Darktable-OpenCL
Apple M1 testing with a Apple MacBook Air (M1 2020) and Apple M1 15GB on Arch Linux ARM via the Phoronix Test Suite.


Athlon3050U/32GBRAM: 

	Processor: AMD Athlon Silver 3050U @ 2.30GHz (2 Cores), Motherboard: HP 879E v84.51 (F.65 BIOS), Chipset: AMD Raven/Raven2, Memory: 30GB, Disk: 1024GB ADATA SX8100NP, Graphics: AMD RAVEN2 (1100/1200MHz), Audio: AMD Raven/Raven2/Fenghuang, Network: Realtek RTL8111/8168/8411 + Realtek RTL8821CE 802.11ac PCIe

	OS: SteamOS rolling, Kernel: 5.18.1-arch1_testHoloISO_20220606.1811 (x86_64), Desktop: KDE Plasma 5.23.5, Display Server: X Server 1.21.1.3, OpenGL: 4.6 Mesa 22.1.2 (LLVM 13.0.0 DRM 3.46), OpenCL: OpenCL 2.1 AMD-APP (3452.0), Vulkan: 1.3.211, Compiler: GCC 11.1.0 + Clang 13.0.0 + LLVM 13.0.0, File-System: ext4, Screen Resolution: 1920x1080

RTX2070Max-Q: 

	Processor: Intel Core i7-9750H @ 4.50GHz (6 Cores / 12 Threads), Motherboard: Dell 0F7T8V (1.14.0 BIOS), Chipset: Intel Cannon Lake PCH, Memory: 32GB, Disk: 2000GB Samsung SSD 970 EVO Plus 2TB + 1000GB CT1000MX500SSD1, Graphics: Intel UHD 630 CFL GT2 8GB (300/405MHz), Audio: Realtek ALC3204, Network: Realtek Device 2502 + Intel-AC 9260

	OS: EndeavourOS rolling, Kernel: 6.2.8-arch1-1 (x86_64), Desktop: KDE Plasma 5.27.3, Display Server: X Server 1.21.1.8, Display Driver: NVIDIA 530.41.03, OpenGL: 4.6 Mesa 23.0.1, OpenCL: OpenCL 3.0 CUDA 12.1.98 + OpenCL 3.0, Compiler: GCC 12.2.1 20230201 + Clang 15.0.7 + LLVM 15.0.7 + CUDA 12.1, File-System: ext4, Screen Resolution: 1920x1080

RTX 3080: 

	Processor: AMD Ryzen 7 5800X 8-Core @ 3.80GHz (8 Cores / 16 Threads), Motherboard: Gigabyte X570 AORUS MASTER (F37b BIOS), Chipset: AMD Starship/Matisse, Memory: 32GB, Disk: 2000GB Samsung SSD 980 PRO with Heatsink 2TB, Graphics: NVIDIA GeForce RTX 3080 10GB, Audio: AMD Tobago HDMI Audio, Monitor: LG ULTRAGEAR + LC27G7xT + SAMSUNG, Network: Intel I211 + Realtek RTL8125 2.5GbE + Intel Wi-Fi 6 AX200

	OS: EndeavourOS rolling, Kernel: 6.3.2-zen1-1-zen (x86_64), Desktop: KDE Plasma 5.27.5, Display Server: X Server 1.21.1.8 + Wayland, Display Driver: NVIDIA 530.41.03, OpenGL: 4.6.0, OpenCL: OpenCL 2.1 AMD-APP (3558.0) + OpenCL 3.0 CUDA 12.1.98, Compiler: GCC 13.1.1 20230429 + PGI Compiler + Clang 15.0.7 + LLVM 15.0.7 + CUDA 12.0, File-System: ext4, Screen Resolution: 7920x1890

5600X3D+Arc_A770_16GB: 

	Processor: AMD Ryzen 5 5600X3D 6-Core (6 Cores / 12 Threads), Motherboard: Gigabyte X570 AORUS MASTER (F37e BIOS), Chipset: AMD Starship/Matisse, Memory: 16GB, Disk: 2000GB Samsung SSD 980 PRO with Heatsink 2TB + 31GB DataTraveler 3.0, Graphics: Intel Arctm A770 DG2 982MB, Audio: NVIDIA GK107 HDMI Audio, Monitor: SAMSUNG + LC27G7xT + LG ULTRAGEAR, Network: Intel I211 + Realtek RTL8125 2.5GbE + Intel Wi-Fi 6 AX200

