The math of the 10D

[Halliday]

Ok, so the 10D changes my lenses by 1.6.
I also have a 1.4 Promaster extender.

Help me with the math, for example does it work like this:

50mm lens x 1.6 x1.4 = 122mm
or
50mm lens x (1.6+1.4) = 150mm

I would like to know because I try not have shutter speeds below the focal length, but I don't know how long my lens is ???

[CyberDyneSystems]

If just the lens,. it's lens times 1.6
or 100mm X 1.6 = 160mm

If it's the lens and T-con,. it's the lens X 1.4X = times 1.6
or 100mm X 1.4 = 140mm X 1.6 = 224mm

[tim]

This subject causes arguments every time it comes up. Personally i'm of the opinion that the 1.6X crop factor should be ignored in this, but I can't back up my opinion with facts. The 1/focal length is only a guideline anyway. The best thing to do is try it for yourself at a number of focal lengths and see what works for you.

[Halliday]

I realize if it's just the lens then 1.6, I just could not wrap my brain around how the adapter fit into the equation.

Testing.... good idea. Most of the times when I could use the adapter I forget I have it :|

[Steven M. Anthony]

This subject causes arguments every time it comes up. Personally i'm of the opinion that the 1.6X crop factor should be ignored in this, but I can't back up my opinion with facts. The 1/focal length is only a guideline anyway. The best thing to do is try it for yourself at a number of focal lengths and see what works for you.

Tim:

The facts that back up your opinion happen to be the laws of physics (as they apply to optics). Indeed, the 1.6 X-Factor is a crop factor, not a magnification factor. BTW, I like your pragmatic approach to the issue. I am amazed at how many people here ask questions that can be easily answered by a little experimentation.

[poke]

I have found that experimentation is far more beneficial to me than the formula. After experimenting a fair bit, I'm starting to get a gut feel for what will work in a given situation, and this seems to be alot quicker and more accurate than trying to calculate the forumla

It also could be just that I suck at maths :D

[Steven M. Anthony]

Well, I have a degree in statistics and used to teach math (as we say it in the USA), and I STILL prefer the old "try it and see" approach! :)

[griff2]

Well, I have a degree in statistics and used to teach math (as we say it in the USA), and I STILL prefer the old "try it and see" approach! :)
My first degree is maths (as we say in the UK) it doesn't make me any better a photographer though..;)

[tim]

I have a degree in technology, which is completely irrelevant, except that I can read the manual, understand it, and work the camera fine! ;) Taking good photos is the hard part... :rolleyes:

[Andy_T]

Halliday,

it's neither of the options given by you.
It's 50 x 1.6 x 1.4, but the result is 112 (according to my calculation :lol: )
This is in metric maths, of course. Don't know what it is in inches.

The good question is if you should really put a pro-master 1.4 extender on a 50 mm lens. It might not be optimal for image quality.

Take a look at this explanation with sample images (http://www.luminous-landscape.com/tutorials/understanding-series/dslr-mag.shtml), it's very helpful in understanding the X factor.

Best regards,
Andy

[Halliday]

I actually use the extender very little. Thinking about crop factor vs magnication factor does make sense. The camera only changes the view, not adding any magnification.

So I think that formula-wise it should be left out. Makes sense.

[kevin_c]

To get you head around this - The optical calculation should be done first (50 x 1.4) this gives you a 70mm lens (on a 35mm camera) BEFORE you even fit the lens to a DSLR.

When fitted to a 10D/20D etc this total (70mm) should then have the 1.6x multiplication factor applied to give the correct DSLR value.

If multiplying various figures it makes no difference what order you put the figures:
50x1.6x1.4 = 112
1.4x50x1.6 = 112
1.6x1.4x50 = 112

I knew my Maths would come in handy one day...

[griff2]

Ah, good old commutivity.

[griff2]

Indeed, the 1.6 X-Factor is a crop factor, not a magnification factor.

It's called a crop factor, but in fact it is a magnification factor since the cropped image, as captured by the smaller sensor, is magnified by 1.6 in order to be viewed at the same size as an image produced by a full-frame sensor; this is analogous to cropping a photograph, then enlarging it - the difference being that with the photograph, you lose resolution, but with light you don't. With a smaller sensor, you also get the added benefit that you're in the len's sweet spot.

[Jon]

It's called a crop factor, but in fact it is a magnification factor since the cropped image, as captured by the smaller sensor, is magnified by 1.6 in order to be viewed at the same size as an image produced by a full-frame sensor; this is analogous to cropping a photograph, then enlarging it - the difference being that with the photograph, you lose resolution, but with light you don't. With a smaller sensor, you also get the added benefit that you're in the len's sweet spot.

It's a crop factor. The "factor" we're referring to describes what happens to the image as captured. What you do with it in post-processing (enlarging, reducing, etc.) is irrelevant. It doesn't add anything that wasn't there in the original capture. Would you crop a 2 mm x 3 mm section out of a full-frame 35 mm slide taken with a 50 mm lens, (for a factor of 12) and say that you used a "600 mm lens"? Even after you blew it up 12x linear to get the same 8x10? And would you expect the same detail as if you'd used the real 600 mm lens to take the picture? Would you expect us to keep straight faces while you claimed that?

[griff2]

I'm not getting into an argument here:rolleyes: , having said that: "the crop factor" has nothing to do with post-processing; it's the difference between what a full-frame sensor "sees" and what a reduced-frame sensor "sees" that gives you the magnification factor or "crop factor". Crop factor is the name that has been given to this type of magnification.

[johneo]

This subject causes arguments every time it comes up. Personally i'm of the opinion that the 1.6X crop factor should be ignored in this, but I can't back up my opinion with facts. The 1/focal length is only a guideline anyway. The best thing to do is try it for yourself at a number of focal lengths and see what works for you.

(highlighted text by me)

Tim, I agree! As far as I'm concerned, my 10D with ANY lens takes photos at 1:1 for a 10D dSLR!
My Elan 7ne with ANY lens takes photos at 1:1 for a 7ne film SLR
I never could understand what the big deal is about this magnification or crop factor ... what comes out of my cameras (dSLR or SLR) is what it is!

[Titus213]

I really like the images from my 300D. I also like the images from my T70. But what I really like is that 'what I see is what I get'. More or less. Unless someone is going to bust my bubble of ignorance and tell me the camera isn't capturing what I see. In which case I may go find a cheap point and shoot because that's all I would be doing with my SLRs.

As too camera shake at longer focal lengths - if I see I'm getting it I either up the shutter speed or decrease the focal length. I have found that doing the math is at best an approximation. Depends on how much coffee and sugar I've consumed before heading out and I can never figure the math on that.

[jimsolt]

I am as far from a math major as you can get, but I do have an opinion about this discussion.
I think I get the point about the difference between a crop factor and a magnification factor, but I also don't think it makes any difference at all what you call it, it's what you see, but that's the whole point, isn't it?
No matter what you call it, when you look at the image you see with a 50mm lens on a 10D sensor it very closely approximates what you would see with an 80 mm lens on a 35mm or "full frame" sensor. Does anyone differ so far?
Now if you add the extender (1.4 magnification) to either camera you get approximately what you would see with a 112mm lens on a 35mm or "full frame" sensor.
If someone will lend me a few bucks I'll go out and buy the stuff necessary to prove that, but I bet it would come out close.
Meanwhile, the Hindenberg has burned and someone else got the shot:D
Jim

[Titus213]

...Meanwhile, the Hindenberg has burned and someone else got the shot:D
Jim

LOL!

[Kennymc]

CDS (first reply) got it right, the teleconverter magnifies the lenses focal length not the crop factor, then the crop factor of the censor comes into play...
50 x 1.4 = 70 ... lens (50) x TC (1.4) = 70
70 x 1.6 = 112 ... lens x TC (70) x CF (1.6) = 112
So a 50mm lens with a 1.4 TC would equal a 112mm lens...
OR...
50 x 1.6 = 80 (lens times crop factor) x 1.4 Teleconverter still equals 112mm...
But the TC actually increases the focal length of the lens nomatter if there is a crop factor or not...
How's that for confusing...

[roanjohn]

OH NO!!!!

A man bought a horse for $60 and sold it for $70. Then he bought the same horse back for $80 and again sold it for $90. How much money did he make in the horse business?


..........not another math problem!!!

Ro1

[Kennymc]

Not enough LOL....

[jimsolt]

CDS (first reply) got it right, the teleconverter magnifies the lenses focal length not the crop factor, then the crop factor of the censor comes into play...
How's that for confusing...

I know we speak different dialects of English, but I still like the "joke" made when you spell sensor differently.
In my version of English, if we had a CENSOR he/she would have censored this thread long ago and sent it to the "does anyone here know what we were talking about when this came up" thread.
I still think even if you were out buying a horse looking through an extender, the Hindenberg would have burned, etc.
Jim

[Kennymc]

Sorry about that but I'm dyslexic and make these terminological in exactitudes due to that reason… Who ever thought of a word like dyslexic for someone who can’t spell?..

[Steven M. Anthony]

Dyslexics of the world untie!!

[mandrillo]

Ok, so the 10D changes my lenses by 1.6.
I also have a 1.4 Promaster extender.

Help me with the math, for example does it work like this:

50mm lens x 1.6 x1.4 = 122mm
or
50mm lens x (1.6+1.4) = 150mm

I would like to know because I try not have shutter speeds below the focal length, but I don't know how long my lens is ???


Forget the maths... go out there and shoot, today with digital it's easier than it used to be when I started 15 years ago, but still, 15 years ago I tried to learn formulas and numbers, but trust me, you don't need mathematical formulas, all you need is your eye and your judgement... nothing works better than experimenting and making mistakes!

Cheers

Wahb

[ijohnson]

This is a really weird thread. Maybe its too early (GMT +9).

I say, if it was a Canon TC, just look at the EXIF info. No maths here.

