface. The Arc B580, by contrast, demands a 190 W TDP, a single 8-pin connector, a suggested 450 W PSU, is dual-slot, and measures 272 mm. If your system can physically and electrically accommodate the B580, the benchmark results make the decision obvious.

FAQ

Q: Which card has higher raw compute performance?

A: The Intel Arc B580. Its FP32 throughput is 13.67 TFLOPS versus 1.828 TFLOPS for the GTX 1630. Its texture rate is 427.2 GTexel/s versus 57.12 GTexel/s, and its pixel rate is 213.6 GPixel/s versus 28.56 GPixel/s.

Q: How do the two compare in memory capacity and bandwidth?

A: The Arc B580 has 12 GB of GDDR6 on a 192-bit bus, delivering 456.0 GB/s. The GTX 1630 has 4 GB of GDDR6 on a 64-bit bus, delivering 96.00 GB/s. The B580 has three times the capacity and nearly five times the bandwidth.

Q: Which card supports newer display outputs?

A: The Arc B580 supports 1x HDMI 2.1a and 3x DisplayPort 2.1. The GTX 1630 supports 1x DVI, 1x HDMI 2.0, and 1x DisplayPort 1.4a. The B580's outputs are more modern.

Q: What are the power requirements for each card?

A: The GTX 1630 has a 75 W TDP, a suggested PSU of 250 W, and requires no power connectors. The Arc B580 has a 190 W TDP, a suggested PSU of 450 W, and requires one 8-pin power connector.

Q: Are there any benchmark tests where the GTX 1630 wins?

A: No. In the shared head-to-head tests (Geekbench OpenCL and Vulkan), the Intel Arc B580 wins both. The GTX 1630 has zero wins in the head-to-head data.

Q: How does the GTX 1630's benchmark score compare to its nearest rivals?

A: The GTX 1630's average score of 24,277 is 0.2% above the GTX 780 Ti (24,236), 0.4% above the RTX 2080 SUPER (24,170), 0.7% below the RX 6600 XT (24,442), and 0.9% above the RX 6800S (24,063).

Architecture Differences

The GTX 1630 uses the TU117 chip built on Turing architecture, manufactured by TSMC on a 12 nm process. It contains 4,700 million transistors on a 200 mm² die, yielding a transistor density of 23.5M per mm². It has no dedicated ray tracing cores and no tensor cores.

The Intel Arc B580 uses the BMG-G21 chip built on Xe2-HPG architecture, manufactured by TSMC on a 5 nm process. It contains 19,600 million transistors on a 272 mm² die, yielding a transistor density of 72.1M per mm². It has 20 ray tracing cores and no tensor cores listed.

The B580 uses a second-generation Intel GPU architecture focused on efficiency and feature completeness. It supports DirectX 12 Ultimate (12_2), while the GTX 1630 supports DirectX 12 (12_1). Both support OpenGL 4.6 and Vulkan 1.4. The B580's transistor density is over three times higher, reflecting the more advanced 5 nm node.

Specification Differences

The two cards differ in nearly every measurable specification:

  • Process node: 12 nm (GTX 1630) vs 5 nm (Arc B580)
  • Transistors: 4,700 million vs 19,600 million
  • Die size: 200 mm² vs 272 mm²
  • Base/boost clock: 1740/1785 MHz vs 2670/2670 MHz
  • Memory size: 4 GB vs 12 GB
  • Memory bus width: 64 bit vs 192 bit
  • Memory bandwidth: 96.00 GB/s vs 456.0 GB/s
  • Memory speed: 12 Gbps effective vs 19 Gbps effective
  • Shading units: 512 vs 2560
  • Texture mapping units: 32 vs 160
  • Render output units: 16 vs 80
  • Ray tracing cores: 0 vs 20
  • Pixel rate: 28.56 GPixel/s vs 213.6 GPixel/s
  • Texture rate: 57.12 GTexel/s vs 427.2 GTexel/s
  • FP32 performance: 1.828 TFLOPS vs 13.67 TFLOPS
  • FP16 performance: 3.656 TFLOPS vs 27.34 TFLOPS (both 2:1 ratio)
  • TDP: 75 W vs 190 W
  • Slot width: Single-slot vs Dual-slot
  • Power connectors: None vs 1x 8-pin
  • Suggested PSU: 250 W vs 450 W
  • Bus interface: PCIe 3.0 x16 vs PCIe 4.0 x8
  • Display outputs: 1x DVI, 1x HDMI 2.0, 1x DisplayPort 1.4a vs 1x HDMI 2.1a, 3x DisplayPort 2.1
  • Physical dimensions: 145 x 69 x 18 mm vs 272 x 115 x 45 mm
  • Production status: End-of-life vs Active
  • Release date: 2022-06-27 vs 2024-12-12

The B580 has five times the shading units, five times the TMUs, and five times the ROPs compared to the GTX 1630.

Head-to-Head Benchmarks

The shared benchmark data shows a one-sided contest. In Geekbench OpenCL, the GTX 1630 scores 24,858 while the Arc B580 scores 92,821. The delta is -73.2%, meaning the B580 performs roughly 3.7 times better. In Geekbench Vulkan, the GTX 1630 scores 23,695 while the B580 scores 109,672. The delta is -78.4%, meaning the B580 performs roughly 4.6 times better.