	OS: Arch Linux, Kernel: 6.4.8-native_amd-xanmod1-1-edge (x86_64), Desktop: GNOME Shell 44.3, Display Server: X Server 1.21.1.8 + Wayland, Display Driver: NVIDIA, OpenGL: 4.6 Mesa 23.1.5, OpenCL: OpenCL 3.0 + OpenCL 3.0 CUDA 11.4.402, Compiler: GCC 13.2.1 20230801 + Clang 15.0.7 + LLVM 15.0.7, File-System: ext4, Screen Resolution: 7920x1440

M1_POCL: 

	Processor: Apple M1 @ 2.06GHz (4 Cores / 8 Threads), Motherboard: Apple MacBook Air (M1 2020), Chipset: Apple Silicon, Memory: 16GB, Disk: 500GB APPLE SSD AP0512Q + 2 x 0GB APPLE SSD AP0512Q, Graphics: Apple M1 15GB, Network: Broadcom BRCM4378 + Broadcom BRCM4378 Bluetooth

	OS: Arch Linux ARM, Kernel: 6.4.0-asahi-10-1-edge-ARCH (aarch64), Desktop: GNOME Shell 44.4, Display Server: X Server 1.21.1.8 + Wayland, OpenGL: 3.1 Mesa 23.3.0-devel, OpenCL: OpenCL 3.0 PoCL 4.1-pre main-0-g0f3e041c Linux RELOC SPIR SPIR-V LLVM 16.0.6 SLEEF POCL_DEBUG + OpenCL 1.1 Mesa 23.1.7-arch1.1 + OpenCL 3.0, Compiler: GCC 12.1.0 + Clang 16.0.6 + LLVM 16.0.6, File-System: ext4, Screen Resolution: 2560x1600


Darktable 3.8.0
Test: Boat - Acceleration: OpenCL
Seconds < Lower Is Better
Athlon3050U/32GBRAM . 25.95 |==================================================


Darktable 3.8.0
Test: Masskrug - Acceleration: OpenCL
Seconds < Lower Is Better
Athlon3050U/32GBRAM . 21.38 |==================================================


Darktable 3.8.0
Test: Server Rack - Acceleration: OpenCL
Seconds < Lower Is Better
Athlon3050U/32GBRAM . 2.551 |==================================================


Darktable 3.8.0
Test: Server Room - Acceleration: OpenCL
Seconds < Lower Is Better
Athlon3050U/32GBRAM . 12.93 |==================================================


Darktable 4.2.1
Test: Boat - Acceleration: OpenCL
Seconds < Lower Is Better
RTX2070Max-Q . 2.198 |=========================================================
RTX 3080 ..... 1.196 |===============================


Darktable 4.2.1
Test: Masskrug - Acceleration: OpenCL
Seconds < Lower Is Better
RTX2070Max-Q . 3.863 |=========================================================
RTX 3080 ..... 2.347 |===================================


Darktable 4.2.1
Test: Server Rack - Acceleration: OpenCL
Seconds < Lower Is Better
RTX2070Max-Q . 0.161 |=========================================================
RTX 3080 ..... 0.120 |==========================================


Darktable 4.2.1
Test: Server Room - Acceleration: OpenCL
Seconds < Lower Is Better
RTX2070Max-Q . 0.813 |=========================================================
RTX 3080 ..... 0.633 |============================================


Darktable 4.4.2
Test: Boat - Acceleration: OpenCL
Seconds < Lower Is Better
5600X3D+Arc_A770_16GB . 3.226 |=====================
M1_POCL ............... 7.225 |================================================
5600X3D+Arc_A770_16GB . 6.200 |=========================================
M1_POCL ............... 5.454 |====================================


Darktable 4.4.2
Test: Masskrug - Acceleration: OpenCL
Seconds < Lower Is Better
5600X3D+Arc_A770_16GB . 3.403 |================================
M1_POCL ............... 4.810 |==============================================
5600X3D+Arc_A770_16GB . 5.069 |================================================
M1_POCL ............... 4.541 |===========================================


Darktable 4.4.2
Test: Server Rack - Acceleration: OpenCL
Seconds < Lower Is Better
5600X3D+Arc_A770_16GB . 0.275 |====================
M1_POCL ............... 0.672 |================================================
5600X3D+Arc_A770_16GB . 0.134 |==========
M1_POCL ............... 0.675 |================================================


Darktable 4.4.2
Test: Server Room - Acceleration: OpenCL
Seconds < Lower Is Better
5600X3D+Arc_A770_16GB . 2.108 |==================================
M1_POCL ............... 2.946 |================================================
5600X3D+Arc_A770_16GB . 2.601 |==========================================
M1_POCL ............... 2.951 |================================================