Intel Arc B770 vs Intel Arc Graphics 32EU Comparison

Intel
GPU

Intel Arc B770

CORE STATE BMG-G31
VRAM 16 GB
CLOCK SPEED 2400 MHz
TDP 225 W
BUS WIDTH 256 bit
ARCHITECTURE Xe2-HPG
nm
PROCESS 5 nm
LAUNCH DATE 2026
VS
Intel
GPU

Arc Graphics 32EU

CORE STATE Arrow Lake-S
VRAM System Shared
CLOCK SPEED 1950 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 B770 vs Intel Arc Graphics 32EU

Where Each One Wins

The Intel Arc B770 and the Intel Arc Graphics 32EU occupy completely different segments of the GPU spectrum, and the data reflects this clearly. The Arc B770 is a discrete, dual-slot add-in board with a 225 W TDP, while the Arc Graphics 32EU is an integrated graphics processor (IGP) with a 65 W TDP that relies on system memory. This fundamental distinction drives every performance comparison.

The Arc B770 has no recorded benchmark scores in the database, meaning its percentile standing is listed at 50, which places it at the midpoint of the database's GPU population. The Arc Graphics 32EU, by contrast, has one recorded measurement: a 3DMark Steel Nomad DX12 score of 733. That score places the IGP at the 3rd percentile of all GPUs, indicating that it sits near the very bottom of the performance distribution.

The wins for each part are therefore defined by their roles rather than direct competition. The Arc B770 wins in raw compute throughput, memory bandwidth, and feature execution. The Arc Graphics 32EU wins in integration efficiency, power draw, and platform simplicity. Its 65 W TDP is under a third of the Arc B770's 225 W envelope, and its system-shared memory approach eliminates the need for dedicated video memory.

In terms of processing resources, the Arc B770 carries 4096 shading units, 256 texture mapping units, and 128 raster operation units. The Arc Graphics 32EU provides only 256 shading units, 16 TMUs, and 8 ROPs. That is a 16x difference in shading units, TMUs, and ROPs. The Arc B770 also includes 32 dedicated ray tracing cores, while the Arc Graphics 32EU lists no ray tracing cores at all.

Pixel and texture throughput follow the same pattern. The Arc B770 delivers 307.2 GPixel/s and 614.4 GTexel/s. The Arc Graphics 32EU manages 15.60 GPixel/s and 31.20 GTexel/s. The discrete card is roughly 19.7x faster in pixel fill and 19.7x faster in texture fill.

Compute figures reinforce the gap. The Arc B770 reaches 19.66 TFLOPS FP32 and 39.32 TFLOPS FP16 (2:1). The integrated part produces 998.4 GFLOPS FP32 and 1.997 TFLOPS FP16 (2:1). The Arc B770 is about 19.7x ahead in FP32 and roughly 19.7x ahead in FP16. The near-identical ratios across shading units, pixel rate, texture rate, and FP32 compute indicate that the architecture scales linearly with execution resource count.

Memory is another decisive differentiator. The Arc B770 uses 16 GB of GDDR6 across a 256-bit bus, yielding 512.0 GB/s of bandwidth. The Arc Graphics 32EU has no dedicated memory; it shares system memory, and bandwidth is listed as system dependent. This makes the discrete card suitable for workloads that demand high sustained bandwidth, while the IGP is limited by platform memory configuration.

Architecture Differences

The two GPUs come from different manufacturing nodes and different architectural generations. The Arc B770 uses the BMG-G31 chip built on Xe2-HPG architecture, the Battlemage generation, fabricated on a 5 nm process at TSMC. The Arc Graphics 32EU uses the Arrow Lake-S chip built on Xe-LPG architecture, the Arc Graphics-M generation for Arrow Lake, fabricated on a 3 nm process at TSMC.

Die size and transistor counts differ substantially, though the Arc B770's transistor count is unknown. The Arc B770's die measures 368 mm². The Arc Graphics 32EU's die measures 243 mm² and contains 17,800 million transistors, giving a transistor density of 73.3M per mm². The integrated part achieves higher density on the 3 nm node, but the discrete part's larger die accommodates far more execution hardware.

Clock behavior also differs. The Arc B770 runs at a 2100 MHz base clock and a 2400 MHz boost clock, with memory clocked at 2000 MHz or 16 Gbps effective. The Arc Graphics 32EU has a 300 MHz base clock and a 1950 MHz boost clock. The integrated GPU's low base clock reflects power management for a 65 W IGP, while its boost clock reaches close to the discrete card's base clock.

