[jct us101]

Something tells me you really don't like squares.

[Steelski]

Something tells me you really don't like squares.

This comes as a surprise to me, I heard it was now "hip" to be square.

[Maffer]

Square sensor sounds useless, at least for landscape people.

[falconeye]

Something tells me you really don't like squares.

This isn't true. I proposed it myself here (more excatly, an octogon-shaped sensor). But it simply doesn't make sense as long as CMOS isn't cheaper than glass. And if it is, than there is no reason for anything smaller than FF anyway (in K mount).

[thibs]

to produce than APS-H (more surface), still having a larger crop factor (narrower FoV with lenses having a larger than APS-C image circle), and most of all, still can only produce images 82% as wide as APS-H.

Is this enough to stop the APS-H square sensor bullshit rumor? Of course, not

Larger crop? No. A larger sensor will record MORE than an APS-C sensor. No crop at all. APS-C would be a crop of that square sensor, yes.

Seems you mixed everything...

[jeffkrol]

Oh dear, this entire thread is a good exercise how an idea, born out of wishful thinking of some guy, and posted at dpr, first becomes rumor and then becomes cited as "news" all over the web. And back-referenced by the original posters as added credential. Very funny!

As for the sensor sizes:

• FF: 36 x 24 mm, diameter 43.3 mm (crop 1.00) , surface 864 mm^2 (100%)
• APS-H (Canon 1Dmk3): 28.7 x 18.7 mm, diameter 34.3 mm (crop 1.26) , surface 537 mm^2 (62%)
• LARGE STUPID SQUARE: 24.25 x 24.25 mm, diameter 34.3 mm (crop 1.26) , surface 588 mm^2 (68%)
• STUPID SQUARE: 23.5 x 23.5 mm, diameter 33.2 mm (crop 1.30) , surface 552mm^2 (64%)
• APS-C (K10D): 23.5 x 15.7 mm, diameter 28.3 mm (crop 1.53) , surface 369 mm^2 (43%)

So, the STUPID SQUARE format is more expensive to produce than APS-H (more surface), still having a larger crop factor (narrower FoV with lenses having a larger than APS-C image circle), and most of all, still can only produce images 82% as wide as APS-H.

The LARGE STUPID SQUARE (identical crop factor to APS-H) is even more expensive to produce and still only 84% as wide.


Discussion here already concluded that a square (or octogon) sensor makes sense only if CMOS cost has become marginal when compared against lens cost. Which will happen some day. After FF sensors became commonplace, not before.


Is this enough to stop the APS-H square sensor bullshit rumor? Of course, not

Your forgetting my Small Stupid Square... 21x21
Chip chart:
http://smithsonianchips.si.edu/ice/cd/CEICM/SECTION7.pdf

[Blue]

Something tells me you really don't like squares.

He doesn't like octagons either!

[creampuff]

Use a 6cm x 6cm medium format film camera long enough and you'll know that there is invariably going to be some cropping of the image to fit most common aspect ratios (for print, media, etc.). So why even bother considering a square sensor? The Hasselblad H3 is definitely not square...

[cateto]

Oh dear, this entire thread is a good exercise how an idea, born out of wishful thinking of some guy, and posted at dpr, first becomes rumor and then becomes cited as "news" all over the web. And back-referenced by the original posters as added credential. Very funny!

Falconeye, I have no idea if all this is a reliable rumour or pure nonsense.

But mind you: filling your post with "stupid", ""bullshit" and similar words don't exactly give any further credibility to your claims.

If your reasons are strong enough to convince everybody, try to use more reasonable writing.

If not, probably is not worth to post anything, as cursing equals to lacking better skills to convince people.

Just saying.

[jay]

Your forgetting my Small Stupid Square... 21x21
Chip chart:
http://smithsonianchips.si.edu/ice/cd/CEICM/SECTION7.pdf

So now we're going SMALLER than APS-C-width? The field-of-view would be reduced, and if we keep the same pixel density (which everyone is bitching about for image quality / dynamic range purposes), we end up with LESS megapixels when we crop 3:2.

And what about lenses with a petal hood?

And who's going to fab this special chip that will only see light in ONE body?

That's prohibitively expensive.



This whole square-format rumor was started by some idiot on dpreview, and doesn't make any sense from a marketing or technological standpoint.

Will someone please explain to me how this even seems plausible? Like, even in the slightest? What is there to gain? And how do they deal with the setbacks I've mentioned? FOV, MP density, petal hoods on lenses, and exorbitant fab process price.

[ewalk]

So now we're going SMALLER than APS-C-width? The field-of-view would be reduced, and if we keep the same pixel density (which everyone is bitching about for image quality / dynamic range purposes), we end up with LESS megapixels when we crop 3:2.

