Viewfinder size and magnification ... you do the math!

Manfr3d

Hello there,

I like numbers and while I was browsing
the-digital-picuture website I came across
this viewfinder comparison chart and started
thinking (yes that actually happens) that
viewfinder magnification must always be
understood in relation to the camera's
sensor size and the viewfinder coverage,
am I right?

For example:

The 5D has a 96% finder and 0.72x magnification relative to fullframe sensor.
The 40D has a 95% finder and 0.95x magnification relative to 1/1.6 fullframe sensor.

Does this mean that in terms of a fullframe camera the 40D would have
a viewfinder with 0.95x 1/1.6 = 0.59x magnification?

I'm curious to know if this is correct.

Viewfinder comparison char (scroll down 1 page):
http://www.the-digital-picture.com/Reviews/Canon-EOS-5D-DSLR-Digital-Camera-Review.aspx

ziggy53

Not sure what you are getting at but the full frame cameras have less viewfinder magnification because they have a larger focussing screen, which tends to make them considerably brighter to view than the crop cameras' viewfinders.

That extra brightness, plus a very large percentage view tend to be features to desire. None of the Canon viewfinders are "that bad", but the 1D series are pretty spectacular.

Manfr3d

ziggy53 wrote:

Not sure what you are getting at but the full frame cameras have less viewfinder magnification because they have a larger focussing screen, which tends to make them considerably brighter to view than the crop cameras' viewfinders.

That extra brightness, plus a very large percentage view tend to be features to desire. None of the Canon viewfinders are "that bad", but the 1D series are pretty spectacular.

Hi Zig,

I know what a good viewfinder specs are
and what magnification and coverage mean
and that one has to trade one for the other.

The viewfinder on my 10D is very small
and dim compared to that of the 5D. So
I was just wondering how to compare two
viewfinders by their attributes (coverage,
magnification, sensor size) in numbers.

I'm not trying to make a buying decision
based on numbers here, I'm only interested
in the math.

ziggy53

I'm not sure you have enough information to make a definitive comparison. You would have to also know things like eye relief and transmissivity comparisons (penta-mirror vs penta-prism for instance.)

Scott_Quier

Isn't viewfinder magnification an expression of the amount of the final image that is viewable in the viewfinder?

The fact that one viewfinder is brighter/dimmer than another is probably a function of the following:

• The 20D/30D/40D use "penta-mirror" configuration rather than the penta-prism that is used in better FF film cameras and in FF digitals (1Ds, for example).
• The amount of image circle transmitted to the viewfinder in a crop camera will be less than that transmitted in FF camera. The result - less light transmitted -> darker viewfinder image.

Manfr3d

Thats what I'm thinking scott. And my assumption is that
because of the smaller sensor the actual viewfinder also
has to be smaller in size. Otherwise the viewfinder
magnification would have to be above 1.0x. Which it isn't.

If I'm correct (see above) then a 5D with 0.59x magnification
would have a viewfinder that has the same size as
the one in the 40D (I'm neglecting the 96% vs. 95%
coverage here for a moment).

And as Ziggy pointed out that doesn't make them look
equal because of the focusing screen and eye-point
differences. But thats not what I'm try to get at.

claudermilk

AFAIK Scott has it right. The viewfinder spec is how much of the actual captured image you see. So the 95% spec is telling you that you see 95% of the captured image.

After having used a 1Ds Mk II side-by side with my 20D, I'm going to find myself in disagreement with most comments on viewfinders now. IMHO it's much ado about nothing & hair-splitting debate with little noticeable effect in practice. For the record, the 1Ds had my 70-200/2.8 on it & the 20D had my 24-70/2.8 on it. I noticed little to no brightness difference or any difference in my ability to see & frame the subject. In theory the FF pentaprism finder should be far superior to the 1.6 crop penta-mirror finder, in practice I did not see that. The one noticeable difference I saw was the info displays on the 1Ds took up less relative room & packed more useful information in--that's a worthwhile improvement by itself & was really the only difference I saw.

Scott_Quier

Manfr3d wrote:

Thats what I'm thinking scott. And my assumption is that
because of the smaller sensor the actual viewfinder also
has to be smaller in size. Otherwise the viewfinder
magnification would have to be above 1.0x. Which it isn't.

No - it's amazing what can be done with optics. If they had wanted to, they could have created a viewfinder that was, physically, twice the size but covered less than 1/2 the captured image. Of course, you have issues with spreading a limited amount of light to viewfinder that size. I would bet that the 95% (or whatever it is) factor is more a design decision, based on how much room they had (wanted to use) for the viewfinder optics.

Manfr3d wrote:

If I'm correct (see above) then a 5D with 0.59x magnification
would have a viewfinder that has the same size as
the one in the 40D (I'm neglecting the 96% vs. 95%
coverage here for a moment).

And as Ziggy pointed out that doesn't make them look
equal because of the focusing screen and eye-point
differences. But thats not what I'm try to get at.

You completely lost me on that second paragraph.