[ron chappel]

This thread is funny

To be serious for a sec-
those doing the 'proper' multiplication of the various factors involved are right.It doesn't matter one bit wether it's called cropping factor or magnification factor,the result is still the same.
It's all about the resulting angular view and how much angular movement can happen in the time it takes the shutter to fire
So 50mm x1.6 x1.4 is right.

Of course those that say you should go out and experiment are kinda MORE right :) but the formula is handy as a starting point.


And the horse trader made $20 :)

[pradeep1]

You will get the image equivalent of 50 x 1.6 x 1.4 = 112, but you apply the hand-holdability rule to the 50 x 1.4 = 70, or 1/80 sec. Simple. :)

[griff2]

You will get the image equivalent of 50 x 1.6 x 1.4 = 112, but you apply the hand-holdability rule to the 50 x 1.4 = 70, or 1/80 sec. Simple. :)

This is not quite correct. For any given shutter speed, the amount of camera movement, and distance of the object you're focused on will determine how blurred that object may appear. When you bring that object closer via magnification - any magnification - the camera movement will also be magnified. E.g. if the object you focusing on is a tree two miles away, with 50mm lens, then, unless there's extreme camera movement, you won't see much blure; on the other hand, if you use a 400mm lens equivalent - which could be achieved by, say, any combination of lens, crop factor and teleconver - to focus in on that same tree, using the same shutter speed and aperture as with the 50mm lens, any camera movement will be multiplied by a factor of eight (length of an arc = rq).

So in a nutshell, the hand holdability rule applies to your effective focal length, since this determines the ultimate magnification factor.

[mkh]

I think the whole thing would be simpler if we forgot about crop and maginification and thought about it in terms of angle of view(AOV).

It takes an 80mm lens to on a 35mm camera to give the same AOV as the 50 will on the smaller digital censor. Since the image the lens transmits doesn't change (i.e a 50mm is always a 50mm) the smaller sensor is only capturing the center area of the image (AOV). For that same area to fill up a fullsize censor (or 35mm film) it would require an 80mm lens.

There really is no multiplication or larger immage. Just a difference in angle of view. .

For people with no 35mm background the whole thing is nonsense as they don't have a reference point to relate it to.

[Steven M. Anthony]

But the Angle of View (AOV) of a 50mm lens on a 10D is the same as the AOV of a 50mm lens on a camera with a full-frame sensor. It's just that on the 10D the sensor is not recording the entire AOV provided by the lens. There is NO optical magnification provided by a smaller (than full frame) sensor. So, 1/(optical focal length) works as a rule of thumb.

From what I know, teleconverters contain optical elements that alter the optical focal length of a lens. When using those, you need to enter the altered optical focal length into the formula--that is, if you can't tell by looking in the viewfinder that camera movement will botch you shot!

[griff2]

There really is no multiplication or larger immage. Just a difference in angle of view. .
To be honest, it's just a different way of explaining the same thing: one man's reduced angle of view is another man's magnification factor, either way you "lose" 37.5% of a full sensor image.


For people with no 35mm background the whole thing is nonsense as they don't have a reference point to relate it to.

True, but the problems arise because people pick up on the reciprocal rules which relay on focal length, and which are meaningless unless you know the true focal length of your camera/lens system. What I find odd is that the EF-S lenses are still marked as per a 35mm body even though they won't actually fit one!

[griff2]

But the Angle of View (AOV) of a 50mm lens on a 10D is the same as the AOV of a 50mm lens on a camera with a full-frame sensor. It's just that on the 10D the sensor is not recording the entire AOV provided by the lens. There is NO optical magnification provided by a smaller (than full frame) sensor. So, 1/(optical focal length) works as a rule of thumb. There isn't any optical magnification but there IS an effective magnification of 1.6, so the total effective focal length of the camera/lens system must be taken into account when using reciprocal rules - just because the camera's adding to the effective focal length of the system, doesn't mean you can ignore it.

[Steven M. Anthony]

There isn't any optical magnification but there IS an effective magnification of 1.6, so the total effective focal length of the camera/lens system must be taken into account when using reciprocal rules - just because the camera's adding to the effective focal length of the system, doesn't mean you can ignore it.

And what creates this magical non-optical, magnification "effect?" If it's not an optical effect, what kind of effect is it?

The smaller (that full) sensor found in many DSLRs simply DOES NOT ALTER THE FOCAL LENGTH OF A LENS. Similarly, it does not alter the "effective focal length" of a lens. All it does is reduce the Angle of View of a given focal length lens compared to the angle of view on a full-frame camera.

It's just like looking through a straw held in your fist. The straw doesn't alter the optics of your eye, or magnify anything in the scene. It just gives you less to look at.

[griff2]

And what creates this magical non-optical, magnification "effect?" If it's not an optical effect, what kind of effect is it?
I give up, what is the point, you either see it or you don't.

It's just like looking through a straw held in your fist. The straw doesn't alter the optics of your eye, or magnify anything in the scene. It just gives you less to look at.
And if you could view this straw image on screen, how would it look, what size would it be?

[Steven M. Anthony]

Don't give up. Describe how it works. I see how it does not alter the effective focal length. Show me how you think it does. By quitting the discussion, it seems like you just don't have an answer and don't want to admit it.

The size of "straw" image would depend on the resolution of the screen one viewed it on.

[rdenney]

Ok, so the 10D changes my lenses by 1.6.
I also have a 1.4 Promaster extender.

Help me with the math, for example does it work like this:

50mm lens x 1.6 x1.4 = 122mm
or
50mm lens x (1.6+1.4) = 150mm

I would like to know because I try not have shutter speeds below the focal length, but I don't know how long my lens is ???

Measure everything in terms of relationship to a "normal" lens, with normal being defined as the diagonal of the film or sensor frame. The normal lens for a 10D is a 28, just as a 43 is normal for full-frame 35mm, 75 is normal for 6x4.5 medium format, and 160 is normal for 4x5 large format.

Thus, with a 10D's 15x23 sensor, a 14mm lens is one-half normal, and a 56mm lens is twice normal. An 85 is three times normal, and a 135 is 4.5 times normal.

To get the same field of view with a full-frame camera, you'd need a 21 (half-normal), an 85 (twice normal), a 135 (three times normal), and a 200 (4.5 times normal).

Notice that in the above lists, the 85 and the 135 appear in both. But the 85 is a three-times-normal medium telephoto on a 10D, and a twice-normal short telephoto on a full-frame camera.

Lenses do not control perspective, they only control field of view. Camera position is ALL that controls perspective. Pick the position, and then select a lens that provides the field of view you need. If, for a given position, you need a half-normal wide-angle lens to sweep up the field of view you want, then you'll need a 14 for the 10D and a 21 for a full-frame camera to see the same thing, from that position. The two images will be identical in every way. Thus, a half-normal lens (in any format) provides the same field of view from a given position, and the given position provides the same perspective.

The 1.4 teleconverter just takes whatever you have and increases the focal length by a 1.4 multiplier. If you need a twice-normal lens (close to your 50), a 1.4 teleconverter will make it a 2.8-times-normal lens. Your 50 will be an 80. And an 80 is 2.8 times normal for a 10D.

In short, don't convert lens focal lengths--that is the path of confusion. Just compare them to the normal focal length for that format.

Believe me, this is ground well trod by multi-format photographers, and I regularly shoot in 15x23mm, 24x36mm, 6x4.5cm, 6x6cm, 6x9cm, and 4x5". How do I know that a 90mm lens on the 4x5 camera is like a 45 on 6x6, a 24 on 35mm, and a 15 on the 10D? Because they are all a little over half the normal focal length for that format.

Shutter speeds below the focal length works only on 35mm. It is conservative on larger formats, and not conservative on smaller formats. On the other hand, smaller formats don't tend to get enlarged as much, so there is more tolerance for minute camera shake. Thus, you can still use that rule of thumb, but there's no need to run it through any equivalence factor. Just use the raw focal length. If you want to be more conservative, cut the shutter speed in half (e.g., 1/500 instead of 1/250).

Rick "whose experience has never lined up very well with that rule of thumb" Denney

[griff2]

Don't give up. Describe how it works. I see how it does not alter the effective focal length. Show me how you think it does. By quitting the discussion, it seems like you just don't have an answer and don't want to admit it.


I haven't given up. I want an answer to the straw question - and "it depends on resolution" is not a satisfactory answer. Or maybe I should answer it myself. You would see a little tiny image on the screen with the straw obscuring most of the view around it. But hey, you don't see that effect with smaller sensor and EF lens, I wonder why?

[rdenney]

But the Angle of View (AOV) of a 50mm lens on a 10D is the same as the AOV of a 50mm lens on a camera with a full-frame sensor.

Angle of view is not very helpful. Consider instead "field of view". You pick your camera position based on the perspective you want--this thing is in the foreground and that thing is in the background. Then, you select the lens based on how much of the scene you want in the picture.

All lenses are related to the normal lens. From a given camera position (and therefore a give perspective), a 14 on a 10D has the same field of view as a 21 on full frame. Both are half the normal lens focal length for their respective formats. Angle of view doesn't matter, because it's not what interests a photographer in the field. As a photographer, I'm only interested only in the relationship of the foreground vs. the background (i.e., the perspective), and the amount of scene shown in the image (i.e., the field of view).

Rick "who shoots in five or six different formats and who is therefore confused by all this confusion" Denney

[Steven M. Anthony]

I haven't given up. I want an answer to the straw question. Or maybe I should answer it myself. You would see a little tiny image on the screen with the straw obscuring most of the view around it. But hey, you don't see that with smaller sensor and EF lens, I wonder why?

That's exactly what happens--and my point (which is why I broought up the analogy in the first place). That's essentially what happens in a 10D relative to a full-frame camera. While the smaller sensor does not obscure the view around it, it just isn't big enough to record the image that falls beyond it's edges. No magnification occurs. There is no change in effective focal length. Just a reduction in field of view. JUST LIKE AN EF LENS ON A 10D.

Sorry if I misunderstood you when you said "I give up." I thought that meant you were giving up...

I am, however, still waiting for your explanation of the non-optical magnification effect...

