# How much can I zoom?

Discussion in 'Digital Photography' started by keskew@comcast.net, Sep 21, 2006.

1. ### Guest

I am trying to find a formula or table that will enable me to predict
how much I can zoom in on a digital photograph until it becomes
unviewable. I know this seems vague so an example of what I mean might
help.

Let's say I have a circuit board that is X inches by Y inches in
dimension. I take a picture of the circuit board using a digital
camera with a resolution of Z mega-pixels so that the board fills the
entire picture. Once I download the picture to my computer I want to
be able to zoom in so that I can see the individual components on the
circuit board. How much can I zoom in until the image becomes too
grainy to make out the individual components?

I know it has to do with the camera resolution and pixels-per-inch
(ppi). I am not necessarily looking for a precise formula, but more of
a reference guide. This is outside my realm of experience so any help
is greatly appreciated.

, Sep 21, 2006

2. ### CliveGuest

I think you need to take some test shots.

Also, megapixels aren't always the key. I could take a photo with a 17
megapixel body without the lens on it. Although the image will be
17megapixels, the image will not show anything. So, lens choice is very
important.

<> wrote in message
news:...
>I am trying to find a formula or table that will enable me to predict
> how much I can zoom in on a digital photograph until it becomes
> unviewable. I know this seems vague so an example of what I mean might
> help.
>
> Let's say I have a circuit board that is X inches by Y inches in
> dimension. I take a picture of the circuit board using a digital
> camera with a resolution of Z mega-pixels so that the board fills the
> entire picture. Once I download the picture to my computer I want to
> be able to zoom in so that I can see the individual components on the
> circuit board. How much can I zoom in until the image becomes too
> grainy to make out the individual components?
>
> I know it has to do with the camera resolution and pixels-per-inch
> (ppi). I am not necessarily looking for a precise formula, but more of
> a reference guide. This is outside my realm of experience so any help
> is greatly appreciated.
>

Clive, Sep 21, 2006

<> wrote in message
news:...

>I am trying to find a formula or table that will enable me to predict
> how much I can zoom in on a digital photograph until it becomes
> unviewable. I know this seems vague so an example of what I mean might
> help.
>
> Let's say I have a circuit board that is X inches by Y inches in
> dimension. I take a picture of the circuit board using a digital
> camera with a resolution of Z mega-pixels so that the board fills the
> entire picture. Once I download the picture to my computer I want to
> be able to zoom in so that I can see the individual components on the
> circuit board. How much can I zoom in until the image becomes too
> grainy to make out the individual components?
>
> I know it has to do with the camera resolution and pixels-per-inch
> (ppi). I am not necessarily looking for a precise formula, but more of
> a reference guide. This is outside my realm of experience so any help
> is greatly appreciated.

Once you go over 100% the picture will simply pixellate so you will not be
able to keep zooming in like they do in spy movies to see tiny details!

You would be better off buying a microscope!

4. ### Guest

Experimentation (test shots) was my last option. I was hoping that a

We do this quit often and you can zoom in pretty far and make out
individual components, wire traces, etc. The problem is that we have
never quantified it. I also know they sell special equipment and
microscopes. These pictures are more for engineering reference than
inspection.

, Sep 21, 2006
5. ### CliveGuest

Why?

Take a memory card into a store along with your circuit board and ask they
if you can take some shots to see how they come out. Then, when you get
home, look at the shots and decide.

What you are asking in your OP is crazy, as there are so many variables.

<> wrote in message
news:...
> Experimentation (test shots) was my last option. I was hoping that a
> rule of thumb already existed.
>
> We do this quit often and you can zoom in pretty far and make out
> individual components, wire traces, etc. The problem is that we have
> never quantified it. I also know they sell special equipment and
> microscopes. These pictures are more for engineering reference than
> inspection.
>

Clive, Sep 21, 2006
6. ### sallyGuest

In article <>,
<> wrote:
>I am trying to find a formula or table that will enable me to predict
>how much I can zoom in on a digital photograph until it becomes
>unviewable.

Depends greatly on the quality of your equipment and also the quality of
your subject matter, you may be able to improve your results substantially
(though it may sometimes guess wrong).

sally, Sep 21, 2006
7. ### Bill FunkGuest

On Thu, 21 Sep 2006 20:38:19 +0100, "Adrian Boliston"
<> wrote:

><> wrote in message
>news:...
>
>>I am trying to find a formula or table that will enable me to predict
>> how much I can zoom in on a digital photograph until it becomes
>> unviewable. I know this seems vague so an example of what I mean might
>> help.
>>
>> Let's say I have a circuit board that is X inches by Y inches in
>> dimension. I take a picture of the circuit board using a digital
>> camera with a resolution of Z mega-pixels so that the board fills the
>> entire picture. Once I download the picture to my computer I want to
>> be able to zoom in so that I can see the individual components on the
>> circuit board. How much can I zoom in until the image becomes too
>> grainy to make out the individual components?
>>
>> I know it has to do with the camera resolution and pixels-per-inch
>> (ppi). I am not necessarily looking for a precise formula, but more of
>> a reference guide. This is outside my realm of experience so any help
>> is greatly appreciated.

