CPU の浮動小数点演算能力の詳細

それぞれの演算命令で、1cycle に実行できる演算の数を割り出したものです。

実際の計測結果からの推定値 (VFP Benchmark を用いて測定した結果から割り出したものです)
実際の測定結果はこちら
CPU FLOPS 理論値一覧

IPC

float 32bit					Scalar (32bit)				SIMD 2 (64bit)				SIMD 4 (128bit)				SIMD 8 (256bit)				SIMD 16 (512bit)
CPU/SoC	CPU core	FPU	SIMD Width		add	mul	mad/fma	total	add	mul	mad/fma	total	add	mul	mad/fma	total	add	mul	mad/fma	total	add	mul	mad/fma	total
BCM2835	ARM1176JZF-S	VFPv2	64bit	64bit mad	0.5	0.5	0.5	0.5	–	–	–	–	–	–	–	–	–	–	–	–	–	–	–	–
S5PC100	Cortex-A8	VFPv3 NEON	128bit	128bit mad	0.1	0.1	0.1	0.1	1	1	1	1	1	1	1	1	–	–	–	–	–	–	–	–
BCM2836	Cortex-A7	VFPv4 NEON	32bit	32bit fma	1	1	1	1	0.5	0.5	0.5	0.5	0.25	0.25	0.25	0.25	–	–	–	–	–	–	–	–
Apple S1	Cortex-A7	VFPv4 NEON	32bit	32bit fma	1	1	1	1	0.5	0.5	0.5	0.5	0.25	0.25	0.25	0.25	–	–	–	–	–	–	–	–
Apple S2	Cortex-A7	VFPv4 NEON	32bit	32bit fma	1	1	1	1	0.5	0.5	0.5	0.5	0.25	0.25	0.25	0.25	–	–	–	–	–	–	–	–
Tegra 2	Cortex-A9	VFPv3	64bit	64bit mad	1	1	1	1	–	–	–	–	–	–	–	–	–	–	–	–	–	–	–	–
Apple A5	Cortex-A9	VFPv3 NEON	128bit	128bit mad	1	1	1	1	1	1	1	1	1	1	1	1	–	–	–	–	–	–	–	–
Tegra 4	Cortex-A15	VFPv4 NEON	128bit	64bit fma x2	1	1	1	1	2	2	2	2	1	1	1	1	–	–	–	–	–	–	–	–
BCM2837	Cortex-A53	AArch64 ASIMD	128bit	64bit fma x2	2	2	2	2	2	2	2	2	1	1	1	1	–	–	–	–	–	–	–	–
Snapdragon 845	(Cortex-A55)	AArch64 ASIMD FP16	128bit	64bit fma x2	2	2	2	2	2	2	2	2	1	1	1	1	–	–	–	–	–	–	–	–
Tegra X1	Cortex-A57	AArch64 ASIMD	128bit	64bit fma x2	2	2	2	2	2	2	2	2	1	1	1	1	–	–	–	–	–	–	–	–
BCM2711	Cortex-A72	AArch64 ASIMD	128bit	64bit fma x2	2	2	2	2	2	2	2	2	1	1	1	1	–	–	–	–	–	–	–	–
Snapdragon 835	(Cortex-A73)	AArch64 ASIMD	128bit	64bit fma x2	2	2	2	2	2	2	2	2	1	1	1	1	–	–	–	–	–	–	–	–
Snapdragon 845	(Cortex-A75)	AArch64 ASIMD FP16	128bit	64bit fma x2	2	2	2	2	2	2	2	2	1	1	1	1	–	–	–	–	–	–	–	–
Apple A6	Swift	VFPv4 NEON	128bit	128bit fma	1	1	1	1	1	1	1	1	1	1	1	1	–	–	–	–	–	–	–	–
Apple A7	Cyclone	AArch64 ASIMD	384bit	128bit add + 128bit fma x2	3	2	2	3	3	2	2	3	3	2	2	3	–	–	–	–	–	–	–	–
Apple A8	Typhoon	AArch64 ASIMD	384bit	128bit add + 128bit fma x2	3	2	2	3	3	2	2	3	3	2	2	3	–	–	–	–	–	–	–	–
Apple A9	Twister	AArch64 ASIMD	384bit	128bit fma x3	3	3	3	3	3	3	3	3	3	3	3	3	–	–	–	–	–	–	–	–
Apple A10	Hurricane	AArch64 ASIMD	384bit	128bit fma x3	3	3	3	3	3	3	3	3	3	3	3	3	–	–	–	–	–	–	–	–
Apple A11	Monsoon	AArch64 ASIMD	384bit	128bit fma x3	3	3	3	3	3	3	3	3	3	3	3	3	–	–	–	–	–	–	–	–
