AMD Radeon RX 9060 XT LP vs NVIDIA RTX PRO 6000 Blackwell Server Comparison

AMD
RADEON

AMD Radeon RX 9060 XT LP

CORE STATE Navi 44
VRAM 16 GB
CLOCK SPEED 3050 MHz
TDP 140 W
BUS WIDTH 128 bit
ARCHITECTURE RDNA 4.0
nm
PROCESS 4 nm
LAUNCH DATE 2025
VS
NVIDIA
GEFORCE

RTX PRO 6000 Blackwell Server

CORE STATE GB202
VRAM 96 GB
CLOCK SPEED 2617 MHz
TDP 600 W
BUS WIDTH 512 bit
ARCHITECTURE Blackwell 2.0
nm
PROCESS 5 nm
LAUNCH DATE 2025

PERFORMANCE BENCHMARKS

geekbench_opencl
88,183
N/A
geekbench_vulkan
39,476
N/A
3dmark_3dmark_steel_nomad_dx12
N/A
5,996

Analysis: AMD Radeon RX 9060 XT LP vs NVIDIA RTX PRO 6000 Blackwell Server

The Verdict

The recorded data presents an unusual comparison. The AMD Radeon RX 9060 XT LP shows a significantly higher average benchmark score of 63,830, placing it in the 89th percentile of all GPUs. The NVIDIA RTX PRO 6000 Blackwell Server, by contrast, records an average score of 5,996, which places it in the 34th percentile. Based strictly on these aggregate numbers, the AMD card is the clear performer in general compute workloads.

However, the benchmark sets are not directly comparable. The AMD card was tested with Geekbench OpenCL and Vulkan, while the NVIDIA card was tested with 3DMark Steel Nomad DX12. The database shows zero head-to-head benchmark wins for either card, meaning no direct comparison exists. The NVIDIA card’s low percentile likely reflects the specific nature of the 3DMark test, which may not represent its full capabilities in server-oriented tasks. The verdict from the data alone: the AMD Radeon RX 9060 XT LP dominates in the recorded general compute metrics, while the NVIDIA card’s data is limited to a single test, making any broader conclusion about its standing premature.

Architecture Differences

The two cards come from fundamentally different design philosophies. The AMD Radeon RX 9060 XT LP uses the Navi 44 chip built on RDNA 4.0 architecture, fabricated on a 4 nm process at TSMC. It contains 29,700 million transistors on a 199 mm² die, yielding a transistor density of 149.2 million per square millimeter. The NVIDIA RTX PRO 6000 Blackwell Server uses the GB202 chip with Blackwell 2.0 architecture, also from TSMC but on a 5 nm process. This chip packs 92,200 million transistors across a much larger 750 mm² die, resulting in a lower density of 122.9 million per square millimeter.

The core configurations diverge sharply. The AMD card has 2,048 shading units, 128 texture mapping units, and 64 raster operation pipelines. It includes 32 ray tracing cores and no dedicated tensor cores. The NVIDIA card has 24,064 shading units, 752 texture mapping units, and 192 raster operation pipelines. It includes 188 ray tracing cores and 752 tensor cores, indicating a design aimed at compute-heavy server workloads. The clock speeds tell a different story: the AMD card boosts to 3,050 MHz, while the NVIDIA card boosts to 2,617 MHz. The AMD card’s higher clock partially compensates for its smaller core count, but the raw throughput gap remains enormous.

Memory architecture also contrasts strongly. The AMD card uses 16 GB of GDDR6 on a 128-bit bus, delivering 322.3 GB/s of bandwidth. The NVIDIA card uses 96 GB of GDDR7 on a 512-bit bus, delivering 1.79 TB/s, more than five times the bandwidth. The process node difference (4 nm versus 5 nm) partially explains why the AMD chip achieves higher transistor density despite having far fewer total transistors. The NVIDIA chip’s massive die size suggests a design prioritizing raw compute capacity over manufacturing efficiency.

Where Each One Wins

The AMD Radeon RX 9060 XT LP wins in the recorded Geekbench tests. Its OpenCL score of 88,183 and Vulkan score of 39,476 contribute to an average of 63,830. These results indicate strong performance in general-purpose compute and graphics workloads that stress raw shader throughput and API responsiveness. The card’s high boost clock of 3,050 MHz and its RDNA 4.0 architecture likely drive these results, though the database does not provide per-test breakdowns beyond the two Geekbench entries.