[Halliday]

Wow. I was just wondering about a math problem and started all this :| I was not lookig for a hard and fast rule, just a rule of thumb. I think I will follow the view to ignore the 1.6 magnification and just go with the lenses focal length.

Rock on!

[Steven M. Anthony]

Angle of view is not very helpful. Consider instead "field of view". You pick your camera position based on the perspective you want--this thing is in the foreground and that thing is in the background. Then, you select the lens based on how much of the scene you want in the picture.

All lenses are related to the normal lens. From a given camera position (and therefore a give perspective), a 14 on a 10D has the same field of view as a 21 on full frame. Both are half the normal lens focal length for their respective formats. Angle of view doesn't matter, because it's not what interests a photographer in the field. As a photographer, I'm only interested only in the relationship of the foreground vs. the background (i.e., the perspective), and the amount of scene shown in the image (i.e., the field of view).

Rick "who shoots in five or six different formats and who is therefore confused by all this confusion" Denney

I get that fov is what a photog thinks about when it comes to composition. But Angle of View is what is related to the focal length of a lens. And the original issue of the thread dealt with lens focal length and shutter speed.

[roanjohn]

OMG!!! Lets all get a 1Ds to end this discussion......my head hurts!!!

:-)

Ro1

[griff2]

I can see where you're coming from, re: FOV. Let me try to explain it from my perspective. Let's say you take a picture, hand held, with say a 50mm lens of a distant tree. You view it on your monitor at say 25% and it looks ok. You then zoom into the picture, to a 100% crop and you can see slight blurring caused by hand shake; that 100% crop is equivalent of using a 200mm lens - but with loss of resolution of course. My point is that whatever you want to call the effect of the crop, the camera and the lens effectively increase the focal length of the combined system, and this increases the susceptibility to hand shake by the same "crop factor", so the effective focal length must be used when using reciprocal rules. Ta da..

[mkh]

The real problem is not the 1.6 (or 1.3) crop/multiplication factor. The real problem is that the camera doesn't wrtie out 1.6% smaller images. :) If it would quit pretending to be a 35mm camera then we wouldn't have this problem. If we all used the same size monitors etc we would be OK but very boring.

For example you don't see point and shoot users worrying about this. And unless you shoot in several different formats (as someone here does) then it shouldn't be an isue. You don't see the larger format guys wasting time trying to figure it out.

Just select the lens that gives you the right FOV (better term than the AOV I suggested) and the right perspective. These cameras are not 35mm SLRs they are a new breed that is essentially a p&s camera in a 35mm like body. We have to start thinking deifferently

[pcasciola]

In Canon's current lineup, the photosites on a full frame camera are further apart than on a cropped sensor, so moving the cropped sensor a given amount (i.e. shake) produces more blur pixel for pixel than the same amount of shake on a full frame. If the pixel density of both sensors was the same, the crop would have no effect on shake, but that's just not the case in the real world.

[rdenney]

I get that fov is what a photog thinks about when it comes to composition. But Angle of View is what is related to the focal length of a lens. And the original issue of the thread dealt with lens focal length and shutter speed.

Perhaps, but that is not what most other respondents were talking about, and that's where the confusion is.

The problem with angle of view is that you also have to consider how much the image is going to be enlarged. Thus, tolerable camera shake is a lot like depth of field, it starts with your acceptable circle of confusion, and works backwards from there. It's more of a standard of acceptance than an arithmetic problem.

Thus, if your maximum circle of confusion is .015 mm on the sensor (which would be hard for the 10D sensor to see in any case), you'll be able to make about an 11X enlargement before the fuzziness inherent in that standard becomes routinely visible in a print. That is based on a standard of 6 lines/mm resolution in a print, which is my personal standard when it's possible. That's about a 6x9" print from a 10D. How much camera shake is that? That's where your angle comes in, but not "angle of view" but rather "angle subtended by .015mm of sensor". And that angle is controlled only by focal length, as you say.

Let's work it the other way. Let's say that the largest print I can tolerate from the 10D is 13x19. That is a 22x enlargement. I will have to tolerate no better than 3 lines/mm acutance in that print to stay within the capabilities of the sensor (and most of the lenses). My circle of confusion would be .015mm. For full-frame 35mm, the degree of enlargement is about 14x to get that 13x19 print. At the same standard (3 lines/mm in the print), my acceptable circle of confusion is .024mm. The angle subtended by that circle of confusion is (at these very small angles) reasonably approximated to 1.6 times larger than the angle subtended by the .015 circle of confusion standard using the smaller sensor (.024/.016 = 1.6). So, I arrive at the 1.6 factor without at all considering the focal length of the lens.

To get the same picture, though, I'll use a lens 1.6 times longer with the full-frame camera (camera position is the same, of course, otherwise the perspective would be different). Thus, the relationship between the focal length of the lens and the acceptable circle of confusion is the same across format, if you intend to make the same size prints of the same scene. If you work to a particular maximum degree of enlargement (say, 11x or 20x), then the 1/focal length = shutter speed rule of thumb only works for 35mm.

One thing any of this arithmetic immediately exposes, though, is that format is still king.

Rick "whose 13x19 prints from the 10D with state-of-the-art lenses look good only at a little distance, compared with medium format using cheap glass" Denney

[Steven M. Anthony]

I can see where you're coming from, re: FOV. Let me try to explain it from my perspective. Let's say you take a picture, hand held, with say a 50mm lens of a distant tree. You view it on your monitor at say 25% and it looks ok. You then zoom into the picture, to a 100% crop and you can see slight blurring caused by hand shake; that 100% crop is equivalent of using a 200mm lens - but with loss of resolution of course. My point is that whatever you want to call the effect of the crop, the camera and the lens effectively increase the focal length of the combined system, and this increases the susceptibility to hand shake by the same "crop factor", so the effective focal length must be used when using reciprocal rules. Ta da..

That's NOT equivalent to using a 200mm lens. Had you shot the same scene with a 200mm, with the same shutter speed, you would have MUCH more blurring due to camera shake than what you would see in the enlargment of the 50mm shot.

[Steven M. Anthony]

Perhaps, but that is not what most other respondents were talking about, and that's where the confusion is.

Well, I can't help it if others stray off topic... :)

...To get the same picture, though, I'll use a lens 1.6 times longer with the full-frame camera (camera position is the same, of course, otherwise the perspective would be different).

Or, one could use one's feet to get the same perspective... :)

[rdenney]

Or, one could use one's feet to get the same perspective... :)

Well, that is the point of my speechifying on the subject, heh, heh.

Perspective is controlled ONLY by camera position. NOT by format, NOT by focal length. You can evaluate perspective simply by using your eyes. Perspective is the relationship in size between foreground and background objects.

Consider this image:

http://www.rickdenney.com/images/american-falls-and-flowers-.jpg

(The wonder of this image is taking a picture of Niagra Falls amidst 40 million tourists on a crowded summer afternoon, and not including any of them in the picture, heh, heh. And it illustrates this point.)

If I move back, the flowers would be smaller with respect to the falls. Now, to get the flowers in that particular arrangement with respect to the distant falls, I have to stand in a particular spot. The film must be at Point X. Now, what I can include in the picture from that spot is a function of format and focal length. In this image, I used a 10D with a 12mm lens. I could also have used a 35mm camera with a 19mm lens, or a 4x5 view camera with a 65mm Super Angulon. But if I moved, it would be a different picture--the flowers would not be the same size or position with respect to the falls.

If I wanted to make this same picture on 35mm film using a 12mm lens, I would have to put the camera in the same spot, and then crop off the excess scenery to get the same field of view. If I wanted to make this same picture with a 10D and a 19mm lens, I would need to replace that 15x23 sensor and replace it with a 24x36 sensor (which I would LOVE to do, heh, heh).

I learned this watching a video of Ansel Adams giving photography lessons at Yosemite. He would walk around with a cardboard crop frame, and when he found the field of view that interested him, he would put a rock on the ground at that spot. Holding the cardboard frame closer was equivalent to a shorter focal length, and he would select the lens based on how much of the scene he wanted in the frame. His exact quote was, "Now, it's just a matter of having the right lens."

That's why I don't do shoe-leather zoom, heh, heh. It changes the perspective.

Okay, I'm hitting a 4-penny nail with a 4-pound sledge hammer, and I hope it's not your thumb that I hit, but when I left this forum a year ago, we were discussing these same things, and I see the same things being discussed now. Obviously, it confuses people!

Here's a better picture just so you won't think my best work is a typical panorama of Niagra Falls:

http://www.rickdenney.com/images/juneau-sunset-2-lores.jpg

Rick "who shot that one on medium format film with a 30mm fisheye" Denney

Rick "

[Steven M. Anthony]

Interesting, Rick. But it still doesn't make an APS-sized sensor magnify an image... :)

Great shots, too. Growing up near Niagara Falls I never thought of it as something to take pictures of! Funny how that works...

[rdenney]

Interesting, Rick. But it still doesn't make an APS-sized sensor magnify an image... :)

Great shots, too. Growing up near Niagara Falls I never thought of it as something to take pictures of! Funny how that works...

Hey, I wasn't getting in the middle of THAT argument. I was hoping to bypass it altogether with alternative explanations that would answer the underlying questions.

You must really hate tourists, heh, heh. After that one afternoon at the Falls, I hated them, and I was one!

Rick "who appreciates the kind words" Denney

[Steven M. Anthony]

I had never thought about the "perspective" perspective before. Most of my shots are made with family in tow--on vacation. I pretty much need to just get the best shot I can from wherever I happen to be! I guess that's why I like my 28 - 300 so much.

RE: Tourists

We have seen the enemy and it is us...

I don't mind tourists as a concept...! In fact, like you point out, you only see them when being a tourist yourself. The ones that bug me are the ones who walk in front of an obvious shot I'm trying to get.

[Jon]

I don't mind tourists as a concept...! In fact, like you point out, you only see them when being a tourist yourself.