Apple M1	Firestorm	AArch64 ASIMD FP16	512bit	128bit add/mul + 128bit fma x3	4	4	3	4	4	4	3	4	4	4	3	4	–	–	–	–	–	–	–	–
Apple S6	Icestorm?	AArch64 ASIMD FP16	256bit	128bit add + 128bit fma	2	1	1	2	2	1	1	2	2	1	1	2	–	–	–	–	–	–	–	–
Tegra K1	Denver	AArch64 ASIMD	256bit	128bit add + 128bit fma	2	1	1	2	2	1	1	2	2	1	1	2	–	–	–	–	–	–	–	–
Snapdragon MSM8250	Scorpion	VFPv3 NEON	128bit	128bit mad	1	1	1	1	1	1	1	1	1	1	1	1	–	–	–	–	–	–	–	–
Snapdragon S4 Pro MSM8264	Krait	VFPv4 NEON	128bit	128bit fma	1	1	1	1	1	1	1	1	1	1	1	1	–	–	–	–	–	–	–	–
Snapdragon 820	Kryo	AArch64 ASIMD	256bit	128bit add + 128bit fma	2	1	1	2	2	1	1	2	2	1	1	2	–	–	–	–	–	–	–	–
Atom Z2560	Saltwell	SSSE3	192bit	128bit add + 64bit mul	1	1	(1)	2	–	–	–	–	1	0.5	(0.7)	0.7	–	–	–	–	–	–	–	–
Celeron J1900	SIlvermont	SSE4.2	192bit	128bit add + 64bit mul	1	1	(1)	2	–	–	–	–	1	0.5	(0.7)	0.7	–	–	–	–	–	–	–	–
Atom x7-Z8700	Airmont	SSE4.2	192bit	128bit add + 64bit mul	1	1	(1)	2	–	–	–	–	1	0.5	(0.7)	0.7	–	–	–	–	–	–	–	–
Core 2	Penryn	SSE4.1	256bit	128bit add + 128bit mul	1	1	(1)	2	–	–	–	–	1	1	(1)	2	–	–	–	–	–	–	–	–
Core i7-2700	SandyBridge	AVX	512bit	256bit add + 256bit mul	1	1	(1)	2	–	–	–	–	1	1	(1)	2	1	1	(1)	2	–	–	–	–
Core i7-3615QM	IvyBridge	AVX	512bit	256bit add + 256bit mul	1	1	(1)	2	–	–	–	–	1	1	(1)	2	1	1	(1)	2	–	–	–	–
Core i7-4790K	Haswell	AVX2/FMA3	512bit	256bit fma/add + 256bit fma/mul	1	2	2	2	–	–	–	–	1	2	2	2	1	2	2	2	–	–	–	–
Core i7-6700K	Skylake	AVX2/FMA3	512bit	256bit fma + 256bit fma	2	2	2	2	–	–	–	–	2	2	2	2	2	2	2	2	–	–	–	–
Core i7-1030NG7	IceLake	AVX512FVLBWDQ	512bit	256bit fma + 256bit fma	2	2	2	2	–	–	–	–	2	2	2	2	2	2	2	2	1	1	1	1
Athlon 5350	Jaguar	AVX	256bit	128bit add + 128bit mul	1	1	(1)	2	–	–	–	–	1	1	(1)	2	0.5	0.5	(0.5)	0.5	–	–	–	–
A10-7870K	Steamroller	AVX/FMA3	256bit	128bit fma + 128bit fma	2	2	2	2	–	–	–	–	2	2	2	2	1	1	1	1	–	–	–	–
Ryzen 7 1800X	Zen	AVX2/FMA3	512bit	128bit add x2 + 128bit mul x2	2	2	2	3	–	–	–	–	2	2	2	3	1	1	1	2	–	–	–	–
Ryzen 5 3400G	Zen+ APU	AVX2/FMA3	512bit	128bit add x2 + 128bit mul x2	2	2	2	3	–	–	–	–	2	2	2	3	1	1	1	2	–	–	–	–
Ryzen 7 PRO 4750G	Zen2 APU	AVX2/FMA3	1024bit	256bit add x2 + 256bit mul x2	2	2	2	4	–	–	–	–	2	2	2	4	2	2	2	4	–	–	–	–
Ryzen 9 3950X	Zen2	AVX2/FMA3	1024bit	256bit add x2 + 256bit mul x2	2	2	2	4	–	–	–	–	2	2	2	4	2	2	2	4	–	–	–	–