>
>Once you go over 100% the picture will simply pixellate so you will not be
>able to keep zooming in like they do in spy movies to see tiny details!
>
>You would be better off buying a microscope!
>

BINGO! We have a winnah!

Just about any image viewer will let you zoom in to 100%; beyond that,
pixelation set in.
--
Bill Funk
replace "g" with "a"

Bill Funk, Sep 21, 2006
8. ### Mark²Guest

wrote:
> I am trying to find a formula or table that will enable me to predict
> how much I can zoom in on a digital photograph until it becomes
> unviewable. I know this seems vague so an example of what I mean
> might help.
>
> Let's say I have a circuit board that is X inches by Y inches in
> dimension. I take a picture of the circuit board using a digital
> camera with a resolution of Z mega-pixels so that the board fills the
> entire picture. Once I download the picture to my computer I want to
> be able to zoom in so that I can see the individual components on the
> circuit board. How much can I zoom in until the image becomes too
> grainy to make out the individual components?
>
> I know it has to do with the camera resolution and pixels-per-inch
> (ppi). I am not necessarily looking for a precise formula, but more
> of a reference guide. This is outside my realm of experience so any
> help is greatly appreciated.

Once you reach the pixel level (100%), you're done.

--
Images (Plus Snaps & Grabs) by Mark² at:
www.pbase.com/markuson

Mark², Sep 22, 2006
9. ### Randy BerbaumGuest

wrote:
: I am trying to find a formula or table that will enable me to predict
: how much I can zoom in on a digital photograph until it becomes
: unviewable. I know this seems vague so an example of what I mean might
: help.

: Let's say I have a circuit board that is X inches by Y inches in
: dimension. I take a picture of the circuit board using a digital
: camera with a resolution of Z mega-pixels so that the board fills the
: entire picture. Once I download the picture to my computer I want to
: be able to zoom in so that I can see the individual components on the
: circuit board. How much can I zoom in until the image becomes too
: grainy to make out the individual components?

: I know it has to do with the camera resolution and pixels-per-inch
: (ppi). I am not necessarily looking for a precise formula, but more of
: a reference guide. This is outside my realm of experience so any help
: is greatly appreciated.

Let me see if I can help you understand why you are getting the responses
you are.

If you are shooting an image of a board that is 8" x 10" and then print it
out as an 8x10 print you would get a 1:1 magnification. But if the numbers
printed on any single component is imaged by a single pixel each number
would be a single dot at most. Very hard to read. So even if you "blow up"
the image to a higher magnification you will still have each number on the
component represented by a single dot. At a minimum you will probably need
each number to be represented by 6 to 9 pixels to have any chance at all
of reading the numbers. And even this will not be easy to read.

Now if you shoot a similar image of a board that is only 4" x 5", each
pixel will only be covering 1/4 the area of the board of the prior image.
And so you will be able to resolve details 4 x smaller than previously.
Notice the number of pixels and the "zoom" on the screen have little to do
with the readability of the number. It is much more a case of how many
pixels will image the number. So it will be the magnification (and
clarity) of the macro lens, plus the size of the area being imaged, plus
the size of the area that is used by each individual number to be read,
and only then the number of pixels in the entire image, that will go into
being able to read the numbers on the components.

BTW the responses of 100% zoom are also mostly correct. When you "zoom in"
on an image with an editor you can only get up to 1 pixel being displayed
by one dot on the screen (for the purists, one "dot" being composed of one
red, one green and one blue phosphor). Once you go above that zoom you
will get a distorted image. Straight lines will zig-zag, and thin lines
will become long rows of blobs. It is possible to zoom slightly beyond
100% which will make the image larger on the screen, but the distortion
will get larger and larger until it is difficult or impossible to
recognize anything. Depending on your ability to mentally decode the blurs
and blobs you may be able to zoom higher than someone else. But in general
you won't be going much beyond 100 to 110%, IMHO. But this zoom
number is a relationship between the number of image pixels being
displayed by each dot on the screen, and thus has no hard and fast
relationship to the size of the object in real life to the size of the
image on the screen. They are related, but not absolute.

So to answer your simple question there would have to be lots of variables
measured first. We would need to know ALL the parameters of the lens (and
there are a LOT of them). Then we would need to know the exact dimensions
of the PC board to be imaged. Then we would have to know if you have the
equipment to exactly position the board and camera to maximize the size of
the board in the image. Then we would need the dimensions of the smallest
number that will need to be read. Then (yes) the number of pixels per
image. And lastly we would have to know (from experimentation) how many
pixels per number will give us a readable image, consistantly. And after
all that, the best we could give you is a "it MIGHT work" or a "probably
not".

My best advice is to experiment. Borrow a camera and see how close you
have to get to get a consistant read of the numbers and measure the size
of that the image covers on the original image. Then divide these
dimensions into the total dimensions of the full sized board and then
multiply that result by the number of pixels in the test image. Work each
dimension (Horiz, Vert) seperately. This would be a good aprox pixel count
target for choosing a possible camera. Then give it a try with your
fingers crossed.

