Intel Arc A310E vs Intel Arc Graphics 64EU Comparison

Intel
GPU

Intel Arc A310E

CORE STATE DG2-128
VRAM 4 GB
CLOCK SPEED 2000 MHz
TDP 75 W
BUS WIDTH 64 bit
ARCHITECTURE Xe-HPG
nm
PROCESS 6 nm
LAUNCH DATE 2024
VS
Intel
GPU

Arc Graphics 64EU

CORE STATE Arrow Lake-S
VRAM System Shared
CLOCK SPEED 1900 MHz
TDP 65 W
BUS WIDTH System Shared
ARCHITECTURE Xe-LPG
nm
PROCESS 3 nm
LAUNCH DATE 2024

PERFORMANCE BENCHMARKS

3dmark_3dmark_steel_nomad_dx12
N/A
733

Analysis: Intel Arc A310E vs Intel Arc Graphics 64EU

FAQ

Q: Which GPU has the higher FP32 compute throughput?

A: The Intel Arc A310E delivers 3.072 TFLOPS FP32, while the Intel Arc Graphics 64EU provides 1.946 TFLOPS FP32. The A310E is roughly 58% ahead in raw single-precision compute.

Q: How do the two compare in memory configuration?

A: The Arc A310E uses 4 GB of dedicated GDDR6 memory on a 64-bit bus with 124.0 GB/s bandwidth. The Arc Graphics 64EU uses System Shared memory, with bus width and bandwidth listed as System Shared and System Dependent respectively.

Q: What is the transistor density difference between the two chips?

A: The Arc A310E, built on TSMC 6 nm, has a transistor density of 45.9M per mm² with 7,200 million transistors on a 157 mm² die. The Arc Graphics 64EU, built on TSMC 3 nm, has a density of 73.3M per mm² with 17,800 million transistors on a 243 mm² die.

Q: Which product has a higher pixel rate?

A: The Arc A310E achieves 32.00 GPixel/s, slightly ahead of the Arc Graphics 64EU's 30.40 GPixel/s.

Q: What is the difference in ray tracing support?

A: The Arc A310E includes 6 dedicated RT cores. The Arc Graphics 64EU does not list any RT cores in its specifications.

Q: How do the two compare in terms of shading units?

A: The Arc A310E has 768 shading units, while the Arc Graphics 64EU has 512 shading units. Both have 32 TMUs and 16 ROPs.

Where Each One Wins

The Intel Arc A310E is the clear winner in every measurable compute and graphics throughput category recorded in the database. Its FP32 throughput of 3.072 TFLOPS exceeds the Arc Graphics 64EU's 1.946 TFLOPS by 58%. The A310E also leads in pixel rate (32.00 GPixel/s versus 30.40 GPixel/s) and texture rate (64.00 GTexel/s versus 60.80 GTexel/s). FP16 performance follows the same pattern, with the A310E delivering 6.144 TFLOPS versus 3.891 TFLOPS for the 64EU.

The Arc A310E also wins on memory architecture. Dedicated 4 GB GDDR6 with 124.0 GB/s bandwidth provides predictable performance for workloads that depend on memory latency and bandwidth. The Arc Graphics 64EU relies on System Shared memory, making its effective bandwidth dependent on the host platform's memory subsystem.

The Arc Graphics 64EU holds advantages in process technology and power envelope. Its 3 nm TSMC node gives it a higher transistor density (73.3M per mm² versus 45.9M per mm²). Its 65 W TDP is lower than the A310E's 75 W TDP. It also has a much lower base clock of 300 MHz, although the boost clock reaches 1900 MHz, close to the A310E's fixed 2000 MHz.

For integrated graphics use cases, the Arc Graphics 64EU occupies a specific niche: it is an IGP solution with a Ring Bus interface, meaning it requires no separate power connectors and uses system memory. The Arc A310E is a discrete single-slot card with 4x mini-DisplayPort 2.0 outputs and a PCIe 4.0 x8 interface.

Architecture Differences

The two GPUs come from different architectural generations within Intel's lineup. The Arc A310E uses the DG2-128 chip based on the Xe-HPG architecture, part of the Alchemist (Arc 3) generation. The Arc Graphics 64EU uses the Arrow Lake-S chip based on the Xe-LPG architecture, part of the Arc Graphics (Arrow Lake) generation.