float 64bit					Scalar (64bit)				SIMD 2 (128bit)				SIMD 4 (256bit)				SIMD 8 (512bit)
CPU/SoC	CPU core	FPU	SIMD Width		add	mul	mad/fma	total	add	mul	mad/fma	total	add	mul	mad/fma	total	add	mul	mad/fma	total
BCM2835	ARM1176JZF-S	VFPv2	64bit	64bit mad	0.5	0.5	0.5	0.5	–	–	–	–	–	–	–	–	–	–	–	–
S5PC100	Cortex-A8	VFPv3 NEON	128bit	128bit mad	0.1	0.1	0.1	0.1	–	–	–	–	–	–	–	–	–	–	–	–
BCM2836	Cortex-A7	VFPv4 NEON	32bit	32bit fma	1	0.2	0.2	0.2	–	–	–	–	–	–	–	–	–	–	–	–
Apple S1	Cortex-A7	VFPv4 NEON	32bit	32bit fma	1	0.2	0.2	0.2	–	–	–	–	–	–	–	–	–	–	–	–
Apple S2	Cortex-A7	VFPv4 NEON	32bit	32bit fma	1	0.2	0.2	0.2	–	–	–	–	–	–	–	–	–	–	–	–
Tegra 2	Cortex-A9	VFPv3	64bit	64bit mad	1	0.5	0.5	0.5	–	–	–	–	–	–	–	–	–	–	–	–
Apple A5	Cortex-A9	VFPv3 NEON	128bit	128bit mad	1	0.5	0.5	0.5	–	–	–	–	–	–	–	–	–	–	–	–
Tegra 4	Cortex-A15	VFPv4 NEON	128bit	64bit fma x2	1	1	1	1	–	–	–	–	–	–	–	–	–	–	–	–
BCM2837	Cortex-A53	AArch64 ASIMD	128bit	64bit fma x2	2	2	2	2	1	1	1	1	–	–	–	–	–	–	–	–
Snapdragon 845	(Cortex-A55)	AArch64 ASIMD FP16	128bit	64bit fma x2	2	2	2	2	1	1	1	1	–	–	–	–	–	–	–	–
Tegra X1	Cortex-A57	AArch64 ASIMD	128bit	64bit fma x2	2	2	2	2	1	1	1	1	–	–	–	–	–	–	–	–
BCM2711	Cortex-A72	AArch64 ASIMD	128bit	64bit fma x2	2	2	2	2	1	1	1	1	–	–	–	–	–	–	–	–
Snapdragon 835	(Cortex-A73)	AArch64 ASIMD	128bit	64bit fma x2	2	2	2	2	1	1	1	1	–	–	–	–	–	–	–	–
Snapdragon 845	(Cortex-A75)	AArch64 ASIMD FP16	128bit	64bit fma x2	2	2	2	2	1	1	1	1	–	–	–	–	–	–	–	–
Apple A6	Swift	VFPv4 NEON	128bit	128bit fma	1	1	1	1	–	–	–	–	–	–	–	–	–	–	–	–
Apple A7	Cyclone	AArch64 ASIMD	384bit	128bit add + 128bit fma x2	3	2	2	3	3	2	2	3	–	–	–	–	–	–	–	–
Apple A8	Typhoon	AArch64 ASIMD	384bit	128bit add + 128bit fma x2	3	2	2	3	3	2	2	3	–	–	–	–	–	–	–	–
Apple A9	Twister	AArch64 ASIMD	384bit	128bit fma x3	3	2	2	3	3	2	2	3	–	–	–	–	–	–	–	–
Apple A10	Hurricane	AArch64 ASIMD	384bit	128bit fma x3	3	3	3	3	3	3	3	3	–	–	–	–	–	–	–	–
Apple A11	Monsoon	AArch64 ASIMD	384bit	128bit fma x3	3	3	3	3	3	3	3	3	–	–	–	–	–	–	–	–
Apple M1	Firestorm	AArch64 ASIMD FP16	512bit	128bit add/mul + 128bit fma x3	4	4	3	4	4	4	3	4	–	–	–	–	–	–	–	–