Good luck

Randy

==========
Randy Berbaum
Champaign, IL

Randy Berbaum, Sep 22, 2006
10. ### Dave MartindaleGuest

writes:
>I am trying to find a formula or table that will enable me to predict
>how much I can zoom in on a digital photograph until it becomes
>unviewable. I know this seems vague so an example of what I mean might
>help.

It's actually not that difficult to do the calculations for this. All
you need is a calculator.

First, you need to know the size of the image produced by the camera in
pixels. There is a formula to calculate X and Y resolution given the
number of megapixels and the image aspect ratio, but most camera reviews
give the image size in pixels directly, so let's assume you can get
that. For an example, assume you have a 6 megapixel DSLR whose output
image is 3072x2048 pixels.

When you photograph a circuit board, you first have to figure out how
the board will fit the image. A square board will fill the image height
but leave 1/3 of the width unused. A long skinny board will fit the
width but leave some of the height unused. Once you decide which
dimension of the image will fit the board, divide the corresponding
width/height of the image in pixels by the width/height of the board in
inches to get a scale factor in pixels/inch.

For example, suppose your board is 8x10 inches. At maximum size, this
will fill the image height but not use the full width, so it is height
that matters. The pixels/inch on the board is 2048/8 = 256. So you get
256 pixels per inch, or every pixel is 0.0039 inches.

So you will be able to *see* all of the parts on the board as long as
they are more than a few thousandths of an imch in size. But to be able
to read lettering on the components, the letters should be somewhere
near 10 pixels high.

If your resistors are "0406" size, 0.040 by 0.060 inches in size, that's
about 10x15 pixels when the image is displayed at "100%" size. You
definitely should be able to see the resistors, but being able to read
the resistance value printed on top could be a challenge.

On the other hand, if the board is 4x5 inches, the pixel density doubles
and reading small text becomes much easier

Dave

Dave Martindale, Sep 22, 2006
11. ### Guest

Thanks to everyone who responded, especially Randy Berbaum and Dave
Martindale. You guys confirmed the conclusion I came to last night
after continuing to work on this. I also did some experimentation.
While it is true that the picture begins to "pixellate" after 100%,
the images are still viewable (at least for my purposes) up to 200%.
In some cases the pictures were viewable up to 250%. I believe this is
because for the camera I was using, (Cannon Power Shot S2 IS - 5
megapixel) at the highest settings it produced 8.3" X 11.7"
pictures at approximately 220 ppi. That allows the picture to be
enlarged to 200%, or greater, and be viewable, though certainly not
photo quality. Thanks again.

, Sep 22, 2006
12. ### John McWilliamsGuest

wrote:
> Thanks to everyone who responded, especially Randy Berbaum and Dave
> Martindale. You guys confirmed the conclusion I came to last night
> after continuing to work on this. I also did some experimentation.
> While it is true that the picture begins to "pixellate" after 100%,
> the images are still viewable (at least for my purposes) up to 200%.
> In some cases the pictures were viewable up to 250%. I believe this is
> because for the camera I was using, (Cannon Power Shot S2 IS - 5
> megapixel) at the highest settings it produced 8.3" X 11.7"
> pictures at approximately 220 ppi. That allows the picture to be
> enlarged to 200%, or greater, and be viewable, though certainly not
> photo quality. Thanks again.
>

I think you'd also find, if your monitor were large enough, and you
could get far enough away from it, that many low ppi counts would look
fine. Same with printing; bill boards seem to be printed with inches per
dot, not dots per inch! Although I exaggerate, a well done bill board
looks great from the right distance.

--
john mcwilliams

John McWilliams, Sep 22, 2006
13. ### dmasterGuest

wrote:
> Thanks to everyone who responded, especially Randy Berbaum and Dave
> Martindale. You guys confirmed the conclusion I came to last night
> after continuing to work on this. I also did some experimentation.
> While it is true that the picture begins to "pixellate" after 100%,
> the images are still viewable (at least for my purposes) up to 200%.
> In some cases the pictures were viewable up to 250%. I believe this is

So far so good. Yes the pictures are going to be "viewable", meaning
recognizable to some arbitrary magnification. But, that does not mean
you are seeing anything new! Once your eyes can see the individual
pixels...there is nothing more to see. Sure, zooming with make the
original

> because for the camera I was using, (Cannon Power Shot S2 IS - 5
> megapixel) at the highest settings it produced 8.3" X 11.7"
> pictures at approximately 220 ppi. That allows the picture to be
> enlarged to 200%, or greater, and be viewable, though certainly not
> photo quality. Thanks again.

Ick. You've run off the rails. The "ppi" of the file the camera
produces
has absolutely nothing to do with your application. Heck, if you would
sit twice as far from the monitor, you'd magically be able to zoom
twice
as much, and the image would still be viewable! But as I said above,
this is pointless. Once you can distinguish every piece of information
(pixels) nothing more can be gained.

Ah... That may be what is misleading you... depending on your eyes
and distance from the monitor, you may need to zoom well beyond
100% to see every pixel.

Dan (Woj...)

dmaster, Sep 22, 2006