RTX 4070 SUPER
Intel Core i9-13900K testing with a ASUS TUF GAMING Z790-PRO WIFI (1401 BIOS) and NVIDIA GeForce RTX 3090 24GB on EndeavourOS rolling via the Phoronix Test Suite.


NVIDIA RTX 4070 SUPER: 

	Processor: Intel Core i9-13900K @ 5.50GHz (24 Cores / 32 Threads), Motherboard: ASUS TUF GAMING Z790-PRO WIFI (1401 BIOS), Chipset: Intel Device 7a27, Memory: 32GB, Disk: 4001GB Seagate ZP4000GP304001, Graphics: ASUS NVIDIA GeForce RTX 4070 SUPER 12GB, Audio: Realtek ALC1220, Monitor: ARZOPA, Network: Intel I226-V + Intel Device 7a70

	OS: EndeavourOS rolling, Kernel: 6.7.1-arch1-1 (x86_64), Desktop: KDE Plasma 5.27.10, Display Server: X Server 1.21.1.11, Display Driver: NVIDIA 550.40.07, OpenGL: 4.6.0, OpenCL: OpenCL 3.0 CUDA 12.4.74, Compiler: GCC 13.2.1 20230801, File-System: ext4, Screen Resolution: 1920x1080

NVIDIA RTX 4070: 

	Processor: Intel Core i9-13900K @ 5.50GHz (24 Cores / 32 Threads), Motherboard: ASUS TUF GAMING Z790-PRO WIFI (1401 BIOS), Chipset: Intel Device 7a27, Memory: 32GB, Disk: 4001GB Seagate ZP4000GP304001, Graphics: MSI NVIDIA GeForce RTX 4070 12GB, Audio: Realtek ALC1220, Monitor: ARZOPA, Network: Intel I226-V + Intel Device 7a70

	OS: EndeavourOS rolling, Kernel: 6.7.1-arch1-1 (x86_64), Desktop: KDE Plasma 5.27.10, Display Server: X Server 1.21.1.11, Display Driver: NVIDIA 550.40.07, OpenGL: 4.6.0, OpenCL: OpenCL 3.0 CUDA 12.4.74, Compiler: GCC 13.2.1 20230801 + CUDA 12.3, File-System: ext4, Screen Resolution: 1920x1080

NVIDIA RTX 4070 TI: 

	Processor: Intel Core i9-13900K @ 5.50GHz (24 Cores / 32 Threads), Motherboard: ASUS TUF GAMING Z790-PRO WIFI (1401 BIOS), Chipset: Intel Device 7a27, Memory: 32GB, Disk: 4001GB Seagate ZP4000GP304001, Graphics: NVIDIA GeForce RTX 4070 Ti 12GB, Audio: Realtek ALC1220, Monitor: ARZOPA, Network: Intel I226-V + Intel Device 7a70

	OS: EndeavourOS rolling, Kernel: 6.7.1-arch1-1 (x86_64), Desktop: KDE Plasma 5.27.10, Display Server: X Server 1.21.1.11, Display Driver: NVIDIA 550.40.07, OpenGL: 4.6.0, OpenCL: OpenCL 3.0 CUDA 12.4.74, Compiler: GCC 13.2.1 20230801 + CUDA 12.3, File-System: ext4, Screen Resolution: 1920x1080

NVIDIA RTX 3090: 

	Processor: Intel Core i9-13900K @ 5.50GHz (24 Cores / 32 Threads), Motherboard: ASUS TUF GAMING Z790-PRO WIFI (1401 BIOS), Chipset: Intel Device 7a27, Memory: 32GB, Disk: 4001GB Seagate ZP4000GP304001, Graphics: NVIDIA GeForce RTX 3090 24GB, Audio: Realtek ALC1220, Monitor: PI-KVM Video, Network: Intel I226-V + Intel Device 7a70

	OS: EndeavourOS rolling, Kernel: 6.7.4-arch1-1 (x86_64), Desktop: KDE Plasma 5.27.10, Display Server: X Server 1.21.1.11, Display Driver: NVIDIA 550.40.07, OpenGL: 4.6.0, OpenCL: OpenCL 3.0 CUDA 12.4.74, Compiler: GCC 13.2.1 20230801 + CUDA 12.3, File-System: ext4, Screen Resolution: 1920x1080


TensorFlow 2.12
Device: GPU - Batch Size: 1 - Model: AlexNet
images/sec > Higher Is Better
NVIDIA RTX 4070 SUPER . 13.92 |=============================================
NVIDIA RTX 4070 ....... 14.04 |==============================================
NVIDIA RTX 4070 TI .... 14.79 |================================================
NVIDIA RTX 3090 ....... 14.45 |===============================================


TensorFlow 2.12
Device: GPU - Batch Size: 16 - Model: VGG-16
images/sec > Higher Is Better
NVIDIA RTX 4070 SUPER . 1.48 |================================================
NVIDIA RTX 4070 ....... 1.50 |=================================================
NVIDIA RTX 4070 TI .... 1.49 |=================================================
NVIDIA RTX 3090 ....... 1.49 |=================================================


TensorFlow 2.12
Device: GPU - Batch Size: 32 - Model: VGG-16
images/sec > Higher Is Better
NVIDIA RTX 4070 SUPER . 1.50 |=================================================
NVIDIA RTX 4070 ....... 1.50 |=================================================
NVIDIA RTX 4070 TI .... 1.50 |=================================================
NVIDIA RTX 3090 ....... 1.50 |=================================================


TensorFlow 2.12
Device: GPU - Batch Size: 64 - Model: VGG-16
images/sec > Higher Is Better
NVIDIA RTX 4070 ....... 1.50 |=================================================
NVIDIA RTX 4070 TI .... 1.50 |=================================================
NVIDIA RTX 3090 ....... 1.51 |=================================================


TensorFlow 2.12
Device: GPU - Batch Size: 16 - Model: AlexNet
images/sec > Higher Is Better
NVIDIA RTX 4070 SUPER . 31.59 |===============================================
NVIDIA RTX 4070 ....... 31.45 |===============================================
NVIDIA RTX 4070 TI .... 31.70 |================================================
NVIDIA RTX 3090 ....... 31.98 |================================================


TensorFlow 2.12
Device: GPU - Batch Size: 32 - Model: AlexNet
images/sec > Higher Is Better
NVIDIA RTX 4070 SUPER . 33.40 |================================================
NVIDIA RTX 4070 ....... 33.32 |================================================
NVIDIA RTX 4070 TI .... 33.29 |================================================
NVIDIA RTX 3090 ....... 33.53 |================================================


ProjectPhysX OpenCL-Benchmark 1.2
Operation: Memory Bandwidth Coalesced Write
GB/s > Higher Is Better
NVIDIA RTX 4070 SUPER . 455.01 |========================
NVIDIA RTX 4070 ....... 459.43 |========================
NVIDIA RTX 4070 TI .... 457.17 |========================
NVIDIA RTX 3090 ....... 887.31 |===============================================


TensorFlow 2.12
Device: GPU - Batch Size: 1 - Model: VGG-16
images/sec > Higher Is Better
NVIDIA RTX 4070 SUPER . 1.35 |================================================
NVIDIA RTX 4070 ....... 1.36 |================================================
NVIDIA RTX 4070 TI .... 1.38 |=================================================
NVIDIA RTX 3090 ....... 1.38 |=================================================


ProjectPhysX OpenCL-Benchmark 1.2
Operation: INT8 Compute
TIOPs/s > Higher Is Better
NVIDIA RTX 4070 SUPER . 14.31 |============================================
NVIDIA RTX 4070 ....... 12.12 |=====================================
NVIDIA RTX 4070 TI .... 15.73 |================================================
NVIDIA RTX 3090 ....... 13.73 |==========================================


ProjectPhysX OpenCL-Benchmark 1.2
Operation: Memory Bandwidth Coalesced Read
GB/s > Higher Is Better
NVIDIA RTX 4070 SUPER . 464.86 |=========================
NVIDIA RTX 4070 ....... 465.18 |=========================
NVIDIA RTX 4070 TI .... 465.07 |=========================
NVIDIA RTX 3090 ....... 864.11 |===============================================


ProjectPhysX OpenCL-Benchmark 1.2
Operation: INT16 Compute
TIOPs/s > Higher Is Better
NVIDIA RTX 4070 SUPER . 17.17 |=============================================
NVIDIA RTX 4070 ....... 14.28 |=====================================
NVIDIA RTX 4070 TI .... 18.28 |================================================
NVIDIA RTX 3090 ....... 17.00 |=============================================


