On Wed, Jun 20, 2001 at 02:09:11PM +0300, Severi Salminen wrote:
> > that an +1/2 stop increment over f1.4 is f1.7 and not f1.8!
> I'd say not a big deal. Actually the right answer is not 1.7, but
> 1.6: sqrt(2)^sqrt(2)=1.63...
Where did you get that formula from?
It should be 2^(3/4) = 1.68...
Incrementing by a stop means multiplying by sqrt(2),
by N stops multiplying by (sqrt(2))^N -
thus half a stop means (sqrt(2))^0.5 = sqrt(sqrt(2)),
a third stop (sqrt(2))^1/3, etc.
Thus, with three decimal places:
full stop +1/3 +1/2 +2/3
1.000 1.122 1.189 1.260
1.414 1.587 1.682 1.782
2.000 2.245 2.378 2.520
2.828 3.175 3.364 3.564
4.000 4.490 4.757 5.040
5.657 6.350 6.727 7.127
8.000 8.980 9.514 10.079
11.314 12.699 13.454 14.254
16.000 17.959 19.027 20.159
22.627 25.398 26.909 28.509
32.000 35.919 38.055 40.317
> So in every case all these numbers are approximations.
Yes - but not only are they occasionally rounded wrong,
they are also are different depending on whether 1/2 or 1/3
stop stepping are being used.
If you have a body where it can be selected like EOS 3,
you can observe that 2.8 plus half stop is 3.5,
but 2.8 plus 2/3 stops is also 3.5!
(I don't have camera on me now so the numbers could
be wrong, but you get the idea.)
So if the camera displays 3.5 you can't know which it means
without checking the CF that selects 1/2 vs. 1/3 step.
Not that it matters any in practice
> Think about the shutter speeds: 1/4, 1/8, 1/15!. 1/60,1/125! So in
> every case your pictures are ruined :)
--
Tapani Tarvainen
*
****
*******
***********************************************************
* For list instructions, including unsubscribe, see:
* http://www.a1.nl/phomepag/markerink/eos_list.htm
***********************************************************