The Xe-HPG architecture in the A310E was designed for discrete graphics, with dedicated hardware resources including 6 RT cores. The Xe-LPG architecture in the 64EU is an integrated graphics solution without listed RT cores. This architectural difference explains the compute gap: the A310E has 768 shading units versus 512 for the 64EU, despite both having 32 TMUs and 16 ROPs.

Process technology differs significantly. The A310E uses TSMC 6 nm with 7,200 million transistors on a 157 mm² die. The 64EU uses TSMC 3 nm with 17,800 million transistors on a 243 mm² die. The 64EU's die is larger because it is part of a broader Arrow Lake-S processor, not a standalone GPU die. The transistor density is higher on 3 nm (73.3M per mm² versus 45.9M per mm²), reflecting the more advanced node.

Clock behavior also differs. The A310E has a fixed clock of 2000 MHz for both base and boost. The 64EU has a base clock of just 300 MHz with a boost clock of 1900 MHz. This suggests the integrated GPU scales its frequency dynamically based on thermal and power headroom.

Memory architecture is fundamentally different. The A310E uses dedicated GDDR6 with a 64-bit bus and a specific memory clock of 1937 MHz (15.5 Gbps effective). The 64EU uses System Shared memory, with all memory parameters marked as System Shared or System Dependent. This makes the 64EU's memory performance entirely dependent on the host system's RAM.

Specification Differences

| Specification | Intel Arc A310E | Intel Arc Graphics 64EU |

|---|---|---|

| Architecture | Xe-HPG | Xe-LPG |

| Generation | Alchemist (Arc 3) | Arc Graphics (Arrow Lake) |

| Process Node | 6 nm | 3 nm |

| Transistors | 7,200 million | 17,800 million |

| Die Size | 157 mm² | 243 mm² |

| Transistor Density | 45.9M / mm² | 73.3M / mm² |

| Base Clock | 2000 MHz | 300 MHz |

| Boost Clock | 2000 MHz | 1900 MHz |

| Memory Size | 4 GB | System Shared |

| Memory Type | GDDR6 | System Shared |

| Memory Bus Width | 64 bit | System Shared |

| Memory Bandwidth | 124.0 GB/s | System Dependent |

| Shading Units | 768 | 512 |

| RT Cores | 6 | null |

| Pixel Rate | 32.00 GPixel/s | 30.40 GPixel/s |

| Texture Rate | 64.00 GTexel/s | 60.80 GTexel/s |

| FP32 | 3.072 TFLOPS | 1.946 TFLOPS |

| FP16 | 6.144 TFLOPS | 3.891 TFLOPS |

| TDP | 75 W | 65 W |

| Slot Width | Single-slot | IGP |

| Power Connectors | None | null |

| Bus Interface | PCIe 4.0 x8 | Ring Bus |

| Display Outputs | 4x mini-DisplayPort 2.0 | Motherboard Dependent |

| Production Status | End-of-life | Active |

| Release Date | 2024-03-31 | 2024-10-23 |

Head-to-Head Benchmarks

The database records one benchmark for the Arc Graphics 64EU: 3DMark Steel Nomad DX12 with a score of 733. The Arc A310E has no recorded benchmark scores in the database, and the head-to-head benchmark list is empty. Consequently, direct comparison must rely on the 64EU's performance percentile and nearest rivals.

The Arc Graphics 64EU sits at the 3rd percentile of all GPUs in the database. Its nearest rival is the Intel Arc Graphics 32EU, which also scores 733, a delta of 0%. The Intel Arc Graphics 24EU also matches at 733, again with a delta of 0%. These three integrated GPUs from Intel's Arc lineup produce identical scores in this test.

The AMD Radeon HD 6470M scores 723, putting the 64EU 1.4% ahead. The NVIDIA GeForce GT 415M scores 751, putting the 64EU 2.4% behind. These deltas are small, indicating that the 64EU's performance class is comparable to legacy mobile discrete GPUs from over a decade ago.

The Arc A310E's absence from the benchmark database means its 50th percentile ranking is based on specification-derived data rather than measured results. Its compute resources, however, suggest it would significantly outperform the 64EU if measured in the same test. The A310E has 50% more shading units (768 versus 512), 58% higher FP32 throughput, and 5.3% higher pixel rate. Its dedicated memory bandwidth of 124.0 GB/s removes the system memory bottleneck that constrains the 64EU.

