Results

Results for the following processors are shown below:

Intel Core2 = Intel Core2 Duo E6400, "Conroe", family 6, model 15, stepping 6

Intel P4 = Intel Xeon 2.80GHz, "Prestonia", family 15, model 2, stepping 7

AMD K8 = AMD Athlon 64 X2 4600+, "Manchester", family 15, model 43, stepping 1

AMD K7 = AMD Athlon XP 2600+, "Thoroughbred", family 6, model 8, stepping 1 (note: this cpu does not support SSE2, but appears to run some SSE2 opcodes that have corresponding MMX opcodes without error, hence the table entries)

Instruction	Latency				Throughput (Reciprocal)
	AMD K7	AMD K8	Intel Core2	Intel P4 (F2)	AMD K7	AMD K8	Intel Core2	Intel P4 (F2)
psignb xmm, xmm	-	-	1	-	-	-	1	-
psignw xmm, xmm	-	-	1	-	-	-	1	-
psignd xmm, xmm	-	-	1	-	-	-	1	-
pshufb xmm, xmm	-	-	3	-	-	-	2	-
pmulhrsw xmm, xmm	-	-	3	-	-	-	1	-
pmaddubsw xmm, xmm	-	-	3	-	-	-	1	-
phsubw xmm, xmm	-	-	6	-	-	-	4	-
phsubsw xmm, xmm	-	-	6	-	-	-	4	-
phsubd xmm, xmm	-	-	5	-	-	-	3	-
phaddw xmm, xmm	-	-	6	-	-	-	4	-
phaddsw xmm, xmm	-	-	6	-	-	-	4	-
phaddd xmm, xmm	-	-	5	-	-	-	3	-
palignr xmm, xmm, imm8	-	-	2	-	-	-	1	-
pabsb xmm, xmm	-	-	1	-	-	-	0.5	-
pabsw xmm, xmm	-	-	1	-	-	-	0.5	-
pabsd xmm, xmm	-	-	1	-	-	-	0.5	-
addsubpd xmm, xmm	-	4	3	-	-	2	1	-
addsubps xmm, xmm	-	4	3	-	-	2	1	-
haddpd xmm, xmm	-	4	5	-	-	2	2	-
haddps xmm, xmm	-	4	9	-	-	2	3	-
hsubpd xmm, xmm	-	4	5	-	-	2	2	-
hsubps xmm, xmm	-	4	9	-	-	2	3	-
movddup xmm, xmm	-	2	1	-	-	1	1	-
movshdup xmm, xmm	-	2	1	-	-	2	1	-
movsldup xmm, xmm	-	2	1	-	-	2	1	-
lddqu xmm, xmm	-	-	-	-	-	-	-	-
cvtdq2ps xmm, xmm	-	49.05	3	14.00	-	46.05	1	2
cvtps2dq xmm, xmm	-	46.12	3	13.93	-	43.13	1	2
cvttps2dq xmm, xmm	-	46.12	3	13.97	-	43.13	1	2
movdqa xmm, xmm	2	2	1	6	0.5	1	0.33	1
movdqu xmm, xmm	2	2	1	6	0.5	1	0.33	1
movq xmm, xmm	-	2	1	2	-	1	0.33	2
packsswb xmm, xmm	2	2	4	4	0.5	1.5	2	2
packssdw xmm, xmm	2	2	4	4	0.5	1.5	2	2
packuswb xmm, xmm	2	2	4	4	0.5	1.5	2	2
paddb xmm, xmm	2.10	2	1	2	0.5	1	1	2
paddw xmm, xmm	2	2	1	2	0.5	1	1	2
paddd xmm, xmm	2	2	1	2	0.5	1	1	2