Apple S6	Icestorm?	AArch64 ASIMD FP16	256bit	128bit add + 128bit fma	2	1	1	2	2	1	1	2	–	–	–	–	–	–	–	–
Tegra K1	Denver	AArch64 ASIMD	256bit	128bit add + 128bit fma	2	1	1	2	2	1	1	2	–	–	–	–	–	–	–	–
Snapdragon MSM8250	Scorpion	VFPv3 NEON	128bit	128bit mad	1	1	1	1	–	–	–	–	–	–	–	–	–	–	–	–
Snapdragon S4 Pro MSM8264	Krait	VFPv4 NEON	128bit	128bit fma	1	1	1	1	–	–	–	–	–	–	–	–	–	–	–	–
Snapdragon 820	Kryo	AArch64 ASIMD	256bit	128bit add + 128bit fma	2	1	1	2	2	1	1	2	–	–	–	–	–	–	–	–
Atom Z2560	Saltwell	SSSE3	192bit	128bit add + 64bit mul	1	0.5	(0.7)	0.7	0.5	0.25	(0.3)	0.3	–	–	–	–	–	–	–	–
Celeron J1900	SIlvermont	SSE4.2	192bit	128bit add + 64bit mul	1	0.5	(0.7)	0.7	0.5	0.25	(0.3)	0.3	–	–	–	–	–	–	–	–
Atom x7-Z8700	Airmont	SSE4.2	192bit	128bit add + 64bit mul	1	0.5	(0.7)	0.7	0.5	0.25	(0.3)	0.3	–	–	–	–	–	–	–	–
Core 2	Penryn	SSE4.1	256bit	128bit add + 128bit mul	1	1	(1)	2	1	1	(1)	2	–	–	–	–	–	–	–	–
Core i7-2700	SandyBridge	AVX	512bit	256bit add + 256bit mul	1	1	(1)	2	1	1	(1)	2	1	1	(1)	2	–	–	–	–
Core i7-3615QM	IvyBridge	AVX	512bit	256bit add + 256bit mul	1	1	(1)	2	1	1	(1)	2	1	1	(1)	2	–	–	–	–
Core i7-4790K	Haswell	AVX2/FMA3	512bit	256bit fma/add + 256bit fma/mul	1	2	2	2	1	2	2	2	1	2	2	2	–	–	–	–
Core i7-6700K	Skylake	AVX2/FMA3	512bit	256bit fma + 256bit fma	2	2	2	2	2	2	2	2	2	2	2	2	–	–	–	–
Core i7-1030NG7	IceLake	AVX512FVLBWDQ	512bit	256bit fma + 256bit fma	2	2	2	2	2	2	2	2	2	2	2	2	1	1	1	1
Athlon 5350	Jaguar	AVX	256bit	128bit add + 128bit mul	1	1	(1)	2	1	1	(1)	2	0.5	0.5	(0.5)	0.5	–	–	–	–
A10-7870K	Steamroller	AVX/FMA3	256bit	128bit fma + 128bit fma	2	2	2	2	2	2	2	2	1	1	1	1	–	–	–	–
Ryzen 7 1800X	Zen	AVX2/FMA3	512bit	128bit add x2 + 128bit mul x2	2	2	2	3	2	2	2	3	1	1	1	2	–	–	–	–
Ryzen 5 3400G	Zen+ APU	AVX2/FMA3	512bit	128bit add x2 + 128bit mul x2	2	2	2	3	2	2	2	3	1	1	1	2	–	–	–	–
Ryzen 7 PRO 4750G	Zen2 APU	AVX2/FMA3	1024bit	256bit add x2 + 256bit mul x2	2	2	2	4	2	2	2	4	2	2	2	4	–	–	–	–
Ryzen 9 3950X	Zen2	AVX2/FMA3	1024bit	256bit add x2 + 256bit mul x2	2	2	2	4	2	2	2	4	2	2	2	4	–	–	–	–