The 64EU's 733 score places it in the bottom 3% of all GPUs. This is consistent with its IGP design and 512 shading units. The A310E, with 768 shading units and a 2000 MHz clock, should score substantially higher, though no direct measurement exists to quantify the gap.

The nearest rival data for the 64EU shows clustering at the 733 score level. The 32EU and 24EU variants, which have fewer shading units by implication of their naming, achieve identical scores. This suggests the 3DMark Steel Nomad test may be memory-bound or otherwise limited for these integrated parts, preventing higher-resource variants from differentiating themselves.

In contrast, the A310E's dedicated 4 GB GDDR6 at 124.0 GB/s provides a memory bandwidth advantage that no integrated GPU can match, regardless of the host system's RAM speed. This architectural separation means the A310E can sustain texture and compute throughput without contending with CPU memory traffic.

The production status difference matters for long-term availability. The A310E is end-of-life, with a successor in Battlemage. The 64EU is active, released in October 2024 as part of the Arrow Lake platform. The A310E was released in March 2024, roughly seven months earlier.

Both GPUs support DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4, so API compatibility is identical. The A310E offers 4x mini-DisplayPort 2.0 outputs, while the 64EU's display capabilities are motherboard dependent, meaning they vary with the host board's implementation.

The A310E requires a 250 W suggested PSU, though it uses no power connectors itself. The 64EU, as an IGP, has no PSU requirement listed. The A310E's 75 W TDP versus the 64EU's 65 W TDP is a modest difference, but the discrete card's dedicated VRM and cooling allow sustained operation at its 2000 MHz clock, whereas the IGP's 1900 MHz boost is subject to the host processor's thermal budget.

DETAILED SPECIFICATIONS

SPECIFICATION
A310E
Graphics 64EU
Core Specs
Shading Units
768
512 -33.3%
Shaders
768
512 -33.3%
TMUs
32
32 0.0%
ROPs
16
16 0.0%
Execution Units
96
64 -33.3%
Clocks
Base Clock
2000 MHz
300 MHz
Boost Clock
2000 MHz
1900 MHz
Memory Clock
1937 MHz 15.5 Gbps effective
System Shared
Memory
Memory Size
4 GB
System Shared
VRAM (MB)
4,096
Memory Type
GDDR6
System Shared
Memory Bus
64 bit
System Shared
Bandwidth
124.0 GB/s
System Dependent
Cache
L2 Cache
4 MB
Performance
Pixel Rate
32.00 GPixel/s
30.40 GPixel/s
Texture Rate
64.00 GTexel/s
60.80 GTexel/s
FP32 (TFLOPS)
3.072 TFLOPS
1.946 TFLOPS
FP64 (TFLOPS)
768.0 GFLOPS (1:4)
FP16 (TFLOPS)
6.144 TFLOPS (2:1)
3.891 TFLOPS (2:1)
AI/RT
RT Cores
6
XMX Cores
96
Power
TDP
75 W
65 W
TDP (W)
75
65 -13.3%
Suggested PSU
250 W
Power Connectors
None
Architecture
Architecture
Xe-HPG
Xe-LPG
GPU Name
DG2-128
Arrow Lake-S
Generation
Alchemist (Arc 3)
Arc Graphics (Arrow Lake)
Process Size
6 nm
3 nm
Transistors
7,200 million
17,800 million
Die Size
157 mm²
243 mm²
Foundry
TSMC
TSMC
Density
45.9M / mm²
73.3M / mm²
API Support
DirectX
12 Ultimate (12_2)
12 Ultimate (12_2)
OpenGL
4.6
4.6
Vulkan
1.4
1.4
OpenCL
3.0
3.0
Shader Model
6.6
6.8
Physical
Slot Width
Single-slot
IGP
Length
168 mm 6.6 inches
Height
69 mm 2.7 inches
Outputs
4x mini-DisplayPort 2.0
Motherboard Dependent
Bus Interface
PCIe 4.0 x8
Ring Bus
Other
Production
End-of-life
Active
Predecessor
Xe Graphics
HD Graphics
Successor
Battlemage
View Arc A310E Details View Arc Graphics 64EU Details