paddsb xmm, xmm	2	2	1	2	0.5	1	1	2
paddsw xmm, xmm	2	2	1	2	0.5	1	1	2
paddusb xmm, xmm	2.10	2	1	2	0.5	1	1	2
paddusw xmm, xmm	2.10	2	1	2	0.5	1	1	2
pand xmm, xmm	2.10	2	1	2	0.5	1	1	2
pandn xmm, xmm	2	2	1	2	0.5	1	1	2
pavgb xmm, xmm	2	2	1	2	0.5	1	1	2
pavgw xmm, xmm	2	2	1	2	0.5	1	1	2
pcmpeqb xmm, xmm	2.10	2	1	2	0.5	1	1	2
pcmpeqd xmm, xmm	2	2	1	2	0.5	1	1	2
pcmpeqw xmm, xmm	2	2	1	2	0.5	1	1	2
pcmpgtb xmm, xmm	2.10	2	1	2	0.5	1	1	2
pcmpgtd xmm, xmm	2	2	1	2	0.5	1	1	2
pcmpgtw xmm, xmm	2	2	1	2	0.5	1	1	2
pmaddwd xmm, xmm	3	3	3	6	1	2	1	2
pmaxub xmm, xmm	2	2	1	2	0.5	1	1	2
pmaxsw xmm, xmm	2	2	1	2	0.5	1	1	2
pminub xmm, xmm	2	2	1	2	0.5	1	1	2
pminsw xmm, xmm	2	2	1	2	0.5	1	1	2
pmulhuw xmm, xmm	3	3	3	6	1	2	1	2
pmulhw xmm, xmm	3	3	3	6	1	2	1	2
pmullw xmm, xmm	3	3	3	6	1	2	1	2
pmuludq xmm, xmm	-	3	3	6	-	2	1	2
por xmm, xmm	2	2	1	2	0.5	1	1	2
psadbw xmm, xmm	3.19	3	3	4	1	2	1	2
pshufd xmm, xmm, imm8	2	3	4	4	0.5	1.5	1	2
pshufhw xmm, xmm, imm8	2	2	1	2	0.5	1	1	2
pshuflw xmm, xmm, imm8	2	2	1	2	0.5	1	1	2
psllw xmm, xmm	2	2	2	2	0.5	1	1	2
pslld xmm, xmm	2	2	2	2	0.5	1	1	2
psllq xmm, xmm	2	2	2	2	0.5	1	1	2
psraw xmm, xmm	2	2	2	2	0.5	1	1	2
psrad xmm, xmm	2	2	2	2	0.5	1	1	2
psrlw xmm, xmm	2	2	2	2	0.5	1	1	2
psrld xmm, xmm	2	2	2	2	0.5	1	1	2
psrlq xmm, xmm	2	2	2	2	0.5	1	1	2
psubb xmm, xmm	2	2	1	2	0.5	1	1	2
psubw xmm, xmm	2	2	1	2	0.5	1	1	2
psubd xmm, xmm	2	2	1	2	0.5	1	1	2
psubsb xmm, xmm	2	2	1	2	0.5	1	1	2
psubsw xmm, xmm	2	2	1	2	0.5	1	1	2
psubusb xmm, xmm	2	2	1	2	0.5	1	1	2
psubusw xmm, xmm	2	2	1	2	0.5	1	1	2
punpckhbw xmm, xmm	2	2	4	4	0.5	1	2	2
punpckhwd xmm, xmm	2.10	2	4	4	0.5	1	2	2
punpckhdq xmm, xmm	2	2	4	4	0.5	1	2	2
punpckhqdq xmm, xmm	-	2	1	4	-	1	1	2
punpcklbw xmm, xmm	2	2	4	2	0.5	1	2	2
punpcklwd xmm, xmm	2	2	4	2	0.5	1	2	2
punpckldq xmm, xmm	2	2	4	2	0.5	1	2	2
punpcklqdq xmm, xmm	-	0.5	1	2	-	0.5	1	2
pxor xmm, xmm	2	2	1	2	0.5	1	1	2
paddq mm, mm	-	2	2	2	-	0.5	1	1
psubq mm, mm	-	2	2	2	-	0.5	1	1
pmuludq mm, mm	-	3	3	6	-	1	1	1
addpd xmm, xmm	-	4	3	4	-	2	1	2
addsd xmm, xmm	-	4	3	4	-	1	1	2