The Arc B580 also has results in other benchmark suites that the GTX 1630 lacks entirely. In Passmark G3D, it scores 15,748. In Passmark GPU Compute, it scores 7,729. In 3DMark Steel Nomad DX12, it scores 3,068. These additional data points confirm the B580 is a much more capable card across multiple testing methodologies.

The GTX 1630's only benchmark entries are the two Geekbench tests, both of which it loses decisively. Its average benchmark score of 24,277 is derived from those two results. The B580's average score of 23,021 is pulled down by its Passmark DirectX tests (9, 10, 11, 12 scores ranging from 76 to 183), which are likely not directly comparable to its other results. Even with that drag, the B580's OpenCL and Vulkan scores are so far ahead that the comparison remains unambiguous.

Where Each One Wins

Intel Arc B580, performance-critical workloads. Every single benchmark in the shared set goes to the B580. Its 12 GB memory and 456.0 GB/s bandwidth suit modern game assets and compute tasks. Its 20 ray tracing cores enable hardware-accelerated ray tracing, a feature the GTX 1630 lacks entirely. The B580's FP32 throughput of 13.67 TFLOPS is roughly 7.5 times that of the GTX 1630. For any workload that stresses GPU compute, texture throughput, or pixel fill rate, the B580 is the clear winner. It also supports DirectX 12 Ultimate, which includes features like mesh shaders and variable rate shading, compared to the GTX 1630's DirectX 12_1 support.

NVIDIA GeForce GTX 1630, constrained legacy systems. The GTX 1630 wins on physical and electrical compatibility. Its 75 W TDP means it runs off the PCIe slot alone, with no auxiliary power connector needed. Its 145 mm length, single-slot profile, and 18 mm width fit in small form factor cases. Its PCIe 3.0 x16 interface works in older motherboards. Its 4 GB memory and 96.00 GB/s bandwidth are modest but sufficient for low-resolution or older titles. The card is end-of-life, so it is not a future-proof option, but for a system that cannot accommodate a dual-slot, 272 mm, 190 W card, the GTX 1630 is the only one of the two that will physically work.

The data does not support any scenario where the GTX 1630 outperforms the B580 in compute or graphics benchmarks. Its only advantages are form factor, power draw, and interface compatibility. If those constraints apply, the GTX 1630 is a functional choice. If they do not, the B580 is the superior card by every measurable metric in the FACT PACK.

DETAILED SPECIFICATIONS

SPECIFICATION
B580
GTX 1630
Core Specs
Shading Units
2,560
512 -80.0%
Shaders
2,560
512 -80.0%
TMUs
160
32 -80.0%
ROPs
80
16 -80.0%
SM Count
8
Execution Units
20
Clocks
Base Clock
2670 MHz
1740 MHz
Boost Clock
2670 MHz
1785 MHz
Memory Clock
2375 MHz 19 Gbps effective
1500 MHz 12 Gbps effective
Memory
Memory Size
12 GB
4 GB
VRAM (MB)
12,288
4,096 -66.7%
Memory Type
GDDR6
GDDR6
Memory Bus
192 bit
64 bit
Bandwidth
456.0 GB/s
96.00 GB/s
Cache
L1 Cache
256 KB (per EU)
64 KB (per SM)
L2 Cache
18 MB
1024 KB
Performance
Pixel Rate
213.6 GPixel/s
28.56 GPixel/s
Texture Rate
427.2 GTexel/s
57.12 GTexel/s
FP32 (TFLOPS)
13.67 TFLOPS
1.828 TFLOPS
FP64 (TFLOPS)
854.4 GFLOPS (1:16)
57.12 GFLOPS (1:32)
FP16 (TFLOPS)
27.34 TFLOPS (2:1)
3.656 TFLOPS (2:1)
AI/RT
RT Cores
20
XMX Cores
160
Power
TDP
190 W
75 W
TDP (W)
190
75 -60.5%
Suggested PSU
450 W
250 W
Power Connectors
1x 8-pin
None
Architecture
Architecture
Xe2-HPG
Turing
GPU Name
BMG-G21
TU117
Generation
Battlemage (Arc 5)
GeForce 16
Process Size
5 nm
12 nm
Transistors
19,600 million
4,700 million
Die Size
272 mm²
200 mm²
Foundry
TSMC
TSMC
Density
72.1M / mm²
23.5M / mm²
API Support
DirectX
12 Ultimate (12_2)
12 (12_1)
OpenGL
4.6
4.6
Vulkan
1.4
1.4
OpenCL
3.0
3.0
CUDA
7.5
Shader Model
6.6
6.8
Physical
Slot Width
Dual-slot
Single-slot
Length
272 mm 10.7 inches
145 mm 5.7 inches
Height
115 mm 4.5 inches
69 mm 2.7 inches
Outputs
1x HDMI 2.1a3x DisplayPort 2.1
1x DVI1x HDMI 2.01x DisplayPort 1.4a
Bus Interface
PCIe 4.0 x8
PCIe 3.0 x16
Other
Launch Price
249 USD
Production
Active
End-of-life
Predecessor
Alchemist
GeForce 10
Successor
GeForce 20
View Arc B580 Details View GeForce GTX 1630 Details