Oh? You haven't lived until you've tried to ride the Washington DC Metrorail system during tourist season (and I use it daily to commute!). Why do they have to stand at the top and bottom of the escalator to re-form the groups that they just nose-counted immediately inside, then outside the train door while everyone else was trying to get off?

[griff2]

Steven M. Anthony wrote:
That's NOT equivalent to using a 200mm lens. Had you shot the same scene with a 200mm, with the same shutter speed, you would have MUCH more blurring due to camera shake than what you would see in the enlargment of the 50mm shot.
Lens bluring, due to motion shake is a function of distance from the lens: http://www.griff2.co.uk/pictures/length_of_arc.gif
As you can see, if hand shake is represented by rotation through an angle q, r1, would represent 50mm and r2 200mm, the length of arc at r2 is four time that at r1. In other words, for any given camera movement, the blur, due to camera movement, at 200mm will be four times that at 50mm, for any lens.

[Steven M. Anthony]

Griff: Not AT 50mm or AT 200mm--I said there would be more visable camera shake with a 200mm LENS versus a 50mm LENS (for a given shutter speed). That's the whole point of the 1/focal length rule of thumb... And I'm not sure camera shake IS represented by rotation through the angle of view.

[griff2]

The 1/focal length rule exists for the following reason: as your focal length increases, so does your susceptibility to blur from camera shake, see figure, it's not lens specific, it's just that the longer the lens is, the more you'll notice it. If you hade a super duper zoom which could go from 17 - 500 say, then you could conduct an experiment that would prove this. And I'm not sure camera shake IS represented by rotation through the angle of view.
Care to put forward a hypothesis?

[Steven M. Anthony]

The 1/focal length rule exists for the following reason: as your focal length increases, so does your susceptibility to blur from camera shake, see figure, it's not lens specific, it's just that the longer the lens is, the more you'll notice it.

Exactly my point, griff2. But what do you mean by it's not lens specific? My point was simply that the longer the focal length, the more suseptable to camera shake. Are you trying to make a different point?

As for my hypothesis on camera shake, I think it's a complex, compound motion. Lots of lateral motion parallel to the sensor plane, probably some rotational motion, too.

As for rotational motion, it also seems that where the pivot point for the rotation is relative to the sensor is important. Angular rotation pivoting at the sensor, for example, seems like it would create more motion blur for distant objects (like in your figure above--or is it on the previous page by now!). Angular rotation pivoting at the far end of the lens seems like it would create more motion blur for close objects. I don't know all the math involved for that one--maybe you do and can draw another figure.

Lateral motion produces quite different results--again, I don't have the math formula to back this up, but I do have a ruler and a photo with lots of lateral motion. Motion blur from lateral movement seems to be the same for distant and close objects. I can't post the photo in question, but you can look at it on my web site (see link below--go to the "emotions gallery;" it's the photo of a city street at dusk, done in portrait orientation. Note, the other night scene in that gallery was taken in a moving car--which adds another motion component...). You'll see the light streaks make the same pattern all over the image. And the length of the streaks is the same for streaks made by near-by lights as they are for streaks made by distant lights.

[slin100]

In response to the original question, yes, you multiply the actual focal length of the lens by 1.6 to get the effective focal length (efl), where "effective" is defined to be the focal length required on a 35mm film camera to capture the same field of view. You multiply the efl by 1.4 if a 1.4X teleconverter is added.

Regarding the 1/focal length rule, I personally use 1/efl. If you think about digital P&S which have crop factors of 5x or more and lenses with focal lengths in the teens, one should see that 1/real focal length won't produce reasonable results. So, while this is not a proof, I hope it's convincing enough.

[Steven M. Anthony]

Regarding the 1/focal length rule, I personally use 1/efl. If you think about digital P&S which have crop factors of 5x or more and lenses with focal lengths in the teens, one should see that 1/real focal length won't produce reasonable results. So, while this is not a proof, I hope it's convincing enough.

I think what that shows, is that the 1/focal length rule works for 35mm, but not all other formats. And you have 2 things going on with P&S cameras. One is the sensor size, like you mentioned. The other is lens construction. Without the need to provide space for a mirror to swing out of the way, a given Angle of View (i.e., one that fills the sensor) can be had with a smaller focal length than is needed for the same AoV on a 35mm format DSLR with a mirror.

[Jon]

You're both right and both wrong. The shake recorded on the film/sensor will be greater on a long focal length lens than on a short one. BUT when you enlarge an original the blur (from camera shake, circle of confusion, or any other source) is also enlarged. If you enlarge two originals by the same amount, the shorter f.l. lens will show less blur, the smaller sensor camera will have a smaller print size, etc. BUT when you enlarge two dissimilarly-sized originals to the same final size, the aberrations, blurs and so forth on the smaller one will be magnified more. This applies whether you're referring to an APS-format vs 35 mm with the "equivalent" lens, or a 50 mm lens picture cropped to correspond to the coverage of a 200 mm. lens, and comparably enlarged. Of course, when you increase the magnification of the tighter crop, you magnify all the blurs, distortions and aberrations that were recorded with it.

[griff2]

At the end of the day, if I mount a 200mm lens on my 300D, hand held, I can either use 1/200 or, taking effective focal length into account 1/320. Assuming I'm happy with my DOF, and there's enough light I'll go fo 1/320, why go for a slower speed when it could ruin the shot? Even if you're not convinced by the effective focal length argument, using the efl will not do any harm, but using lens focal length could lose you the shot.

[pcasciola]

Regarding the 1/focal length rule, I personally use 1/efl. If you think about digital P&S which have crop factors of 5x or more and lenses with focal lengths in the teens, one should see that 1/real focal length won't produce reasonable results. So, while this is not a proof, I hope it's convincing enough.And you are correct in doing so (1/effective focal length). It's extreme cases like this that prove the point that the crop factor has to be taken into account. My Coolpix 5700 had a 72mm lens with a 4x crop factor (280 effective FOV). If I used the 1/focal length rule (1/72), every shot was obviously very blurred. I had to shoot 1/250th or faster to get crisp shots.

[Jon]

Without the need to provide space for a mirror to swing out of the way, a given Angle of View (i.e., one that fills the sensor) can be had with a smaller focal length than is needed for the same AoV on a 35mm format DSLR with a mirror.

So a 28 mm lens on a Leica RF covers a larger area than a 28 mm lens on a Leica SLR? The Leica RF doesn't need a mirror. I don't think so.

Angle of view (which, BTW, is what Canon quotes)/field of view (which, as a linear measure is only meaningful when a range is also specified) is a function of lens focal length and sensor dimension. Retrofocus/telephoto lens designs have absolutely nothing to do with it.

[slin100]

I think what that shows, is that the 1/focal length rule works for 35mm, but not all other formats. And you have 2 things going on with P&S cameras. One is the sensor size, like you mentioned. The other is lens construction. Without the need to provide space for a mirror to swing out of the way, a given Angle of View (i.e., one that fills the sensor) can be had with a smaller focal length than is needed for the same AoV on a 35mm format DSLR with a mirror.
Steven,

I think you're way off base. I've seen no basis that lens construction matters. A P&S lens can be designed with a much shorter lens registration but that has no bearing on camera shake.

[rdenney]

So a 28 mm lens on a Leica RF covers a larger area than a 28 mm lens on a Leica SLR? The Leica RF doesn't need a mirror. I don't think so.

Angle of view (which, BTW, is what Canon quotes)/field of view (which, as a linear measure is only meaningful when a range is also specified) is a function of lens focal length and sensor dimension. Retrofocus/telephoto lens designs have absolutely nothing to do with it.

Yup. The light rays cross at the same point with a symmetrical 28 and a retrofocus 28. The only difference is where the glass sits with respect to that crossing point. The retrofocus design puts the glass mostly in front of the crossing point. Some lenses, like the Russian Jupiter 12 35mm rangefinder lens, puts most of the glass behind the crossing point, with the result that it won't fit in rangefinders than have a swing-away meter cell. Telephoto designs, of course, put the glass mostly behind the nodal point, to keep the lens barrel from being so long. But the effect is the same when it comes to camera shake.

Frankly, I don't find the 1/focal length rule to be accurate enough to support the level of precision being applied here. It assumes a particular standard of sharpness, which may be higher for you or lower for me, and it assumes that the ergonomics of holding the camera is the same for lenses of different focal lengths. I find I do better than 1/FL (or EFL) with short lenses that are easy to brace against my forehead, and not as well (in my shaky old age) with long lenses that are too light. Heavier cameras do better than lighter cameras, which is why I can REALLY cheat that rule of thumb using my five-pound Exakta 66 with the three-pound Sonnar lens, heh, heh. But I can't come close to good enough using that rule, even adjusted for the relationship to the normal focal length, using lightweight plastic lenses on the 10D. That, more than anything, was the reason I traded in the 75-300 for the 70-200/4L--the former could not be held or braced and I was too shaky.

Rick "who thinks developing your own rule of thumb through experience will yield better results" Denney

[slin100]

You're both right and both wrong.
Jon,

Can you point where I was wrong? I don't believe I said anything that contradicted what you said. In fact, I agree with you.

[Jon]

Jon,

Can you point where I was wrong? I don't believe I said anything that contradicted what you said. In fact, I agree with you.

Sorry. Not you - SMA and griff2, who've been hammering at each other for a couple of pages now.

[Steven M. Anthony]

Sorry. Not you - SMA and griff2, who've been hammering at each other for a couple of pages now.

Well, I hope griff2 doesn't feel hammered by my comments. I certainly don't feel hammered by his. I thought we were having an interesting discussion on focal length and camera shake. True, we don't agree (yet) on a position. But I've learned a lot from his drawing and comments. Can't people discuss an issue from 2 points of view without others seeing it as some sort of battle or "hammering?"

[Steven M. Anthony]

Steven,

I think you're way off base. I've seen no basis that lens construction matters. A P&S lens can be designed with a much shorter lens registration but that has no bearing on camera shake.

