AMD Instinct MI350X vs NVIDIA GeForce RTX 5060 Ti 8 GB Comparison

AMD
RADEON

AMD Instinct MI350X

CORE STATE MI350 256CU
VRAM 288 GB
CLOCK SPEED 2200 MHz
TDP 1000 W
BUS WIDTH 8192 bit
ARCHITECTURE CDNA 4.0
nm
PROCESS 3 nm
LAUNCH DATE 2025
VS
NVIDIA
GEFORCE

GeForce RTX 5060 Ti 8 GB

CORE STATE GB206
VRAM 8 GB
CLOCK SPEED 2572 MHz
TDP 180 W
BUS WIDTH 128 bit
ARCHITECTURE Blackwell 2.0
nm
PROCESS 5 nm
LAUNCH DATE 2025

PERFORMANCE BENCHMARKS

3dmark_3dmark_steel_nomad_dx12
N/A
4,019
passmark_directx_10
N/A
142
passmark_directx_11
N/A
220
passmark_directx_12
N/A
82
passmark_directx_9
N/A
228
passmark_g2d
N/A
1,169
passmark_g3d
N/A
21,981
passmark_gpu_compute
N/A
11,365

Analysis: AMD Instinct MI350X vs NVIDIA GeForce RTX 5060 Ti 8 GB

Head-to-Head Benchmarks

The database contains no shared benchmark results for the AMD Instinct MI350X and the NVIDIA GeForce RTX 5060 Ti 8 GB. The Instinct MI350X has no recorded benchmarks, an average benchmark score of 0, and a percentile ranking of 50 among all GPUs. The RTX 5060 Ti, by contrast, has eight recorded benchmark scores, an average of 4901, and a percentile ranking of 29.

The RTX 5060 Ti's strongest results come from compute-oriented tests. In Passmark GPU Compute, it scores 11365, which is its highest individual benchmark. Its Passmark G3D score is 21981, and its Passmark G2D score is 1169. In 3DMark Steel Nomad DX12, it records 4019 points. Older DirectX API tests show lower scores: Passmark DirectX 9 yields 228, DirectX 11 yields 220, DirectX 10 yields 142, and DirectX 12 yields 82.

Because the MI350X has no benchmark entries, the head-to-head comparison cannot rely on direct measurements. The nearest rivals for the RTX 5060 Ti provide context for its standing. The AMD FirePro W5130M averages 4904, a delta of -0.1% relative to the RTX 5060 Ti. The NVIDIA GeForce GTS 450 averages 4893, a delta of 0.2%. The AMD Radeon R7 M265 averages 4929, a delta of -0.6%, and the AMD Radeon R7 M360 averages 4931, also a delta of -0.6%. These deltas place the RTX 5060 Ti within a narrow band around 4900 points, indicating that its average score sits nearly identical to several older or lower-tier parts.

The MI350X's percentile of 50 places it at the median of all GPUs in the database, while the RTX 5060 Ti's percentile of 29 puts it below that midpoint. However, the absence of MI350X benchmark scores means this percentile likely reflects the card's absence from testing rather than a measured performance level. The RTX 5060 Ti's average score of 4901 with a single recorded run in each test gives a concrete, if limited, picture of its capabilities. The MI350X, with avgBenchmarkScore of 0, has no comparable data point.

FAQ

Q: What is the MI350X's average benchmark score?

A: The MI350X has an average benchmark score of 0, with no individual benchmark entries recorded in the database.

Q: What is the RTX 5060 Ti's average benchmark score across all tests?

A: The RTX 5060 Ti has an average benchmark score of 4901, derived from eight separate test results.

Q: Which GPU has the higher percentile ranking among all GPUs?

A: The MI350X ranks at the 50th percentile, while the RTX 5060 Ti ranks at the 29th percentile. The MI350X holds the higher percentile, though this is not supported by any actual benchmark scores.

Q: What are the closest rivals to the RTX 5060 Ti by average score?