ProjectPhysX OpenCL-Benchmark 1.2
Operation: INT32 Compute
TIOPs/s > Higher Is Better
NVIDIA RTX 4070 SUPER . 19.89 |=============================================
NVIDIA RTX 4070 ....... 16.38 |=====================================
NVIDIA RTX 4070 TI .... 21.05 |================================================
NVIDIA RTX 3090 ....... 20.03 |==============================================


ProjectPhysX OpenCL-Benchmark 1.2
Operation: INT64 Compute
TIOPs/s > Higher Is Better
NVIDIA RTX 4070 SUPER . 4.214 |==============================================
NVIDIA RTX 4070 ....... 3.443 |=====================================
NVIDIA RTX 4070 TI .... 4.420 |================================================
NVIDIA RTX 3090 ....... 3.135 |==================================


TensorFlow 2.12
Device: GPU - Batch Size: 256 - Model: VGG-16
images/sec > Higher Is Better
NVIDIA RTX 4070 TI .... 1.50 |=================================================
NVIDIA RTX 3090 ....... 1.51 |=================================================


TensorFlow 2.12
Device: GPU - Batch Size: 512 - Model: VGG-16
images/sec > Higher Is Better


TensorFlow 2.12
Device: GPU - Batch Size: 64 - Model: AlexNet
images/sec > Higher Is Better
NVIDIA RTX 4070 SUPER . 33.97 |================================================
NVIDIA RTX 4070 ....... 33.93 |================================================
NVIDIA RTX 4070 TI .... 34.06 |================================================
NVIDIA RTX 3090 ....... 33.93 |================================================


GpuOwl 7.2.1
Exponent: 77936867
Iterations / Second > Higher Is Better
NVIDIA RTX 4070 SUPER . 646.41 |=============================================
NVIDIA RTX 4070 ....... 530.32 |=====================================
NVIDIA RTX 4070 TI .... 676.59 |===============================================
NVIDIA RTX 3090 ....... 645.99 |=============================================


GpuOwl 7.2.1
Exponent: 332220523
Iterations / Second > Higher Is Better
NVIDIA RTX 4070 SUPER . 137.44 |============================================
NVIDIA RTX 4070 ....... 112.61 |====================================
NVIDIA RTX 4070 TI .... 145.84 |===============================================
NVIDIA RTX 3090 ....... 137.32 |============================================


TensorFlow 2.12
Device: GPU - Batch Size: 1 - Model: GoogLeNet
images/sec > Higher Is Better
NVIDIA RTX 4070 SUPER . 12.62 |===============================================
NVIDIA RTX 4070 ....... 12.78 |================================================
NVIDIA RTX 4070 TI .... 12.79 |================================================
NVIDIA RTX 3090 ....... 12.82 |================================================


GpuOwl 7.2.1
Exponent: 57885161
Iterations / Second > Higher Is Better
NVIDIA RTX 4070 SUPER . 869.07 |============================================
NVIDIA RTX 4070 ....... 714.80 |=====================================
NVIDIA RTX 4070 TI .... 919.13 |===============================================
NVIDIA RTX 3090 ....... 866.31 |============================================


TensorFlow 2.12
Device: GPU - Batch Size: 1 - Model: ResNet-50
images/sec > Higher Is Better
NVIDIA RTX 4070 SUPER . 4.35 |=================================================
NVIDIA RTX 4070 ....... 4.34 |=================================================
NVIDIA RTX 4070 TI .... 4.32 |=================================================
NVIDIA RTX 3090 ....... 4.35 |=================================================


TensorFlow 2.12
Device: GPU - Batch Size: 256 - Model: AlexNet
images/sec > Higher Is Better
NVIDIA RTX 4070 SUPER . 34.16 |===============================================
NVIDIA RTX 4070 TI .... 34.61 |================================================
NVIDIA RTX 3090 ....... 34.46 |================================================


TensorFlow 2.12
Device: GPU - Batch Size: 512 - Model: AlexNet
images/sec > Higher Is Better
NVIDIA RTX 4070 SUPER . 35.10 |===============================================
NVIDIA RTX 4070 ....... 35.21 |================================================
NVIDIA RTX 4070 TI .... 35.44 |================================================
NVIDIA RTX 3090 ....... 35.58 |================================================


TensorFlow 2.12
Device: GPU - Batch Size: 16 - Model: GoogLeNet
images/sec > Higher Is Better
NVIDIA RTX 4070 SUPER . 15.67 |================================================
NVIDIA RTX 4070 ....... 15.66 |================================================
NVIDIA RTX 4070 TI .... 15.69 |================================================
NVIDIA RTX 3090 ....... 15.68 |================================================


TensorFlow 2.12
Device: GPU - Batch Size: 16 - Model: ResNet-50
images/sec > Higher Is Better
NVIDIA RTX 4070 SUPER . 5.46 |=================================================
NVIDIA RTX 4070 ....... 5.49 |=================================================
NVIDIA RTX 4070 TI .... 5.46 |=================================================
NVIDIA RTX 3090 ....... 5.49 |=================================================


TensorFlow 2.12
Device: GPU - Batch Size: 32 - Model: GoogLeNet
images/sec > Higher Is Better
NVIDIA RTX 4070 SUPER . 15.61 |===============================================
NVIDIA RTX 4070 ....... 15.63 |===============================================
NVIDIA RTX 4070 TI .... 15.81 |================================================
NVIDIA RTX 3090 ....... 15.67 |================================================


TensorFlow 2.12
Device: GPU - Batch Size: 32 - Model: ResNet-50
images/sec > Higher Is Better
NVIDIA RTX 4070 SUPER . 5.51 |================================================
NVIDIA RTX 4070 ....... 5.55 |=================================================
NVIDIA RTX 4070 TI .... 5.50 |================================================
NVIDIA RTX 3090 ....... 5.57 |=================================================


TensorFlow 2.12
Device: GPU - Batch Size: 64 - Model: GoogLeNet
images/sec > Higher Is Better
NVIDIA RTX 4070 SUPER . 15.52 |================================================
NVIDIA RTX 4070 ....... 15.54 |================================================
NVIDIA RTX 4070 TI .... 15.50 |================================================
NVIDIA RTX 3090 ....... 15.63 |================================================


TensorFlow 2.12
Device: GPU - Batch Size: 64 - Model: ResNet-50
images/sec > Higher Is Better
NVIDIA RTX 4070 SUPER . 5.55 |=================================================
NVIDIA RTX 4070 ....... 5.55 |=================================================
NVIDIA RTX 4070 TI .... 5.53 |=================================================
NVIDIA RTX 3090 ....... 5.57 |=================================================


PlaidML
FP16: No - Mode: Training - Network: Mobilenet - Device: OpenCL
Examples Per Second > Higher Is Better


PlaidML
FP16: No - Mode: Inference - Network: IMDB LSTM - Device: OpenCL
Examples Per Second > Higher Is Better


PlaidML
FP16: No - Mode: Inference - Network: Mobilenet - Device: OpenCL
Examples Per Second > Higher Is Better


PlaidML
FP16: Yes - Mode: Inference - Network: Mobilenet - Device: OpenCL
Examples Per Second > Higher Is Better


PlaidML
FP16: No - Mode: Inference - Network: DenseNet 201 - Device: OpenCL
Examples Per Second > Higher Is Better


LeelaChessZero 0.30
Backend: OpenCL
Nodes Per Second > Higher Is Better


PyTorch 2.1
Device: NVIDIA CUDA GPU - Batch Size: 1 - Model: ResNet-50
batches/sec > Higher Is Better
NVIDIA RTX 4070 SUPER . 557.73 |===============================================
NVIDIA RTX 4070 ....... 546.76 |==============================================
NVIDIA RTX 4070 TI .... 535.39 |=============================================
NVIDIA RTX 3090 ....... 525.12 |============================================


PyTorch 2.1
Device: NVIDIA CUDA GPU - Batch Size: 1 - Model: ResNet-152
batches/sec > Higher Is Better
NVIDIA RTX 4070 SUPER . 201.94 |===============================================
NVIDIA RTX 4070 ....... 198.18 |==============================================
NVIDIA RTX 4070 TI .... 201.19 |===============================================
NVIDIA RTX 3090 ....... 197.12 |==============================================


PyTorch 2.1
Device: NVIDIA CUDA GPU - Batch Size: 16 - Model: ResNet-50
batches/sec > Higher Is Better
NVIDIA RTX 4070 SUPER . 509.45 |===============================================
NVIDIA RTX 4070 ....... 458.39 |==========================================
NVIDIA RTX 4070 TI .... 502.92 |==============================================
NVIDIA RTX 3090 ....... 419.76 |=======================================