No bearing on camera shake--but that's why you can't use the 1/focal length formula for P&S cameras. Converting the P&S lens fl to 35mm gives a good rule of thumb. And like griff2 pointed out, there's no harm if it's over-conservative. But what you are accounting for there is the equivalent angle of view, not sensor crop factor.

Pcasciola mentioned his Coolpix 5700 "had a 72mm lens with a 4x crop factor (280 effective FOV)." I don't know the innards of that camera, but my guess is that the image circle created by its lens is not "cropped" by the sensor--like it is in a 10D. Rather, the image circle likely fills the sensor--the lens is just constructed in a way that creates a smaller image circle that fills the 2/3" CCD. Because of the way the lens is constructed, its 72mm focal length has the AoV of a 280mm lens on a 35mm format camera.

So why not apply the "factor" to the 10D? Because the lens system is already a 35mm format system. It's just that the sensor captures less of the image circle created by that system.

Of course, when enlarging the image one enlarges the camera shake. But the 10D camera/lens system does not magnify camera shake more-so than a full-frame sensor (of equal pixel density) would in the same camera.

[slin100]

Pcasciola mentioned his Coolpix 5700 "had a 72mm lens with a 4x crop factor (280 effective FOV)." I don't know the innards of that camera, but my guess is that the image circle created by its lens is not "cropped" by the sensor--like it is in a 10D. Rather, the image circle likely fills the sensor--the lens is just constructed in a way that creates a smaller image circle that fills the 2/3" CCD. Because of the way the lens is constructed, its 72mm focal length has the AoV of a 280mm lens on a 35mm format camera.

Respectfully, I don't understand your point. I don't see how whether or not the image circle of a lens is cropped has any bearing on camera shake.

So why not apply the "factor" to the 10D? Because the lens system is already a 35mm format system. It's just that the sensor captures less of the image circle created by that system.

I believe your assertion falls apart if you consider an EF-S lens, where the image circle is not cropped.

Of course, when enlarging the image one enlarges the camera shake. But the 10D camera/lens system does not magnify camera shake more-so than a full-frame sensor (of equal pixel density) would in the same camera.
True, the 10D/lens system does not inherently magnify camera shake, but one must compenstate for the extra magnification of the final print, compared to a comparable image taken from a 35mm system. You must consider the final print magnification when selecting a shutter speed.

As far as the absoluteness of the 1/efl rule of thumb, I agree that it should be tailored to one's abilities. The point is that whatever rule of thumb works for an individual on a 35mm system should be compensated by the sensor size. If 1/50 works for you on a 35mm system, then 1/80 will produce similar results on a 10D.

[Steven M. Anthony]

Respectfully, I don't understand your point. I don't see how whether or not the image circle of a lens is cropped has any bearing on camera shake.

It doesn't. It has bearing on THE USE OF THE RULE OF THUMB FORMULA for avoiding camera shake.

I believe your assertion falls apart if you consider an EF-S lens, where the image circle is not cropped.

The 10D does not accept EF-S lenses.

True, the 10D/lens system does not inherently magnify camera shake, but one must compenstate for the extra magnification of the final print, compared to a comparable image taken from a 35mm system. You must consider the final print magnification when selecting a shutter speed.

But say you took 2 photos using a 50mm lens, from the same spot, 1 with a 10D and 1 with a full-frame digital with equal pixel density. If you take the the respective files and print a 16 X 24" print of each, the one from the 10D will look "magnified" compared to the other. But if you crop the full frame shot to have the same content as that of the 10D image, and print THAT to 16 X 24", it will look JUST LIKE the image from the 10D. If you still don't see what I'm talking about, check out the link below for a great treatment of the topic.

www.luminous-landscape.com/tutorials/understanding-series/dslr-mag.shtml

As far as the absoluteness of the 1/efl rule of thumb, I agree that it should be tailored to one's abilities. The point is that whatever rule of thumb works for an individual on a 35mm system should be compensated by the sensor size. If 1/50 works for you on a 35mm system, then 1/80 will produce similar results on a 10D.

But the 10D IS a 35mm system--it just has a smaller format sensor--so 1/50 will work on the 10D as well as it works on any 35mm system...

[pcasciola]

As far as the absoluteness of the 1/efl rule of thumb, I agree that it should be tailored to one's abilities. The point is that whatever rule of thumb works for an individual on a 35mm system should be compensated by the sensor size. If 1/50 works for you on a 35mm system, then 1/80 will produce similar results on a 10D.I completely agree.

[Kennymc]

There is another factor to consider as well, some people have steadier hands than others... IMHO if your images are blurred due to camera shake use a faster shutter speed... Experiment with all your lenses and take note at what shutter speed your images show signs of blur then always use a faster speed... General rules are that, just general, they don't fit everyone... There is no good saying you should use 1/60 for a 50mm in 35mm format if you can't hold the camera steady, then again you might be able to hold it steady at 1/15, only experimentation will determine what speed you need to hold a particular lens steady...

[Titus213]

Wow! My head really does hurt. I think I'll go take some pictures.

[slin100]

It doesn't. It has bearing on THE USE OF THE RULE OF THUMB FORMULA for avoiding camera shake.

That's what I meant.

The 10D does not accept EF-S lenses.

Sigh. Ok, let's switch to the 300D, which does. So, what's your point? My point was that your assertion that a P&S lens that has an image circle optimized to cover the sensor somehow makes the rule of thumb different holds no water. And my counterexample was EF-S lenses. Obviously, the image circle is irrelevant. Therefore, P&S cameras are subject to the same rule of thumb: 1/efl.

But say you took 2 photos using a 50mm lens, from the same spot, 1 with a 10D and 1 with a full-frame digital with equal pixel density. If you take the the respective files and print a 16 X 24" print of each, the one from the 10D will look "magnified" compared to the other. But if you crop the full frame shot to have the same content as that of the 10D image, and print THAT to 16 X 24", it will look JUST LIKE the image from the 10D. If you still don't see what I'm talking about, check out the link below for a great treatment of the topic.

I do understand this point. It's irrelevant as far as the rule of thumb is concerned. The following exercise illustrates my point.

Take 2 photos, one with a 50mm lens on a 10D and one with a 80mm lens on a film camera from the same spot. If you take the respective files and print a 16"x24" print of each, the one from the 10D will exhibit more shake unless you shorten the shutter speed by 1.6x.

But the 10D IS a 35mm system--it just has a smaller format sensor--so 1/50 will work on the 10D as well as it works on any 35mm system...The 10D is a 22.7mm x 15.1mm system that happens to take 35mm lenses. There's a huge difference.

[slin100]

I completely agree.Thanks, Phil.

[slin100]

Wow! My head really does hurt. I think I'll go take some pictures.But, but, what shutter speed will you use? :)

[Steven M. Anthony]

That's what I meant.

The bearing it has on the use of the formula (sigh) is what you use for the denominator...

Sigh. Ok, let's switch to the 300D, which does. So, what's your point? My point was that your assertion that a P&S lens that has an image circle optimized to cover the sensor somehow makes the rule of thumb different holds no water. And my counterexample was EF-S lenses. Obviously, the image circle is irrelevant. Therefore, P&S cameras are subject to the same rule of thumb: 1/efl.

My point is that P&S camera/lens systems are designed differently than DSLR systems that use an APS-sized sensor in a camera built for a larger format--and that the 1/fl rule of thumb was designed with the 35mm format in mind. That is, the 1/fl rule of thumb WORKS for 35mm format systems, but doesn't work for other formats. SO, if you are using another format, and want to use the 1/fl rule of thumb, you FIRST have to convert your non-35mm-format lens focal length to the 35mm format equivalent, THEN use the rule of 1/efl. One need not do this when using a 10D because the lenses that fit that camera ARE ALREADY 35mm-format lenses! (Read the luminous landscape article--it should clear things up for you.)



I do understand this point. It's irrelevant as far as the rule of thumb is concerned. The following exercise illustrates my point.

Take 2 photos, one with a 50mm lens on a 10D and one with a 80mm lens on a film camera from the same spot. If you take the respective files and print a 16"x24" print of each, the one from the 10D will exhibit more shake unless you shorten the shutter speed by 1.6x.
The 10D is a 22.7mm x 15.1mm system that happens to take 35mm lenses. There's a huge difference.

Nope, you have it backwards. If take your shots at the same shutter speed, and use the 1/fl rule based on the 10D's 50mm (not converting it, because the 50mm is ALREADY a 35mm format lens), you'd be shooting both at about 1/50--say 1/60th. That's slower than the rule of thumb would suggest for an 80mm on a film camera. Now do you see?

[pcasciola]

My point is that P&S camera/lens systems are designed differently than DSLR systems that use an APS-sized sensor in a camera built for a larger format--and that the 1/fl rule of thumb was designed with the 35mm format in mind. That is, the 1/fl rule of thumb WORKS for 35mm format systems, but doesn't work for other formats. SO, if you are using another format, and want to use the 1/fl rule of thumb, you FIRST have to convert your non-35mm-format lens focal length to the 35mm format equivalent, THEN use the rule of 1/efl. One need not do this when using a 10D because the lenses that fit that camera ARE ALREADY 35mm-format lenses! (Read the luminous landscape article--it should clear things up for you.)So, by your own admission, since any system smaller than 35mm has to be adjusted to 1/efl, EF-S lenses do in fact have to be converted, because then the entire system is APS sized, not 35mm. EF-S lenses do NOT fill a 35mm frame, and only shares the EF mounting system so these camera's can use both sets of lenses. That's no different than a P&S camera, just larger. But, if EF-S lenses have to be adjusted for because the entire system is smaller then 35mm, then why not a smaller system that is just "borrowing" a larger 35mm lens and using the center part of it?

[Steven M. Anthony]

So, by your own admission, since any system smaller than 35mm has to be adjusted to 1/efl, EF-S lenses do in fact have to be converted, because then the entire system is APS sized, not 35mm.

Right! And it's ANY (not just smaller) format that is not 35mm that has to be converted before the 1/fl rule makes sense (in a general, rule of thumb way).


