Please Explain the 4/3 Sensor Size to Me?
 

I understand that 4/3 is the aspect ratio of the width and height of the sensor.

What I don't understand is why it is seen as a viable size for modern motion work. Everything is 16X9 or wider, FWICT. So why create a sensor that uses an old SD aspect ratio?

I could understand if one were going to use anamorphic lenses, but I highly doubt that's the intention.

So I have to believe I'm missing something, but I don't know what it is.

(While this post was inspired by the specs of the soon to be released Panasonic AF100, I think there something more generally positive about the 4/3 sensor size that I'm overlooking.)

Thanks much.

The 4/3rds sensor size is nothing more than a compromise in reality. This is no bad thing as full frame 35mm was seen in just the same light when 6 x 6 was all the rage.

4/3rds chips means that DSLRs can be lighter, smaller, maybe cheaper (Olympus), and bridge cameras (Panasonic) can flourish now that electronic v'finders are getting so good.

The 4/3rds chip is big - very big - when compared to to the 1/3" chip fould in many camcorders, and masking it down to 16:9 is no big loss or deal. Of course bigger chips generally work better in low light, they allow creative control of dof and can give very acceptable stills as well as movies out of the same camera.

So you're not missing anything, it's just yet another format, somewhere between the 1/8" cip in your phone cam to the full frame in the 5DII.

tom.

Micro Four Thirds system - Wikipedia, the free encyclopedia

The AF 100 has a 16;9 sensor fitted. Given that the 2/3" sensor has been used on major feature films, the larger sensor being used on this camera shouldn't be an issue for that reason alone.

No, the 4/3 refers to four thirds of an inch, an inch and one third ...

The imaging chip as used in an AF100 is not 4:3 at all -- it is (as best we can know before the device is released) 19mm x 11mm, a ratio of 16:9.

I believe that the GH2, as another example, has a 'flexible' chip size that allows for different windows of use depending on the shooting mode, so it too has a 16:9 mode in addition to various photo aspect ratios.

4/3 is not the same as 4:3

Cheers,
GB

There's four thirds, and micro four thirds.

IIRC, the term "four thirds" is a referral to old school analog video camera tubes. The name refers to the external size of the physical tube, not the image area. As it stands, the term has nothing to do with the size or aspect ratio of the image it records. It's just a name. "That which we call a rose, by any other name would smell as sweet" [Shakespeare] and all that implies.

Geoff,

I believe this is not the case. FWIU, 4/3 is both an aspect ratio and the "size" of the chip (using the outdated way that chip size is designated).

The imaging surface of the chip has an aspect ratio of 4/3. So to shoot 16/9, the camera essentially letterboxes the chip.

What I don't understand is why not just make the chip 16/9 to begin with? Perhaps the chip is also used in other offerings, so it's easier/cheaper to use it in the AF100 than make a new chip?

Bruce from your first link:

In an interview John Knaur, a Senior Product Manager at Olympus, stated that "The FourThirds refers to both the size of the imager and the aspect ratio of the sensor."

I believe 4/3 is the aspect ratio of the sensor, but the sensor can be cropped to shoot other aspect ratios.

The specs for the Panasonic AF-100 indicate it is not a 4:3 chip -- it is a 16:9 chip. Or at least the only implementation is as a 16:9 chip ... as the same chip in the GH2 is a 'variable' chip, and offers 4:3 as well as 16:9. This is not 'cropping' in the perjorative sense, as the 4:3 option uses less than the full width of the chip.

Olympus is alone in describing the chip as a 4:3 chip -- and my guess is that was a marketing choice, not a technical one. But I'm speculating about Olympus -- Panasonic has been very clear in indicating no special relationship to 4:3. The dimensions listed for the AF-100 are 18.9mm x 10.6mm, pretty much exactly 16:9.

See the dimensions of the images as described towards the end of this Panasonic summary, note that 4:3 does not represent the pixel width of 16:9
http://panasonic.net/avc/lumix/popup...lease/gh2.html

Cheers,
GB

Peter, a slash is usually not used to designate aspect ratio, only size. Aspect ratio uses a colon. So 4:3, 16:9, 1.85:1 and 2.35:1 are all aspect ratios, which have nothing to do with size. 1/4, 1/3, 1/2 and 4/3 are all sizes, which have nothing to do with aspect ratio. 4/3 is not the same as 4:3, and 16/9 is not the same as 16:9.