PyTorch 2.1
Device: NVIDIA CUDA GPU - Batch Size: 32 - Model: ResNet-50
batches/sec > Higher Is Better
NVIDIA RTX 4070 SUPER . 501.50 |===============================================
NVIDIA RTX 4070 ....... 459.94 |===========================================
NVIDIA RTX 4070 TI .... 505.55 |===============================================
NVIDIA RTX 3090 ....... 420.29 |=======================================


PyTorch 2.1
Device: NVIDIA CUDA GPU - Batch Size: 64 - Model: ResNet-50
batches/sec > Higher Is Better
NVIDIA RTX 4070 SUPER . 507.45 |===============================================
NVIDIA RTX 4070 ....... 458.36 |==========================================
NVIDIA RTX 4070 TI .... 505.62 |===============================================
NVIDIA RTX 3090 ....... 419.03 |=======================================


PyTorch 2.1
Device: NVIDIA CUDA GPU - Batch Size: 16 - Model: ResNet-152
batches/sec > Higher Is Better
NVIDIA RTX 4070 SUPER . 195.40 |===============================================
NVIDIA RTX 4070 ....... 187.26 |=============================================
NVIDIA RTX 4070 TI .... 194.29 |===============================================
NVIDIA RTX 3090 ....... 164.14 |=======================================


PyTorch 2.1
Device: NVIDIA CUDA GPU - Batch Size: 256 - Model: ResNet-50
batches/sec > Higher Is Better
NVIDIA RTX 4070 SUPER . 504.67 |===============================================
NVIDIA RTX 4070 ....... 459.93 |===========================================
NVIDIA RTX 3090 ....... 416.89 |=======================================


PyTorch 2.1
Device: NVIDIA CUDA GPU - Batch Size: 32 - Model: ResNet-152
batches/sec > Higher Is Better
NVIDIA RTX 4070 SUPER . 195.39 |==============================================
NVIDIA RTX 4070 ....... 187.69 |============================================
NVIDIA RTX 4070 TI .... 198.82 |===============================================
NVIDIA RTX 3090 ....... 163.74 |=======================================


PyTorch 2.1
Device: NVIDIA CUDA GPU - Batch Size: 512 - Model: ResNet-50
batches/sec > Higher Is Better
NVIDIA RTX 4070 SUPER . 504.27 |===============================================
NVIDIA RTX 4070 ....... 459.27 |===========================================
NVIDIA RTX 4070 TI .... 504.66 |===============================================
NVIDIA RTX 3090 ....... 416.20 |=======================================


PyTorch 2.1
Device: NVIDIA CUDA GPU - Batch Size: 64 - Model: ResNet-152
batches/sec > Higher Is Better
NVIDIA RTX 4070 SUPER . 196.07 |===============================================
NVIDIA RTX 4070 ....... 186.63 |=============================================
NVIDIA RTX 4070 TI .... 197.02 |===============================================
NVIDIA RTX 3090 ....... 164.14 |=======================================


PyTorch 2.1
Device: NVIDIA CUDA GPU - Batch Size: 256 - Model: ResNet-152
batches/sec > Higher Is Better
NVIDIA RTX 4070 SUPER . 194.58 |===============================================
NVIDIA RTX 4070 ....... 187.27 |=============================================
NVIDIA RTX 4070 TI .... 195.86 |===============================================
NVIDIA RTX 3090 ....... 161.01 |=======================================


PyTorch 2.1
Device: NVIDIA CUDA GPU - Batch Size: 512 - Model: ResNet-152
batches/sec > Higher Is Better
NVIDIA RTX 4070 SUPER . 195.30 |===============================================
NVIDIA RTX 4070 ....... 187.51 |=============================================
NVIDIA RTX 4070 TI .... 194.87 |===============================================
NVIDIA RTX 3090 ....... 164.35 |========================================


PyTorch 2.1
Device: NVIDIA CUDA GPU - Batch Size: 1 - Model: Efficientnet_v2_l
batches/sec > Higher Is Better
NVIDIA RTX 4070 SUPER . 106.37 |==============================================
NVIDIA RTX 4070 ....... 107.59 |===============================================
NVIDIA RTX 4070 TI .... 108.59 |===============================================
NVIDIA RTX 3090 ....... 105.55 |==============================================


PyTorch 2.1
Device: NVIDIA CUDA GPU - Batch Size: 16 - Model: Efficientnet_v2_l
batches/sec > Higher Is Better
NVIDIA RTX 4070 ....... 103.68 |===============================================
NVIDIA RTX 4070 TI .... 103.45 |===============================================
NVIDIA RTX 3090 ....... 98.11  |============================================


PyTorch 2.1
Device: NVIDIA CUDA GPU - Batch Size: 32 - Model: Efficientnet_v2_l
batches/sec > Higher Is Better
NVIDIA RTX 4070 SUPER . 102.60 |===============================================
NVIDIA RTX 4070 ....... 102.90 |===============================================
NVIDIA RTX 4070 TI .... 96.50  |============================================
NVIDIA RTX 3090 ....... 99.05  |=============================================


PyTorch 2.1
Device: NVIDIA CUDA GPU - Batch Size: 64 - Model: Efficientnet_v2_l
batches/sec > Higher Is Better
NVIDIA RTX 4070 SUPER . 102.60 |===============================================
NVIDIA RTX 4070 ....... 101.55 |==============================================
NVIDIA RTX 4070 TI .... 103.20 |===============================================
NVIDIA RTX 3090 ....... 99.84  |=============================================


PyTorch 2.1
Device: NVIDIA CUDA GPU - Batch Size: 256 - Model: Efficientnet_v2_l
batches/sec > Higher Is Better
NVIDIA RTX 4070 SUPER . 103.17 |===============================================
NVIDIA RTX 4070 ....... 101.24 |==============================================
NVIDIA RTX 4070 TI .... 103.24 |===============================================
NVIDIA RTX 3090 ....... 99.43  |=============================================


PyTorch 2.1
Device: NVIDIA CUDA GPU - Batch Size: 512 - Model: Efficientnet_v2_l
batches/sec > Higher Is Better
NVIDIA RTX 4070 SUPER . 103.57 |===============================================
NVIDIA RTX 4070 ....... 101.43 |==============================================
NVIDIA RTX 4070 TI .... 103.50 |===============================================
NVIDIA RTX 3090 ....... 99.25  |=============================================


Caffe 2020-02-13
Model: AlexNet - Acceleration: NVIDIA CUDA  - Iterations: 100
Milli-Seconds < Lower Is Better


Caffe 2020-02-13
Model: AlexNet - Acceleration: NVIDIA CUDA  - Iterations: 200
Milli-Seconds < Lower Is Better


Caffe 2020-02-13
Model: AlexNet - Acceleration: NVIDIA CUDA  - Iterations: 1000
Milli-Seconds < Lower Is Better


Caffe 2020-02-13
Model: GoogleNet - Acceleration: NVIDIA CUDA  - Iterations: 100
Milli-Seconds < Lower Is Better


Caffe 2020-02-13
Model: GoogleNet - Acceleration: NVIDIA CUDA  - Iterations: 200
Milli-Seconds < Lower Is Better


Caffe 2020-02-13
Model: GoogleNet - Acceleration: NVIDIA CUDA  - Iterations: 1000
Milli-Seconds < Lower Is Better


NCNN 20230517
Target: Vulkan GPU
ms < Lower Is Better


NCNN 20230517
Target: Vulkan GPU - Model: mobilenet
ms < Lower Is Better
NVIDIA RTX 4070 ....... 7.20  |=============================
NVIDIA RTX 4070 TI .... 7.45  |==============================
NVIDIA RTX 3090 ....... 6.92  |============================
NVIDIA RTX 3090 ....... 7.27  |=============================
NVIDIA RTX 4070 SUPER . 8.62  |==================================
NVIDIA RTX 4070 ....... 10.14 |========================================
NVIDIA RTX 4070 TI .... 8.43  |==================================
NVIDIA RTX 3090 ....... 12.07 |================================================


NCNN 20230517
Target: Vulkan GPU-v2-v2 - Model: mobilenet-v2
ms < Lower Is Better
NVIDIA RTX 4070 ....... 2.48 |==========================
NVIDIA RTX 4070 TI .... 2.54 |===========================
NVIDIA RTX 3090 ....... 2.67 |============================
NVIDIA RTX 3090 ....... 2.34 |========================
NVIDIA RTX 4070 SUPER . 3.03 |================================
NVIDIA RTX 4070 ....... 4.69 |=================================================
NVIDIA RTX 4070 TI .... 2.43 |=========================
NVIDIA RTX 3090 ....... 2.65 |============================