Memory architecture is entirely different. The Arc B770 uses 16 GB of GDDR6 with a 256-bit bus. The Arc Graphics 32EU uses system-shared memory with a system-dependent bus width and bandwidth. The discrete card's 512.0 GB/s bandwidth is fixed and dedicated, while the IGP's bandwidth depends entirely on the host platform's memory subsystem.

Both GPUs support DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. The Arc B770 supports PCIe 4.0 x16, while the Arc Graphics 32EU connects via Ring Bus. Display outputs also differ: the Arc B770 provides 1x HDMI 2.1a and 3x DisplayPort 2.1, while the Arc Graphics 32EU's outputs are motherboard dependent.

The Arc B770 is a dual-slot card requiring a 1x 6-pin and 1x 8-pin power connector, with a suggested 550 W power supply. The Arc Graphics 32EU is an IGP with no power connectors and no suggested PSU, drawing power directly from the motherboard. The Arc B770's predecessor is Alchemist, and its release date is recorded as 2025-12-31. The Arc Graphics 32EU's predecessor is HD Graphics-M, it is marked as Active in production status, and it was released on 2024-10-23.

The Arc Graphics 32EU's nearest rivals in the database are all similarly low-power parts. It scores exactly the same 733 as the Intel Arc Graphics 24EU and the Intel Arc Graphics 64EU, with a 0% delta. It sits 1.4% above the AMD Radeon HD 6470M (723) and 2.4% below the NVIDIA GeForce GT 415M (751). These comparisons show that the 32EU's performance tier is extremely clustered; small changes in driver or platform can reshuffle the order.

The Verdict

The recorded data presents a straightforward hierarchy. The Intel Arc B770 is a high-throughput discrete GPU built for rendering, compute, and bandwidth-intensive tasks. Its 50th percentile ranking places it at the middle of the database's GPU population, a position consistent with a mainstream discrete part. The Intel Arc Graphics 32EU, with its 3rd percentile ranking and 733 Steel Nomad score, is a low-power integrated solution for basic graphics output.

The Arc B770 should be selected when the workload requires dedicated memory, high fill rates, ray tracing, or sustained FP32 throughput. Its 16 GB GDDR6 pool and 512.0 GB/s bandwidth support large textures and high-resolution buffers. Its 32 ray tracing cores provide hardware acceleration that the Arc Graphics 32EU lacks entirely.

The Arc Graphics 32EU should be selected when power efficiency and platform integration are the priorities. Its 65 W TDP fits within a standard desktop CPU package, and its system-shared memory eliminates the cost and complexity of discrete VRAM. Its 733 benchmark score places it directly alongside other entry-level IGPs, and its performance delta against the Arc Graphics 24EU and 64EU is exactly 0%, indicating that the 32EU sits at the same measured performance level despite having more execution units than the 24EU.

The data does not support treating these as competitors. They are complementary tiers in Intel's product stack. The Arc B770 targets a discrete graphics market; the Arc Graphics 32EU targets integrated platforms where the CPU's built-in graphics must suffice.

FAQ

Q: How much faster is the Intel Arc B770 than the Intel Arc Graphics 32EU in FP32 compute?

A: The Arc B770 delivers 19.66 TFLOPS FP32, while the Arc Graphics 32EU delivers 998.4 GFLOPS FP32. That puts the discrete card roughly 19.7x ahead.

Q: What is the memory configuration of each GPU?

A: The Arc B770 uses 16 GB of GDDR6 on a 256-bit bus with 512.0 GB/s bandwidth. The Arc Graphics 32EU uses system-shared memory, with system-dependent bus width and bandwidth.

Q: Does the Arc Graphics 32EU support ray tracing?

A: No. The Arc Graphics 32EU lists no ray tracing cores, while the Arc B770 includes 32 dedicated RT cores.

Q: How does the Arc Graphics 32EU compare to its nearest rivals?

A: It scores 733, identical to the Intel Arc Graphics 24EU and Intel Arc Graphics 64EU (0% delta), 1.4% above the AMD Radeon HD 6470M (723), and 2.4% below the NVIDIA GeForce GT 415M (751).

Q: What are the power requirements for each GPU?

A: The Arc B770 has a 225 W TDP, requires a 1x 6-pin and 1x 8-pin power connector, and needs a suggested 550 W power supply. The Arc Graphics 32EU has a 65 W TDP and uses no power connectors.

Q: Which API features do both GPUs share?

A: Both support DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4.

Head-to-Head Benchmarks

There are no direct head-to-head benchmark entries in the database for these two products, and the Arc B770 has no individual benchmark scores recorded. The comparison must therefore rely on architectural specifications and the Arc Graphics 32EU's single 3DMark Steel Nomad DX12 score of 733.