So while a 4/3 chip can indeed be 4:3, it can also be 16:9.

You gotta knows your punctuationses.

My hesitance about the damn thing is that it appears unless you want to buy all new lenses or deal with some kind of adaptor and crop factor mess - you're gonna have to dump what you've got and buy all new kit.

THAT was the big draw for the 5d to me.

Same lens system could be used for excellent stills AND video - and there was a robust market for 35mm glass already in place.

Not sure what a bag full of those early 4/3 lenses will bring in another 5 years.

But that's a bet you either take - or avoid as you make your call.

Some good points raised, and Geoff Baker's right in that the GH1 has a 4/3rds chip that allows 4:3 as well as 16:9 with both aspect ratios having the same diagonal dimension, so no copping necessary.

But this brings me to crop factors and Bill's point about 4/3s lenses. At the moment we have full frame, 1.2x, 1.5x and 1.6x crop factors (and possibly more, I don't know). Lenses made for the 1.5 and 1.6 crops vignette on the full frame and 1.2x crops, right?

So are the lenses made for the Micro 4/3rds system (a 2x crop) no good on the 1.6 and 1.5 chipped (APS) cameras? I'm getting confused here, as you see. Will Sigma, Tokina and all make different focal length lenses designed for these different chip sizes?

tom.

I think Bruce Watson said it best with "the term "four thirds" is a referral to old school analog video camera tubes" - it should be seen as referring to the lens coverage, not any defined sensor shape.

In this case, a "four thirds lens" should be seen as having a usable circular coverage area with a diameter of roughly 4/3". You can put any shape of sensor in it you like - as long as the diagonal is no more than 4/3".

If the camera is likely to be used for multiple aspect ratios (say 4:3 for stills, 16:9 for video) it would make sense to use an oversized sensor (such that the corners fall outside the lens coverage area) and crop accordingly, so that the used area of sensor always has the same (and optimum) diagonal. That's a lot easier to describe with a diagram.....!

But the key is to think of "four thirds" in terms of circular lens coverage - not any fixed sensor size or shape.
In short, no. The APS chip size camera have a chip diagonal bigger than the lens coverage of a four thirds lens. Try to use such a lens and it will corner cut very badly.

The 4/3 standard was originally a stills and not a video system and yes it was the ratio as well as sensor size. 4:3 was the old PC display standard, as oppose to the 3:2 stills standard in 35mm film, aps-c and aps-h.

Now that we've moved into the video world, the 4:3 ratio bit is quickly forgotten, except for the fact that you can still select that as the in-camera stills ratio.

Also from that first link is this:

"The name of the system stems from the size of the image sensor used in the cameras, which is commonly referred to as a 4/3" type or 4/3 type sensor. The common inch-based sizing system is derived from vacuum image-sensing video camera tubes, which are now obsolete. The imaging area of a Four Thirds sensor is equal to that of a video camera tube of 4/3" diameter."

My intent was for you to read the information at the links, since you asked for clarification. You did this, which is good. Now if you'll do a little math, you find that the sensors themselves have no dimensions that relate to four thirds of anything, either imperial or metric measurements. I don't have any idea why Mr. Knaur said what he did. But experience has taught me to take engineering's word over marketing's any time when it comes to stuff like this.

Keep in mind that Marketing, not Engineering, designs the product and specifies what it will be. Mr. Knaur said what he did because that's what 4/3 is in this case -- a reference at the time to both the sensor size (no matter how obsolete and inaccurate that particular nomenclature is) as well as the aspect ratio of the *still images* for which this format was *originally intended* by a group of *still camera manufacturers.* The fact that the system has transitioned on to include 16:9 HD video isn't a compelling reason to change the name of the original format standard, especially since several manufacturers are involved... it would take a year's worth of committee meetings and an expensive re-branding campaign to do so.