mad/fma = カッコ付き表記は積和命令を持っていない代わりに add + mul の 2 命令を実行したときの値です。
total = 同時に実行できる命令数の最大値

FOP

Scalar

注意: この数値は実行可能な命令数 (IPC) ではなく FLOPS の算出に用いる 1 cycle で実行可能な浮動小数点演算数 (FOP) です。例えば fma の値に CPU clock を乗算すると FLOPS が求まります。

Scalar		single float (32bit x1)				double float (64bit x1)
CPU	FPU	mul	add	mad	fma	mul	add	mad	fma
XBurst JZ4775	FPU F32	0.1	0.1	0.07	–	0.06	0.1	0.06	–	ZWatch
ARM1176JZF-S	VFPv2	0.5	0.5	1	–	0.5	0.5	1	–	iPhone 3G , Raspberry Pi
Cortex-A7	VFPv4 + NEON	1	1	2	2	0.25	1	0.5	0.4	Raspberry Pi 2
Cortex-A8	VFPv3 + NEON	0.14	0.14	0.18	–	0.1	0.1	0.1	–	iPhone 3GS, Nexus S
Cortex-A9	VFPv3 + NEON	1	1	2	–	0.5	1	1	–	Nexus 7 (2012), iPhone 4, Galaxy Nexus
Cortex-A15	VFPv4 + NEON	1	1	1.4	2	1	1	1.4	1.4	Nexus 10
Cortex-A53 64	AArch64 NEON	2	2	-	2	2	2	–	2	Dragonboard 410c
Cortex-A57 64	AArch64 NEON	2	2	-	2	2	2	–	2	SHIELD Android TV
Cortex-A72 64	AArch64 NEON	2	2	-	2.3	2	2	–	2.3	Fire TV 2015
Scorpion	VFPv3 + NEON	1	1	2	–	0.5	1	1	–	Nexus One
Krait (400)	VFPv4 + NEON	1	1	2	2	1	1	1.6	2	Nexus 4/5, Nexus 7 (2013)
Kryo 64	AArch64 NEON	1	2	-	2	1	2	–	2	HTC 10
A6 Swift	VFPv4 + NEON	1	1	1	1	1	1	1	1	iPhone 5, iPad 4
A7 Cyclone 32	AArch32 NEON	1	1	2	2	2	3	3	3	iPhone 5s, iPad Air
A7 Cyclone 64	AArch64 NEON	2	3	–	4	2	3	–	1.6	iPhone 5s, iPad Air
A8X Typhoon 64	AArch64 NEON	2	3	–	4	2	3	–	4	iPad Air 2
A9 Twister 64	AArch64 NEON	3	3	–	6	2	3	–	4	iPhone SE
Denver 64	AArch64 NEON	1	2	–	2	1	2	–	2	Nexus 9
Atom Bonnell 32	SSSE3	1	1	(2)	–	0.5	1	(1.5)	–
Atom Silvermont 64	SSE4.2	1	1	?	–	0.5	1	?	–	BayTrail
AMD Jaguar	SSE4.2/AVX	1	1	(2)	–	0.5	1	?	–	Athlon 5350 (Kabini)
Core2 Penryn 64	SSE4.1	1	1	(2)	–	1	1	(2)	–
Core i7 Sandy 64	SSE4.2/AVX	1	1	(2)	–	1	1	(2)	–
Core i7 Ivy 64	SSE4.2/AVX	1	1	(2)	–	1	1	(2)	–
Core i7 Haswell 64	SSE4.2/AVX2/FMA3	1.6	1	–	3.2	1.6	1	–	3.2	Core i7-4790K
Celeron Haswell 64	SSE4.2	1.6	1	(1.6)	–	1.6	1	(1.6)	–	Celeron 2955U
Core i7 Skylake 64	SSE4.2/AVX2/FMA3	2	2	–	4	2	2	–	4	Core i7-6700K
Ryzen 7 1800X 64	SSE4.2/AVX2/FMA3	2	2	–	3.2	2	2	–	3.2	Ryzen 7 1800X

