Lens Mathmatics for dummies

A 35-100mm lens is a 35-100mm lens, whether it's on a Nikon or an Olympus. For any given lighting situation, and any given subject-to-camera distance, exposure and depth of field do not change. The "equivalence" simply means that Olympus's smaller sensor crops the edges away from what a larger sensor would see, so that the field of view is different. 35mm is a moderate wide-angle on a "full-frame" sensor, and the 4/3 sensor crops that down to a moderate tele.

However, if you try to get the same field of view (for example, a head-and-shoulders portrait) with a 4/3 sensor as you get with a full frame sensor, you have to change the camera-to-subject distance, and that DOES change the depth of field for any given aperture. The closer you are, the less depth of field.

--
Dean Bergmann
Olympus: OM-1, OM-2, XA, E-510
Mamiya C330
http://www.pbase.com/installer/galleries
 
Gidday Illy
once again this descends into a brand slanging match, funny thing is
all brands take equally good pictures.........funny that!
--
http://illy.smugmug.com
EXACTLY the point I keep trying to make in the face of outrageous bashing, flaming and baiting that goes on by the trolls here ...

My E-510 doesn't limit my photography; I limit its capabilities ...

Regards, john from Melbourne, Australia.
-----

The Camera doth not make the Man (or Woman) ...
Perhaps being kind to cats, dogs & children does ...

http://canopuscomputing.com.au/gallery2/main.php



Bird Control Officers on active service.
 
With the 4/3 sized sensors (or any "smaller" sensor when comparing two different sized sensors), you are using a smaller focal length to achieve the same angle of view (setting aside aspect ratio issues).

Remember the very old DOF formula: all other things held constant, a shorter focal length yields larger DOF.

It's that simple.
I notice many people her have a good understanding of the
mathematical/ scientific concept of digital photography. Now as being
fairly new to slr photography, I'd like to get an understanding of
4/3 lens with reference to f stop.

As we know to maintain exposure if we step down our aperature we must
increase our shutter speed, proportionally.

I also have read how with 4/3 there seems to be a greater dof at a
given f stop than 2/3 or 135 sensors. the question I have is does the
35-100 f2 in effective terminology become 70-200 f2, 70-200 f4( since
you double one do you halve the other?), or something in between?

Sorry if I ran on a bit. I'm not looking for all the technical
aspects, just a general understanding.
--
Galleries: http://www.dheller.net

I am one of the few who decry elitism.
 
Wrong. The lens projects an image circle. The sensor intercepts a
portion of the projected light.
That is correct.
The larger the sensor, the larger
portion is intercepted.
This statement is not correct as such. Based upon your logic if I were to put a sensor that is 20 times as large as FF into the camera then I would get 20 times as much light. However, this is not the case as you are not throwing away the vast majority of the light that you get.

So you are trying to tell me that 4/3 throws away more than 75% of the light coming into the lens? Oh yeah, and a small compact camera has an even smaller sensor and then throws away 90%+ of the light? :-)

Of course you are not capturing all the light as the image circle is round, but things are not as simple as you make it sound.

I think there is a bit more to camera/lens design than just sticking some type of sensor behind some type of lens and claiming that the bigger the better. Unfortunately, that is what you argue here.
 
You obviously ignore my graph that shows the way that any unit area
is affected by the same amount of light (of course the total area
does not matter).
Part of the reason I ignored your graph is because your example with the earth is so flawed. You also ignore my post to the tutorial that explicitly stated that you need longer exposure time on FF than on smaller sensors for the same DOF. If you accepted that then we would not need to argue here :-)
You need some "good desire" to accept my example about earth as good.
The lens does not collect light. It directs the light wich pas
through the lens.
Yes, but the amount of light that the lens passes on depends upon the design. That is why a f2 lens is "brighter" than a f4 lens, because it passes on four times as much light. That is also the reason why you have a brighter image in the viewfinder with a brighter lens.
 
If the image circle is large enough to illuminate the sensor, it's exactly as I've described simply in my last post. For example, if you mount a legacy 35mm format f3.5 lens on a 4/3 camera (using an adapter), it will meter and expose just like a f3.5 4/3 lens. The smaller 4/3 sensor will intercept the center portion of the image circle. (Really fast legacy lenses are reputed to meter more erratically, due perhaps to the smaller 4/3 mirror box baffling (blocking) some of the light trying to reach the meter, but this is a digression...)