NCNN 20230517
Target: Vulkan GPU-v3-v3 - Model: mobilenet-v3
ms < Lower Is Better
NVIDIA RTX 4070 ....... 2.15 |============
NVIDIA RTX 3090 ....... 2.20 |============
NVIDIA RTX 3090 ....... 2.21 |============
NVIDIA RTX 4070 SUPER . 2.25 |=============
NVIDIA RTX 4070 ....... 8.71 |=================================================
NVIDIA RTX 4070 TI .... 2.09 |============
NVIDIA RTX 3090 ....... 3.19 |==================


NCNN 20230517
Target: Vulkan GPU - Model: shufflenet-v2
ms < Lower Is Better
NVIDIA RTX 4070 ....... 2.08 |========================
NVIDIA RTX 4070 TI .... 2.01 |========================
NVIDIA RTX 3090 ....... 2.09 |=========================
NVIDIA RTX 3090 ....... 2.04 |========================
NVIDIA RTX 4070 SUPER . 2.31 |===========================
NVIDIA RTX 4070 ....... 2.11 |=========================
NVIDIA RTX 4070 TI .... 2.03 |========================
NVIDIA RTX 3090 ....... 4.17 |=================================================


NCNN 20230517
Target: Vulkan GPU - Model: mnasnet
ms < Lower Is Better
NVIDIA RTX 4070 ....... 2.24 |===========================
NVIDIA RTX 4070 TI .... 4.14 |=================================================
NVIDIA RTX 3090 ....... 2.30 |===========================
NVIDIA RTX 3090 ....... 2.16 |==========================
NVIDIA RTX 4070 SUPER . 3.85 |==============================================
NVIDIA RTX 4070 ....... 2.22 |==========================
NVIDIA RTX 4070 TI .... 2.30 |===========================
NVIDIA RTX 3090 ....... 2.24 |===========================


NCNN 20230517
Target: Vulkan GPU - Model: efficientnet-b0
ms < Lower Is Better
NVIDIA RTX 4070 ....... 3.59  |============
NVIDIA RTX 4070 TI .... 3.46  |============
NVIDIA RTX 3090 ....... 3.54  |============
NVIDIA RTX 3090 ....... 3.34  |============
NVIDIA RTX 4070 SUPER . 5.07  |==================
NVIDIA RTX 4070 ....... 3.46  |============
NVIDIA RTX 4070 TI .... 3.49  |============
NVIDIA RTX 3090 ....... 13.87 |================================================


NCNN 20230517
Target: Vulkan GPU - Model: blazeface
ms < Lower Is Better
NVIDIA RTX 4070 TI .... 0.82 |==============================================
NVIDIA RTX 3090 ....... 0.84 |===============================================
NVIDIA RTX 3090 ....... 0.87 |=================================================
NVIDIA RTX 4070 SUPER . 0.84 |===============================================
NVIDIA RTX 4070 ....... 0.84 |===============================================
NVIDIA RTX 4070 TI .... 0.81 |==============================================
NVIDIA RTX 3090 ....... 0.86 |================================================


NCNN 20230517
Target: Vulkan GPU - Model: googlenet
ms < Lower Is Better
NVIDIA RTX 4070 TI .... 7.37  |================================
NVIDIA RTX 3090 ....... 6.11  |===========================
NVIDIA RTX 3090 ....... 6.14  |===========================
NVIDIA RTX 4070 SUPER . 11.04 |================================================
NVIDIA RTX 4070 ....... 6.06  |==========================
NVIDIA RTX 4070 TI .... 5.87  |==========================
NVIDIA RTX 3090 ....... 7.49  |=================================


NCNN 20230517
Target: Vulkan GPU - Model: vgg16
ms < Lower Is Better
NVIDIA RTX 4070 ....... 45.52  |===============
NVIDIA RTX 4070 TI .... 34.49  |===========
NVIDIA RTX 3090 ....... 24.45  |========
NVIDIA RTX 3090 ....... 17.88  |======
NVIDIA RTX 4070 SUPER . 117.81 |======================================
NVIDIA RTX 4070 ....... 54.54  |==================
NVIDIA RTX 4070 TI .... 32.05  |==========
NVIDIA RTX 3090 ....... 145.72 |===============================================


NCNN 20230517
Target: Vulkan GPU - Model: resnet18
ms < Lower Is Better
NVIDIA RTX 4070 ....... 5.11  |==============
NVIDIA RTX 4070 TI .... 7.74  |=====================
NVIDIA RTX 3090 ....... 8.94  |=========================
NVIDIA RTX 3090 ....... 4.12  |===========
NVIDIA RTX 4070 SUPER . 8.97  |=========================
NVIDIA RTX 4070 ....... 8.58  |========================
NVIDIA RTX 4070 TI .... 5.47  |===============
NVIDIA RTX 3090 ....... 17.41 |================================================


NCNN 20230517
Target: Vulkan GPU - Model: alexnet
ms < Lower Is Better
NVIDIA RTX 4070 ....... 5.78  |=================
NVIDIA RTX 4070 TI .... 6.07  |==================
NVIDIA RTX 3090 ....... 6.20  |==================
NVIDIA RTX 3090 ....... 3.60  |===========
NVIDIA RTX 4070 SUPER . 16.17 |================================================
NVIDIA RTX 4070 ....... 9.33  |============================
NVIDIA RTX 4070 TI .... 3.74  |===========
NVIDIA RTX 3090 ....... 3.69  |===========


NCNN 20230517
Target: Vulkan GPU - Model: resnet50
ms < Lower Is Better
NVIDIA RTX 4070 ....... 8.72  |=========
NVIDIA RTX 4070 TI .... 12.25 |=============
NVIDIA RTX 3090 ....... 8.20  |=========
NVIDIA RTX 3090 ....... 12.70 |=============
NVIDIA RTX 4070 SUPER . 46.26 |================================================
NVIDIA RTX 4070 ....... 8.24  |=========
NVIDIA RTX 4070 TI .... 14.32 |===============
NVIDIA RTX 3090 ....... 27.77 |=============================


NCNN 20230517
Target: Vulkan GPU - Model: yolov4-tiny
ms < Lower Is Better
NVIDIA RTX 4070 ....... 20.74 |================
NVIDIA RTX 4070 TI .... 16.37 |============
NVIDIA RTX 3090 ....... 13.31 |==========
NVIDIA RTX 3090 ....... 11.29 |========
NVIDIA RTX 4070 SUPER . 63.82 |================================================
NVIDIA RTX 4070 ....... 25.11 |===================
NVIDIA RTX 4070 TI .... 16.47 |============
NVIDIA RTX 3090 ....... 26.85 |====================


NCNN 20230517
Target: Vulkan GPU - Model: squeezenet_ssd
ms < Lower Is Better
NVIDIA RTX 4070 ....... 5.18 |=====================================
NVIDIA RTX 4070 TI .... 6.13 |============================================
NVIDIA RTX 3090 ....... 5.20 |=====================================
NVIDIA RTX 3090 ....... 4.90 |===================================
NVIDIA RTX 4070 SUPER . 6.86 |=================================================
NVIDIA RTX 4070 ....... 5.27 |======================================
NVIDIA RTX 4070 TI .... 5.36 |======================================
NVIDIA RTX 3090 ....... 6.63 |===============================================


NCNN 20230517
Target: Vulkan GPU - Model: regnety_400m
ms < Lower Is Better
NVIDIA RTX 4070 ....... 6.21  |===========================
NVIDIA RTX 4070 TI .... 5.89  |=========================
NVIDIA RTX 3090 ....... 6.47  |============================
NVIDIA RTX 3090 ....... 6.73  |=============================
NVIDIA RTX 4070 SUPER . 11.11 |================================================
NVIDIA RTX 4070 ....... 6.50  |============================
NVIDIA RTX 4070 TI .... 5.97  |==========================
NVIDIA RTX 3090 ....... 8.06  |===================================


NCNN 20230517
Target: Vulkan GPU - Model: vision_transformer
ms < Lower Is Better
NVIDIA RTX 4070 ....... 382.82 |=====================
NVIDIA RTX 4070 TI .... 497.66 |============================
NVIDIA RTX 3090 ....... 327.82 |==================
NVIDIA RTX 3090 ....... 354.57 |====================
NVIDIA RTX 4070 SUPER . 844.61 |===============================================
NVIDIA RTX 4070 ....... 281.56 |================
NVIDIA RTX 4070 TI .... 390.18 |======================
NVIDIA RTX 3090 ....... 663.24 |=====================================


