• Please use real names.

    Greetings to all who have registered to OPF and those guests taking a look around. Please use real names. Registrations with fictitious names will not be processed. REAL NAMES ONLY will be processed

    Firstname Lastname

    Register

    We are a courteous and supportive community. No need to hide behind an alia. If you have a genuine need for privacy/secrecy then let me know!
  • Welcome to the new site. Here's a thread about the update where you can post your feedback, ask questions or spot those nasty bugs!

Describing sensor sizes

Doug Kerr

Well-known member
[Part 1]

I'm sure most of you are well familiar with most of this, but I thought some might enjoy being reminded of some of the history.

Introduction

The subject here is how we describe the sizes of the sensors in digital cameras.

It turns out that different conventions are used in three "bands' of sensor sizes:

A. Larger than 36 mm × 24 mm

B. In the range of about 22 mm × 14 mm through about 36 mm × 24 mm

C. Smaller than about 22 mm × 14 mm

I will discuss these in a peculiar order.

Band B sensors

The basic concept

Cameras with sensors in the "band B" size range have an "ancestry", direct or indirect, in the "35 mm" film still camera (to be more explicit, in the "full frame 35 mm" still camera). Some of these digital cameras (notably those from Canon and Nikon) are tied to their full-frame 35 mm ancestors in that they are interchangeable lens cameras (almost all the single-lens reflex type) that can use lenses in the substantial repertories from their manufacturers developed for use with the full-frame 35 mm cameras, or further lenses added to those series.

Others (notably from Sony) have no such direct ancestry, but nevertheless, even of only by virtue of their sensor sizes, occupy a parallel stable.

The "full frame" sensor size

Of course the frame size of the full-frame 35-mm film cameras was (nominally) 36 mm × 24 mm. Because that is the same as the largest frame size (thus, sensor size) of this band of digital cameras, a digital sensor frame size of nominally 36 mm × 24 mm is spoken of as the "full frame" sensor size. (Yes, of course this is sort of a circular definition!).

Smaller sensors in this band

Photographers utilize lenses of varying focal length primarily to attain varying field of view for their images. The field of view provided by a lens is the result of both the focal length of the lens and the size of the image format (sensor, in digital cameras). The larger the format size, the greater the field of view; the greater the focal length, the less the field of view.

Because of the widespread use of the 36 mm × 24 mm format in film cameras, even moderately-sophisticated photographers came to realize what common focal lengths would produce fields of view suited for various work. A focal length of about 50 mm was considered "normal" (another fascinating story), while we might use a focal length of perhaps 135 mm for "telephoto" work (that term itself being the subject of another fascinating story) or a focal length of perhaps 24 mm for "wide-angle" work.

But of course a 50 mm lens mounted to a digital camera with a sensor size of about 22.5 mm × 15 mm would yield the same field of view as would be given by a lens of focal length 80 mm mounted on a camera with a sensor size of 36 mm × 24 mm (a so-called "full frame" sensor size).

Thus came into use the often misunderstood notion of "equivalent focal length" (more thoroughly described as "full-fame 35-mm equivalent focal length").

In our example camera, with a sensor size of 22.5 mm × 15 mm, for a lens of any given focal length, its "full-frame 35-mm equivalent focal length" when mounted on that camera would be about 1.6 times its focal length.

Thus, the description "1.6 X" came to be associated with sensors of size approximately 22.5 mm × 15 mm (as their sensor dimensions were 1/1.6 times the dimensions of the "full frame 35-mm" size sensor). The number was often said to be the "focal length multiplier". A more apt, if clumsy, term would have been "full-frame 35-mm equivalent focal length factor".

"Cropped" sensors

Because a sensor of size 22.5 mm × 15 mm (for example) was smaller that a sensor of size 46 mm × 24 mm, which has come to be called "full frame" (people not bothering with the rest of the description, "35-mm"), some people came to consider such "smaller" sensors as "cropped" from the "full frame size". So we may see someone, seeing a colleague's new camera, evidently a "band B" camera, ask, "Is that full frame or cropped?" (or sometimes just "crop").

Sadly, the "full-frame 35-mm equivalent focal length factor" often came to be stated as the "crop factor" of the camera. Thus a camera whose sensor had dimensions of 0.625 times the dimensions of the "full-frame 35-mm" frame (1/1.6) would be said to have a "1.6 crop factor."

"APS" notation

"APS" refers to the Advanced Photo System, a new film system introduced in 1996 by Eastman Kodak, intended to make film use more convenient for a range of users than the traditional "135-type" cartridge used for 35-mm film. The film was nominally 24 mm wide.

Almost invariably, the camera took a frame of dimensions 30.2 mm × 16.7 mm. But the user, at the time of taking the shot (usually), would instruct the camera that the frame size he really wanted to use was one of there, designated by three letters;

H (high definition): 30.2 mm × 16.7 mm. This has an aspect ratio of almost exactly 16:9, consistent with the format then coming into use for "high-definition" television.

