NVIDIA RTX 6000 Ada
vs
PNY GeForce RTX 5090 Overclocked Triple Fan


GPU comparison with benchmarks

The NVIDIA RTX 6000 Ada has 18176 shader units and these clock a maximum of 2.54 GHz which results in an FP32 computing power of 92.15 TFLOPS.

The PNY GeForce RTX 5090 Overclocked Triple Fan has 21760 shader units and these clock at a maximum of 2.53 GHz and achieve an FP32 computing power of 87.82 TFLOPS.

GPU

The NVIDIA RTX 6000 Ada is equipped with a graphics chip of the Ada Lovelace architecture, which has 142 streaming multiprocessors.

The graphics card PNY GeForce RTX 5090 Overclocked Triple Fan is based on the Blackwell architecture and is equipped with 170 execution units.
NVIDIA RTX 6000 Ada GPU PNY GeForce RTX 5090 Overclocked Triple Fan
NVIDIA RTX 6000 Ada Based on NVIDIA GeForce RTX 5090
AD102 GPU Chip GB202-300-A1
Ada Lovelace Architecture Blackwell
142 Streaming Multiprocessors 170
18176 Shader 21760
192 Render Output Units 176
568 Texture Units 680
142 Raytracing Cores 170

Memory

The NVIDIA RTX 6000 Ada is equipped with a 48 GB large GDDR6 graphics memory, which is equipped with 2.500 GHz clocks.

The PNY GeForce RTX 5090 Overclocked Triple Fan has a 32 GB large GDDR7 graphics memory and the memory clock is 1.750GHz.

48 GB Memory Size 32 GB
GDDR6 Memory Type GDDR7
2.500 GHz Memory Clock 1.750 GHz
20.0 Gbps Memory Speed 28.0 Gbps
960 GB/s Memory bandwidth 1792 GB/s
384 bit Memory Interface 512 bit

Clock Speeds

The NVIDIA RTX 6000 Ada has a turbo clock of 2.535 GHz. The turbo frequency of the PNY GeForce RTX 5090 Overclocked Triple Fan is 2.527 GHz.

2.175 GHz Base Clock 2.017 GHz
2.535 GHz Boost Clock 2.527 GHz
Avg (Game) Clock
Yes Overclocking Yes

Thermal Design

The NVIDIA RTX 6000 Ada has 1 x 16-Pin connectors through which it is supplied with power. The manufacturer specifies the maximum operating temperature of the graphics card as 93 °C.

The PNY GeForce RTX 5090 Overclocked Triple Fan is equipped with a 1 x 16-Pin PCIe power connector that supplies it with power. The maximum operating temperature of the card is 90 °C.

300 W (Reference) TDP 600 W
-- TDP (up) --
93 °C Tjunction max 90 °C
1 x 16-Pin PCIe-Power 1 x 16-Pin

Cooler & Fans

The graphics processor and graphics memory of the NVIDIA RTX 6000 Ada are cooled with a Air cooling.

The PNY GeForce RTX 5090 Overclocked Triple Fan is cooled by a Air cooling.

Radial Fan-Type Axial
1 x 70 mm Fan 1 3 x 100 mm
-- Fan 2 --
Air cooling Cooler-Type Air cooling
0 dB Noise (Idle) 0 dB
-- Noise (Load) --

Connectivity

The NVIDIA RTX 6000 Ada supports the downward compatible copy protection HDCP in version 2.3.

The PNY GeForce RTX 5090 Overclocked Triple Fan supports HDCP in version 2.3 and all previous versions.

4 Max. Displays 4
2.3 HDCP-Version 2.3
-- HDMI Ports 1x HDMI v2.1b
4x DP v1.4a DP Ports 3x DP v2.1b
-- DVI Ports --
-- VGA Ports --
-- USB-C Ports --

Featureset

The NVIDIA RTX 6000 Ada supports the DirectX standard from Microsoft in version 12_2 (Ultimate).

The PNY GeForce RTX 5090 Overclocked Triple Fan supports Microsoft's DirectX version 12_2 (Ultimate).

7680x4320 Max. resolution 7680x4320
12_2 (Ultimate) DirectX 12_2 (Ultimate)
Yes Raytracing Yes
Yes DLSS / FSR Yes
No LED lighting LED No LED lighting

Supported Video Codecs

This area lists which video codecs can be decoded or encoded by the graphics cards in hardware. As a result, the power consumption can be reduced due to a lower processor load.

Decode / Encode h264 Decode / Encode
Decode / Encode h265 / HEVC Decode / Encode
Decode / Encode AV1 Decode / Encode
Decode VP8 Decode / Encode
Decode VP9 Decode / Encode

Dimensions

The NVIDIA RTX 6000 Ada is -- wide and therefore needs the space of 4 PCIe-Slots in the case.

The width of the PNY GeForce RTX 5090 Overclocked Triple Fan is -- and thus requires the space of 4 PCIe-Slots in the PC case.