A: The nearest rivals are the AMD FirePro W5130M (4904, -0.1% delta), NVIDIA GeForce GTS 450 (4893, 0.2% delta), AMD Radeon R7 M265 (4929, -0.6% delta), and AMD Radeon R7 M360 (4931, -0.6% delta).

Q: How does the RTX 5060 Ti perform in Passmark DirectX 12 versus DirectX 9?

A: The RTX 5060 Ti scores 82 in Passmark DirectX 12 and 228 in Passmark DirectX 9, showing a significantly higher result in the older API test.

Q: Does the MI350X have any recorded 3DMark or Passmark scores?

A: No, the MI350X has no recorded benchmarks in any test, leaving its measured performance entirely unspecified.

Architecture Differences

The two GPUs target entirely different segments, and their architectures reflect that split. The MI350X uses AMD's CDNA 4.0 architecture, designed for compute acceleration, while the RTX 5060 Ti uses NVIDIA's Blackwell 2.0 architecture, oriented toward graphics and consumer workloads.

The MI350X is built on a 3 nm process at TSMC, while the RTX 5060 Ti uses a 5 nm process, also from TSMC. The MI350X integrates 185,000 million transistors on a die size of 2380 mm², yielding a transistor density of 77.7M per mm². The RTX 5060 Ti packs 21,900 million transistors into a 181 mm² die, giving a density of 121.0M per mm². The MI350X's die is more than thirteen times larger by area, but the RTX 5060 Ti achieves a higher transistor density.

Shading unit counts diverge sharply. The MI350X has 16,384 shading units and 1,024 texture mapping units, while the RTX 5060 Ti has 4,608 shading units and 144 TMUs. The MI350X has no ROPs listed, whereas the RTX 5060 Ti has 48 ROPs. The RTX 5060 Ti also includes 36 ray tracing cores and 144 tensor cores; the MI350X lists neither RT cores nor tensor cores.

Memory architecture differs fundamentally. The MI350X uses 288 GB of HBM3e memory across an 8192-bit bus, delivering 8.19 TB/s of bandwidth. The RTX 5060 Ti uses 8 GB of GDDR7 memory on a 128-bit bus, providing 448.0 GB/s. The MI350X's memory bandwidth is roughly eighteen times higher, and its capacity is 36 times larger.

The MI350X's pixel rate is listed as 0 MPixel/s, reflecting its lack of display outputs, while the RTX 5060 Ti outputs 123.5 GPixel/s. Texture rate also differs: the MI350X reaches 2,252.8 GTexel/s versus 370.4 GTexel/s for the RTX 5060 Ti. FP32 compute is 72.09 TFLOPS for the MI350X and 23.70 TFLOPS for the RTX 5060 Ti, with both listing FP16 at the same figures (1:1 ratio). The MI350X supports no DirectX, OpenGL, or Vulkan APIs, whereas the RTX 5060 Ti supports DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4.

Specification Differences

The MI350X and RTX 5060 Ti differ across nearly every specification field. The MI350X uses a 3 nm node, while the RTX 5060 Ti uses 5 nm. Transistor counts are 185,000 million versus 21,900 million. Die sizes are 2380 mm² versus 181 mm². Transistor densities are 77.7M per mm² versus 121.0M per mm².

Clocks diverge significantly. The MI350X runs a base clock of 1000 MHz and a boost clock of 2200 MHz, with memory at 2000 MHz (8 Gbps effective). The RTX 5060 Ti has a base clock of 2407 MHz and a boost clock of 2572 MHz, with memory at 1750 MHz (28 Gbps effective). The RTX 5060 Ti's higher clock speeds are offset by its much narrower memory bus.

Memory capacity is 288 GB versus 8 GB. Memory type is HBM3e versus GDDR7. Bus width is 8192 bit versus 128 bit. Bandwidth is 8.19 TB/s versus 448.0 GB/s. Shading units are 16,384 versus 4,608. TMUs are 1,024 versus 144. ROPs are 0 versus 48. The RTX 5060 Ti has 36 RT cores and 144 tensor cores; the MI350X has none listed.