C (classic): 25.1 mm × 16.7 mm. This has an aspect ratio of almost exactly 3:2, the same as for the "classic" full-frame 35-mm cameras.

P (panoramic): 30.2 mm × 9.5 mm. This has an aspect ratio of almost exactly 3:1, and is of course intended for "panoramic" format images.

But regardless of the format chosen by the photographer, the camera (almost always) exposed a frame of dimensions 30.2 mm × 16.7 mm. The desired frame size, set by the photographer, was encoded magnetically on the film. In the laboratory, prints of the selected aspect ratio for each frame (if it were marked for the C or P format) were produced by cropping the actual exposed frame (here a proper use of the term "crop"). (For the H format, the full frame was printed - here a proper use of the term "full frame".)

Thus the three aspect ratio options in the APS system came to be known as the H, C, and P "crops" (even though the H format did not require cropping).

Now back to our digital camera sensor size story. A popular sensor size in "band B" cameras was either
about 22.5 mm × 15 mm (Canon) or about 24 mm × 16 mm (Nikon). Because of the anxiety of writers about digital photography for ways to express these sensor size "ranges", it became common to speak of wither of these sizes as "APS-C" (since the "C crop" of an APS frame had dimensions of 25.1 mm × 16.7 mm, and these were "something like that").

Now, for quite a while, a common digital sensor size in the Canon camera repertoire was in the area of 27.7 mm × 18.5 mm. (That would be a "1.3X" size.)

And in the same vein I discussed above, this general sensor size range became known as the "APS-H" size (since the "H crop" of an APS frame had dimensions of 30.2 mm × 16.7 mm, and these were "almost something like that").

************
The reader with a sensitive stomach may wish to take some Pepto-Bismol here.

************

As for me, I'm going to have breakfast.

[to be continued]
 

Doug Kerr

Well-known member
[Continued - Part 2]

Band C sensors

Introduction

Cameras in sensor size band C have sensors whose dimensions are typically less than 22 mm × 14 mm.

Before I can describe the notation commonly used to describe the size of these sensors, I will describe a separate but related topic.

The Vidicon TV pickup tube

Many different types of "pickup tube" were used during the evolution of television, but the most prominent one was the image orthicon. This tube was itself large, complex, and costly, and depended upon a complicated supporting infrastructure, both in the camera proper and in the associated camera control unit.

These factors made it impractical to deploy a television camera to provide surveillance of a parking lot, to embed a camera in a production machine to allow the operator to confirm its proper operation in a way that could not be done by direct observation, or to establish low-cost television studios for educational purposes.

These limits were relieved by the development, in the early 1950s, of the vidicon pickup tube. This was smaller, less complex of itself, required a much less complicated supporting infrastructure, and did not require the attention of a "flight engineer" to keep it tooting properly.

One of the most important early vidicon tubes had a target (we would today say "sensor") whose dimensions were nominally 8.8 mm × 6.6 mm (following the 4:3 aspect ratio then universally used for television operation).

Shortly, to provide for better resolution, a new series of vidicon tubes was introduced, these having a target whose dimensions were approximate 1.5 times those of the other tubes (I don't recall the actual nominal dimensions at the moment).

This of course then led to need for a way to indicate which of these families was used in a particular camera system. Of course the technicians involved in actual use rarely had any awareness of the actual target dimensions. (The technicians who used image orthicon cameras rarely knew the dimension of the targets there, either.)

Now, with respect to the cathode ray tubes (CRTs) then used in television receivers (and more to our point, in television monitors as well) the sizes were designated in terms of the nominal diameter (in inches) of the glass envelope of the tube. This was not the diagonal dimension of a rectangular image the tube could produce, although the two numbers were generally not too different.

In fact, the industry designations of these CRTs began with the nominal envelope dimension in inches. For example, a popular tube used in monitors (and in fact in small television receivers - I had one of those once) was the 7JP4.

Its envelope diameter was 7 inches (± 1/8 inch) and its "image circle" (the diameter of the largest image it could properly support) was 6 inches (a bit larger in some variants). And this was called a "7-inch" CRT.

So, with this perspective in mind, the technicians came to distinguish the two popular families of vidicon tube in terms of their glass envelope diameters. The smaller type (8.8 mm × 6.6 mm target) had an envelope diameter of just about 2/3 inch, and the larger one an envelope diameter of just about one inch. So of course these came to be called "2/3 inch" and "one inch" vidicons.

Hold that thought.

Embarrassingly-small sensors

When the first "consumer" digital cameras emerged, marketing folks were reticent to state, in advertisements and the like, their sensor sizes, which might be on the order of 4 mm × 3 mm (0.16 in × 0.12 in) or even less. They were concerned that this would seem very tiny to those with even a rudimentary knowledge of photographic technology.