NCNN 20230517
Target: Vulkan GPU - Model: FastestDet
ms < Lower Is Better
NVIDIA RTX 4070 ....... 2.67 |=====================
NVIDIA RTX 4070 TI .... 3.04 |=======================
NVIDIA RTX 3090 ....... 2.50 |===================
NVIDIA RTX 3090 ....... 2.65 |====================
NVIDIA RTX 4070 SUPER . 2.86 |======================
NVIDIA RTX 4070 ....... 2.34 |==================
NVIDIA RTX 4070 TI .... 2.84 |======================
NVIDIA RTX 3090 ....... 6.38 |=================================================


Rodinia 3.1
Test: OpenCL Particle Filter
Seconds < Lower Is Better
NVIDIA RTX 4070 SUPER . 3.480 |=========================================
NVIDIA RTX 4070 ....... 4.098 |================================================
NVIDIA RTX 4070 TI .... 3.291 |=======================================
NVIDIA RTX 3090 ....... 3.844 |=============================================


ArrayFire 3.9
Test: Conjugate Gradient OpenCL


ProjectPhysX OpenCL-Benchmark 1.2
Operation: FP32 Compute
TFLOPs/s > Higher Is Better
NVIDIA RTX 4070 SUPER . 38.59 |=============================================
NVIDIA RTX 4070 ....... 31.77 |=====================================
NVIDIA RTX 4070 TI .... 40.91 |================================================
NVIDIA RTX 3090 ....... 39.40 |==============================================


Blender 4.0
Blend File: BMW27 - Compute: NVIDIA OptiX
Seconds < Lower Is Better
NVIDIA RTX 4070 SUPER . 5.57 |===========================================
NVIDIA RTX 4070 ....... 6.21 |================================================
NVIDIA RTX 4070 TI .... 5.43 |==========================================
NVIDIA RTX 3090 ....... 6.31 |=================================================


Blender 4.0
Blend File: Classroom - Compute: NVIDIA OptiX
Seconds < Lower Is Better
NVIDIA RTX 4070 SUPER . 12.60 |========================================
NVIDIA RTX 4070 ....... 14.86 |===============================================
NVIDIA RTX 4070 TI .... 12.30 |=======================================
NVIDIA RTX 3090 ....... 15.26 |================================================


Blender 4.0
Blend File: Fishy Cat - Compute: NVIDIA OptiX
Seconds < Lower Is Better
NVIDIA RTX 4070 SUPER . 9.45  |=========================================
NVIDIA RTX 4070 ....... 11.03 |================================================
NVIDIA RTX 4070 TI .... 9.02  |=======================================
NVIDIA RTX 3090 ....... 10.64 |==============================================


Blender 4.0
Blend File: Barbershop - Compute: NVIDIA OptiX
Seconds < Lower Is Better
NVIDIA RTX 4070 SUPER . 51.30 |==========================================
NVIDIA RTX 4070 ....... 58.44 |================================================
NVIDIA RTX 4070 TI .... 50.73 |==========================================
NVIDIA RTX 3090 ....... 54.30 |=============================================


Blender 4.0
Blend File: Pabellon Barcelona - Compute: NVIDIA OptiX
Seconds < Lower Is Better
NVIDIA RTX 4070 SUPER . 14.29 |========================================
NVIDIA RTX 4070 ....... 16.55 |==============================================
NVIDIA RTX 4070 TI .... 13.97 |=======================================
NVIDIA RTX 3090 ....... 17.30 |================================================


NeatBench 5
Acceleration: GPU
FPS > Higher Is Better
NVIDIA RTX 4070 SUPER . 4070 |=================================================
NVIDIA RTX 4070 ....... 4070 |=================================================
NVIDIA RTX 4070 TI .... 4070 |=================================================
NVIDIA RTX 3090 ....... 3090 |=====================================


IndigoBench 4.4
Acceleration: OpenCL GPU - Scene: Bedroom
M samples/s > Higher Is Better
NVIDIA RTX 4070 SUPER . 19.80 |=============================================
NVIDIA RTX 4070 ....... 18.20 |==========================================
NVIDIA RTX 4070 TI .... 20.26 |==============================================
NVIDIA RTX 3090 ....... 20.96 |================================================


IndigoBench 4.4
Acceleration: OpenCL GPU - Scene: Supercar
M samples/s > Higher Is Better
NVIDIA RTX 4070 SUPER . 52.81 |===============================================
NVIDIA RTX 4070 ....... 48.52 |===========================================
NVIDIA RTX 4070 TI .... 53.59 |================================================
NVIDIA RTX 3090 ....... 52.01 |===============================================


LuxCoreRender 2.6
Scene: DLSC - Acceleration: GPU
M samples/sec > Higher Is Better
NVIDIA RTX 4070 SUPER . 13.59 |===============================================
NVIDIA RTX 4070 ....... 11.74 |========================================
NVIDIA RTX 4070 TI .... 13.95 |================================================
NVIDIA RTX 3090 ....... 12.99 |=============================================


LuxCoreRender 2.6
Scene: Danish Mood - Acceleration: GPU
M samples/sec > Higher Is Better
NVIDIA RTX 4070 SUPER . 10.56 |==============================================
NVIDIA RTX 4070 ....... 8.89  |=======================================
NVIDIA RTX 4070 TI .... 10.99 |================================================
NVIDIA RTX 3090 ....... 10.20 |=============================================


LuxCoreRender 2.6
Scene: Orange Juice - Acceleration: GPU
M samples/sec > Higher Is Better
NVIDIA RTX 4070 SUPER . 11.72 |==============================================
NVIDIA RTX 4070 ....... 10.40 |=========================================
NVIDIA RTX 4070 TI .... 11.89 |===============================================
NVIDIA RTX 3090 ....... 12.14 |================================================


LuxCoreRender 2.6
Scene: LuxCore Benchmark - Acceleration: GPU
M samples/sec > Higher Is Better
NVIDIA RTX 4070 SUPER . 12.82 |===============================================
NVIDIA RTX 4070 ....... 10.92 |========================================
NVIDIA RTX 4070 TI .... 13.23 |================================================
NVIDIA RTX 3090 ....... 13.12 |================================================


LuxCoreRender 2.6
Scene: Rainbow Colors and Prism - Acceleration: GPU
M samples/sec > Higher Is Better
NVIDIA RTX 4070 SUPER . 27.67 |========================================
NVIDIA RTX 4070 ....... 23.26 |==================================
NVIDIA RTX 4070 TI .... 27.71 |========================================
NVIDIA RTX 3090 ....... 33.29 |================================================


FAHBench 2.3.2

Ns Per Day > Higher Is Better
NVIDIA RTX 4070 SUPER . 366.06 |=============================================
NVIDIA RTX 4070 ....... 317.20 |=======================================
NVIDIA RTX 4070 TI .... 382.16 |===============================================
NVIDIA RTX 3090 ....... 343.02 |==========================================


Hashcat 6.2.4
Benchmark: MD5
H/s > Higher Is Better
NVIDIA RTX 4070 SUPER . 67583033333 |=======================================
NVIDIA RTX 4070 ....... 56147866667 |================================
NVIDIA RTX 4070 TI .... 73312233333 |==========================================
NVIDIA RTX 3090 ....... 67177300000 |======================================


Hashcat 6.2.4
Benchmark: SHA1
H/s > Higher Is Better
NVIDIA RTX 4070 SUPER . 22132600000 |========================================
NVIDIA RTX 4070 ....... 18202466667 |================================
NVIDIA RTX 4070 TI .... 23532400000 |==========================================
NVIDIA RTX 3090 ....... 21323733333 |======================================


Hashcat 6.2.4
Benchmark: 7-Zip
H/s > Higher Is Better
NVIDIA RTX 4070 SUPER . 1176467 |===========================================
NVIDIA RTX 4070 ....... 976967  |====================================
NVIDIA RTX 4070 TI .... 1262633 |==============================================
NVIDIA RTX 3090 ....... 1056000 |======================================


Hashcat 6.2.4
Benchmark: SHA-512
H/s > Higher Is Better
NVIDIA RTX 4070 SUPER . 3232733333 |========================================
NVIDIA RTX 4070 ....... 2673300000 |=================================
NVIDIA RTX 4070 TI .... 3462500000 |===========================================
NVIDIA RTX 3090 ....... 3081866667 |======================================


Hashcat 6.2.4
Benchmark: TrueCrypt RIPEMD160 + XTS
H/s > Higher Is Better
NVIDIA RTX 4070 SUPER . 802967 |============================================
NVIDIA RTX 4070 ....... 660967 |====================================
NVIDIA RTX 4070 TI .... 858600 |===============================================
NVIDIA RTX 3090 ....... 797833 |============================================