Manfr3d

Scott_Quier wrote:

No - it's amazing what can be done with optics. If they had wanted to, they could have created a viewfinder that was, physically, twice the size but covered less than 1/2 the captured image. Of course, you have issues with spreading a limited amount of light to viewfinder that size. I would bet that the 95% (or whatever it is) factor is more a design decision, based on how much room they had (wanted to use) for the viewfinder optics.

I know they could do it. But then the viewfinder magnification would be
larger than 1 right? Since all cameras have lower magnification the
images in the camera's viewfinder cannot be larger than the size of
the cameras sensor.

Scott_Quier wrote:

You completely lost me on that second paragraph.

Hehe. I'm saying that one cannot derive the quality of a viewfinder
by only looking at how large the image in the viewfinder is. The
eyepoint (~distance from the eye to the viewfinder image) and
the focussing screen also important factors to be considered.

Manfr3d

claudermilk wrote:

AFAIK Scott has it right. The viewfinder spec is how much of the actual captured image you see. So the 95% spec is telling you that you see 95% of the captured image.

yes thats why they call it "coverage"

claudermilk wrote:

IMHO it's much ado about nothing & hair-splitting debate with little noticeable effect in practice.

This is exactly the discussion I'm trying to avoid here. I'm not interested
in which viewfinder is better I just want to understand the math behind
viewfinders.

claudermilk

No problem.

Looking at Canon's site it is interesting to note the differences.

40D, 95% coverage, 0.95x magnification on a 22.2x14.8mm sensor
5D, 96% coverage, 0.71x magnification on a 35.8x23.9mm sensor
1D Mk III, 100% coverage, 0.76x magnification on a 28.1x18.7mm sensor
1Ds Mk III, 100% coverage, 0.76x magnification on a 36x24mm sensor

I'll have to put a ruler to my viewfinder to see if some playing on a calculator means anything, but it appears they are trying to keep the viewfinders approximately similar sizes.

Manfr3d

claudermilk wrote:

No problem.

Looking at Canon's site it is interesting to note the differences.

40D, 95% coverage, 0.95x magnification on a 22.2x14.8mm sensor
5D, 96% coverage, 0.71x magnification on a 35.8x23.9mm sensor
1D Mk III, 100% coverage, 0.76x magnification on a 28.1x18.7mm sensor
1Ds Mk III, 100% coverage, 0.76x magnification on a 36x24mm sensor

If you mean the size of the image inside the viewfinder by viewfinder size
then I think you're wrong. Look at the 40D for example:

First of all if you have 0.95 magnification on a 22.2x14.8mm sensor.
This means that the image showing in the viewfinder (thats what I
refer to as viewfinder size) cannot be larger than 0.95 times the sensor
size. Multiplying the sensor size (height and width) with the magnification
value gives the maximum size of view viewfinder (height and width).
Here:

0.95x22.2mm = 21.09mm and
0.95x13.8mm = 13,11mm

As a second factor the coverage has to be considered. If we have a
viewfinder with 100% coverge we are done with the above calculation.

If the coverage however is below 100% then we're not seeing the whole
image in the viewfinder ... the image we see is cropped from the
original image the sensor would record. This means that the viewfinder
size shrinks again to the % value of the coverage. Continuing the above
example one gets:

95% of 21.09mm = 20.04mm and
95% of 13,11mm =12,46mm

So the viewfinder size of the 40D would be 20.04mm by 12,46mm.

Similarly we get:

5D, 96% coverage, 0.71x magnification on a 35.8x23.9mm sensor:

96% of 0,71x35.8mm = 96% of 25,42mm = 24,40mm and
96% of 0,71x23,9mm = 96% of 16,97mm = 16,29mm

1D Mk III, 100% coverage, 0.76x magnification on a 28.1x18.7mm sensor


100% of 0,76x28.1mm = 21,36mm and
100% of 0,76x18.7mm = 14.21mm

1Ds Mk III, 100% coverage, 0.76x magnification on a 36x24mm sensor:

100% of 0,76x36mm = 27,36mm and
100% of 0,76x24mm = 18,24mm

if all this is correct then I'm now able to compare the viewfinder of different
cameras such the one of the 40D and the 5D:

The 5D's viewfindersize is close to 4mm larger in height and width that the
one of the 40D. A huge difference!

claudermilk

That's kind of how I was looking at it. I figured sensor size x magnification factor x coverage should give you the size. But what makes me think this isn't right is the eyecup is the same size on them all...isn't it?

I think we've gotten to the point of a couple of blind men fumbling around in a doorless darkened room...

Manfr3d

claudermilk wrote:

That's kind of how I was looking at it. I figured sensor size x magnification factor x coverage should give you the size. But what makes me think this isn't right is the eyecup is the same size on them all...isn't it?

I think we've gotten to the point of a couple of blind men fumbling around in a doorless darkened room...

The eyecup is only the window to the viewfinder image ..
it doesn't change the size of the image you see in the
viewfinder.