The f ratio determines the light intensity (per unit area) projected onto the sensors. Regardless of what lens you're discussing, assuming of course the aperatures are correctly calibrated, f ratio determines that light intensity. A larger sensor within the image circle captures more total light than a smaller sensor, but exactly the same light per unit area (assuming of course the uniform illumination throughout the image circle, which is never literally true but close enough for practical exposure purposes).

You'll note that compact cameras have smaller sensors but they also have smaller lenses (shorter focal lengths, often with similar f ratios). These smaller lenses are capturing and projecting less light (smaller aperature as measured in mm, not f ratio, and smaller image circles) but the same light per unit area within their smaller image circles as DSLR lenses with same f ratio.

There is a lot of confusion around this topic. It comes up once a month I think. The discussion takes on brand connotations. People get defensive and start throwing up unsubstantiated arguments.

Larger formats DO have advantages, specifically in flexibility for shallow DOF and and high ISO. It's not just 35mm vs APS and 4/3. There's medium format, large format... There are reasons for larger formats besides increased resolution. Why do Olympus owners want to argue that point?

4/3 has some terrific advantages too, specifically size, weight, and cost. And this translates mostly into lens performance as I see it. Call it what you will (telecentricity?). I believe, given accepted norms in lens size and weight, especially in the "pro" line and higher, Olympus is able to build larger, heavier, costlier lenses (than would traditionally be required to deliver the image circle required; still in the normal range vs competition) and this provides design freedom to deliver better corner-to-corner performance even wide open, often with larger than normal zoom ranges, than is typical of its competition.

On balance, I prefer Olympus 4/3. I like the bodies but this is mostly because of the lens offerings - great focal length ranges with fewer faster high quality lenses (reduced number of lenses and weight) than I think I could afford (or may even be available) with other brands).
 
Yes, but the amount of light that the lens passes on depends upon the
design. That is why a f2 lens is "brighter" than a f4 lens, because
it passes on four times as much light. That is also the reason why
you have a brighter image in the viewfinder with a brighter lens.
It's absolutely right. I just have (maybe my poor english) the impression that you said something different on previous posts.
Anyway that you mention above is right.

Remember that a 50mm lens passes 4X times more light than a 25mm lens for given aperture, but the exposure time is the same for the same gray sceen. So, there is a huge amount of physics behind this "strange" phainomenon. That means and explains my opinion about the importance of light/unit area.
 
Gidday Jonas
Tjena John,
NOTE** that neither of these sizes are 'imaging sizes'; you either
have to compare sensor size or imaging size, Jonas. It is simply not
acceptable to compare other than known dimensions.
I compare the active area used for the image only. I'm not interested in counting the chip's are for connectors and such things, just as the thickness is of little interest. I'm discussing practical photography, not chip design or circuit layouts.

We have the official imaging size of the 4/3 sensor and it seems as we can agree on that one, being 17.3x13mm. The Pentax/Nikon 1.5 crop sensor I have measured myself. I had a long discussion with Riley back when doing that. He likes to use the chip size and I like to use the imaging area. You can decide for yourself which area that is of interest for a photographer.
Nikon's sensors are the same size as Pentax and they are 60% larger
than the 4/3 sensors (area).
Not exactly 60%, nor even approximately correct ...
Indeed there is a round off error but the area is 60% larger.
The crop factor is not 1.6 but 1.5.
So, I made an error of one decimal point ... I admit it, your Honour,
I will never do it again, just don't send me to jail forever ...
You used the Canon number. No big deal but if we talk facts we can try to get things correct.
In relation to DOF this is 1.33 f-stops. Then from an 1.5 crop factor
sensor to a FF sensor it is another step of 1.5 stops.
This adds up to 2.83 f-stops, Jonas. Since when did 2.83 = 2?
Since... well, never. It was me expressing myself in a bad way. Here goes, out of a DOF point of view, from 4/3 via 1.5x to FF:

4/3 - use f/2 in this sample

1.5 crop factor sensor - multiply with 1.33 and get f-stop 2.67. Expressed in f-stops the difference between 2 and 2.67 is 0.82 stops

FF sensor - multiply again with 1.5 and get f-stop 4. Now the difference between FF and the 1.5 crop factor sensor is 1.18 stops

From 4/3 to FF we get this:

4/3 - use f/2 in this sample

FF sensor - multiply with 2 and get f-stop 4. Expressed in f-stops the difference between f/4 and f/2 is 2 stops.