NAMD CUDA 2.14
ATPase Simulation - 327,506 Atoms
days/ns < Lower Is Better
NVIDIA RTX 4070 SUPER . 0.06791 |=============================
NVIDIA RTX 4070 ....... 0.07498 |================================
NVIDIA RTX 4070 TI .... 0.06788 |=============================
NVIDIA RTX 3090 ....... 0.10822 |==============================================


OctaneBench 2020.1
Total Score
Score > Higher Is Better
NVIDIA RTX 4070 SUPER . 720.97 |==============================================
NVIDIA RTX 4070 ....... 648.00 |=========================================
NVIDIA RTX 4070 TI .... 735.94 |===============================================
NVIDIA RTX 3090 ....... 674.25 |===========================================


FinanceBench 2016-07-25
Benchmark: Black-Scholes OpenCL
ms < Lower Is Better
NVIDIA RTX 4070 SUPER . 5.912 |=========================================
NVIDIA RTX 4070 ....... 6.906 |================================================
NVIDIA RTX 4070 TI .... 5.226 |====================================
NVIDIA RTX 3090 ....... 5.741 |========================================


cl-mem 2017-01-13
Benchmark: Copy
GB/s > Higher Is Better
NVIDIA RTX 4070 SUPER . 331.8 |============================================
NVIDIA RTX 4070 ....... 330.3 |============================================
NVIDIA RTX 4070 TI .... 333.3 |============================================
NVIDIA RTX 3090 ....... 360.8 |================================================


cl-mem 2017-01-13
Benchmark: Read
GB/s > Higher Is Better
NVIDIA RTX 4070 SUPER . 446.2 |==========================
NVIDIA RTX 4070 ....... 446.3 |==========================
NVIDIA RTX 4070 TI .... 446.3 |==========================
NVIDIA RTX 3090 ....... 825.8 |================================================


cl-mem 2017-01-13
Benchmark: Write
GB/s > Higher Is Better
NVIDIA RTX 4070 SUPER . 407.5 |==========================
NVIDIA RTX 4070 ....... 406.7 |==========================
NVIDIA RTX 4070 TI .... 412.2 |==========================
NVIDIA RTX 3090 ....... 753.8 |================================================


clpeak 1.1.2
OpenCL Test: Integer Compute INT
GIOPS > Higher Is Better
NVIDIA RTX 4070 SUPER . 18170.54 |=========================================
NVIDIA RTX 4070 ....... 14555.19 |=================================
NVIDIA RTX 4070 TI .... 19821.10 |=============================================
NVIDIA RTX 3090 ....... 17923.33 |=========================================


clpeak 1.1.2
OpenCL Test: Single-Precision Float
GFLOPS > Higher Is Better
NVIDIA RTX 4070 SUPER . 35492.69 |=========================================
NVIDIA RTX 4070 ....... 28479.39 |=================================
NVIDIA RTX 4070 TI .... 38691.73 |=============================================
NVIDIA RTX 3090 ....... 34906.79 |=========================================


clpeak 1.1.2
OpenCL Test: Double-Precision Double
GFLOPS > Higher Is Better
NVIDIA RTX 4070 SUPER . 630.11 |============================================
NVIDIA RTX 4070 ....... 515.17 |====================================
NVIDIA RTX 4070 TI .... 667.05 |===============================================
NVIDIA RTX 3090 ....... 642.23 |=============================================


clpeak 1.1.2
OpenCL Test: Global Memory Bandwidth
GBPS > Higher Is Better
NVIDIA RTX 4070 SUPER . 437.65 |=========================
NVIDIA RTX 4070 ....... 437.21 |=========================
NVIDIA RTX 4070 TI .... 437.63 |=========================
NVIDIA RTX 3090 ....... 816.55 |===============================================


MandelGPU 1.3pts1
OpenCL Device: GPU
Samples/sec > Higher Is Better
NVIDIA RTX 4070 SUPER . 587219538.2 |========================================
NVIDIA RTX 4070 ....... 516770131.2 |===================================
NVIDIA RTX 4070 TI .... 619106132.5 |==========================================
NVIDIA RTX 3090 ....... 484098913.8 |=================================


ViennaCL 1.7.1
Test: CPU BLAS - sCOPY
GB/s > Higher Is Better
NVIDIA RTX 4070 SUPER . 132 |==================================================
NVIDIA RTX 4070 ....... 131 |==================================================
NVIDIA RTX 4070 TI .... 132 |==================================================
NVIDIA RTX 3090 ....... 132 |==================================================


ViennaCL 1.7.1
Test: CPU BLAS - sAXPY
GB/s > Higher Is Better
NVIDIA RTX 4070 SUPER . 156 |==================================================
NVIDIA RTX 4070 ....... 153 |=================================================
NVIDIA RTX 4070 TI .... 156 |==================================================
NVIDIA RTX 3090 ....... 154 |=================================================


ViennaCL 1.7.1
Test: CPU BLAS - sDOT
GB/s > Higher Is Better
NVIDIA RTX 4070 SUPER . 165.0 |===============================================
NVIDIA RTX 4070 ....... 166.0 |===============================================
NVIDIA RTX 4070 TI .... 168.0 |================================================
NVIDIA RTX 3090 ....... 132.1 |======================================


ViennaCL 1.7.1
Test: CPU BLAS - dCOPY
GB/s > Higher Is Better
NVIDIA RTX 4070 SUPER . 70.8 |=================================================
NVIDIA RTX 4070 ....... 71.0 |=================================================
NVIDIA RTX 4070 TI .... 71.3 |=================================================
NVIDIA RTX 3090 ....... 70.2 |================================================


ViennaCL 1.7.1
Test: CPU BLAS - dAXPY
GB/s > Higher Is Better
NVIDIA RTX 4070 SUPER . 87.2 |=================================================
NVIDIA RTX 4070 ....... 86.8 |=================================================
NVIDIA RTX 4070 TI .... 87.3 |=================================================
NVIDIA RTX 3090 ....... 86.2 |================================================


ViennaCL 1.7.1
Test: CPU BLAS - dDOT
GB/s > Higher Is Better
NVIDIA RTX 4070 SUPER . 96.8 |=================================================
NVIDIA RTX 4070 ....... 96.7 |=================================================
NVIDIA RTX 4070 TI .... 96.4 |=================================================
NVIDIA RTX 3090 ....... 95.2 |================================================


ViennaCL 1.7.1
Test: CPU BLAS - dGEMV-N
GB/s > Higher Is Better
NVIDIA RTX 4070 SUPER . 102 |==================================================
NVIDIA RTX 4070 ....... 103 |==================================================
NVIDIA RTX 4070 TI .... 103 |==================================================
NVIDIA RTX 3090 ....... 103 |==================================================


ViennaCL 1.7.1
Test: CPU BLAS - dGEMV-T
GB/s > Higher Is Better
NVIDIA RTX 4070 SUPER . 109.0 |================================================
NVIDIA RTX 4070 ....... 109.0 |================================================
NVIDIA RTX 4070 TI .... 102.7 |=============================================
NVIDIA RTX 3090 ....... 110.0 |================================================


ViennaCL 1.7.1
Test: CPU BLAS - dGEMM-NN
GFLOPs/s > Higher Is Better
NVIDIA RTX 4070 SUPER . 119 |=================================================
NVIDIA RTX 4070 ....... 122 |==================================================
NVIDIA RTX 4070 TI .... 117 |================================================
NVIDIA RTX 3090 ....... 113 |==============================================


ViennaCL 1.7.1
Test: CPU BLAS - dGEMM-NT
GFLOPs/s > Higher Is Better
NVIDIA RTX 4070 SUPER . 117 |================================================
NVIDIA RTX 4070 ....... 122 |==================================================
NVIDIA RTX 4070 TI .... 118 |================================================
NVIDIA RTX 3090 ....... 119 |=================================================


ViennaCL 1.7.1
Test: CPU BLAS - dGEMM-TN
GFLOPs/s > Higher Is Better
NVIDIA RTX 4070 SUPER . 115 |==============================================
NVIDIA RTX 4070 ....... 121 |================================================
NVIDIA RTX 4070 TI .... 125 |==================================================
NVIDIA RTX 3090 ....... 121 |================================================


ViennaCL 1.7.1
Test: CPU BLAS - dGEMM-TT
GFLOPs/s > Higher Is Better
NVIDIA RTX 4070 SUPER . 122 |=================================================
NVIDIA RTX 4070 ....... 118 |================================================
NVIDIA RTX 4070 TI .... 124 |==================================================
NVIDIA RTX 3090 ....... 113 |==============================================


