AMD Radeon RX 6750 GRE 10 GB vs NVIDIA B200 SXM6 Comparison

AMD
RADEON

AMD Radeon RX 6750 GRE 10 GB

CORE STATE Navi 22
VRAM 10 GB
CLOCK SPEED 2450 MHz
TDP 170 W
BUS WIDTH 160 bit
ARCHITECTURE RDNA 2.0
nm
PROCESS 7 nm
LAUNCH DATE 2023
VS
NVIDIA
GEFORCE

B200 SXM6

CORE STATE GB100
VRAM 180 GB
CLOCK SPEED 1830 MHz
TDP 1000 W
BUS WIDTH 8192 bit
ARCHITECTURE Blackwell
nm
PROCESS 5 nm
LAUNCH DATE 2024

PERFORMANCE BENCHMARKS

3dmark_3dmark_steel_nomad_dx12
1,995
N/A

Analysis: AMD Radeon RX 6750 GRE 10 GB vs NVIDIA B200 SXM6

The Verdict

The recorded data presents an unusual comparison. The AMD Radeon RX 6750 GRE 10 GB is a client-side graphics card with one recorded benchmark score of 1995 in the 3DMark Steel Nomad DX12 test, placing it in the 13th percentile of all GPUs. The NVIDIA B200 SXM6 is a server accelerator with no recorded gaming or standard graphics benchmarks, holding a 50th percentile position based on its inclusion in the database. The RX 6750 GRE is positioned for rasterized gaming at 1080p or 1440p, while the B200 SXM6 is a compute-oriented module with no display outputs, making it unsuitable for conventional gaming use. The data indicates these two products target entirely different workloads. The RX 6750 GRE delivers a measurable graphics score, while the B200 SXM6 shows no comparable score in the same test suite.

FAQ

Q: Why does the NVIDIA B200 SXM6 have a higher percentile than the AMD Radeon RX 6750 GRE despite having no benchmark scores?

A: The B200 SXM6 holds a 50th percentile ranking versus the RX 6750 GRE's 13th percentile. The B200's percentile is derived from its presence in the database rather than from the same 3DMark Steel Nomad DX12 test, for which it has no recorded score.

Q: What is the memory capacity difference between the two GPUs?

A: The AMD Radeon RX 6750 GRE 10 GB uses 10 GB of GDDR6 memory on a 160-bit bus, delivering 320.0 GB/s of bandwidth. The NVIDIA B200 SXM6 uses 180 GB of HBM3e memory on an 8192-bit bus, delivering 8.19 TB/s of bandwidth.

Q: Which GPU has higher FP32 floating-point performance?

A: The NVIDIA B200 SXM6 reaches 69.34 TFLOPS of FP32 compute, while the AMD Radeon RX 6750 GRE 10 GB delivers 11.29 TFLOPS. The B200 is roughly six times higher in this metric.

Q: Can the NVIDIA B200 SXM6 output video to a display?

A: No. The B200 SXM6 has no display outputs and lists DirectX, OpenGL, and Vulkan APIs as N/A. The AMD Radeon RX 6750 GRE 10 GB includes 1x HDMI 2.1 and 3x DisplayPort 1.4a outputs.

Q: What is the power connector requirement for each card?

A: The AMD Radeon RX 6750 GRE 10 GB uses a single 8-pin power connector with a suggested PSU of 450 W and a TDP of 170 W. The NVIDIA B200 SXM6 has no listed power connectors but requires a suggested PSU of 1400 W and has a TDP of 1000 W.

Q: How do the transistor counts compare?

A: The AMD Radeon RX 6750 GRE 10 GB contains 17,200 million transistors on a 335 mm² die. The NVIDIA B200 SXM6 contains 208,000 million transistors on a 1628 mm² die.

Architecture Differences

The AMD Radeon RX 6750 GRE 10 GB uses the Navi 22 chip built on RDNA 2.0 architecture, fabricated on a 7 nm process at TSMC. It belongs to the Navi II generation within the Radeon RX 6000 series. The NVIDIA B200 SXM6 uses the GB100 chip built on Blackwell architecture, fabricated on a 5 nm process at TSMC, belonging to the Server Blackwell generation.

Transistor density differs substantially. The RX 6750 GRE packs 17,200 million transistors into a 335 mm² die, yielding a density of 51.3 million transistors per square millimeter. The B200 SXM6 contains 208,000 million transistors on a 1628 mm² die, yielding a density of 127.8 million transistors per square millimeter. The B200's higher density reflects the tighter 5 nm process.

The RX 6750 GRE implements 2304 shading units, 144 texture mapping units, 64 render output units, and 36 ray tracing cores. It has no dedicated tensor cores. The B200 SXM6 uses 18,944 shading units, 592 texture mapping units, 24 render output units, and 592 tensor cores. The B200 lists no ray tracing core count in the database.