Easy there...no need to raise the old blood pressure.

I don't think anyone is claiming Pentax is going smaller than APS-C width. The poster you quoted was just pointing out that there were other, smaller square sensors.

I'm pretty sure, given the use of the wink smilie, that his intentions were anything but tongue-in-cheek.

[jeffkrol]

So now we're going SMALLER than APS-C-width? The field-of-view would be reduced, and if we keep the same pixel density (which everyone is bitching about for image quality / dynamic range purposes), we end up with LESS megapixels when we crop 3:2.

And what about lenses with a petal hood?

And who's going to fab this special chip that will only see light in ONE body?

That's prohibitively expensive.



This whole square-format rumor was started by some idiot on dpreview, and doesn't make any sense from a marketing or technological standpoint.

Will someone please explain to me how this even seems plausible? Like, even in the slightest? What is there to gain? And how do they deal with the setbacks I've mentioned? FOV, MP density, petal hoods on lenses, and exorbitant fab process price.

Well a 21mm sq sensor has a surface area of 441mm squared. APS-c is 384mm squared. Rough guessitmate, keeping pixel density the same, gives you a 16MP sensor.
Increasing to 24 (sorry don't want to do the math) should give you an equivalent 14mp+ if cropped.
21mm may also fit the image circle of current lenses w/ just a tad of viginetting at wide angle large aperatures.
Samsung currently ONLY makes the CMOS for Pentax so it's not unreasonable for them to do a square sensor if yields are close to aps-c.
Pentax contribution to Sony's bottom line is probably minimal at best. Fab plants will probably be scrounging for customers. Pentax can design their own chips ect...
Increasing pixel density is not counterproductive if support circuitry is of high quality. Daylight landscapes suffer little (and actually gain) w/ higher pixel densities.
This also gets you to the same pixel count even cropped so it won't be smaller..
Matter of taste. I have nothing against higher pixel densities.
Do I want a square sensor? Doesn't really matter to me if I don't lose density in a crop.....
Anyways it IS a way to be different and do not see (if people actually buy into it) it hurting Pentax.
aps-H or square based on the current long axis WOULD hurt Pentax in all areas... I just picked 21 based on the chip yield on old fab equipment.....
IS Pentrax going this route... guess nobody but Pentax REALLY knows. I would be surprised if it wasn't at least kicked about on a saki break........
As to pro's and con's of density.................fun read
http://forums.dpreview.com/forums/read.asp?forum=1018&message=30236775&q=pixel+density&qf=m

[Torphoto]

Its BS, just like 1.3 crop rumours.

For one thing it would be narrower and deeper (if it was intended to fit in the same APSC image circle) so your wide angle lenses would not be so wide any more (it would be closer to 1.7X crop).

For another, I really dont fancy cropping every image I make when they already fit near perfectly on A4 and A3 sized paper. Rather than crop a square image to 3:2, I would be actually be easier to crop a 3:2 image to square.

Then you have all the folks who want 16:9 format!

The only reason it was used on large format cameras is that most of them could not be used in portrait mode because they were too blinking heavy.

Hey whats wrong with us who want a 16x9 crop

[falconeye]

Falconeye [...] filling your post with "stupid", ""bullshit" and similar words don't exactly give any further credibility to your claims.

I think I have a reputation for serious posting most of the time.
And when I see bullshit, most of the time, I stay silent.
But if I see bullshit and cannot stay silent, I say what I see: bullshit. Maybe, I have a clearer vision than others, maybe others have other names for bullshit. I don't know. But don't expect lies from my part. This is bullshit.

But you're right, I really should relax and ridicularize the entire discussion. So, this is my take:
It isn't true that round sensors are more difficult to produce. The wafers are round in the first place, aren't they? Just make the sensor large enough to fil lthe wafers!! This is full frame!

[jeffkrol]

I think I have a reputation for serious posting most of the time.
And when I see bullshit, most of the time, I stay silent.
But if I see bullshit and cannot stay silent, I say what I see: bullshit. Maybe, I have a clearer vision than others, maybe others have other names for bullshit. I don't know. But don't expect lies from my part. This is bullshit.

But you're right, I really should relax and ridicularize the entire discussion. So, this is my take:
It isn't true that round sensors are more difficult to produce. The wafers are round in the first place, aren't they? Just make the sensor large enough to fil lthe wafers!! This is full frame!

So you want a 100mm dia sensor? Cutting round sensors out of a larger round sensor would probably cost more..
I do have an 80mm wafer on my desk currently (scrap from ages ago) ... heck of a big sensor.