EF-S lenses do NOT fill a 35mm frame, and only shares the EF mounting system so these camera's can use both sets of lenses.

Some of "these cameras" can use them--not the 10D. And to my knowledge, they won't work on any Canon 35mm film cameras.

That's no different than a P&S camera, just larger.

Right. Each has lenses designed for the sensor size used by the camera. Someone, earlier, brought up how odd it was that Canon converts the focal lengeth to the 35mm-format equivalent. Maybe it's so people don't over-convert...


But, if EF-S lenses have to be adjusted for because the entire system is smaller then 35mm, then why not a smaller system that is just "borrowing" a larger 35mm lens and using the center part of it?

Because EF lenses are designed for 35mm-format cameras and the 300D and 10D are based on a 35mm platform. So, with a 35mm lens on a 35mm camera, no conversion is needed. But when you put an EF-S lens on, you should convert before calculating the 1/fl rule.

[pcasciola]

Canon converts the focal length on EF-S lenses? That's news to me.

My point is, I think your argument does not hold water, because if an EF-S lens at 50mm on say a 20D has to be figured as 1/efl (1/80th), then you have to apply the same rule when using a normal EF 50/1.8 on the same 20D, no?

BTW, EF-S lenses DO work on film cameras with a small modification. I've seen shots taken with the EF-S on film as well as 10D's.

I keep saying I am not going to get involved in these crop factor discussions, but I am inevitably sucked in for some reason. They seem more like arguments about the density of our skulls. :D

[Steven M. Anthony]

Canon converts the focal length on EF-S lenses? That's news to me.

My point is, I think your argument does not hold water, because if an EF-S lens at 50mm on say a 20D has to be figured as 1/efl (1/80th), then you have to apply the same rule when using a normal EF 50/1.8 on the same 20D, no?

BTW, EF-S lenses DO work on film cameras with a small modification. I've seen shots taken with the EF-S on film as well as 10D's.

I keep saying I am not going to get involved in these crop factor discussions, but I am inevitably sucked in for some reason. They seem more like arguments about the density of our skulls. :D

Well, maybe I just don't know enough about EF-S lenses. My understanding is that they don't work on a 10D. I was working on the assumption that they were designed to fill the sensor on a 300d/20d the way an EF lens fills a full frame. Maybe that is a false assumption.

So if the EF-S lenses are designed to "fit" the aps-sized sensor, then you need to convert the fl before using the rule of thumb. If they are just constructed in such a way as to not operate on some 35mm cameras due to mirror size (my understanding is that the read element of the EF-S lenses sticks too far into the camera body for the 10D's mirror), and are designed to 35mm-format specs otherwise, then you DO NOT need to convert the fl before applying the rule of thumb.

[pcasciola]

Well, maybe I just don't know enough about EF-S lenses. My understanding is that they don't work on a 10D. I was working on the assumption that they were designed to fill the sensor on a 300d/20d the way an EF lens fills a full frame. Maybe that is a false assumption.

So if the EF-S lenses are designed to "fit" the aps-sized sensor, then you need to convert the fl before using the rule of thumb. If they are just constructed in such a way as to not operate on some 35mm cameras due to mirror size (my understanding is that the read element of the EF-S lenses sticks too far into the camera body for the 10D's mirror), and are designed to 35mm-format specs otherwise, then you DO NOT need to convert the fl before applying the rule of thumb.Well, you're mostly right. EF-S lenses do not work on the 10D "as is" and need to be heavily modified to fit, but modified they do work, so why would that change the 1/efl rule?

What about the 50/1.8 on the 20D compared to the EF-S at 50mm on the 20D. Are you saying I have to shoot the EF-S at 1/80th, but I can shoot the EF 50/1.8 at 1/50 even though the FOV is the same?

Boy, I really wish we all just had full frame sensors.

[Steven M. Anthony]

Well, you're mostly right. EF-S lenses do not work on the 10D "as is" and need to be heavily modified to fit, but modified they do work, so why would that change the 1/efl rule?

What about the 50/1.8 on the 20D compared to the EF-S at 50mm on the 20D. Are you saying I have to shoot the EF-S at 1/80th, but I can shoot the EF 50/1.8 at 1/50 even though the FOV is the same?

Boy, I really wish we all just had full frame sensors.

I thought I just said I don't know enough about EF-S lenses to have a definite answer for you...

I also laid out 2 scenarios that should cover the possible scenarios of the EF-S lens construction: If they are essentially 35mm-format lenses, you do NOT need to convert before using the 1/fl rule. If they are NOT 35mm-format lenses, you DO need to convert. Is there another scenario I'm missing?

You can apply these contingent statements to your question about the 50mm EF vs. the 50mm EF-S.

[slin100]

I thought I just said I don't know enough about EF-S lenses to have a definite answer for you...

I also laid out 2 scenarios that should cover the possible scenarios of the EF-S lens construction: If they are essentially 35mm-format lenses, you do NOT need to convert before using the 1/fl rule. If they are NOT 35mm-format lenses, you DO need to convert. Is there another scenario I'm missing?

You can apply these contingent statements to your question about the 50mm EF vs. the 50mm EF-S.
EF-S lenses are not 35mm format lenses. They are designed to only cover an APS-C sensor. So, what law of physics makes it necessary to apply a conversion for EF-S lenses but not for EF lenses? Answering with, "Because the EF lens is a 35mm format lens," is not sufficient. You must explain to us how an EF-S with an image circle optimized to cover an APS-C sensor somehow magnifies shake when an EF lens does not.

I posit that no such law exists, and that the conversion must be applied to both EF and EF-S lenses. If it were somehow possible to do a controlled test, say with a 18-55 EF-S lens and a 50mm EF lens, you would see this is true.

[Jim_T]

EF-S lenses are also prone to the 1.6 crop so they can't just cover the APS-C sized sensor.
The 18mm - 55mm of the Rebel kit lens still converts to a 28.8mm - 88mm field of vew. If the lens were designed specifically for the smaller sensor, then there would be no crop and an 18mm EF-S would provide the field of view of an 18mm lens.

The only difference between an EF-S and an EF lens is they set the back element closer to the sensor (or more correctly the image plane), in the EF-S lens. This allows for a smaller cheaper lens.. Especially for a cheap wide angle lens.

Because the back element is set further back, it won't work on a standard EF mount camera. The mirror of an EF mount camera will strike the back of an EF-S lens.

Read the first paragraph in this link where they talk about the 1.6 crop. There would be no 1.6 crop if the lenses focused exactly on the 10D/20D/Rebel sensor.

http://photo.net/equipment/canon/efs/efs_lenses.html

[pcasciola]

EF-S lenses are also prone to the 1.6 crop so they can't just cover the APS-C sized sensor.If this is true, how come pictures taken with the 18-55 EF-S on a film camera are round?

[Steven M. Anthony]

EF-S lenses are not 35mm format lenses. They are designed to only cover an APS-C sensor. So, what law of physics makes it necessary to apply a conversion for EF-S lenses but not for EF lenses? Answering with, "Because the EF lens is a 35mm format lens," is not sufficient. You must explain to us how an EF-S with an image circle optimized to cover an APS-C sensor somehow magnifies shake when an EF lens does not.

I posit that no such law exists, and that the conversion must be applied to both EF and EF-S lenses. If it were somehow possible to do a controlled test, say with a 18-55 EF-S lens and a 50mm EF lens, you would see this is true.

You've lost me know. NONE of the things we've been talking about magnify camera shake.

To summarize: The 1/fl rule of thumb works well for 35mm format. Why? Who knows (or cares)? It just does. Maybe it's evidence of "intelligent design." If you use it, and have relatively stable hands, your shots will come out without a lot of camera shake.

If the camera/lens system IS NOT 35mm, using the 1/fl rule leads to either relatively "shakier" results, or crisper results, depending on which way the conversion goes. Example: Use a P&S with a 17mm lens and shoot at 1/17th, and you'll end up with shaky photos. Convert that 17mm to its 35mm-format equivalent and THEN apply the 1/(e)fl rule and you should be okay. If EF-S lenses are truly NOT 35mm format, then you should apply the relevant conversion factor before using the 1/fl rule. Since the 28 - 300mm I use on my 10D IS a 35mm-format lens, I don't do any conversion--and my shots come out great.

[slin100]

You've lost me know. NONE of the things we've been talking about magnify camera shake.

But that's what you're intimating. You said that an EF-S lens must be converted and an EF lens need not be. That suggests using an EF-S makes camera shake more evident and that must be countered by shorter shutter speeds.

To summarize: The 1/fl rule of thumb works well for 35mm format. Why? Who knows (or cares)? It just does. Maybe it's evidence of "intelligent design." If you use it, and have relatively stable hands, your shots will come out without a lot of camera shake.

If the camera/lens system IS NOT 35mm, using the 1/fl rule leads to either relatively "shakier" results, or crisper results, depending on which way the conversion goes. Example: Use a P&S with a 17mm lens and shoot at 1/17th, and you'll end up with shaky photos. Convert that 17mm to its 35mm-format equivalent and THEN apply the 1/(e)fl rule and you should be okay. If EF-S lenses are truly NOT 35mm format, then you should apply the relevant conversion factor before using the 1/fl rule. Since the 28 - 300mm I use on my 10D IS a 35mm-format lens, I don't do any conversion--and my shots come out great.
I'm sorry, Steven, but I've run out of ways to explain myself. I'll just end with this. A 10D with an EF lens is not a 35mm system. The sensor defines the format, not the lens. The focal length on the lens must be converted in order to apply the 1/fl rule of thumb. I'm, frankly, dumbfounded that you believe that a 300D with an EF-S lens at 50mm would be subject to different rules than a 10D with an EF lens at 50mm. You don't even have to understand EF-S lens designs to know that this can't be the case. The image will look identical in each camera's viewfinder. Therefore, they must adhere to the same rules.

That you get great results without applying any conversion only means you are steadier than the rule of thumb suggests. That suggests that if you used the same lens on a film camera that you could probably achieve the same results at shutterspeeds 1.6x longer than 1/fl.