Yup - I'm guilty as well of sloppy presentation and I sure hope I got it right above.
The Pentax/Nikon sensors active area are 360mm^2 (23.2x15.5 mm)
The 4/3 sensor is 225mm^2 (17.3x13 mm)
Please give precise reference for this assertion, since "we" are
getting picky, in the extreme.
The reference you are asking for is unavailable as the search engine here times out. I'll get back with it later. You can however get an idea about how the measurement was done by looking at this image:


On total sensor area the Nikon is 23.6 x 15.8 = 372.88 mm²; the
Olympus is 18.0 x 13.5 = 243 mm². This gives the areal ratio as
1.5345; not 1.60, as you suggest.
Come on. That's the way Riley does it. We are talking images. The imaging area is of interest. The sides, connectors and holes for mounting or whatever are of no interest when discussing how to achieve an image.
I had to cut some paragraphs talking about the confusion over "APS-C" and the different sensor sizes.
So, with all the B/S being spruiked here, I am not the ultimate
expert on Nikon sensor sizes (I threw myself on the mercy of the
court - see above - rotflmho) ... compared with the absolutely
outrageous assertions made here, I am out by a whopping single
decimal point and somewhat less than 1/3 of a stop and you are
hopping into ME ? What about the "gentlemen" who are out by a factor
of two** or more stops in their assertions, Jonas? How about
actually correcting some of their arithmetic?
I think they are corrected at the same time.
Secondly it gives the smaller sensor what is often desperately
needed, MORE** DoF at wider apertures. In order to get any decent
DoF with the larger sensor, it is necessary to stop the lens down to
heck and ramp the ISO as well. Often by at least 4 stops between the
two. Diffraction sets in and the ISO performance of the larger sensor
is absolutely needed as there is no other choice .
Not 4 stops John. As long as we keep to 4/3 and FF there is a 2 stop difference.
and the smaller sensors an advantage when it
comes to reach.
Yes. Big time. AND decent lens design for UWAs. And transportable**
lenses for telephoto work ...

So we reach a point where the "shortcomings" of 4/3rds are actually
an advantage, at both ends of the lens range.
Yup. The important point is set by the individual. I'm the sort that use lenses (35mm equivalences here) between 20 and 100mm with the wast majority of shots taken around 50mm.

Anyone at an extreme end may have other preferences. Many working in the same range as I do may have other preferences.
AND the lenses are faster, cheaper (relative to grade), smaller
(comparing like to like, i.e. an f2 lens with an EFL f2 lens etc -
oh, sorry there aren't any ... - well if there were ... , lighter,
better optical quality ... DID I mention just better all round?
Sometimes, sometimes not. Look at the weather shielded EF24-70/2.8. It is one stop slower but has one stop better DOF control. It starts at 24mm and it is around half the price or so compared to the ZD14-35/2. It is also 50grams heavier (taken from memory).
If I have made any other 0.1 errors, Jonas, please forgive me ...
Of course.

Cheers,

--
Jonas

I had to cut stuff away from my reply in order to make it fit in one post. Sorry.
 
while you bandy my reasoning about Jonas
i dont mind being 'wrong' or misguided
all i want is proof, not some guy with a ruler and a flashlight

i have provided from time to time excerpts from kodak documents about 4/3rds imager sizes (you may have a few yourself), and provided enough to field the idea that the sensors at least, vary somewhat. Im happy with the notion that the current panasonic batch have an imaging size of 17.3x13, i think we can both accept that the sensor is larger, maybe not.

so heres the shot, i have asked others for the same and looked myself but come up dry, I have also asked for E1 owners to do a FOV test to see if its different. No takers...