267 mm Length 329 mm
112 mm Height 138 mm
-- Width 71 mm
2 PCIe-Slots Width (Slots) 4 PCIe-Slots
-- Weight --
PCIe 4.0 x 16 GPU Interface PCIe 5 x 16
No SFF-Ready No

Additional data

Manufactured using the 5 nanometer process, NVIDIA RTX 6000 Ada was released in Q1/2023.

The PNY GeForce RTX 5090 Overclocked Triple Fan published in Q1/2025 is manufactured with a structure width of 5 nanometers.

-- Part-no VCG509032TFXPB1-O
Q1/2023 Release date Q1/2025
6,799 $ (Reference) Launch Price 2,329 $ (Reference)
5 nm Structure size 5 nm
data sheet Documents data sheet

Rate these graphics cards now

Here you can rate the NVIDIA RTX 6000 Ada to help other visitors make their purchasing decisions. The average rating is 3.8 stars (6 ratings). Rate now:
Here you can rate the PNY GeForce RTX 5090 Overclocked Triple Fan to help other visitors make their purchasing decisions. The average rating is 5.0 stars (1 ratings). Rate now:

Benchmark results

3DMark Benchmark (DirectX, Raytracing)

3DMark is a benchmark program that determines the performance of certain components of a computer and then reports the performance as a numerical value.

Time Spy Extreme Graphics score

Steel Nomad Light Graphics score

Steel Nomad Graphics score

Port Royal (Raytracing)

Speed Way Graphics Score (Raytracing)

Solar Bay (Raytracing)

Cyberpunk 2077

Cyberpunk 2077 was developed by CD Projekt Red and is based on the development studio's own REDengine 4. Ray tracing is switched off in our benchmarks.

3840x2160 (2160p)

NVIDIA RTX 6000 Ada NVIDIA RTX 6000 Ada
48 GB GDDR6
81 fps (avg)

Battlefield 5

Battlefield 5 is a visually stunning game that is ideal as a graphics card benchmark. We test the game with maximum details on Windows 10.

1920x1080 (1080p)

NVIDIA RTX 6000 Ada NVIDIA RTX 6000 Ada
48 GB GDDR6
200 fps (avg)

Geekbench 6 (OpenCL, Vulkan, Metal)

Geekbench 6 is a cross-platform benchmark for main processors, which also carries out 3 different graphics benchmarks and outputs them in the form of a numerical value.

Geekbench 6 - OpenCL

NVIDIA RTX 6000 Ada NVIDIA RTX 6000 Ada
48 GB GDDR6
290282

Geekbench 6 - Vulkan

NVIDIA RTX 6000 Ada NVIDIA RTX 6000 Ada
48 GB GDDR6
236468

FP32 Performance (Single-precision TFLOPS)

The theoretical computing power of the graphics card with single precision (32 bit) in TFLOPS indicates how many trillion FP32 floating point operations the graphics card (GPU) can perform per second.

FP32 (TFLOPS)

Popular comparisons

1. NVIDIA RTX A6000 NVIDIA RTX 6000 Ada NVIDIA RTX A6000 vs NVIDIA RTX 6000 Ada
2. AMD Radeon Pro W7900 NVIDIA RTX 6000 Ada AMD Radeon Pro W7900 vs NVIDIA RTX 6000 Ada
3. NVIDIA RTX 6000 Ada NVIDIA RTX 5000 Ada NVIDIA RTX 6000 Ada vs NVIDIA RTX 5000 Ada
4. NVIDIA RTX 6000 Ada NVIDIA GeForce RTX 4090 Founders Edition NVIDIA RTX 6000 Ada vs NVIDIA GeForce RTX 4090 Founders Edition
5. NVIDIA RTX 6000 Ada ASUS ROG Astral GeForce RTX 5090 NVIDIA RTX 6000 Ada vs ASUS ROG Astral GeForce RTX 5090
6. NVIDIA GeForce RTX 5090 Founders Edition NVIDIA RTX 6000 Ada NVIDIA GeForce RTX 5090 Founders Edition vs NVIDIA RTX 6000 Ada
7. NVIDIA RTX 5880 Ada NVIDIA RTX 6000 Ada NVIDIA RTX 5880 Ada vs NVIDIA RTX 6000 Ada
8. Apple M2 Ultra 76-Core GPU NVIDIA RTX 6000 Ada Apple M2 Ultra 76-Core GPU vs NVIDIA RTX 6000 Ada
9. NVIDIA RTX 6000 Ada NVIDIA RTX 4000 Ada NVIDIA RTX 6000 Ada vs NVIDIA RTX 4000 Ada
10. NVIDIA RTX 6000 Ada MSI GeForce RTX 4090 SUPRIM 24G NVIDIA RTX 6000 Ada vs MSI GeForce RTX 4090 SUPRIM 24G


back to index
Facebook
Twitter