The largest wins for the Arc B770 come from its execution resource advantage. With 4096 shading units versus 256, the discrete card holds a 16x lead in raw shader count. Its 256 TMUs versus 16 give another 16x lead in texture operations. Its 128 ROPs versus 8 produce the same 16x margin in raster work. These ratios translate directly into throughput: 307.2 GPixel/s versus 15.60 GPixel/s, and 614.4 GTexel/s versus 31.20 GTexel/s.

Memory bandwidth is the most lopsided category. The Arc B770's 512.0 GB/s dedicated GDDR6 bandwidth is fixed and independent of the rest of the system. The Arc Graphics 32EU's bandwidth is listed as system dependent, meaning it inherits whatever memory bandwidth the host CPU platform provides. In a typical desktop configuration, shared memory bandwidth is far lower than dedicated GDDR6, and it is also contended with CPU workloads.

Compute throughput follows the same pattern. The Arc B770's 19.66 TFLOPS FP32 is roughly 19.7x the Arc Graphics 32EU's 998.4 GFLOPS. FP16 performance shows 39.32 TFLOPS versus 1.997 TFLOPS, a similar ratio. The consistency of this multiplier across shading units, pixels, texels, and FP32 indicates both designs scale execution resources linearly with no architectural efficiency gap at the specification level.

Clock speeds tell a more nuanced story. The Arc B770 boosts to 2400 MHz, while the Arc Graphics 32EU boosts to 1950 MHz. The discrete card's 2100 MHz base clock is higher than the IGP's 1950 MHz boost clock. However, the IGP's 300 MHz base clock is extremely low, suggesting aggressive power gating at idle or light load, which is appropriate for a 65 W integrated part.

The Arc Graphics 32EU's nearest rival data provides context for its performance tier. Its 733 Steel Nomad score is identical to both the 24EU and 64EU variants, meaning the database records no measurable performance difference between those three configurations. It sits 1.4% above the AMD Radeon HD 6470M and 2.4% below the NVIDIA GeForce GT 415M. These deltas are small enough that driver updates or platform memory differences could flip the ordering.

The Arc B770's 50th percentile ranking versus the Arc Graphics 32EU's 3rd percentile ranking places them far apart in the overall GPU distribution. A GPU at the 50th percentile sits at the median of all recorded GPUs, while the 3rd percentile places the IGP among the slowest parts in the database. No benchmark data exists to refine the Arc B770's position further, but the specification sheet alone establishes a consistent 16x to 19.7x resource and throughput advantage across every measurable execution category.

DETAILED SPECIFICATIONS

SPECIFICATION
B770
Graphics 32EU
Core Specs
Shading Units
4,096
256 -93.8%
Shaders
4,096
256 -93.8%
TMUs
256
16 -93.8%
ROPs
128
8 -93.8%
Execution Units
32
32 0.0%
Clocks
Base Clock
2100 MHz
300 MHz
Boost Clock
2400 MHz
1950 MHz
Memory Clock
2000 MHz 16 Gbps effective
System Shared
Memory
Memory Size
16 GB
System Shared
VRAM (MB)
16,384
Memory Type
GDDR6
System Shared
Memory Bus
256 bit
System Shared
Bandwidth
512.0 GB/s
System Dependent
Cache
L2 Cache
16 MB
Performance
Pixel Rate
307.2 GPixel/s
15.60 GPixel/s
Texture Rate
614.4 GTexel/s
31.20 GTexel/s
FP32 (TFLOPS)
19.66 TFLOPS
998.4 GFLOPS
FP64 (TFLOPS)
2.458 TFLOPS (1:8)
FP16 (TFLOPS)
39.32 TFLOPS (2:1)
1.997 TFLOPS (2:1)
AI/RT
RT Cores
32
XMX Cores
256
Power
TDP
225 W
65 W
TDP (W)
225
65 -71.1%
Suggested PSU
550 W
Power Connectors
1x 6-pin + 1x 8-pin
Architecture
Architecture
Xe2-HPG
Xe-LPG
GPU Name
BMG-G31
Arrow Lake-S
Generation
Battlemage (Arc 7)
Arc Graphics-M (Arrow Lake)
Process Size
5 nm
3 nm
Transistors
unknown
17,800 million
Die Size
368 mm²
243 mm²
Foundry
TSMC
TSMC
Density
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
Dual-slot
IGP
Outputs
1x HDMI 2.1a3x DisplayPort 2.1
Motherboard Dependent
Bus Interface
PCIe 4.0 x16
Ring Bus
Other
Production
Active
Predecessor
Alchemist
HD Graphics-M
View Arc B770 Details View Arc Graphics 32EU Details