andnpd xmm, xmm	-	2	1	2	-	2	1	2
andpd xmm, xmm	-	2	1	2	-	2	1	2
cmppd xmm, xmm, imm8	-	2	3	4	-	2	1	2
cmpsd xmm, xmm, imm8	-	2	3	4	-	1	1	2
comisd xmm, xmm	-	1	1	3	-	1	1	3
cvtdq2pd xmm, xmm	-	50.10	4	10	-	46.12	1	4
cvtpd2dq xmm, xmm	-	7	4	10	-	3.24	1	2
cvtpd2ps xmm, xmm	-	8	4	10	-	3.24	1	2
cvtps2pd xmm, xmm	-	43.07	2	13.32	-	41.07	2	2
cvtsd2ss xmm, xmm	-	8	4	12.01	-	2	1	4
cvtss2sd xmm, xmm	-	41.56	2	8	-	39.57	2	2
cvttpd2dq xmm, xmm	-	7	4	10	-	3.24	1	2
divpd xmm, xmm	-	33.88	6	69.07	-	33.88	5	69.07
divsd xmm, xmm	-	19.93	6	38.03	-	16.94	5	38.04
maxpd xmm, xmm	-	2	3	4	-	2	1	2
maxsd xmm, xmm	-	2	3	4	-	1	1	2
minpd xmm, xmm	-	2	3	4	-	2	1	2
minsd xmm, xmm	-	2	3	4	-	1	1	2
movapd xmm, xmm	-	2	1	6	-	1	0.33	1
movsd xmm, xmm	-	2	1	4	-	0.5	1	2
movupd xmm, xmm	-	2	1	6	-	1	0.33	1
mulpd xmm, xmm	-	4	5	6	-	2	1	2
mulsd xmm, xmm	-	4	5	6	-	1	1	2
orpd xmm, xmm	-	2	1	2	-	2	1	2
shufpd xmm, xmm, imm8	-	2.49	1	4	-	2.49	1	2
sqrtpd xmm, xmm	-	47.83	6	69.07	-	47.83	5	69.07
sqrtsd xmm, xmm	-	26.90	6	38.04	-	23.91	5	38.04
subpd xmm, xmm	-	4	3	4	-	2	1	2
subsd xmm, xmm	-	4	3	4	-	1	1	2
ucomisd xmm, xmm	-	1	1	3	-	1	1	3
unpckhpd xmm, xmm	-	2	1	4	-	1	1	2
unpcklpd xmm, xmm	-	0.5	1	2	-	0.5	1	2
xorpd xmm, xmm	-	2	1	2	-	2	1	2
addps xmm, xmm	4	4	3	4	2.12	2	1	2
addss xmm, xmm	4	4	3	4	1	1	1	2
andnps xmm, xmm	2	2	1	2	2	2	1	2
andps xmm, xmm	2	2	1	2	2	2	1	2
cmpps xmm, xmm	-	-	-	-	-	-	-	-
cmpss xmm, xmm	-	-	-	-	-	-	-	-
comiss xmm, xmm	1.14	1	1	3	1.14	1	1	3
divps xmm, xmm	32.19	29.89	6	39.04	32.45	29.89	5	39.04
divss xmm, xmm	17.31	15.94	6	23.02	13.47	12.95	5	23.02
maxps xmm, xmm	2	2	3	4	2	2	1	2
maxss xmm, xmm	2.10	2	3	4	1	1	1	2
minps xmm, xmm	2	2	3	4	2	2	1	2
minss xmm, xmm	2	2	3	4	1	1	1	2
movaps xmm, xmm	2	2	1	6	1.15	1	0.33	1
movhlps xmm, xmm	0.5	0.5	1	4	0.5	0.5	1	2
movlhps xmm, xmm	0.5	0.5	1	2	0.5	0.5	1	2
movss xmm, xmm	2	2	1	2	0.5	0.5	1	2
movups xmm, xmm	2	2	1	6	1.14	1	0.33	1
mulps xmm, xmm	4	4	4	6	2	2	1	2
mulss xmm, xmm	4.21	4	4	6	1	1	1	2
orps xmm, xmm	2	2	1	2	2	2	1	2
