A Stop is not a Stop

[iND]

Traditionally full f stops reflect the aperature opening that is needed to half (or double) the amount of light passing to the sensor. As we progress up the f stops 0.5- 128 we decrease the light by half with each smaller opening.
<table style="text-align: center;" class="wikitable"]
<tbody>
<tr bgcolor="#ccffcd"]
<td>


0.5
</td>
<td>0.7</td>
<td>1.0</td>
<td>1.4</td>
<td>2</td>
<td>2.8</td>
<td>4</td>
<td>5.6</td>
<td>8</td>
<td>11</td>
<td>16</td>
<td>22</td>
<td>32</td>
<td>45</td>
<td>64</td>
<td>90</td>
<td>128</td>
</tr>
</tbody>
</table>


So 1.4 - 2.2 is a 'stop' and 16-22 is a 'stop' but each stop does not representthe same


amount of light being blocked.


We also use the term 'stop' to refer to a change in speed settings


1 sec 1/2 1/4 1/8 1/16 1/32 1/64 1/128 1/256 1/512


likewise were refer to changing the speed setting from 1/4-1/8 as a 'stop' decrease


and also a change from 1/128 - 1/256 as a stop decrease.


Each will decrease the amoung of light passing through to the sensor.


The 1/128-1/256 represnts much less light stopped that 1/2- 1/4.


Just like the aperature stops the speed stops decreae the light by half (or double)


but this is where the simularity ends.


Certainly each stop is not stopping the same amoung of light.


This leads usalso to the realizationthat time stops and aperature stops can not be directly


compared. Changing and aperature by one stop has no direct translation to changing


speed by one stop.


To make this also more confusing as our camera are marked in partial f stops and partial


timechanges.


So why all this:


Comments are beign made like:


A stop is a stop


or


IS improves hand held by 3 stops (stops of speed not aperature)


do we cosider this true over the entire stop range ie 1 sec exposure is improved by 3 stops


as well as a 1/60 exposure improved by 3 stops. NO


A stop is not a stop. It is relative.



I think we throw the term stop around a little too loosely lately.


Comments?

[Dave Johnston]

My head just exploded.


.... rebuilding.....rebuilding...." A fatal error has just occured"...

[ddt0725]

STOP! Just when I think I've got it ...I find myself more confused! [*-)]

Good topic, iND. This thread has Neuro written all over it! [:P] Where are you, John!?

Denise

[Jon Ruyle]

So 1.4 - 2.2 is a 'stop' and 16-22 is a 'stop' but each stop does not representthe same


amount of light being blocked.

That's right. A stop does not refer to a quantity of light being blocked but a fraction of light being blocked. Exposure values are an exponential scale (in which one notch on the scale represents multiplication by a constant factor), not an arithmetic one (in which one notch on the scale represents an addition or subtraction of the same amount).

Changing and aperature by one stop has no direct translation to changing


speed by one stop.

I think you're confused. The idea is that closing the lens one stop (cutting area in half) has the same effect on exposure as doubling the shutter speed. A stop, in this sense, actually is a stop.

IS improves hand held by 3 stops (stops of speed not aperature)

Speed *and* aperture must change to get the same exposure. Most people in most situations fixate on one or the other, but of course if you change one without changing the other, you'll get a different exposure.

do we cosider this true over the entire stop range ie 1 sec exposure is improved by 3 stops


as well as a 1/60 exposure improved by 3 stops. NO

IS improvement is an estimate. The amount of improvement depends on the way you shake. If your hand shakes with a super high frequency but low amplitude, IS would provide no benefit. But this isn't how people really shake.


There may be some shake patterns for which IS can correct at higher shutter speeds better than at lower shutter speeds (and some people have said IS does not give the full benefit at low shutter speeds, ie, short focal lengths), but my experience is that a 3 stop IS does give 3 stops of improvement. With my 24-105, I can shoot about 1/4 second exposures at 24mm, or about 1/16 second at 105mm. But this is just an empirical thing, YMMV. (Eg, may be that the 24-105 does not have enough travel in the IS system to give the full 3-stop benefit at the wide end for some people's way of shaking).

[neuroanatomist]

So 1.4 - 2.2 is a 'stop'

Actually, 1.4 - 2.2 is one and a third stops.

Each stop does not representthe sameamount of light being blocked.

In absolute terms, no. In relative terms, yes. Half of 100 is 50, an absolute difference of 50. Half of 4 is 2, an absolute difference of 2. Is "half of..." not "half of..."? The term "half of...," just like the term "stop of light" is relative.

Traditionally full f stops reflect the aperature opening that is needed to half (or double) the amount of light passing to the sensor.

Not exactly. An f-number is also relative - in this case, to focal length. The f-number is the ratio of focal length to aperture(iris diaphragm) diameter. So, a 100mm lens at f/4 has an aperture diameter of 25mm, whereas a <s>30mm</s>20mm lens at f/4 has an aperture diameter of 5mm. But, those two lenses would let in the same absolute amount of light. The aperture is 1/5 the diameter because the focal length is 1/5 as long.