Pixel rate is 0 MPixel/s versus 123.5 GPixel/s. Texture rate is 2,252.8 GTexel/s versus 370.4 GTexel/s. FP32 is 72.09 TFLOPS versus 23.70 TFLOPS. FP16 is 72.09 TFLOPS versus 23.70 TFLOPS. TDP is 1000 W versus 180 W. Slot width is OAM Module versus Dual-slot. Power connectors are None versus 1x 8-pin. Suggested PSU is 1400 W versus 450 W. Bus interface is PCIe 5.0 x16 versus PCIe 5.0 x8.

Display outputs: the MI350X has none, while the RTX 5060 Ti has 1x HDMI 2.1b and 3x DisplayPort 2.1b. API support is N/A for the MI350X, while the RTX 5060 Ti supports DirectX 12 Ultimate, OpenGL 4.6, and Vulkan 1.4. Dimensions: the MI350X measures 102 mm in length and 165 mm in width; the RTX 5060 Ti measures 241 mm in length, 111 mm in height, and 40 mm in width. Release dates are 2025-06-11 for the MI350X and 2025-04-15 for the RTX 5060 Ti. The MI350X has no launch MSRP; the RTX 5060 Ti has a launch MSRP of 379 USD.

Where Each One Wins

The MI350X wins decisively in raw compute throughput. Its FP32 figure of 72.09 TFLOPS is more than three times the RTX 5060 Ti's 23.70 TFLOPS. Its texture rate of 2,252.8 GTexel/s dwarfs the 370.4 GTexel/s of the RTX 5060 Ti. Memory bandwidth of 8.19 TB/s versus 448.0 GB/s makes the MI350X suited for data-intensive workloads, while the RTX 5060 Ti's 8 GB capacity limits it to far smaller datasets. The MI350X's 288 GB capacity and HBM3e memory type indicate a design for large-scale compute, not interactive graphics.

The RTX 5060 Ti wins in areas tied to display and graphics output. It has 48 ROPs, a 123.5 GPixel/s pixel rate, and three DisplayPort 2.1b outputs plus one HDMI 2.1b. The MI350X offers no display outputs and a 0 MPixel/s pixel rate, meaning it cannot drive monitors. The RTX 5060 Ti also supports DirectX 12 Ultimate, OpenGL 4.6, and Vulkan 1.4, while the MI350X lists N/A for all APIs. For gaming or any visual application, the RTX 5060 Ti is the only option with functional graphics pipelines.

The RTX 5060 Ti also wins on physical practicality. Its TDP is 180 W versus 1000 W for the MI350X, and its suggested PSU is 450 W versus 1400 W. It uses a single 8-pin power connector, while the MI350X has no power connectors listed and relies on an OAM module slot. The RTX 5060 Ti is a dual-slot card measuring 241 mm in length, 111 mm in height, and 40 mm in width, whereas the MI350X is an OAM module at 102 mm by 165 mm, a form factor intended for servers.

The RTX 5060 Ti has recorded benchmark scores, including a Passmark GPU Compute score of 11365 and a Passmark G3D score of 21981. The MI350X has no benchmark scores at all, so its measured performance in standard tests remains unknown. The RTX 5060 Ti's nearest rivals, all hovering near 4900 average points, show that its standing in the database is mid-tier at best. The MI350X's percentile of 50 versus the RTX 5060 Ti's 29 is the only comparative ranking available, but without any MI350X benchmark results, that percentile carries limited interpretive weight.