rcpps xmm, xmm	3	3	3	4	2	2	2	4
rcpss xmm, xmm	3	3	3	4	1	1	2	3
rsqrtps xmm, xmm	3	3	3	4	2.12	2	2	4
rsqrtss xmm, xmm	3	3	3	4	1	1	2	3
shufps xmm, xmm, imm8	2.61	2.49	4	4	2.61	2.49	1	2
sqrtps xmm, xmm	38.34	35.87	6	39.04	38.59	35.87	5	39.04
sqrtss xmm, xmm	20.04	18.93	6	23.02	16.68	15.94	5	23.02
subps xmm, xmm	4	4	3	4	2	2	1	2
subss xmm, xmm	4	4	3	4	1	1	1	2
ucomiss xmm, xmm	1.14	1	1	3	1.14	1	1	3
unpckhps xmm, xmm	3	3	2	4	2	2	1	2
unpcklps xmm, xmm	3	3	2	2	2.12	2	1	2
xorps xmm, xmm	2	2	1	2	2	2	1	2
pavgb mm, mm	2	2	1	2	0.5	0.5	0.5	1
pavgw mm, mm	2.10	2	1	2	0.5	0.5	0.5	1
pmaxub mm, mm	2	2	1	2	0.5	0.5	0.5	1
pmaxsw mm, mm	2	2	1	2	0.5	0.5	0.5	1
pminub mm, mm	2	2	1	2	0.5	0.5	0.5	1
pminsw mm, mm	2	2	1	2	0.5	0.5	0.5	1
pmulhuw mm, mm	3.19	3	3	6	1	1	1	1
psadbw mm, mm	3	3	3	4	1	1	1	1
movq mm, mm	2	2	1	6	0.5	0.5	0.33	1
packsswb mm, mm	2	2	1	2	0.5	0.5	1	1
packssdw mm, mm	2	2	1	2	0.5	0.5	1	1
packuswb mm, mm	2.10	2	1	2	0.5	0.5	1	1
paddb mm, mm	2	2	1	2	0.5	0.5	0.5	1
paddw mm, mm	2	2	1	2	0.5	0.5	0.5	1
paddd mm, mm	2	2	1	2	0.5	0.5	0.5	1
paddsb mm, mm	2.10	2	1	2	0.5	0.5	0.5	1
paddsw mm, mm	2.18	2	1	2	0.5	0.5	0.5	1
paddusb mm, mm	2	2	1	2	0.5	0.5	0.5	1
paddusw mm, mm	2	2	1	2	0.5	0.5	0.5	1
pand mm, mm	2	2	1	2	0.5	0.5	0.36	1
pandn mm, mm	2.10	2	1	2	0.5	0.5	0.36	1
pcmpeqb mm, mm	2	2	1	2	0.5	0.5	0.5	1
pcmpeqd mm, mm	2	2	1	2	0.5	0.5	0.5	1
pcmpeqw mm, mm	2	2	1	2	0.5	0.5	0.5	1
pcmpgtb mm, mm	2	2	1	2	0.5	0.5	0.5	1
pcmpgtd mm, mm	2.10	2	1	2	0.5	0.5	0.5	1
pcmpgtw mm, mm	2	2	1	2	0.5	0.5	0.5	1
pmaddwd mm, mm	3.19	3	3	6	1	1	1	1
pmulhw mm, mm	3	3	3	6	1	1	1	1
pmullw mm, mm	3	3	3	6	1	1	1	1
por mm, mm	2	2	1	2	0.5	0.5	0.36	1
psllq mm, mm	2	2	1	2	0.5	0.5	1	1
psllw mm, mm	2	2	1	2	0.5	0.5	1	1
pslld mm, mm	2	2	1	2	0.5	0.5	1	1
psraw mm, mm	2	2	1	2	0.5	0.5	1	1
psrad mm, mm	2	2	1	2	0.5	0.5	1	1
psrlq mm, mm	2	2	1	2	0.5	0.5	1	1
psrlw mm, mm	2	2	1	2	0.5	0.5	1	1
psrld mm, mm	2	2	1	2	0.5	0.5	1	1
psubb mm, mm	2.10	2	1	2	0.5	0.5	0.5	1
psubw mm, mm	2	2	1	2	0.5	0.5	0.5	1
psubd mm, mm	2	2	1	2	0.5	0.5	0.5	1
