Почему стандартный list быстрее массива Numpy?

import numpy as np
import time
 
fields = [6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6]
fields2 = np.array([6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6])
 
time1 = time.time()
i = 0
for _ in range(1000001):
    fields[i] += 1
    i += 1
    if i > 11:
       i  = 0
print(f'List: {time.time()- time1}')
print(fields)
 
time1 = time.time()
i = 0
for _ in range(1000001):
    fields2[i] += 1
    i += 1
    if i > 11:
       i  = 0
print(f'Numpy: {time.time()- time1}')
print(fields2)

#import time
#import numpy as np
 
fields = [6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6]
 
def f1(fields):
    i = 0
    for _ in range(1000001):
        fields[i] += 1
        i += 1
        if i > 11:
           i = 0
 
def f2(fields):
    l = len(fields)
    i = 0
    while i <= 1_000_000:
        fields[i % l] += 1
        i += 1
    # assert fields == [83340, 83340, 83340, 83340, 83340, 83339, 83339, 83339, 83339, 83339, 83339, 83339]
 
 
def f3(fields):
    n = 1_000_001
    l = len(fields)
    s = n // l
    for i in range(l):
        fields[i] += s
    for i in range(n % l):
        fields[i] += 1
    #assert fields == [83340, 83340, 83340, 83340, 83340, 83339, 83339, 83339, 83339, 83339, 83339, 83339]
 
import timeit
 
print(timeit.timeit('f1(fields.copy())', setup='from __main__ import f1, fields', number=10))
print(timeit.timeit('f2(fields.copy())', setup='from __main__ import f2, fields', number=10))
print(timeit.timeit('f3(fields.copy())', setup='from __main__ import f3, fields', number=10))

>>> print(sum([83340, 83340, 83340, 83340, 83340, 83339, 83339, 83339, 83339, 83339, 83339, 83339]))
1000073
>>> 1000073 - sum([6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6])
1000001
>>>