Architecturally, the MI350X's CDNA 4.0 design with 16,384 shading units and no RT cores targets compute acceleration, while the RTX 5060 Ti's Blackwell 2.0 design with 36 RT cores and 144 tensor cores targets consumer graphics and ray tracing. The MI350X's 3 nm process and 185,000 million transistors indicate a flagship-scale compute part, while the RTX 5060 Ti's 5 nm process and 21,900 million transistors represent a mainstream consumer GPU. The MI350X's 8192-bit memory bus and 288 GB capacity far exceed the RTX 5060 Ti's 128-bit bus and 8 GB, confirming that the MI350X is built for memory-heavy workloads like AI training or large-scale simulation, not for rasterization or real-time rendering.

The RTX 5060 Ti's higher clock speeds (2572 MHz boost versus 2200 MHz) help it in latency-sensitive tasks, but its narrower memory bus and smaller memory pool limit sustained throughput. The MI350X's lower clocks are compensated by massive parallelism and memory bandwidth, giving it an edge in throughput-bound scenarios. The RTX 5060 Ti's 48 ROPs and display outputs make it the only one of the two that can produce a visual output, and its API support means it can run standard graphics software. The MI350X cannot run any standard graphics API, so its utility is confined to compute-only environments.

DETAILED SPECIFICATIONS

SPECIFICATION
Instinct MI350X
RTX 5060 Ti 8 GB
Core Specs
Shading Units
16,384
4,608 -71.9%
Shaders
16,384
4,608 -71.9%
TMUs
1,024
144 -85.9%
ROPs
0
48 +∞%
Compute Units
256
—
SM Count
—
36
Clocks
Base Clock
1000 MHz
2407 MHz
Boost Clock
2200 MHz
2572 MHz
Memory Clock
2000 MHz 8 Gbps effective
1750 MHz 28 Gbps effective
Memory
Memory Size
288 GB
8 GB
VRAM (MB)
294,912
8,192 -97.2%
Memory Type
HBM3e
GDDR7
Memory Bus
8192 bit
128 bit
Bandwidth
8.19 TB/s
448.0 GB/s
Cache
L1 Cache
16 KB (per CU)
128 KB (per SM)
L2 Cache
16 MB
32 MB
L3 Cache
256 MB
—
Performance
Pixel Rate
0 MPixel/s
123.5 GPixel/s
Texture Rate
2,252.8 GTexel/s
370.4 GTexel/s
FP32 (TFLOPS)
72.09 TFLOPS
23.70 TFLOPS
FP64 (TFLOPS)
36.04 TFLOPS (1:2)
370.4 GFLOPS (1:64)
FP16 (TFLOPS)
72.09 TFLOPS (1:1)
23.70 TFLOPS (1:1)
AI/RT
RT Cores
—
36
Tensor Cores
—
144
Matrix Cores
1,024
—
Power
TDP
1000 W
180 W
TDP (W)
1,000
180 -82.0%
Suggested PSU
1400 W
450 W
Power Connectors
None
1x 8-pin
Architecture
Architecture
CDNA 4.0
Blackwell 2.0
GPU Name
MI350 256CU
GB206
Generation
Instinct (MIx)
GeForce 50
Process Size
3 nm
5 nm
Transistors
185,000 million
21,900 million
Die Size
2380 mm²
181 mm²
Foundry
TSMC
TSMC
Density
77.7M / mm²
121.0M / mm²
AMD MCM
MCM
2
—
API Support
DirectX
—
12 Ultimate (12_2)
OpenGL
—
4.6
Vulkan
—
1.4
OpenCL
3.0
3.0
CUDA
—
12.0
Shader Model
—
6.9
Physical
Slot Width
OAM Module
Dual-slot
Length
102 mm 4 inches
241 mm 9.5 inches
Height
—
111 mm 4.4 inches
Outputs
No outputs
1x HDMI 2.1b3x DisplayPort 2.1b
Bus Interface
PCIe 5.0 x16
PCIe 5.0 x8
Other
Launch Price
—
379 USD
Production
—
Active
Predecessor
Radeon Instinct
GeForce 40
Successor
—
GeForce 60
View Instinct MI350X Details View GeForce RTX 5060 Ti 8 GB Details