↑ core あたりの演算能力 (Scalar)
数値は 1 cycle で実行できる演算個数。数値が大きい方が高速
- 例: mad/fma は 1命令 = 2fop なので、2 のとき mad/fma 命令を 1cycle で実行できることになります。
- 同じように mad/fma が 4 のとき、2命令同時実行可能であることを意味しています。
ARM: mad は旧積和命令、fma は Fused multiply add 命令です。fma 対応は VFPv4 以降、AArch64 では fma のみとなっています。
Intel: mad は単独の積和命令ではなく add, mul の interleave 時の数値となっています。区別するため括弧がついています。

SIMD sp

注意: この数値は実行可能な命令数 (IPC) ではなく FLOPS の算出に用いる 1 cycle で実行可能な浮動小数点演算数 (FOP) です。例えば fma の値に CPU clock を乗算すると FLOPS が求まります。

SIMD (Vector) sp		SIMD2 single fp (32bit x2)				SIMD4 single fp (32bit x4)				SIMD8 single fp (32bit x8)
CPU	FPU	mul	add	mad	fma	mul	add	mad	fma	mul	add	mad	fma
Cortex-A7	VFPv4 + NEON	1	1	2	2	1	1	2	2	–	–	–	–
Cortex-A8	VFPv3 + NEON	2	2	4	–	2	2	4	–	–	–	–	–
Cortex-A9	VFPv3 + NEON	2	2	4	–	2	2	4	–	–	–	–	–
Cortex-A15	VFPv4 + NEON	4	4	8	8	4	4	8	8	–	–	–	–
Cortex-A53	AArch64 NEON	4	4	–	8	4	4	–	8	–	–	–	–
Cortex-A57	AArch64 NEON	4	4	–	8	4	4	–	8	–	–	–	–
Cortex-A72	AArch64 NEON	4	4	–	8	4	4	–	8	–	–	–	–
Scorpion	VFPv3 + NEON	2	2	4	–	4	4	8	–	–	–	–	–
Krait 400	VFPv4 + NEON	2	2	4	4	4	4	8	8	–	–	–	–
Kyro	AArch64 NEON	2	4	–	4	2	4	–	4	–	–	–	–
A6 Swift	VFPv4 + NEON	2	2	4	4	4	4	8	8	–	–	–	–
A7 Cyclone 32	AArch32 NEON	4	6	8	8	8	12	16	16	–	–	–	–
A7 Cyclone 64	AArch64 NEON	4	6	–	8	8	12	–	16	–	–	–	–
A8X Typhoon 64	AArch64 NEON	4	6	–	8	8	12	–	16	–	–	–	–
A9 Twister 64	AArch64 NEON	6	6	–	12	12	12	–	24	–	–	–	–
Denver 64	AArch64 NEON	2	3	–	4	4	6	–	8	–	–	–	–
Atom Bonnell 32	SSSE3	–	–	–	–	2	4	(6)	–	–	–	–	–
Atom Silvermont 64	SSE4.2	–	–	–	–	2	4	(6)	–	–	–	–	–
AMD Jaguar 64	SSE4.2/AVX	–	–	–	–	4	4	(8)	–	4	4	(8)	–
Core2 Penryn 64	SSE4.1	–	–	–	–	4	4	(8)	–	–	–	–	–
Core i7 Sandy 64	SSE4.2/AVX	–	–	–	–	4	4	(8)	–	8	8	(16)	–
Core i7 Ivy 64	SSE4.2/AVX	–	–	–	–	4	4	(8)	–	8	8	(16)	–
Core i7 Haswell 64	SSE4.2/AVX2/FMA3	–	–	–	–	8	4	(8)	16	16	8	(16)	32
Celeron Haswell 64	SSE4.2	–	–	–	–	8	4	(8)	–	–	–	–	–
Core i7 Skylake 64	SSE4.2/AVX2/FMA3	–	–	–	–	8	8	(8)	16	16	16	(16)	32
Ryzen 7 1800X 64	SSE4.2/AVX2/FMA3	–	–	–	–	8	8	(12)	12	8	8	(16)	16