The NVIDIA RTX PRO 6000 Blackwell Server wins in the 3DMark Steel Nomad DX12 test, scoring 5,996. This test likely stresses DirectX 12 rendering pipelines, ray tracing, and geometry processing. The card’s 188 ray tracing cores and 752 tensor cores suggest it should excel in such workloads, but the score itself is modest. The database shows no wins for either card in head-to-head benchmarks, so the division of victories is entirely based on the separate test suites. The AMD card’s data indicates broad compute strength, while the NVIDIA card’s data indicates a specific, perhaps less favorable, result in one modern gaming-oriented test.

The use-case split is clear from the numbers. The AMD card suits workloads measured by Geekbench, such as OpenCL compute and Vulkan graphics. The NVIDIA card, despite its low percentile, likely targets server environments where its 96 GB memory and tensor cores matter more than the recorded benchmark suggests. The data does not support a claim that the NVIDIA card is weak overall; it only shows weakness in the one test recorded.

FAQ

Q: Which card has the higher average benchmark score?

A: The AMD Radeon RX 9060 XT LP has an average benchmark score of 63,830, while the NVIDIA RTX PRO 6000 Blackwell Server has an average score of 5,996.

Q: How do their memory configurations differ?

A: The AMD card uses 16 GB of GDDR6 on a 128-bit bus with 322.3 GB/s bandwidth. The NVIDIA card uses 96 GB of GDDR7 on a 512-bit bus with 1.79 TB/s bandwidth.

Q: Which card has more shading units?

A: The NVIDIA RTX PRO 6000 Blackwell Server has 24,064 shading units, compared to the AMD card’s 2,048 shading units.

Q: What is the process node difference?

A: The AMD card is built on a 4 nm process, while the NVIDIA card is built on a 5 nm process. Both use TSMC as the foundry.

Q: Do the cards have tensor cores?

A: The AMD card has no tensor cores. The NVIDIA card has 752 tensor cores.

Q: What are the power requirements?

A: The AMD card has a TDP of 140 W and uses a single 8-pin connector, with a suggested 300 W power supply. The NVIDIA card has a TDP of 600 W and uses a single 16-pin connector, with a suggested 1000 W power supply.

Head-to-Head Benchmarks

The database lists no head-to-head benchmarks between these two cards, so a direct comparison of identical tests is impossible. The wins each card records come from different test suites, which complicates interpretation. The AMD card’s Geekbench OpenCL score of 88,183 is its strongest recorded result, placing it well above its own Vulkan score of 39,476. This suggests OpenCL compute workloads favor the AMD architecture, likely due to its high boost clock of 3,050 MHz and efficient RDNA 4.0 design.

The NVIDIA card’s only recorded benchmark is 3DMark Steel Nomad DX12 with a score of 5,996. This result places it in the 34th percentile, far below the AMD card’s 89th percentile. The nearest rivals for the NVIDIA card include the NVIDIA GeForce GTX 770M with a delta of -0.1% and the AMD Radeon RX 6400 with a delta of -0.1%, indicating the score is tightly clustered with older, lower-tier cards. The AMD card’s nearest rivals include the NVIDIA CMP 30HX with a delta of 0% and the AMD Radeon RX 7600M with a delta of 0.1%, showing it sits alongside mid-range mobile parts.

The biggest win for the AMD card is its average score, which is 10.6 times higher than the NVIDIA card’s average. The NVIDIA card’s win, if it can be called one, is its 3DMark score being its only data point, which does not indicate superiority in any shared workload. The absence of head-to-head tests means the database cannot confirm which card would win in a direct match, but the recorded averages strongly favor the AMD card in general compute.

Specification Differences

The two cards differ across nearly every major specification. The AMD card uses RDNA 4.0 architecture, while the NVIDIA card uses Blackwell 2.0. The process node differs: 4 nm for AMD, 5 nm for NVIDIA. Transistor counts are 29,700 million for AMD versus 92,200 million for NVIDIA. Die sizes are 199 mm² for AMD versus 750 mm² for NVIDIA, with transistor densities of 149.2M per mm² for AMD and 122.9M per mm² for NVIDIA.

Clock speeds differ in both base and boost. The AMD card has a base clock of 1,380 MHz and a boost of 3,050 MHz, with a game clock of 2,450 MHz. The NVIDIA card has a base clock of 1,590 MHz and a boost of 2,617 MHz, with no game clock listed. Memory speeds also differ: the AMD card runs at 2,518 MHz with 20.1 Gbps effective, while the NVIDIA card runs at 1,750 MHz with 28 Gbps effective.