so please
come up with a technical paper on any apsc sensor, long as we know what it is
then we can see the size of the image area against dpreview quoted 'sensor size'

it just could be that canon apsc is 22.2mm x 14.8mm imaging size
or, if it isnt what is it
or nikon/pentax or whatever

when we have definitive proof that x=x and y=y we can have apples with apples representation. All i want here is procedural fairness, thats not an unfair call

if one is quoting sensor size and the other imager size it all goes to u know what

that way we can resolve this forever
i am honestly interested in a solution

in an ideal world i want to know what they all are by size, im ok with discussing definitive evidence of anything at this point
but i aint backing down until that is done
thats as good as it gets...

just an aside check this out



thats the E300 sensor and it has an SHQ of 3264x2448
sometimes it just (just) gets harder doesnt it ?
--
Riley

in my home, the smoke alarm is the dinner bell (just)
 
(...)
just an aside check this out

http://i194.photobucket.com/albums/z179/realink_album/review/KAF-8300_internet.jpg
thats the E300 sensor and it has an SHQ of 3264x2448
sometimes it just (just) gets harder doesnt it ?
No. it doesn't get harder. It is still very easy Riley:

In my comparison I use real life numbers. Until now I have used the official Olympus imagener size of 17.3x13mm. If you want to change the numbers it doesn't change anything in reality. As my comparison is based on the result from normal shooting it is valid for any practical use. As everything is the same in the relative way it doesn't matter what the fact sheet says.

I guess you remember our last discussion on this. It is pretty meaningless to repeat everything but when thinking of the method: same lens, same distance, same raw converter and in the end the resulting differences in FOV... well, then the numbers I presented are very much real.

What does it matter? Nothing of course, as nothing has changed. In real life we all already know how it works. It is just this disinformation I'm so tired off.

--
Jonas
 
once again this descends into a brand slanging match, funny thing is
all brands take equally good pictures.........funny that!
--
http://illy.smugmug.com
EXACTLY the point I keep trying to make in the face of outrageous
bashing, flaming and baiting that goes on by the trolls here ...

My E-510 doesn't limit my photography; I limit its capabilities ...
You do not sound like you are coming here to learn and not to limit your camera's capabilities. Your profile says you have been taking photos for 52 years, since you were 8 years old, but the images in your gallery are of quality my 5 year old would make. And I am not saying badly post processed, they are absolutely tasteless. Biggest bashing of the system I have ever seen.

wba
 
wba
once again this descends into a brand slanging match, funny thing is
all brands take equally good pictures.........funny that!
--
http://illy.smugmug.com
EXACTLY the point I keep trying to make in the face of outrageous
bashing, flaming and baiting that goes on by the trolls here ...

My E-510 doesn't limit my photography; I limit its capabilities ...
You do not sound like you are coming here to learn and not to limit
your camera's capabilities. Your profile says you have been taking
photos for 52 years, since you were 8 years old, but the images in
your gallery are of quality my 5 year old would make. And I am not
saying badly post processed, they are absolutely tasteless. Biggest
bashing of the system I have ever seen.

wba
AND people with:

1) no real name;

2) no web site;

3) no gear;

4) no email address;

5) no images;

6) no manners; etc

Qualify as A-Grade internet super heroes ...

At least I have the guts to :

1) admit that I am not the world's greatest photographer;

2) display my work;

3) put up my honest thoughts on various subjects (even though I am a cr@p photographer, I know just a little bit more than that about technical things), even if I may be wrong. At least I then find put that I am wrong ...

UNLIKE a drive-by shooter, who is an anonymous no-body, with a (maybe, but maybe not) talented five year old. Actually He probably is talented, being the son a thirteen year old father ...

Grow up, sport - or you will not last long around here.

You should at least learn to treat others with dignity, courtesy and respect; especially when engaged in robust discussion with one another. Jonas and I actually respect one another, and merely robustly question each other's point of view and facts. Out of this comes "truth", as best it can be ascertained. Ditto, Riley. Perhaps YOU can learn something from this ... .

Regards, john from Melbourne, Australia.
-----

The Camera doth not make the Man (or Woman) ...
Perhaps being kind to cats, dogs & children does ...

http://canopuscomputing.com.au/gallery2/main.php



Bird Control Officers on active service.
 
I was talking about 16mm vs. 35mm for cinema earlier.