psubsb mm, mm	2	2	1	2	0.5	0.5	0.5	1
psubsw mm, mm	2	2	1	2	0.5	0.5	0.5	1
psubusb mm, mm	2.10	2	1	2	0.5	0.5	0.5	1
psubusw mm, mm	2	2	1	2	0.5	0.5	0.5	1
punpckhbw mm, mm	2	2	1	2	0.5	0.5	1	1
punpckhwd mm, mm	2	2	1	2	0.5	0.5	1	1
punpckhdq mm, mm	2	2	1	2	0.5	0.5	1	1
punpcklbw mm, mm	2.10	2	1	2	0.5	0.5	1	1
punpcklwd mm, mm	2	2	1	2	0.5	0.5	1	1
punpckldq mm, mm	2	2	1	2	0.5	0.5	1	1
pxor mm, mm	2	2	1	2	0.5	0.5	0.36	1
pavgusb mm, mm	2	2	-	-	0.5	0.5	-	-
pf2id mm, mm	5	5	-	-	1	1	-	-
pfacc mm, mm	4	4	-	-	1	1	-	-
pfadd mm, mm	4	4	-	-	1	1	-	-
pfcmpeq mm, mm	2	2	-	-	1	1	-	-
pfcmpge mm, mm	2	2	-	-	1	1	-	-
pfcmpgt mm, mm	2	2	-	-	1	1	-	-
pfmax mm, mm	2.10	2	-	-	1	1	-	-
pfmin mm, mm	2	2	-	-	1	1	-	-
pfmul mm, mm	4	4	-	-	1	1	-	-
pfrcp mm, mm	3	3	-	-	1	1	-	-
pfrcpit1 mm, mm	4.20	4	-	-	1	1	-	-
pfrcpit2 mm, mm	4	4	-	-	1	1	-	-
pfrsqit1 mm, mm	4.20	4	-	-	1	1	-	-
pfrsqrt mm, mm	3	3	-	-	1	1	-	-
pfsub mm, mm	4	4	-	-	1	1	-	-
pfsubr mm, mm	4.20	4	-	-	1	1	-	-
pi2fd mm, mm	5	5	-	-	1	1	-	-
pmulhrw mm, mm	3	3	-	-	1	1	-	-
pf2iw mm, mm	5.30	5	-	-	1	1	-	-
pfnacc mm, mm	4.21	4	-	-	1	1	-	-
pfpnacc mm, mm	4.21	4	-	-	1	1	-	-
pi2fw mm, mm	5	5	-	-	1	1	-	-
pswapd mm, mm	-	2	-	-	-	0.5	-	-
bswap r64	-	0.33	1	-	-	0.33	1	-
bswap r32	0.38	0.33	1	1	0.38	0.33	1	1
cmove r64, r64	-	1	2	-	-	0.33	2	-
cmove r32, r32	1	1	2	6	0.45	0.33	1	1.25
cmovne r64, r64	-	1	2	-	-	0.33	2	-
cmovne r32, r32	1	1	2	6	0.45	0.33	1	1.25
cmpxchg r64, r64	-	2	8	-	-	2	7	-
cmpxchg r32, r32	3.19	2.16	8	9	3	2.16	7.53	9
mov r64, r64	-	1	1	-	-	0.33	0.33	-
mov r32, r32	1	1	1	0.5	0.38	0.33	0.33	0.33
xadd r64, r64	-	2	2.12	-	-	1	2	-
xadd r32, r32	2	2	2	1	1.14	1	2	1
xchg r64, r64	-	1.07	2	-	-	2	1.33	-
xchg r32, r32	1.18	1	2.14	1	1.87	2	1	1
adc r64, r64	-	1	2	-	-	1	2	-
adc r32, r32	1	1	2	7	1	1	2	7
add r64, r64	-	1	1	-	-	0.33	1	-
add r32, r32	1	1	1	0.5	0.38	0.33	0.47	0.35
cmp r64, r64	-	0.33	1	-	-	0.33	1	-
cmp r32, r32	0.38	0.33	1	0.33	0.38	0.33	1	0.33
dec r64	-	0.33	0.33	-	-	0.33	0.33	-
dec r32	0.33	0.33	0.33	0.66	0.33	0.33	0.33	0.66