↑ core あたりの演算能力 (Vector) sp
数値は 1 cycle で実行できる演算数。数値が大きいほうが高速
- 例: SIMD4 add は 1命令 = 4fop なので、4 のとき 1cycle で実行できることになります。
- 例: SIMD4 mad/fma は 1命令 = 8fop なので、8 のとき 1cycle で実行できることになります。
括弧は専用の積和命令を持っていないが加算と乗算命令をペアリングなことを意味しています。

SIMD dp

注意: この数値は実行可能な命令数 (IPC) ではなく FLOPS の算出に用いる 1 cycle で実行可能な浮動小数点演算数 (FOP) です。例えば fma の値に CPU clock を乗算すると FLOPS が求まります。

SIMD (Vector) dp		SIMD2 double fp (64bit x2)				SIMD4 double fp (64bit x4)
CPU	FPU	mul	add	mad	fma	mul	add	mad	fma
Cortex-A53	AArch64 NEON	2	2	–	4	–	–	–	–
Cortex-A57	AArch64 NEON	2	2	–	4	–	–	–	–
Cortex-A72	AArch64 NEON	2	2	–	4	–	–	–	–
A7 Cyclone 64	AArch64 NEON	4	6	–	8	–	–	–	–
A8 Typhoon 64	AArch64 NEON	4	6	–	8	–	–	–	–
A9 Twister 64	AArch64 NEON	4	6	–	8	–	–	–	–
Kyro 64	AArch64 NEON	1	2	–	2	–	–	–	–
Denver 64	AArch64 NEON	2	3	–	4	–	–	–	–
Atom Bonnell 32	SSSE3	0.4	0.5	–	–	–	–	–	–
Atom Silvermont 64	SSE4.2	0.5	1	(1.5)	–	–	–	–	–
AMD Jaguar	SSE4.2/AVX	1	2	(3)	–	1	2	(3)	–
Core2 Penryn 64	SSE4.1	2	2	(3?)	–	–	–	–	–
Core i7 Sandy 64	SSE4.2/AVX	2	2	(4)	–	4	4	(8)	–
Core i7 Ivy 64	SSE4.2/AVX	2	2	(4)	–	4	4	(8)	–
Core i7 Haswell 64	SSE4.2/AVX2/FMA3	4	2	(4)	8?	8	4	(8)	16?
Celeron Haswell 64	SSE4.2	4	2	(4)	–	–	–	–	–
Core i7 Skylake 64	SSE4.2/AVX2/FMA3	4	4	–	8	8	8	–	16
Ryzen 7 1800X 64	SSE4.2/AVX2/FMA3	4	4	–	6.3	4	4	(8)	8

↑ core あたりの演算能力 (Vector) dp
ARM の SIMD2 double fp 対応は ARM64 以降

HYPERでんち

目次

CPU の浮動小数点演算能力の詳細

IPC

FOP

Scalar

SIMD sp

SIMD dp