[pcasciola]

I give up. I just don't see why a "digital" lens at 200mm that produces a 25-27mm image circle would have to be adjusted for, but not a 200mm lens that covers a 35mm image circle, while both will have nearly identical FOVs on a 10D.

[Steven M. Anthony]

But that's what you're intimating. You said that an EF-S lens must be converted and an EF lens need not be. That suggests using an EF-S makes camera shake more evident and that must be countered by shorter shutter speeds.

I said IF THE EF-S series IS NOT 35mm format, it needs to be converted before applying the 1/fl rule... If they are indeed NOT 35mm format, you need to apply the conversion just like you need to for P&S digicams.

I'm sorry, Steven, but I've run out of ways to explain myself. I'll just end with this. A 10D with an EF lens is not a 35mm system. The sensor defines the format, not the lens.

Now I see your error. What defines the format is NOT the sensor size. What defines the format is the size of the image circle the system produces. With your logic, putting a 10D's sensor in a 8X10 format camera converts it to an asp-c format camera. That's just not how it works. The properties of the 35mm format system do not magically change when you pop in a smaller or larger sensor. The 10D's sensor records a fraction of the image a full-frame sensor would record. In effect, the 10D's sensor produces a "cropped" version of the full-frame--hence the term "crop factor."[/QUOTE]

[slin100]

I said IF THE EF-S series IS NOT 35mm format, it needs to be converted before applying the 1/fl rule... If they are indeed NOT 35mm format, you need to apply the conversion just like you need to for P&S digicams.

I already said that an EF-S lens is not a 35mm format lens.

Now I see your error. What defines the format is NOT the sensor size. What defines the format is the size of the image circle the system produces. With your logic, putting a 10D's sensor in a 8X10 format camera converts it to an asp-c format camera. That's just not how it works. The properties of the 35mm format system do not magically change when you pop in a smaller or larger sensor. The 10D's sensor records a fraction of the image a full-frame sensor would record. In effect, the 10D's sensor produces a "cropped" version of the full-frame--hence the term "crop factor."
Sorry, the size of the image circle does not define the format. That is your error, but it seems that neither I nor anyone else can convince you.

[Steven M. Anthony]

I already said that an EF-S lens is not a 35mm format lens.

Then what didn't you understand in my response?

Sorry, the size of the image circle does not define the format. That is your error, but it seems that neither I nor anyone else can convince you.

So putting a roll of 35mm film in an 8X10 camera makes it a 35mm camera?! I'd like to see the math on that one! Really. The reason no one can convince me that the sensor size defines the properties and characteristics of a lens is because it's simply not true.

[KJGradwell]

Halliday,

it's neither of the options given by you.
It's 50 x 1.6 x 1.4, but the result is 112 (according to my calculation :lol: )
This is in metric maths, of course. Don't know what it is in inches.

The good question is if you should really put a pro-master 1.4 extender on a 50 mm lens. It might not be optimal for image quality.

Take a look at this explanation with sample images (http://www.luminous-landscape.com/tutorials/understanding-series/dslr-mag.shtml), it's very helpful in understanding the X factor.

Best regards,
Andy

Is this a fair summary so far?

35mm: 36mmX 24 DSLR 22.7X15 ((36x24)/(22.7X15))1/2 = 1.593.
The image (DSLR) in the focal plane of the lens is cropped by this factor.
The uncropped image ( on the sensor ) is the same size in each case.
An object in the uncropped image (DSLR) occupies a larger area of the field.
Conseqently gving the same "view" as a lens with a FL 1.6 x larger.
In order to print at the same size (say 6X4 ) the 35mm image is magnified 4.25 times and the DSLR image by 6.8 times.
So on the print the digital image is magnified by 1.6 times.
WYSIWYG holds in all cases.
Questions

I've (very recently) returned to photography after an abscence of 20 years with a 300D. I have had no reasons to compare what i get noe to what I got then so this discussion is a bit academic? Yes? No?
One issue that the discussion does raise is wide angle lenses the 28 - 55 mm lens that came with my camera is 45 to 88 in old money is it not? I suppose I need to look for an 18mm lens for my Camera? Has anyone any recommendations or comments as to where I might start please?

[griff2]

I can answer your last question first. If money's tight you could get the kit lens: the 18-55mm, it's ok, but a bit soft even in the "sweet spot", with more money to spare, you could get the 17-40 f/4L; I've just purchased this lens to replace the 18-50mm kit lens and am expecting great things from it:D

The 1.6 "crop factor" will multiply the focal length of any EF(-S) lens by 1.6, so (now for the contentious bit) you should take this into account when applying any reciprocal (1/focal length) rules.

I won't make any more comments on you're bulleted list since it will, no doubt, open up a whole new can of worms, and I want to spend some quality time with my new 17-40 f/4L:mrgreen:

[rdenney]

The bearing it has on the use of the formula (sigh) is what you use for the denominator...



Steven, the problem I have with your way of explaining things is that it really doesn't matter what the lens was designed for.

Let's say, for example, that I mount an adaptor on my 10D, and a Zeiss 50mm wide-angle lens from an Exakta 66 (which is a medium-format camera). How would that be different, from a camera shake perspective, than using a Canon 50mm lens? The medium-format 50 has an image circle twice as big as the Canon lens, but it will behave in exactly the same way. The only reason that Zeiss 50 is a wide-angle lens is because it is usually used to light up a 6x6 frame.

Or, what if I take that APS-sized sensor and install it in the back of my Pentax 645, and then set my 45-85 zoom on 50? It will still behave exactly the same way as a 10D with a Canon 50, even though it's in a camera "designed for 6x4.5".

Now, if I use that medium-format 50 to light up a medium-format frame, then I won't enlarge it as much, because the frame is bigger. But if you ignore that, then I have to worry about camera shake as much with the 50 when it's used as a wide-angle as I do with a 50 when it's used as a telephoto (as it is with the 10D). If I enlarged that 50mm wide-angle 6x4.5 frame to 24x30" (a roughly 14x enlargement), and look at the print from six inches, then camera shake will affect the apparent sharpness exactly the same as a 10D image, made with a 50, enlarged to 8x10" (also 14x) and viewed from six inches. The point is, though, that you can't ignore the degree of enlargement. Thus, the smaller sensor magnifies camera movement because its images are enlarged more.

(As an aside, you could take that 8x10 print from the 10D/50 and lay it on the 24x30 print from the 645/50, and the image on the edge of the 8x10 would line up with the image on the 24x30 exactly. That's where the notion of "cropping" comes in.)

When you introduce the concept of "35mm lenses in a body designed for 35mm camera", you cause unnecessary confusion. There is focal length and degree of enlargement (which brings in format). Nothing else matters when it comes to camera shake. The only reason the smaller sensor is any different is because we enlarge the resulting image more. The angle subtended by the edge of the frame isn't important until you decide how big a print you need. The angle subtended by a given camera movement is important only when you decide your degree of enlargement. The different between the two (degree of enlargement and print size) is format, and that's where the equivalency factor comes into play.

Rick "who uses medium-format lenses on his 10D all the time and treats them just the same, except they are heavier and easier to hold" Denney

[Steven M. Anthony]

Steven, the problem I have with your way of explaining things is that it really doesn't matter what the lens was designed for.

Doesn't matter with regards to camera shake? That's exactly my point!!!

Let's say, for example, that I mount an adaptor on my 10D, and a Zeiss 50mm wide-angle lens from an Exakta 66 (which is a medium-format camera). How would that be different, from a camera shake perspective, than using a Canon 50mm lens? The medium-format 50 has an image circle twice as big as the Canon lens, but it will behave in exactly the same way. The only reason that Zeiss 50 is a wide-angle lens is because it is usually used to light up a 6x6 frame.

YES!!! That's why one need NOT take the crop factor into account when using the 1/fl rule!!! The CROP FACTOR has NO impact on CAMERA SHAKE.

Now, if I use that medium-format 50 to light up a medium-format frame, then I won't enlarge it as much, because the frame is bigger. But if you ignore that, then I have to worry about camera shake as much with the 50 when it's used as a wide-angle as I do with a 50 when it's used as a telephoto (as it is with the 10D).

Yes, yes, a thousand times yes!!!!!!!!!!!!!! And this is why one need NOT take the crop factor into account when using the 1/fl rule!!! The CROP FACTOR has NO impact on CAMERA SHAKE.


If I enlarged that 50mm wide-angle 6x4.5 frame to 24x30" (a roughly 14x enlargement), and look at the print from six inches, then camera shake will affect the apparent sharpness exactly the same as a 10D image, made with a 50, enlarged to 8x10" (also 14x) and viewed from six inches. The point is, though, that you can't ignore the degree of enlargement. Thus, the smaller sensor magnifies camera movement because its images are enlarged more.

Agreed!!! But that doesn't mean that the crop factor maginfies camera shake--it means the enlargment process magnifies camera shake, right?

(As an aside, you could take that 8x10 print from the 10D/50 and lay it on the 24x30 print from the 645/50, and the image on the edge of the 8x10 would line up with the image on the 24x30 exactly. That's where the notion of "cropping" comes in.)

YES!!!! And I think THAT's what others are missing here.

[griff2]

rdenney wrote: If I enlarged that 50mm wide-angle 6x4.5 frame to 24x30" (a roughly 14x enlargement), and look at the print from six inches, then camera shake will affect the apparent sharpness exactly the same as a 10D image, made with a 50, enlarged to 8x10" (also 14x) and viewed from six inches. The point is, though, that you can't ignore the degree of enlargement. Thus, the smaller sensor magnifies camera movement because its images are enlarged more.

Steven M. Anthony wrote: Agreed!!! But that doesn't mean that the crop factor maginfies camera shake--it means the enlargment process magnifies camera shake, right?


Halleluiah!


rdenney wrote: (As an aside, you could take that 8x10 print from the 10D/50 and lay it on the 24x30 print from the 645/50, and the image on the edge of the 8x10 would line up with the image on the 24x30 exactly. That's where the notion of "cropping" comes in.)

