NVIDIA GeForce RTX 4090 vs ATI Radeon HD 5770 Mac Full Specs

Summary

98,304 Shaders 2.52GHz	800 Shaders 850MHz
144GB GDDR6X6.05TB/s	1GB GDDR576.8GB/s
Peak·FP32·FP16 495.5 TFLOPS	Peak·FP32·FP16 1.36 TFLOPS
Form Factor PCIe Card	Form Factor PCIe Card
TDP 2700W	TDP 108W
Power Connectors 6x 16-Pin -	Power Connectors - 1x 6-Pin

Highlights

Peak OPS

Peak OPS FP32 FP16 Tensor FP8-16 FP8-16 Tensor Sparse Tensor FP8-32 FP8-32 Tensor Sparse Tensor FP16-16 FP16-16 Tensor Sparse Tensor FP16-32 FP16-32 Tensor Sparse FP64 BF16 Tensor BF16 BF16 Tensor Sparse Tensor TF32 Tensor INT8 Tensor INT4 Memory Bandwidth TDP

6x GeForce RTX 409015.85 POPSINT4 Tensor Sparse

x11657.70

Radeon HD 5770 Mac1.36 TFLOPSFP32

x1

Show Top Score

Theoretical Performance

Clock Speed2520MHz UC·Base·Boost·OC	Clock Speed850MHz UC·Boost·OC
Peak OPS 15.85 POPSINT4 Tensor Sparse	Peak OPS 1.36 TFLOPSFP32
Tensor FP8-16 3.96 PFLOPS FP8-16 Tensor Sparse 7.93 PFLOPS Tensor FP8-32 1.98 PFLOPS FP8-32 Tensor Sparse 3.96 PFLOPS	-
FP16 495.5 TFLOPS Tensor FP16-16 1.98 PFLOPS FP16-16 Tensor Sparse 3.96 PFLOPS Tensor FP16-32 990.9 TFLOPS FP16-32 Tensor Sparse 1.98 PFLOPS	-
BF16 495.5 TFLOPS Tensor BF16 990.9 TFLOPS BF16 Tensor Sparse 1.98 PFLOPS	-
Tensor TF32 495.5 TFLOPS	Tensor TF32 -
FP32 495.5 TFLOPS	FP32 1.36 TFLOPS
FP64 7.74 TFLOPS	-
Tensor INT4 7.93 POPS	Tensor INT4 -
Tensor INT8 3.96 POPS	-
Ray 1.15 POPS	Ray -
Pixel Rate 443.5 GPixel/s	Pixel Rate 13.6 GPixel/s
Texture Rate 1.29 TTexel/s	Texture Rate 34 GTexel/s

Core Config

Shaders 98,304 Shaders	Shaders 800 Shaders
TMUs 3,072 TMUs	TMUs 40 TMUs
ROPs 1,056 ROPs	ROPs 16 ROPs
Tensor Cores 3,072 T-Cores	Tensor Cores -
RT Cores 768 RT-Cores	RT Cores -
SMs 768 SMs	CUs 10 CUs

Clocks

Base Clock 2.23GHz	Base Clock -
Boost Clock 2.52GHz	Boost Clock 850MHz

Cache

L1 64KB/SM Tex	L1 8KB/CU
L2 Cache 72MB shared	L2 Cache 256KB shared

Memory

144GB GDDR6X	1GB GDDR5
Memory Bus 384-bit	Memory Bus 128-bit
Memory Speed 21GT/s	Memory Speed 4.8GT/s
Memory Bandwidth 6.05TB/s	Memory Bandwidth 76.8GB/s
ECC No	ECC No

Power & Thermals

TDP 2700W	TDP 108W
Max Temp 90°C Max	Max Temp -

Video Output

Max Resolution 7680x4320	Max Resolution 2560x1600
Refresh Rate Calculator 7680x4320 Resolution8K UHD Refresh Rate 60Hz	Refresh Rate Calculator 2560x1440 Resolution1440p QHD Refresh Rate 66Hz
Multi-Monitor 24	Multi-Monitor 2
Adaptive Sync G-Sync FreeSync	Adaptive Sync -
DSC Supported	DSC Not Supported
HDCP HDCP 2.3	HDCP -

Ports

18x DisplayPort 1.4 - 6x HDMI 2.1 -

- 2x Mini DisplayPort 1.1 - 1x DVI-I Dual-Link

API Support

Shader Model 6.6	Shader Model 5
DirectX DirectX 12 Direct3D 12_3	DirectX DirectX 11 Direct3D 11_0
OpenGL 4.6 OpenCL 3 Vulkan 1.3	OpenGL 4.5 OpenCL 1.2 Vulkan -
CUDA 8.9 GFX - PureVideo HD VP12 VDPAU Feature Set L	CUDA - GFX 4 PureVideo HD - VDPAU -

Encoder

Encoder 2x NVENC 8	-
Codec AVC (H.264) HEVC (H.265) AV1	Codec -

Decoder

Decoder NVDEC 5	Decoder UVD 2.2
Codec MPEG-1 MPEG-2 MPEG-4 VC-1 VP8 VP9 AVC (H.264) HEVC (H.265) AV1	Codec - - - VC-1 - - AVC (H.264) - -

Card

Form Factor PCIe Card	Form Factor PCIe Card
PCIe PCIe 4.0 x96 18-Slots	PCIe PCIe 2.1 x16 2-Slots
Height 137mm (5.39") Width 304mm (11.97") Depth 61mm (2.4")	Height 111mm (4.37") Width 208mm (8.19") Depth 37mm (1.46")
Cooling Open-Air 2x Fans	Cooling Blower 1x Fan
Power Connectors 6x 16-Pin	Power Connectors 1x 6-Pin
Multi-GPU -	Multi-GPU Multi-GPU Support Supported Multi-GPU Type 4-way CrossFireX

Chip

Manufacturer NVIDIA	Manufacturer ATI
Chip Designer NVIDIA	Chip Designer ATI
Architecture Ada Lovelace	Architecture TeraScale 2
Family GeForce 40	Family Radeon HD 5000
Branding	Branding
Codename NV182	Codename Juniper
Chip Variant AD102-200-A1	Chip Variant Juniper Pro
Market Segment Desktop	Market Segment Desktop
Release Date Oct 12, 2022	Release Date Oct 13, 2009

Fabrication

Foundry TSMC	Foundry TSMC
Fabrication Node 4N	Fabrication Node 40nm
Die Size 608mm²	Die Size 170mm²
Transistor Count 76.3B	Transistor Count 1B
Transistor Density 125.4 MTr/mm²	Transistor Density 6.12 MTr/mm²

Images