Memory architecture separates the two fundamentally. The RX 6750 GRE uses 10 GB of GDDR6 with a 160-bit bus and 320.0 GB/s bandwidth. The B200 SXM6 uses 180 GB of HBM3e with an 8192-bit bus and 8.19 TB/s bandwidth. The B200's memory bandwidth is over 25 times higher, and its capacity is 18 times larger.

Pixel and texture rates also diverge. The RX 6750 GRE achieves 156.8 GPixel/s and 352.8 GTexel/s. The B200 SXM6 achieves 43.92 GPixel/s and 1,083.4 GTexel/s. The B200's lower pixel rate but higher texture rate indicates a compute-oriented design rather than a rasterization-focused one.

The RX 6750 GRE includes full graphics API support with DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. The B200 SXM6 lists DirectX as N/A, OpenGL as N/A, and Vulkan as N/A. This confirms the B200 is not intended for traditional graphics rendering workloads.

Specification Differences

The process node differs: the RX 6750 GRE uses 7 nm, while the B200 SXM6 uses 5 nm. Both use TSMC as the foundry. The B200's die is larger at 1628 mm² versus 335 mm², and its transistor count is higher at 208,000 million versus 17,200 million.

Clock behavior differs sharply. The RX 6750 GRE has a base clock of 1941 MHz, a boost clock of 2450 MHz, and a game clock of 2189 MHz. The B200 SXM6 has a base clock of 120 MHz and a boost clock of 1830 MHz, with no game clock listed. The memory clocks show 2000 MHz with 16 Gbps effective for the RX 6750 GRE, versus 2000 MHz with 8 Gbps effective for the B200.

The power envelope is dramatically different. The RX 6750 GRE has a TDP of 170 W, uses a dual-slot cooler, one 8-pin power connector, and a suggested PSU of 450 W. The B200 SXM6 has a TDP of 1000 W, uses an SXM module form factor, no listed power connectors, and a suggested PSU of 1400 W.

The RX 6750 GRE uses PCIe 4.0 x16, while the B200 SXM6 uses PCIe 6.0 x16. Display outputs exist only on the RX 6750 GRE with 1x HDMI 2.1 and 3x DisplayPort 1.4a. The B200 lists no outputs.

Dimensions are only recorded for the RX 6750 GRE: 267 mm length, 110 mm height, and 40 mm width. The B200 SXM6 has no recorded dimensions.

The RX 6750 GRE was released on 2023-10-16, while the B200 SXM6 was released on 2024-10-31. The RX 6750 GRE's predecessor is Navi and successor is Navi III. The B200's predecessor is Server Hopper and successor is Server Rubin. Both are listed as Active in production status.

Head-to-Head Benchmarks

The head-to-head benchmark table contains no entries. No common test scores exist for both GPUs. The only recorded benchmark belongs to the AMD Radeon RX 6750 GRE 10 GB, which scored 1995 in 3dmark_3dmark_steel_nomad_dx12. The NVIDIA B200 SXM6 has an empty benchmark array and an average benchmark score of 0.

The RX 6750 GRE's nearest rivals provide context for its single score. It sits 1% above the NVIDIA GRID K1 (1976), 1.1% above the NVIDIA GeForce GTX 470 (1973), 1.6% below the Intel UHD Graphics 770 (2027), and 1.7% below the NVIDIA GeForce GTX 660M (2029). These deltas indicate the RX 6750 GRE's score places it in a narrow performance band around the 2000-point mark in this particular test.

The B200 SXM6 has no nearest rivals listed and no benchmark scores, so no direct numerical comparison against the RX 6750 GRE is possible from the recorded data. The B200's 50th percentile reflects its overall database position, but without a shared workload, the two cannot be ranked against each other in any measured test.

The FP32 compute figures offer the only arithmetic comparison. The B200 SXM6 delivers 69.34 TFLOPS, which is 6.14 times the RX 6750 GRE's 11.29 TFLOPS. The B200 also delivers FP16 at 69.34 TFLOPS with a 1:1 ratio, while the RX 6750 GRE delivers FP16 at 22.58 TFLOPS with a 2:1 ratio. Memory bandwidth comparison shows the B200 at 8.19 TB/s versus 320.0 GB/s for the RX 6750 GRE, a 25.6 times difference.

