Canon 85 1.2 is in a different league man.
The league of slooow focus. Only advantage being a stop faster. With
a D3 you make up that stop.
Actually, 1.2 vs 1.4 is only a third of a stop faster. If the best
shutter speed with the 1.4 is 1/60, then with the 1.2 you'll be able
to get 1/80. Not much of a difference.
I stand corrected. Minor DOF difference too.
Typical one-third-stop f-number scale
f/# 1.0 1.1 1.2 1.4 1.6 1.8 2 2.2 2.5 2.8 3.3 3.5 4 4.5
5.0 5.6 6.3 7 8 9 10 11 12.5 14 16 18 20 22
Very interesting article on Wikipedia.
The article you cite mentions that
Notice that sometimes a number is ambiguous; for example,
f/1.2 may be used in either a half-stop or a one-third-stop
system
But it doesn't say why. The reasons is that it's the rounding
used. Showing only two significant digits gets you
into trouble. With only two digits, you can't tell the
difference between the one-half-stop sequence and the
one-third-stop sequence.
Taken at five significant digits, the first half-stop past f/1
is 1.1892, while the first (one-)third-stop past f/1 is 1.2599.
At only two significant digits, both those figures round to 1.2,
but they aren't the same.
Between f/1 and f/2, taken at full, (one-)half-stop, (one-)third-stop,
and (one-)sixth-stop intervals, we find these numbers to five significant
digits:
1/1: 1.0000 1.4142 2.0000
1/2: 1.0000 1.1892 1.4142 1.6818 2.0000
1/3: 1.0000 1.1225 1.2599 1.4142 1.5874 1.7818 2.0000
1/6: 1.0000 1.0595 1.1225 1.1892 1.2599 1.3348 1.4142
1.4983 1.5874 1.6818 1.7818 1.8877 2.0000
The formula is simple: just raise the square root of two
to the power of the your stop, like to 1.5 or to 1.33333etc.
Thus sqrt(2) 1.5 yields something different than
sqrt(2) 1.33333 yeilds. (Or sqrt(2) 4/3 if you prefer.)
Here's a simple and little tiny program that shows this progression:
- ! usr/bin/perl
$LO = -1; # start at -1st full stop, or change to 0
$HI = 9; # go until the 9th full stop (f/16)
for $delta (1, 2, 3, 6) {
print "1/$delta:\t";
for ($stop = $LO; $stop
printf "%5.4f ", sqrt(2) $stop;
}
print "\n";
}
That shows you a (one-)sixth-stop sequence of
0.7071 0.7492 0.7937 0.8409 0.8909 0.9439
1.0000 1.0595 1.1225 1.1892 1.2599 1.3348
1.4142 1.4983 1.5874 1.6818 1.7818 1.8877
2.0000 2.1189 2.2449 2.3784 2.5198 2.6697
2.8284 2.9966 3.1748 3.3636 3.5636 3.7755
4.0000 4.2379 4.4898 4.7568 5.0397 5.3394
5.6569 5.9932 6.3496 6.7272 7.1272 7.5510
8.0000 8.4757 8.9797 9.5137 10.0794 10.6787
11.3137 11.9865 12.6992 13.4543 14.2544 15.1020
16.0000 16.9514 17.9594 19.0273 20.1587 21.3574
You wonder what rounding Canon was using with their
old 50mm f/0.95 lens. Did they round 0.9439 to 0.95?
Or where they using a one-seventh-stop system? The
last seventh before f/1 is f/0.9517 (that is, sqrt(2) (-1/7)
is 0.95169515301062...), while the last sixth before f/1
is only 0.943874312681694.... Seems like Marketing's
idea of rounding to me.
--tom