In video right now there's a gap that I'd love to see 4/3 fill. Right now there are 2/3" sensors that are pretty much the broadcast standard. Then it jumps to "Super35" sized sensors, which are essentially APS-sized sensors.

Well, that's a big jump.

Super-8 is 5.68mm x 4.23mm. Super-16mm is 11.76mm by 7.08mm. 35mm is 24.89mm by 18.67mm. These were good alternatives to each other.

2/3" chips are 8.8mm x 6.6mm, which is kind of in-between Super-8 and 16mm. From there the jump is to Super35 at 24.4mm x 13.7mm. Now I know this is based on lens availability. 2/3" lenses have been around for quite a while, so it makes sense that the format has become a broadcast standard. Likewise, there are a ton of 35mm cinema lenses out there, so Super-35 makes sense. But the variety of standards have expanded below 2/3" and not above it. There are 1/3" cameras and 1/2" cameras, but still nothing between 2/3" and Super-35.

4/3 at 18mm x 13.5mm would be a really nice video standard, IMO.
 
(...)
just an aside check this out

http://i194.photobucket.com/albums/z179/realink_album/review/KAF-8300_internet.jpg
thats the E300 sensor and it has an SHQ of 3264x2448
sometimes it just (just) gets harder doesnt it ?
No. it doesn't get harder. It is still very easy Riley:

In my comparison I use real life numbers. Until now I have used the
official Olympus imagener size of 17.3x13mm. If you want to change
the numbers it doesn't change anything in reality. As my comparison
is based on the result from normal shooting it is valid for any
practical use. As everything is the same in the relative way it
doesn't matter what the fact sheet says.

I guess you remember our last discussion on this. It is pretty
meaningless to repeat everything but when thinking of the method:
same lens, same distance, same raw converter and in the end the
resulting differences in FOV... well, then the numbers I presented
are very much real.

What does it matter? Nothing of course, as nothing has changed. In
real life we all already know how it works. It is just this
disinformation I'm so tired off.
yes i can see it might be tiring, but please appreciate that some people are interested in the minutia of tech detail

does it matter? well 'if' it creates conflict, then yes it does, b/se the truth to it will eliminate that...
--
Riley

in my home, the smoke alarm is the dinner bell (just)
 
I was talking about 16mm vs. 35mm for cinema earlier.

In video right now there's a gap that I'd love to see 4/3 fill. Right
now there are 2/3" sensors that are pretty much the broadcast
standard. Then it jumps to "Super35" sized sensors, which are
essentially APS-sized sensors.

Well, that's a big jump.

Super-8 is 5.68mm x 4.23mm. Super-16mm is 11.76mm by 7.08mm. 35mm is
24.89mm by 18.67mm. These were good alternatives to each other.

2/3" chips are 8.8mm x 6.6mm, which is kind of in-between Super-8 and
16mm. From there the jump is to Super35 at 24.4mm x 13.7mm. Now I
know this is based on lens availability. 2/3" lenses have been around
for quite a while, so it makes sense that the format has become a
broadcast standard. Likewise, there are a ton of 35mm cinema lenses
out there, so Super-35 makes sense. But the variety of standards have
expanded below 2/3" and not above it. There are 1/3" cameras and 1/2"
cameras, but still nothing between 2/3" and Super-35.

4/3 at 18mm x 13.5mm would be a really nice video standard, IMO.
i dont understand for the life of me why Panaosonic of all ppl havent done this with 2/3" sensors and a bayonet mount. Just imagine how it would aid their video business. They even got part of the way there with LC1 and the F2-2.4/ 28-90 (EFL) lens

OTOH, as a what if, a 4/3" sensor would just (just) blow away the competition for video production, and they could use all that lovely glass, with very little vignetting etc. You might 'get' my drift....

you could even use one of these



--
Riley

in my home, the smoke alarm is the dinner bell (just)
 
(...)
just an aside check this out

http://i194.photobucket.com/albums/z179/realink_album/review/KAF-8300_internet.jpg
thats the E300 sensor and it has an SHQ of 3264x2448
sometimes it just (just) gets harder doesnt it ?
No. it doesn't get harder. It is still very easy Riley:

In my comparison I use real life numbers. Until now I have used the
official Olympus imagener size of 17.3x13mm. If you want to change
the numbers it doesn't change anything in reality. As my comparison
is based on the result from normal shooting it is valid for any
practical use. As everything is the same in the relative way it
doesn't matter what the fact sheet says.