So somebody got the bright idea of drawing upon the notion used for the size of vidicon tubes and their targets. Since we called a vidicon tube whose target dimensions were 8.8 mm × 6.6 mm a "2/3 inch" (0.67 inch) size, then a digital camera sensor with dimensions 4 mm × 3 mm could be called a "0.30 inch" sensor.

But that still sounded pretty wee. So the next step was based on the fact that 0.30 could be expressed, mathematically, as 1/3.3. So then, reasoned the marketing guys, we could speak of this camera as having a "1/3.3 inch" sensor. And "3.3 inch" (of course that was not what the notation said) didn't sound all that small.

So this is how we end up with cameras whose sensors are described as "1/2.8 inch" or "1/1.8 inch" sizes.

The four thirds system

Now a new system for digital cameras is called the "four thirds" system (it has an advanced cousin called the "micro four-thirds" system). The "four-thirds" has two meanings:

• The standard aspect ratio is 4:3 (not the 3:2 we find in most "Band B" cameras, so this deserves emphasis). Of course a ratio of 4:3 is in fact, mathematically, the fraction 4/3 ("four thirds")

• The standard sensor size (18 mm × 13.5 mm) would be described, in the "vidicon envelope diamater" scheme, as a "1.37 inch" sensor (no need to state it as its inverse - it is not that tiny-sounding). But this is very close to 4/3 (1.33), or "four thirds".

"One inch" sensors

Another popular genre today in the "band C" sensor size range is he "one inch" sensor size camera, such as my Panasonic FZ1000. Again, this notation is under the "vidicon envelope diameter" convention. Accordingly, the designation "one inch" would imply a sensor size of 13.2 mm × 9.9 mm.

However, many of these cameras offer a 3:2 aspect ratio, which muddles the situation somewhat.

As an example, in the Panasonic FZ1000, the overall sensor dimensions are 13.2 mm × 8.8 mm. If we worked from the sensor width only, that would come in, under the "vidicon envelope diameter" scheme, exactly at "one inch". And if we compare sensor diagonal dimensions, it is still very close (considering that this is nominal at best and nonsensical at worst).

[to be continued]
 

Doug Kerr

Well-known member
[Continued - Part 3]

Band A sensor sizes

Introduction

I have arbitrarily defined "band A" sensor sizes as those larger in size than 36 mm × 24 mm.

Here we don't have well-established conventions (not even silly ones).

MF - does that stand for a whole word?

Film cameras with formats larger than 36 mm × 24 mm, but smaller than 4 in × 5 in, are generally considered medium format (MF) cameras.

A popular genre in that range is cameras using 120-type (or 220-type) film, which is nominally 61 mm wide. Common formats are 56 mm × 41.5 mm and 45 mm × 56 mm, and 56 mm × 84 mm. It is most common to state those by name which is an approximation of the format size in centimeters, "6 × 4.5", "6 × 6", and "6 × 9", respectively.

There have been a wide range of cameras using those formats, from inexpensive "snapshot" cameras to some of the most high-performance, high-cost "hand-held" cameras available.

And often, especially when the preferred format is 56 mm × 41.5 mm ("6 × 4.5"), it is common to have the digits 645 appear in the family name of the cameras.

MF digital cameras

Digital cameras having sensor sizes larger than 36 mm × 24 mm (and I suppose smaller than 4 in × 5 in!) are spoken of as "MF cameras".

Sometimes, because the cameras involved are, in some sense, adaptations of film cameras with a 56 mm × 41.5 mm format (6 × 4.5), the camera designations involve the digits 645.

Does that mean that their format sizes are 60 mm × 45 mm? Not usually.

So how do we describe the format sizes of these cameras? Well, we usually don't. We mention the model, and if anyone cares, they should know. Or look it up.

But when all else fails, we can give the actual dimensions. In millimeters, please.

************

Best regards,

Doug
 

Jerome Marot

Well-known member
MF digital cameras

Digital cameras having sensor sizes larger than 36 mm × 24 mm (and I suppose smaller than 4 in × 5 in!) are spoken of as "MF cameras".

Sometimes, because the cameras involved are, in some sense, adaptations of film cameras with a 56 mm × 41.5 mm format (6 × 4.5), the camera designations involve the digits 645.

Does that mean that their format sizes are 60 mm × 45 mm? Not usually.

So how do we describe the format sizes of these cameras? Well, we usually don't.

Actually, the conventions "x1.3", "x1.1" and "full frame 645" are used. The recently issued Hasselblad camera is "x1.3", for example.
 

Doug Kerr

Well-known member
Hi, Jerome,

Actually, the conventions "x1.3", "x1.1" and "full frame 645" are used. The recently issued Hasselblad camera is "x1.3", for example.

Ah, yes.

And I see there are some who compare the frame size of the new Hasselblad to "full-frame 35-mm" format. That makes it "x0.78" or so.

And the beat goes on.

Thanks.

Best regards,

Doug
 
Top