ViennaCL 1.7.1
Test: OpenCL BLAS - sCOPY
GB/s > Higher Is Better
NVIDIA RTX 4070 SUPER . 334 |==============================================
NVIDIA RTX 4070 ....... 330 |=============================================
NVIDIA RTX 4070 TI .... 336 |==============================================
NVIDIA RTX 3090 ....... 363 |==================================================


ViennaCL 1.7.1
Test: OpenCL BLAS - sAXPY
GB/s > Higher Is Better
NVIDIA RTX 4070 SUPER . 392 |=======================================
NVIDIA RTX 4070 ....... 389 |=======================================
NVIDIA RTX 4070 TI .... 393 |=======================================
NVIDIA RTX 3090 ....... 498 |==================================================


ViennaCL 1.7.1
Test: OpenCL BLAS - sDOT
GB/s > Higher Is Better
NVIDIA RTX 4070 SUPER . 370 |=================================================
NVIDIA RTX 4070 ....... 362 |================================================
NVIDIA RTX 4070 TI .... 365 |=================================================
NVIDIA RTX 3090 ....... 376 |==================================================


ViennaCL 1.7.1
Test: OpenCL BLAS - dCOPY
GB/s > Higher Is Better
NVIDIA RTX 4070 SUPER . 423 |===================================
NVIDIA RTX 4070 ....... 423 |===================================
NVIDIA RTX 4070 TI .... 424 |===================================
NVIDIA RTX 3090 ....... 605 |==================================================


ViennaCL 1.7.1
Test: OpenCL BLAS - dAXPY
GB/s > Higher Is Better
NVIDIA RTX 4070 SUPER . 437 |==============================
NVIDIA RTX 4070 ....... 455 |===============================
NVIDIA RTX 4070 TI .... 437 |==============================
NVIDIA RTX 3090 ....... 724 |==================================================


ViennaCL 1.7.1
Test: OpenCL BLAS - dDOT
GB/s > Higher Is Better
NVIDIA RTX 4070 SUPER . 458 |===================================
NVIDIA RTX 4070 ....... 456 |===================================
NVIDIA RTX 4070 TI .... 457 |===================================
NVIDIA RTX 3090 ....... 659 |==================================================


ViennaCL 1.7.1
Test: OpenCL BLAS - dGEMV-N
GB/s > Higher Is Better
NVIDIA RTX 4070 SUPER . 210 |==================================================
NVIDIA RTX 4070 ....... 209 |==================================================
NVIDIA RTX 4070 TI .... 211 |==================================================
NVIDIA RTX 3090 ....... 187 |============================================


ViennaCL 1.7.1
Test: OpenCL BLAS - dGEMV-T
GB/s > Higher Is Better
NVIDIA RTX 4070 SUPER . 389 |==================================================
NVIDIA RTX 4070 ....... 387 |=================================================
NVIDIA RTX 4070 TI .... 391 |==================================================
NVIDIA RTX 3090 ....... 374 |================================================


ViennaCL 1.7.1
Test: OpenCL BLAS - dGEMM-NN
GFLOPs/s > Higher Is Better
NVIDIA RTX 4070 SUPER . 577 |================================================
NVIDIA RTX 4070 ....... 473 |=======================================
NVIDIA RTX 4070 TI .... 604 |==================================================
NVIDIA RTX 3090 ....... 592 |=================================================


ViennaCL 1.7.1
Test: OpenCL BLAS - dGEMM-NT
GFLOPs/s > Higher Is Better
NVIDIA RTX 4070 SUPER . 584 |================================================
NVIDIA RTX 4070 ....... 477 |=======================================
NVIDIA RTX 4070 TI .... 612 |==================================================
NVIDIA RTX 3090 ....... 595 |=================================================


ViennaCL 1.7.1
Test: OpenCL BLAS - dGEMM-TN
GFLOPs/s > Higher Is Better
NVIDIA RTX 4070 SUPER . 599 |===============================================
NVIDIA RTX 4070 ....... 494 |=======================================
NVIDIA RTX 4070 TI .... 634 |==================================================
NVIDIA RTX 3090 ....... 594 |===============================================


ViennaCL 1.7.1
Test: OpenCL BLAS - dGEMM-TT
GFLOPs/s > Higher Is Better
NVIDIA RTX 4070 SUPER . 613 |===============================================
NVIDIA RTX 4070 ....... 502 |=======================================
NVIDIA RTX 4070 TI .... 648 |==================================================
NVIDIA RTX 3090 ....... 593 |==============================================


Libplacebo 5.229.1

FPS > Higher Is Better


Libplacebo 5.229.1
Test: deband_heavy
FPS > Higher Is Better
NVIDIA RTX 4070 ....... 1844.08 |=====================================
NVIDIA RTX 4070 ....... 1843.26 |=====================================
NVIDIA RTX 4070 TI .... 2306.56 |==============================================
NVIDIA RTX 3090 ....... 2015.93 |========================================
NVIDIA RTX 3090 ....... 2024.61 |========================================
NVIDIA RTX 3090 ....... 2020.16 |========================================
NVIDIA RTX 4070 SUPER . 2186.70 |============================================
NVIDIA RTX 4070 ....... 1847.98 |=====================================
NVIDIA RTX 4070 TI .... 2306.67 |==============================================
NVIDIA RTX 3090 ....... 2017.75 |========================================


Libplacebo 5.229.1
Test: polar_nocompute
FPS > Higher Is Better
NVIDIA RTX 4070 ....... 1969.19 |=====================================
NVIDIA RTX 4070 ....... 1968.37 |=====================================
NVIDIA RTX 4070 TI .... 2459.03 |==============================================
NVIDIA RTX 3090 ....... 2116.50 |========================================
NVIDIA RTX 3090 ....... 2126.31 |========================================
NVIDIA RTX 3090 ....... 2116.79 |========================================
NVIDIA RTX 4070 SUPER . 2327.55 |============================================
NVIDIA RTX 4070 ....... 1972.78 |=====================================
NVIDIA RTX 4070 TI .... 2461.23 |==============================================
NVIDIA RTX 3090 ....... 2119.89 |========================================


Libplacebo 5.229.1
Test: hdr_peakdetect
FPS > Higher Is Better
NVIDIA RTX 4070 ....... 3452.43 |===============================
NVIDIA RTX 4070 ....... 3329.26 |==============================
NVIDIA RTX 4070 TI .... 3475.06 |===============================
NVIDIA RTX 3090 ....... 5104.10 |==============================================
NVIDIA RTX 3090 ....... 4969.74 |=============================================
NVIDIA RTX 3090 ....... 5055.88 |==============================================
NVIDIA RTX 4070 SUPER . 3292.37 |==============================
NVIDIA RTX 4070 ....... 3310.02 |==============================
NVIDIA RTX 4070 TI .... 3544.60 |================================
NVIDIA RTX 3090 ....... 4997.08 |=============================================


Libplacebo 5.229.1
Test: hdr_lut
FPS > Higher Is Better
NVIDIA RTX 4070 ....... 3940.40 |==============================================
NVIDIA RTX 4070 ....... 3946.90 |==============================================
NVIDIA RTX 4070 TI .... 3976.04 |==============================================
NVIDIA RTX 3090 ....... 3376.85 |=======================================
NVIDIA RTX 3090 ....... 3333.77 |=======================================
NVIDIA RTX 3090 ....... 3369.88 |=======================================
NVIDIA RTX 4070 SUPER . 3905.98 |=============================================
NVIDIA RTX 4070 ....... 3927.11 |=============================================
NVIDIA RTX 4070 TI .... 3971.61 |==============================================
NVIDIA RTX 3090 ....... 3313.26 |======================================


Libplacebo 5.229.1
Test: av1_grain_lap
FPS > Higher Is Better
NVIDIA RTX 4070 ....... 4126.40 |==============================================
NVIDIA RTX 4070 ....... 4152.41 |==============================================
NVIDIA RTX 4070 TI .... 4143.96 |==============================================
NVIDIA RTX 3090 ....... 4096.48 |=============================================
NVIDIA RTX 3090 ....... 4120.27 |=============================================
NVIDIA RTX 3090 ....... 4100.36 |=============================================
NVIDIA RTX 4070 SUPER . 4171.00 |==============================================
NVIDIA RTX 4070 ....... 4103.40 |=============================================
NVIDIA RTX 4070 TI .... 4140.87 |==============================================
NVIDIA RTX 3090 ....... 4126.89 |==============================================


RealSR-NCNN 20200818
Scale: 4x - TAA: No
Seconds < Lower Is Better
NVIDIA RTX 4070 SUPER . 6.323 |===========================================
NVIDIA RTX 4070 ....... 7.092 |================================================
NVIDIA RTX 4070 TI .... 5.962 |========================================
NVIDIA RTX 3090 ....... 5.556 |======================================