I guess you remember our last discussion on this. It is pretty
meaningless to repeat everything but when thinking of the method:
same lens, same distance, same raw converter and in the end the
resulting differences in FOV... well, then the numbers I presented
are very much real.

What does it matter? Nothing of course, as nothing has changed. In
real life we all already know how it works. It is just this
disinformation I'm so tired off.
yes i can see it might be tiring, but please appreciate that some
people are interested in the minutia of tech detail

does it matter? well 'if' it creates conflict, then yes it does, b/se
the truth to it will eliminate that...
Hi Riley,

Maybe it isn't very tiring and thinking about it it really isn't meaningless either. And yes, I appreciate people interested in the tech facts and fictions. Maybe it's the fact that you have dismissed my real life sensor size investigation repeatedly with a negligent "I don't need a guy with a ruler" that put me off.

I'll try to stay cool.

So the DOF control steps are, when going from 4/3 via 1.5 crop sensor to FF, 0.82 stops and 1.18 stops. Can we agree about 0.8 and 1.2? The max estimated error in my reading was calculated to 2% and perhaps we don't need more but one decimal. As we are living with estimations only and different aspects of ratio it may be more correct to say "a little less than one stop" and "a little more than one stop" respectively.

And a personal note; I like the 4/3 aspect of ratio. I often even use 1:1 so for me the difference is even a little bit less, but essential. Odd?

regards,

--
Jonas
 
OTOH, as a what if, a 4/3" sensor would just (just) blow away the
competition for video production, and they could use all that lovely
glass, with very little vignetting etc. You might 'get' my drift....
Well, Olympus would probably have to re-house their lenses for video. Probably re-design some of them. You can't have "breathing" when you zoom in and out with a video lens, where it's not even an issue in still photography. Also, most people wouldn't use an autofocus lens for video production. But if they wanted to I have no doubt that Olympus could deliver an amazing video system with optics that would be hard to beat.

4/3 really does seem like a very nice size for a video sensor. Obviously the aspect ratio isn't quite right for HD. 4/3 is 1.3:1, which is the old standard definition NTSC aspect ratio (as well as VGA, etc). HD is 1.7:1 and then anamorphic widescreen is 2.35:1. Using part of the sensor, though, the 4/3 chip could deliver all three major aspect ratios.

If Olympus wanted to do a 4K 4/3 video camera it could really be amazing. The spec for 4K widescreen delivery (at 1.85:1) is 3996 × 2160. That's about 8.5 megapixels. The widescreen spec (at 2.39:1) is 4096 × 1714, which is about seven megapixels. And unlike still photography, most of the time that's going to be at 1/48th of a second. Even for television it's only usually 1/60th of a second. And ISOs over 400 are seldom heard of. So really, if Olympus delivered a camera that could record a clean ISO 320 at those resolutions it would be an indie filmmaker's dream. Documentary guys would be in heaven. If they could re-house the 14-35mm f/2 and the 35-100mm f/2 as geared manual-focus cinema lenses it would be an unbeatable camera system.
 
Seems a simple question by the OP (not really sure of his motives), seems to attract every blinkered fanboy around, with the biggest troll being Sergey

I can't understand why it's difficult for anyone to understand that f2 is always f2, that you multiply 4/3 lens focal lengths by 2 to equate to the same FOV as 35mm lens on a 35mm sensor, and that you also multiply the 4/3 aperture value by 2 to equate the 4/3 DOF to the equivalent 35mm aperture DOF. To get bigger DOF at a lower f value is a real benefit of 4/3 especially when using zoom lenses. Yes, there are times when you want paper thin DOF but that's rare and an acquired taste.

And that's why 4/3 is the way to go if you are shooting long lenses,..... and that's before you take the back breaking weight of 35mm sensor bodies and long lenses into account.

Introducing ISO values and noise into the discussion is just smoke intended to create FUD.

Rob

--
Give a wise man instruction and he is yet wiser !

Olympus E500, 14-45mm, 40-150mm, 11-22mm, 14-54mm
 

Keyboard shortcuts

Back
Top