The amount of light exiting a lens is proportional to the focal length and the area of the iris diaphragm. That's why the f-number scale progresses as it does - the area of a circle is given by&pi;r², so to halve the area of a circle, the factor used is&radic;2 (&pi; is factored out by the ratio). Thus, the full stop progression starts at 1 and goes up (approximately) by multiplying the preceding number by&radic;2 (=1.414).

A stop is a stop

This comment is true, in practice. Changing aperture by one stop doubles or halves the amount of light reaching the sensor. Changing shutter speed by one stopdoubles or halves the amount of light reaching the sensor. It does break down a little at a couple of places in the shutter speed progression, due to rounding (1/8 is not half of 1/15, for example).

IS improves hand held by 3 stops

An IS system rated for 3 stops of improvement applies to either aperture or shutter speed, in terms of intent, and provided that one understands what IS does and doesn't do. You're correct that what IS does is allow hand-holding at slower shutter speeds. But for the same exposure, IS can allow me to use a narrower aperture than I could otherwise use for a handheld shot, to increase the depth of field. However, it's an important consideration that the IS improvement means slower shutter speeds, since they can become too slow to stop subject motion (with the exception of IS on really long lenses).

1 sec exposure is improved by 3 stopsas well as a 1/60 exposure improved by 3 stops

True, but sort of irrelevant. I can't handhold a lens at 1 second exposure anyway. Oh wait - I can, sort of...


[img]/cfs-file.ashx/__key/CommunityServer.Components.UserFiles/00.00.00.35.15/IMG_5F00_2981.jpg[/img]


This was shot handheld with an EF-S 17-55mm f/2.8 IS USM, at 24mm, f/10, and a 1.6 second exposure. Zoomed to 1:1, the camera shake is there, but judging by something like the railing at the left side of the image, it's not bad. In this case, the 3-stop IS improved my 0.2 (1/5) s exposure to 1.6 s, which allowed me to stop down the aperture from f/3.5 to f/10 and increase the DoF to get more of the waterfall in focus - and in this case, subject motion (the water) was part of the desired effect for the shot.

This leads usalso to the realizationthat time stops and aperature stops can not be directlycompared. Changing and aperature by one stop has no direct translation to changingspeed by one stop.

Stops of shutter speed and f-stops can most certainly be directly compared. If you take an exposure reading, then set a one-stop smaller aperture and a one stop slower shutter speed, you'll have the same amount of light. That's why we talk in terms of 'stops' and not aperture diameters, fractions of seconds, etc. Likewise, the light detector is also calibrated in 'stops' - ISO sensitivity progresses as a doubling also. This means we can double or halve the amount of light by a one-stop change in any 'side' of the 'exposure triangle'.

[Jon Ruyle]

Actually, 1.4 - 2.2 is one and a third stops.

Good point

This was shot handheld with an EF-S 17-55mm f/2.8 IS USM, at 24mm, f/10, and a 1.6 second exposure.

Steady hands. Ever consider a career as a neurosurgeon?

[Richard Lane]

Not exactly. An f-number is also relative - in this case, to focal length. The f-number is the ratio of focal length to aperture(iris diaphragm) diameter. So, a 100mm lens at f/4 has an aperture diameter of 25mm, whereas a 30mm lens at f/4 has an aperture diameter of 5mm. But, those two lenses would let in the same absolute amount of light. The aperture is 1/5 the diameter because the focal length is 1/5 as long.

Very Interesting discussion!


Neuro..., I think you may have a typo or you lost me, doesn't a 30mm lens at f/4 have an aperture diameter of 7.5mm, and then shouldn't it say;The aperture is 1/3 the diameter because the focal length is 1/3 as long.


In other words, that these two lenses will let in the same amount of light because their respective focal lengths are approx. 1/3 each other (100mm to 30mm) and their respective aperture diameters (25mm to 7.5mm) are also approx. 1/3 of each other, therefore the relative ratio of light they let in will be the same?


Stop me If I'm off track, Pun intended.


Rich


PS. I love the golden waterfall.

[neuroanatomist]

Neuro..., I think you may have a typo or you lost me, doesn't a 30mm lens at f/4 have an aperture diameter of 7.5mm, and then shouldn't it say;The aperture is 1/3 the diameter because the focal length is 1/3 as long.

Good catch - it was a typo, should have read 20mm, so f/4 would be 5mm diameter.


But the bottom line is that because f-number is a ratio, any lens of any focal length, when set to the a particular f-number, will let the same amount of light through the lens as any other lens of any other focal length set to that same f-number.


Thanks for the compliment on the waterfall - that's at Yerba Buena Gardens in San Francisco.

[Richard Lane]

OK, cool, that makes sense. What are those purple objects towards the top of the photo?


Is that a water valve?

[neuroanatomist]

What are those purple objects towards the top of the photo?

They are lights running up the side of a building (a hotel, I think) a few blocks in the background.