RealSR-NCNN 20200818
Scale: 4x - TAA: Yes
Seconds < Lower Is Better
NVIDIA RTX 4070 SUPER . 34.89 |=======================================
NVIDIA RTX 4070 ....... 42.85 |================================================
NVIDIA RTX 4070 TI .... 33.63 |======================================
NVIDIA RTX 3090 ....... 30.31 |==================================


VkFFT 1.2.31
Test: FFT + iFFT R2C / C2R
Benchmark Score > Higher Is Better
NVIDIA RTX 4070 SUPER . 54794 |===============================================
NVIDIA RTX 4070 ....... 47097 |=========================================
NVIDIA RTX 4070 TI .... 55446 |================================================
NVIDIA RTX 3090 ....... 48418 |==========================================


VkFFT 1.2.31
Test: FFT + iFFT C2C 1D batched in half precision
Benchmark Score > Higher Is Better
NVIDIA RTX 4070 SUPER . 131705 |=======================
NVIDIA RTX 4070 ....... 137762 |========================
NVIDIA RTX 4070 TI .... 136210 |=======================
NVIDIA RTX 3090 ....... 273221 |===============================================


VkFFT 1.2.31
Test: FFT + iFFT C2C Bluestein in single precision
Benchmark Score > Higher Is Better
NVIDIA RTX 4070 SUPER . 15166 |================================================
NVIDIA RTX 4070 ....... 13714 |===========================================
NVIDIA RTX 4070 TI .... 15125 |================================================
NVIDIA RTX 3090 ....... 14205 |=============================================


VkFFT 1.2.31
Test: FFT + iFFT C2C 1D batched in double precision
Benchmark Score > Higher Is Better
NVIDIA RTX 4070 SUPER . 24317 |======================================
NVIDIA RTX 4070 ....... 22390 |===================================
NVIDIA RTX 4070 TI .... 25431 |=======================================
NVIDIA RTX 3090 ....... 30912 |================================================


VkFFT 1.2.31
Test: FFT + iFFT C2C 1D batched in single precision
Benchmark Score > Higher Is Better
NVIDIA RTX 4070 SUPER . 73929  |========================
NVIDIA RTX 4070 ....... 77774  |==========================
NVIDIA RTX 4070 TI .... 73942  |========================
NVIDIA RTX 3090 ....... 141876 |===============================================


VkFFT 1.2.31
Test: FFT + iFFT C2C multidimensional in single precision
Benchmark Score > Higher Is Better
NVIDIA RTX 4070 SUPER . 50299 |===============================================
NVIDIA RTX 4070 ....... 47212 |============================================
NVIDIA RTX 4070 TI .... 51528 |================================================
NVIDIA RTX 3090 ....... 50856 |===============================================


VkFFT 1.2.31
Test: FFT + iFFT C2C Bluestein benchmark in double precision
Benchmark Score > Higher Is Better
NVIDIA RTX 4070 SUPER . 4451 |===============================================
NVIDIA RTX 4070 ....... 3886 |=========================================
NVIDIA RTX 4070 TI .... 4647 |=================================================
NVIDIA RTX 3090 ....... 4195 |============================================


VkFFT 1.2.31
Test: FFT + iFFT C2C 1D batched in single precision, no reshuffling
Benchmark Score > Higher Is Better
NVIDIA RTX 4070 SUPER . 75078  |========================
NVIDIA RTX 4070 ....... 79057  |==========================
NVIDIA RTX 4070 TI .... 75141  |========================
NVIDIA RTX 3090 ....... 144311 |===============================================


vkpeak 20230730

GFLOPS > Higher Is Better


vkpeak 20230730
fp32-scalar
GFLOPS > Higher Is Better
NVIDIA RTX 3090 . 20319.80 |===================================================
NVIDIA RTX 3090 . 20317.63 |===================================================
NVIDIA RTX 3090 . 20353.95 |===================================================
NVIDIA RTX 3090 . 20263.13 |===================================================


vkpeak 20230730
fp32-vec4
GFLOPS > Higher Is Better
NVIDIA RTX 3090 . 26630.59 |===================================================
NVIDIA RTX 3090 . 26767.21 |===================================================
NVIDIA RTX 3090 . 26699.66 |===================================================
NVIDIA RTX 3090 . 26563.72 |===================================================


vkpeak 20230730
fp16-scalar
GFLOPS > Higher Is Better
NVIDIA RTX 3090 . 20113.01 |===================================================
NVIDIA RTX 3090 . 20134.06 |===================================================
NVIDIA RTX 3090 . 20151.44 |===================================================
NVIDIA RTX 3090 . 20080.47 |===================================================


vkpeak 20230730
fp16-vec4
GFLOPS > Higher Is Better
NVIDIA RTX 3090 . 39835.21 |===================================================
NVIDIA RTX 3090 . 39746.91 |===================================================
NVIDIA RTX 3090 . 39860.80 |===================================================
NVIDIA RTX 3090 . 39771.97 |===================================================


vkpeak 20230730
fp64-scalar
GFLOPS > Higher Is Better
NVIDIA RTX 3090 . 638.75 |=====================================================
NVIDIA RTX 3090 . 638.77 |=====================================================
NVIDIA RTX 3090 . 638.84 |=====================================================
NVIDIA RTX 3090 . 638.70 |=====================================================


vkpeak 20230730
fp64-vec4
GFLOPS > Higher Is Better
NVIDIA RTX 3090 . 638.77 |=====================================================
NVIDIA RTX 3090 . 639.52 |=====================================================
NVIDIA RTX 3090 . 638.74 |=====================================================
NVIDIA RTX 3090 . 638.72 |=====================================================


vkpeak 20230730
int32-scalar
GIOPS > Higher Is Better
NVIDIA RTX 3090 . 20290.30 |===================================================
NVIDIA RTX 3090 . 20315.10 |===================================================
NVIDIA RTX 3090 . 20295.27 |===================================================
NVIDIA RTX 3090 . 20280.33 |===================================================


vkpeak 20230730
int32-vec4
GIOPS > Higher Is Better
NVIDIA RTX 3090 . 20017.06 |===================================================
NVIDIA RTX 3090 . 20005.52 |===================================================
NVIDIA RTX 3090 . 20009.73 |===================================================
NVIDIA RTX 3090 . 19996.92 |===================================================


vkpeak 20230730
int16-scalar
GIOPS > Higher Is Better
NVIDIA RTX 3090 . 13273.53 |===================================================
NVIDIA RTX 3090 . 13225.17 |===================================================
NVIDIA RTX 3090 . 13264.91 |===================================================
NVIDIA RTX 3090 . 13259.97 |===================================================


vkpeak 20230730
int16-vec4
GIOPS > Higher Is Better
NVIDIA RTX 3090 . 16338.23 |===================================================
NVIDIA RTX 3090 . 16302.58 |===================================================
NVIDIA RTX 3090 . 16329.72 |===================================================
NVIDIA RTX 3090 . 16331.16 |===================================================


ProjectPhysX OpenCL-Benchmark 1.2
Operation: FP64 Compute
TFLOPs/s > Higher Is Better
NVIDIA RTX 4070 SUPER . 0.621 |=============================================
NVIDIA RTX 4070 ....... 0.510 |=====================================
NVIDIA RTX 4070 TI .... 0.660 |================================================
NVIDIA RTX 3090 ....... 0.637 |==============================================


VkResample 1.0
Upscale: 2x - Precision: Double
ms < Lower Is Better
NVIDIA RTX 4070 SUPER . 339.59 |======================================
NVIDIA RTX 4070 ....... 415.16 |===============================================
NVIDIA RTX 4070 TI .... 322.06 |====================================
NVIDIA RTX 3090 ....... 333.64 |======================================


VkResample 1.0
Upscale: 2x - Precision: Single
ms < Lower Is Better
NVIDIA RTX 4070 SUPER . 18.49 |================================================
NVIDIA RTX 4070 ....... 18.02 |===============================================
NVIDIA RTX 4070 TI .... 18.46 |================================================
NVIDIA RTX 3090 ....... 10.32 |===========================


Waifu2x-NCNN Vulkan 20200818
Scale: 2x - Denoise: 3 - TAA: No
Seconds < Lower Is Better


Waifu2x-NCNN Vulkan 20200818
Scale: 2x - Denoise: 3 - TAA: Yes
Seconds < Lower Is Better
NVIDIA RTX 4070 SUPER . 2.855 |===========================================
NVIDIA RTX 4070 ....... 3.168 |===============================================
NVIDIA RTX 4070 TI .... 2.854 |===========================================
NVIDIA RTX 3090 ....... 3.202 |================================================