Where Each One Wins

The AMD Radeon RX 6750 GRE 10 GB wins in any workload requiring conventional graphics output. It has display connectivity with HDMI 2.1 and DisplayPort 1.4a, full DirectX 12 Ultimate support, OpenGL 4.6, and Vulkan 1.4. Its 1995 score in the 3DMark Steel Nomad DX12 test demonstrates functional rasterization capability. Its pixel rate of 156.8 GPixel/s is 3.6 times higher than the B200's 43.92 GPixel/s, indicating stronger traditional fill-rate performance. The card consumes 170 W TDP with a 450 W suggested PSU, making it suitable for desktop systems with standard power supplies. Its dual-slot form factor and 267 mm length fit conventional chassis designs.

The NVIDIA B200 SXM6 wins in raw compute throughput. Its FP32 output of 69.34 TFLOPS is roughly six times the RX 6750 GRE's 11.29 TFLOPS. Its texture rate of 1,083.4 GTexel/s is 3.1 times higher. Its memory system provides 8.19 TB/s of bandwidth and 180 GB of capacity, vastly exceeding the RX 6750 GRE's 320.0 GB/s and 10 GB. The 592 tensor cores give it dedicated hardware for matrix operations, which the RX 6750 GRE lacks entirely.

Server deployment favors the B200 SXM6. It uses PCIe 6.0 x16, a newer bus standard than the RX 6750 GRE's PCIe 4.0 x16. Its SXM module form factor suits dense accelerator installations. The 1000 W TDP and 1400 W suggested PSU indicate a system designed for high-power compute nodes, not desktop use.

The database shows no overlap in recorded workloads. The RX 6750 GRE's single benchmark entry and the B200's empty benchmark list mean there is no measured scenario where both compete directly. The 50th percentile of the B200 versus the 13th percentile of the RX 6750 GRE does not constitute a head-to-head result, as percentiles derive from different database contexts. Each device wins in its own domain: the RX 6750 GRE for graphics-facing client workloads, the B200 SXM6 for compute-heavy server tasks.

DETAILED SPECIFICATIONS

SPECIFICATION
RX 6750 GRE 10 GB
B200 SXM6
Core Specs
Shading Units
2,304
18,944 +722.2%
Shaders
2,304
18,944 +722.2%
TMUs
144
592 +311.1%
ROPs
64
24 -62.5%
Compute Units
36
—
SM Count
—
148
Clocks
Base Clock
1941 MHz
120 MHz
Boost Clock
2450 MHz
1830 MHz
Game Clock
2189 MHz
—
Memory Clock
2000 MHz 16 Gbps effective
2000 MHz 8 Gbps effective
Memory
Memory Size
10 GB
180 GB
VRAM (MB)
10,240
184,320 +1700.0%
Memory Type
GDDR6
HBM3e
Memory Bus
160 bit
8192 bit
Bandwidth
320.0 GB/s
8.19 TB/s
Cache
L1 Cache
128 KB per Array
256 KB (per SM)
L2 Cache
3 MB
126 MB
L3 Cache
80 MB
—
L0 Cache
32 KB per WGP
—
Performance
Pixel Rate
156.8 GPixel/s
43.92 GPixel/s
Texture Rate
352.8 GTexel/s
1,083.4 GTexel/s
FP32 (TFLOPS)
11.29 TFLOPS
69.34 TFLOPS
FP64 (TFLOPS)
705.6 GFLOPS (1:16)
34.67 TFLOPS (1:2)
FP16 (TFLOPS)
22.58 TFLOPS (2:1)
69.34 TFLOPS (1:1)
AI/RT
RT Cores
36
—
Tensor Cores
—
592
Power
TDP
170 W
1000 W
TDP (W)
170
1,000 +488.2%
Suggested PSU
450 W
1400 W
Power Connectors
1x 8-pin
—
Architecture
Architecture
RDNA 2.0
Blackwell
GPU Name
Navi 22
GB100
Generation
Navi II (RX 6000)
Server Blackwell (Bxx)
Process Size
7 nm
5 nm
Transistors
17,200 million
208,000 million
Die Size
335 mm²
1628 mm²
Foundry
TSMC
TSMC
Density
51.3M / mm²
127.8M / mm²
API Support
DirectX
12 Ultimate (12_2)
—
OpenGL
4.6
—
Vulkan
1.4
—
OpenCL
2.1
3.0
CUDA
—
10.0
Shader Model
6.8
—
Physical
Slot Width
Dual-slot
SXM Module
Length
267 mm 10.5 inches
—
Height
110 mm 4.3 inches
—
Outputs
1x HDMI 2.13x DisplayPort 1.4a
No outputs
Bus Interface
PCIe 4.0 x16
PCIe 6.0 x16
Other
Launch Price
309 USD
34,999 USD
Production
Active
Active
Predecessor
Navi
Server Hopper
Successor
Navi III
Server Rubin
View Radeon RX 6750 GRE 10 GB Details View B200 SXM6 Details