The memory systems are entirely different. The AMD card has 16 GB of GDDR6, a 128-bit bus, and 322.3 GB/s bandwidth. The NVIDIA card has 96 GB of GDDR7, a 512-bit bus, and 1.79 TB/s bandwidth. The compute units differ: the AMD card has 2,048 shading units, 128 TMUs, 64 ROPs, and 32 RT cores. The NVIDIA card has 24,064 shading units, 752 TMUs, 192 ROPs, and 188 RT cores. The NVIDIA card adds 752 tensor cores, which the AMD card lacks.

Pixel and texture rates reflect the core differences. The AMD card delivers 195.2 GPixel/s and 390.4 GTexel/s. The NVIDIA card delivers 502.5 GPixel/s and 1,968.0 GTexel/s. FP32 and FP16 performance are both 24.99 TFLOPS for AMD, while the NVIDIA card delivers 126.0 TFLOPS for both. Power consumption differs substantially: 140 W TDP for AMD versus 600 W for NVIDIA. Power connectors are 1x 8-pin for AMD and 1x 16-pin for NVIDIA, with suggested power supplies of 300 W and 1000 W, respectively.

Display outputs also vary. The AMD card has 1x HDMI 2.1b and 2x DisplayPort 2.1a. The NVIDIA card has 4x DisplayPort 2.1b. Both use PCIe 5.0 x16 and support DirectX 12 Ultimate, OpenGL 4.6, and Vulkan 1.4. The NVIDIA card has recorded dimensions of 267 mm in length, 111 mm in height, and 40 mm in width; the AMD card’s dimensions are not listed. The NVIDIA card has a successor listed as Server Rubin, while the AMD card has none.

DETAILED SPECIFICATIONS

SPECIFICATION
RX 9060 XT LP
RTX PRO 6000 Blackwell Server
Core Specs
Shading Units
2,048
24,064 +1075.0%
Shaders
2,048
24,064 +1075.0%
TMUs
128
752 +487.5%
ROPs
64
192 +200.0%
Compute Units
32
—
SM Count
—
188
Clocks
Base Clock
1380 MHz
1590 MHz
Boost Clock
3050 MHz
2617 MHz
Game Clock
2450 MHz
—
Memory Clock
2518 MHz 20.1 Gbps effective
1750 MHz 28 Gbps effective
Memory
Memory Size
16 GB
96 GB
VRAM (MB)
16,384
98,304 +500.0%
Memory Type
GDDR6
GDDR7
Memory Bus
128 bit
512 bit
Bandwidth
322.3 GB/s
1.79 TB/s
Cache
L1 Cache
—
128 KB (per SM)
L2 Cache
4 MB
128 MB
L3 Cache
32 MB
—
L0 Cache
32 KB per WGP
—
Performance
Pixel Rate
195.2 GPixel/s
502.5 GPixel/s
Texture Rate
390.4 GTexel/s
1,968.0 GTexel/s
FP32 (TFLOPS)
24.99 TFLOPS
126.0 TFLOPS
FP64 (TFLOPS)
780.8 GFLOPS (1:32)
1.968 TFLOPS (1:64)
FP16 (TFLOPS)
24.99 TFLOPS (1:1)
126.0 TFLOPS (1:1)
AI/RT
RT Cores
32
188 +487.5%
Tensor Cores
—
752
Matrix Cores
64
—
Power
TDP
140 W
600 W
TDP (W)
140
600 +328.6%
Suggested PSU
300 W
1000 W
Power Connectors
1x 8-pin
1x 16-pin
Architecture
Architecture
RDNA 4.0
Blackwell 2.0
GPU Name
Navi 44
GB202
Generation
Navi IV (RX 9000)
Server Blackwell (Bxx)
Process Size
4 nm
5 nm
Transistors
29,700 million
92,200 million
Die Size
199 mm²
750 mm²
Foundry
TSMC
TSMC
Density
149.2M / mm²
122.9M / mm²
API Support
DirectX
12 Ultimate (12_2)
12 Ultimate (12_2)
OpenGL
4.6
4.6
Vulkan
1.4
1.4
OpenCL
2.2
3.0
CUDA
—
12.0
Shader Model
6.9
6.9
Physical
Slot Width
Dual-slot
Dual-slot
Length
—
267 mm 10.5 inches
Height
—
111 mm 4.4 inches
Outputs
1x HDMI 2.1b2x DisplayPort 2.1a
4x DisplayPort 2.1b
Bus Interface
PCIe 5.0 x16
PCIe 5.0 x16
Other
Production
Active
Active
Predecessor
Navi III
Server Hopper
Successor
—
Server Rubin
View Radeon RX 9060 XT LP Details View RTX PRO 6000 Blackwell Server Details