div r64, r64	-	-	-	-	-	-	-	-
div r32, r32	-	-	-	-	-	-	-	-
idiv r64, r64	-	-	-	-	-	-	-	-
idiv r32, r32	-	-	-	-	-	-	-	-
imul r64, r64	-	4	5	-	-	4	5	-
imul r32, r32	4	3	3	14.13	4	3	3	14.13
inc r64	-	0.33	0.33	-	-	0.33	0.33	-
inc r32	0.33	0.33	0.33	0.66	0.33	0.33	0.33	0.66
mul r64, r64	-	-	-	-	-	-	-	-
mul r32, r32	-	-	-	-	-	-	-	-
neg r64, r64	-	-	-	-	-	-	-	-
neg r32, r32	-	-	-	-	-	-	-	-
sbb r64, r64	-	1	2	-	-	1	2	-
sbb r32, r32	1	1	2	7	1	1	2	7
sub r64, r64	-	1	1	-	-	0.33	1	-
sub r32, r32	1	1	1	0.5	0.38	0.33	0.47	0.35
and r64, r64	-	1	1	-	-	0.33	1	-
and r32, r32	1	1	1	0.5	0.38	0.33	0.47	0.5
not r64	-	0.33	0.33	-	-	0.33	0.33	-
not r32	0.38	0.33	0.33	0.5	0.38	0.33	0.33	0.5
or r64, r64	-	1	1	-	-	0.33	1	-
or r32, r32	1	1	1	0.5	0.38	0.33	0.47	0.5
xor r64, r64	-	1	1	-	-	0.33	1	-
xor r32, r32	1	1	1	0.5	0.38	0.33	0.47	0.5
rcl r64, imm8	-	3	10	-	-	3	10	-
rcl r32, imm8	4	3	10	18.98	4	3	10	18.98
rcr r64, imm8	-	3	10	-	-	3	10	-
rcr r32, imm8	3.62	3	10	18.98	3.56	3	10	18.98
rol r64, imm8	-	0.33	1	-	-	0.33	1	-
rol r32, imm8	0.44	0.33	1	4	0.44	0.33	1	4
ror r64, imm8	-	0.33	1	-	-	0.33	1	-
ror r32, imm8	0.44	0.33	1	4	0.44	0.33	1	4
sal r64, imm8	-	0.33	0.5	-	-	0.33	0.5	-
sal r32, imm8	0.44	0.33	0.5	1	0.44	0.33	0.5	1
sar r64, imm8	-	0.33	0.5	-	-	0.33	0.5	-
sar r32, imm8	0.44	0.33	0.5	1	0.44	0.33	0.5	1
shl r64, imm8	-	0.33	0.5	-	-	0.33	0.5	-
shl r32, imm8	0.44	0.33	0.5	1	0.44	0.33	0.5	1
shr r64, imm8	-	0.33	0.5	-	-	0.33	0.5	-
shr r32, imm8	0.44	0.33	0.5	1	0.44	0.33	0.5	1
bsf r64, r64	-	9	2	-	-	9	1	-
bsf r32, r32	7	8	2	8	7.51	8	1	2
bsr r64, r64	-	10.96	2	-	-	10	1	-
bsr r32, r32	9	10.96	2	8	9	10	1	2
bt r64, r64	-	0.33	1	-	-	0.33	1	-
bt r32, r32	0.44	0.33	1	6	0.44	0.33	1	6
btc r64, r64	-	2	1	-	-	0.66	1	-
btc r32, r32	2.10	2	1	6	1.14	0.66	1	6
btr r64, r64	-	2	1	-	-	0.66	1	-
btr r32, r32	2	2	1	6	1.14	0.66	1	6
bts r64, r64	-	2	1	-	-	0.66	1	-
bts r32, r32	2	2	1	6	1.14	0.66	1	6
test r64, r64	-	0.33	1	-	-	0.33	1	-
test r32, r32	0.38	0.33	1	0.5	0.38	0.33	1	0.5