GPUs

AMD Radeon RX 7900 GRE vs NVIDIA GeForce RTX 3070 Full Specs

5,120 Shaders
2.25GHz
5,888 Shaders
1.73GHz
16GB GDDR6576GB/s
8GB GDDR6448GB/s
··
45.98 TFLOPS
··
20.31 TFLOPS
Form Factor
PCIe Card
Form Factor
PCIe Card
TDP
260W
TDP
220W
Power Connectors
-
2x 8-Pin
Power Connectors
1x 12-Pin
-

Radeon RX 7900 GRERadeon RX 7900 GRE183.9 TOPSINT4 Tensor
x1
GeForce RTX 3070GeForce RTX 3070650 TOPSINT4 Tensor Sparse
x3.53

Clock Speed
···
Clock Speed
···
Peak OPS
183.9 TOPSINT4 Tensor
Peak OPS
650 TOPSINT4 Tensor Sparse
Tensor FP16-16
45.98 TFLOPS
FP16-16 Tensor Sparse
-
Tensor FP16-32
45.98 TFLOPS
FP16-32 Tensor Sparse
-
Tensor FP16-16
81.25 TFLOPS
FP16-16 Tensor Sparse
162.5 TFLOPS
Tensor FP16-32
40.63 TFLOPS
FP16-32 Tensor Sparse
81.25 TFLOPS
BF16
91.96 TFLOPS
Tensor BF16
45.98 TFLOPS
BF16 Tensor Sparse
-
BF16
20.31 TFLOPS
Tensor BF16
40.63 TFLOPS
BF16 Tensor Sparse
81.25 TFLOPS
Tensor TF32
-
Tensor TF32
20.31 TFLOPS
FP32
45.98 TFLOPS
FP32
20.31 TFLOPS
FP64
1.44 TFLOPS
FP64
317.4 GFLOPS
Tensor INT4
183.9 TOPS
Tensor INT4
325 TOPS
Tensor INT8
45.98 TOPS
Tensor INT8
162.5 TOPS
Ray
-
Ray
30.64 TOPS
Pixel Rate
359.2 GPixel/s
Pixel Rate
165.6 GPixel/s
Texture Rate
718.4 GTexel/s
Texture Rate
317.4 GTexel/s

Shaders
5,120 Shaders
Shaders
5,888 Shaders
TMUs
320 TMUs
TMUs
184 TMUs
ROPs
160 ROPs
ROPs
96 ROPs
Tensor Cores
160 T-Cores
Tensor Cores
184 T-Cores
RT Cores
80 RT-Cores
RT Cores
46 RT-Cores
CUs
80 CUs
SMs
46 SMs

Base Clock
1.88GHz
Base Clock
1.5GHz
Boost Clock
2.25GHz
Boost Clock
1.73GHz

L2 Cache
6.1MB shared
L2 Cache
4.1MB shared
L3 Cache
64MB shared
L3 Cache
-
L3 Bandwidth
2.25TB/s
L3 Bandwidth
-

16GB GDDR6
8GB GDDR6
Memory Bus
256-bit
Memory Bus
256-bit
Memory Speed
18GT/s
Memory Speed
14GT/s
Memory Bandwidth
576GB/s
Memory Bandwidth
448GB/s

TDP
260W
TDP
220W
Max Temp
-
Max Temp
93°C Max

Multi-Monitor
3
Multi-Monitor
4
HDCP
HDCP 2.3
HDCP
HDCP 2.3

2x DisplayPort 2.1
-
-
3x DisplayPort 1.4
1x HDMI 2.1
1x HDMI 2.1
1x USB-C
-

Encoder Model
VCN 4.0
Encoder Model
NVENC 7

Decoder Model
VCN 4.0
Decoder Model
NVDEC 5

Form Factor
PCIe Card
Form Factor
PCIe Card
PCIe
2.5-Slots
PCIe
2.1-Slots
Height
110 mm (4.33")
Width
276 mm (10.87")
Depth
51 mm (2.01")
Height
112 mm (4.41")
Width
242 mm (9.53")
Depth
43 mm (1.69")
Cooling
Open-Air
3x Fans
Cooling
Open-Air
2x Fans
Power Connectors
2x 8-Pin
Power Connectors
1x 12-Pin

Manufacturer
Manufacturer
Chip Designer
Chip Designer
Architecture
Architecture
Family
Family
Branding
Radeon 2023 branding
Branding
GeForce RTX 2018 branding
Codename
Plum Bonito
Codename
NV174
Chip Variant
Navi 31 XL
Chip Variant
GA104-150-A1
Market Segment
Desktop
Market Segment
Desktop
Release Date
Jul 27, 2023
Release Date
Oct 29, 2020

Foundry
TSMC
Foundry
Samsung
Other Foundries
TSMCMCD
Other Foundries
-
Fabrication Node
N5
N6MCD
Fabrication Node
8N
-
Die Size
304mm²
150mm²MCD
Die Size
393mm²
-
Transistor Count
45.4 Billion
8.2 BillionMCD
Transistor Count
17.4 Billion
-
Transistor Density
149.2 MTr/mm²
54.67 MTr/mm²MCD
Transistor Density
44.33 MTr/mm²
-

No images available
No images available