Steven M. Anthony wrote: YES!!!! And I think THAT's what others are missing here.

Thank you and good night.

[rdenney]

Agreed!!! But that doesn't mean that the crop factor maginfies camera shake--it means the enlargment process magnifies camera shake, right?

YES!!!! And I think THAT's what others are missing here.

Yes. Of course, the difference in enlargement and the ill-named "crop factor" (really it should be called an equivalency factor) is..THE SAME!

So, we multiply the focal length by the equivalency factor when we use the 1/fl rule of thumb not because the lens is different, but because we will enlarge the image more (and by that same multiplier). And we enlarge the image more because the sensor is smaller.

I must gently suggest that this is what others have been saying, though perhaps they don't type as fast as me, heh, heh.

Rick "a professional explainer" Denney

[slin100]

Rick, I have been trying to say what you said. I think you did a great job of explaning it. It's crystal clear to me.

To reiterate what Russ said, the focal length must be multiplied by the crop factor before applying the 1/fl rule of thumb.

[pcasciola]

And we enlarge the image more because the sensor is smaller.
I agree with everything you just said, but just to clarify, it's not enlarged just because the sensor is smaller, but because the smaller sensors generally have a higher pixel density than the larger sensor cameras (i.e. smaller, but still 8MP for example).

That's why on a P&S the effects of camera shake are even more dramatic, because they are still producing 6-8MP effectively "enlarged" images, so the effects of shake are therefore enlarged as well. The image circle being produced by the lens is completely unrelated. It's effective angle of view that is important in my opinion.

Good job though, Rick. Hopefully this thread can be put to bed now, and added to the archive of x-factor discussions that went on for too long. ;)

[Kennymc]

Thnk goodness for that, I thought it was going to be 50mm lenses at dawn... ;)

[rdenney]

I agree with everything you just said, but just to clarify, it's not enlarged just because the sensor is smaller, but because the smaller sensors generally have a higher pixel density than the larger sensor cameras (i.e. smaller, but still 8MP for example).

Sez you, heh, heh. It's not pixel density that controls my print size, but rather image quality. Pixel density is only one element in image quality, and, in fact, I find that it is not the critical element with the 10D.

Other elements in image quality include lens quality, technique (which includes camera shake issues, aperture selection, etc.), and a difficult to define quality that I call "information density". Information density is only partly related to pixel density, it also has to do with pixel accuracy.

Thus, a 6MP point-n-shoot image isn't as enlargeable (in degree of enlargement) as is a 10D image, because the lens faults are significant and there the smaller pixels are not as accurate.

Allow me an example from a different workflow to illustrate the point. I have a Minolta Multi-II film scanner, for which I paid more than I did my 10D. I use it for scanning medium and small-format film (or I would if I still used small-format film, which i don't much anymore, heh, heh). For medium format, the scanning resolution is 1128 pixels per inch, yielding an image of 2600x2600 when I scan a 6x6 negative.

That same scanner will scan 35mm film at 2800 pixels per inch, given me images about 2600 by 4000 pixels. If I take a square out of the rectangular frame, I end up with an image about 2600 by 2600 pixels--just the same as with medium format. But it still has over twice the pixel density.

But the medium format image can still be enlarged more. Why? Because each pixel is more accurate. They each integrate more film information than the smaller pixels on the 35mm scan. Thus, the pixels can be larger on the print without looking like pixels.

It's the same comparing 6-MP images from a digicam with a tiny sensor and a 10D. Each pixel on the 10D integrates more light information, because it sees more light information.

This is also why telescopes that are bigger not only see fainter light, but also resolve with more detail. The bigger telescopes, just like the bigger photosites, are less consumed by diffraction and other faults.

That's why the 8MP APS-sized sensor doesn't do it for me. Pixel density doesn't limit my image quality. I want more pixels, but I want them in a larger frame.

Rick "who thinks print size for most people is controlled only by how they want to display it--sadly" Denney

[roanjohn]

ITS $20!!!!!

Man, everytime a crop factor thread starts, it always ends in a mumble jumble of maths with different equations that leads to different conclusions about pixel density, focal length and maximum shutter speeds to avoid motion blur.

Talk about sensor overload!!! :lol: :lol: :lol: :lol: :lol:

There should be a sticky for "CROP FACTOR" alone.....or maybe a whole different section of the forum.

Ro1

[pcasciola]

I was trying to keep it simple, but ok, I see your point. ;)

I am also of the belief, though, that as long as the smaller sensor does not exceed the true optical resolution of the lens, while the smaller pixels are not as accurate (as you put it) it is still in fact enlarged, and more detail can been seen. So, it is effectively magnified to a certain extent. Some say this is not true optical magnification, but a digital sensor is an optical device, so as long as it remains within the limits of what the lens can resolve, it has to be somewhere in between a pure crop and what a 1.4x TC does.

(Just when I thought I was out, they suck me back in). :D

[pcasciola]

There should be a sticky for "CROP FACTOR" alone.....or maybe a whole different section of the forum. Ha. Crop factor section. I like it. Maybe with a warning, "Enter at your own risk". :lol:

[Jon]

There is: 10,000,000 Posts on the X-Factor, Or,... the "Sticky of Shame" (http://photography-on-the.net/forum/showthread.php?t=45388)

[Steven M. Anthony]

Yes. Of course, the difference in enlargement and the ill-named "crop factor" (really it should be called an equivalency factor) is..THE SAME!

So, we multiply the focal length by the equivalency factor when we use the 1/fl rule of thumb not because the lens is different, but because we will enlarge the image more (and by that same multiplier). And we enlarge the image more because the sensor is smaller.

I must gently suggest that this is what others have been saying, though perhaps they don't type as fast as me, heh, heh.

Rick "a professional explainer" Denney

Ugh... While some might enlarge their images more because the aps-c sensor starts you off with a smaller print, the physical size of the sensor DOES NOT CONTRIBUTE TO CAMERA SHAKE. PERIOD.

It might start a series of choices that magnify the existing, recorded camera shake contained in any given image, but it does not induce or otherwise impact the amount of recorded camera shake.

Many of you seem to be trying to take this issue beyond that statement. And if you tend to make large prints of your images, 1/fl probably doesn't work for you anyway. But as the focal length of my 50mm 1.8 lens on my 10D remains 50mm--even though the sensor only records the field of view of an 80mm lens--1/fl is appropriate. You see, the focal length of the lens DOES NOT CHANGE.

[rdenney]

Many of you seem to be trying to take this issue beyond that statement. And if you tend to make large prints of your images, 1/fl probably doesn't work for you anyway. But as the focal length of my 50mm 1.8 lens on my 10D remains 50mm--even though the sensor only records the field of view of an 80mm lens--1/fl is appropriate. You see, the focal length of the lens DOES NOT CHANGE.

No, people are trying to give advice to a fellow who wants to know what shutter speed he ought to use with a Canon 10d. Does he limit his prints to a given enlargement factor or to a given print size? Who knows? So, we answer it both ways and provide explanations that consider both points of view.

Or, we can argue all the live-long day about formulas.

It's all BS anyway--1/fl doesn't work in any format, depending on your sharpness standards, expected print size, technique, and equipment.

If you want to choose shutter speeds with no regard to your expected print size, then be my guest.

But frankly, your whole approach is working from the framework of there being only two formats in the the whole of photography. One who understands the principles--what the formulas mean, not just what they say--can work in ANY format.

Rick "who knows the formulas but who chooses to address the issue" Denney

[roanjohn]

There is: 10,000,000 Posts on the X-Factor, Or,... the "Sticky of Shame" (http://photography-on-the.net/forum/showthread.php?t=45388)

:lol: :lol: :lol: :lol: :lol: :lol: :lol: :lol: :lol: :lol: :lol:

.........well I'll be damned!!!

Ro1

[slin100]

Oh man, we really did have this exact, same discussion before.

X-Factor and 1/1 focal length/shutter speed rule (http://photography-on-the.net/forum/showthread.php?t=48748)

I respectfully submit that everyone give up trying to change Steven Anthony's opinion. He seems to agree that a smaller sensor requires larger print magnification, which will make the effects of camera shake more evident. That's enough for me say that we're on the same page though not the same line. :)

[Steven M. Anthony]

No, people are trying to give advice to a fellow who wants to know what shutter speed he ought to use with a Canon 10d. Does he limit his prints to a given enlargement factor or to a given print size? Who knows? So, we answer it both ways and provide explanations that consider both points of view.

And I'm just pointing out that your explanations are factually incorrect. The advice is good (1/fl probably works great for most making 4X6 prints, but probably not for most making larger prints)--but the "crop factor magnifies camera shake" rationale for it isn't.

[rdenney]

And I'm just pointing out that your explanations are factually incorrect. The advice is good (1/fl probably works great for most making 4X6 prints, but probably not for most making larger prints)--but the "crop factor magnifies camera shake" rationale for it isn't.

Which facts are those that are incorrect?

I said that we enlarge images from smaller sensors more, and that extra enlargement magnifies camera shake effects. And, to get the same print size, that extra degree of enlargement happens to be the same factor as the equivalency factor--1.6.

I also said that a 50 is a 50 is a 50, to which you seem to agree, and that a 50 is wide angle, normal, or telephoto depending on the format (which is ALL that defines "normal").

These are facts, and I've already gone through the arithmetic of it.

I also said that you can't ignore degree of enlargement in your determination of camera shake effects. Thus, the size of the sensor affects camera shake effects because its images are enlarged more.

That is an opinion, and also advice. Use it if you want. Just quit telling people they are wrong when they base their advice on it.

Rick "who is not guessing about his facts" Denney

[griff2]

rdenny wrote: I also said that a 50 is a 50 is a 50, to which you seem to agree, and that a 50 is wide angle, normal, or telephoto depending on the format (which is ALL that defines "normal").
Agreed.

This is the last page of this thread, place you're bets to see who gets the last word;)