Why not a 21.6mm x 21.6mm sensor?

Started Jan 23, 2013 | Discussions
Shop cameras & lenses ▾
Great Bustard Forum Pro • Posts: 35,667
Why not a 21.6mm x 21.6mm sensor?
4

The entire image circle for mFT lenses could fit within such a sensor.  While such a sensor would be 25% larger than an APS-C sensor (468 mm² vs 372 mm²), it would still be barely more than half the size of a FF sensor, so I don't imagine the cost of the sensor itself would be an issue.

The advantages of such a sensor, however, are substantial:  you would not need to rotate the camera from landscape to portrait orientation (especially convenient if using on-camera bounced flash) and you could crop to any aspect ratio you like with no IQ penalty.

So, what are the downsides that negate the upsides?

Klarno
Klarno Veteran Member • Posts: 4,111
Re: Why not a 21.6mm x 21.6mm sensor?
1

Great Bustard wrote:

The entire image circle for mFT lenses could fit within such a sensor. While such a sensor would be 25% larger than an APS-C sensor (468 mm² vs 372 mm²), it would still be barely more than half the size of a FF sensor, so I don't imagine the cost of the sensor itself would be an issue.

The advantages of such a sensor, however, are substantial: you would not need to rotate the camera from landscape to portrait orientation (especially convenient if using on-camera bounced flash) and you could crop to any aspect ratio you like with no IQ penalty.

So, what are the downsides that negate the upsides?

Lens hoods, especially petal or square lens hoods, are effectively configured for one aspect ratio; they should have some leeway, but going to a sensor shape that is different by that much you might get vignetting when trying to use supplied lens hoods at the lens' wider focal lengths (or only focal length, in the case of primes).

-- hide signature --
 Klarno's gear list:Klarno's gear list
Olympus PEN E-PM2 A3000 Olympus E-M1 Olympus Zuiko Digital 14-54mm 1:2.8-3.5 II Olympus Zuiko Digital 11-22mm 1:2.8-3.5 +4 more
MaxB1 Contributing Member • Posts: 899
Re: Why not a 21.6mm x 21.6mm sensor?
1

Having worked in the Semiconductor Industry (Eng'r Mgr) for 40+ years, I think that mfg. COST** of a larger sensor is the biggest problem followed by increased software issues. Also, the internal needs of the camera would be different.

** Mfg cost includes the decreased number of sensors on a wafer as well as the yield (which decreases dramatically with chip size.)  After production testing also increases with chip size.

-- hide signature --
 MaxB1's gear list:MaxB1's gear list
Olympus OM-D E-M5 Olympus Zuiko Digital ED 12-60mm 1:2.8-4.0 SWD Olympus Zuiko Digital ED 50-200mm 1:2.8-3.5 SWD Olympus Zuiko Digital ED 7-14mm 1:4.0 Olympus Zuiko Digital 1.4x Teleconverter EC-14 +8 more
s_grins
s_grins Forum Pro • Posts: 11,017
Re: Why not a 21.6mm x 21.6mm sensor?
1

Why not? because it is square.

Disk would better match image circle, disk  would be logical,  disk would be cheaper, disk idea is 3 years old



-- hide signature --

Looking for equilibrium...

 s_grins's gear list:s_grins's gear list
Panasonic Lumix DMC-GH2 Panasonic Lumix G 20mm F1.7 ASPH Sigma 30mm F2.8 EX DN Sigma 60mm F2.8 DN Art Panasonic Lumix G Vario 14-140mm F3.5-5.6 ASPH. / Power O.I.S +2 more
Great Bustard OP Forum Pro • Posts: 35,667
Re: Why not a 21.6mm x 21.6mm sensor?

s_grins wrote:

Why not? because it is square.

Disk would better match image circle, disk would be logical, disk would be cheaper, disk idea is 3 years old

Don't the disks have to be cut from a square wafer?  So why not save the cutting and just put in the square sensor the circumscribes the image circle?



Great Bustard OP Forum Pro • Posts: 35,667
Interesting angle.

Klarno wrote:

Great Bustard wrote:

The entire image circle for mFT lenses could fit within such a sensor. While such a sensor would be 25% larger than an APS-C sensor (468 mm² vs 372 mm²), it would still be barely more than half the size of a FF sensor, so I don't imagine the cost of the sensor itself would be an issue.

The advantages of such a sensor, however, are substantial: you would not need to rotate the camera from landscape to portrait orientation (especially convenient if using on-camera bounced flash) and you could crop to any aspect ratio you like with no IQ penalty.

So, what are the downsides that negate the upsides?

Lens hoods, especially petal or square lens hoods, are effectively configured for one aspect ratio; they should have some leeway, but going to a sensor shape that is different by that much you might get vignetting when trying to use supplied lens hoods at the lens' wider focal lengths (or only focal length, in the case of primes).

Then again, how hard could it be to redesign some hoods and sell them for $20 each?

Great Bustard OP Forum Pro • Posts: 35,667
So...

MaxB1 wrote:

Having worked in the Semiconductor Industry (Eng'r Mgr) for 40+ years, I think that mfg. COST** of a larger sensor is the biggest problem followed by increased software issues. Also, the internal needs of the camera would be different.

...you think the cost of sensor 25% larger than APS-C would be the largest problem?  I wonder what an APS-C sensor costs compared to an mFT sensor, as APS-C sensors are 65% larger.

** Mfg cost includes the decreased number of sensors on a wafer as well as the yield (which decreases dramatically with chip size.) After production testing also increases with chip size.

Promit Senior Member • Posts: 1,989
Re: Interesting angle.

Great Bustard wrote:

Then again, how hard could it be to redesign some hoods and sell them for $20 each?

Ask Olympus...or try that on the 7-14 or the 7.5, let me know how that works out. At least none of the lenses seem to have internal/rear hoods.

 Promit's gear list:Promit's gear list
Sony a77 II Sony Alpha 7R II Sony 24-70mm F2.8 ZA SSM Carl Zeiss Vario-Sonnar T* Tokina AT-X Pro 11-16mm f/2.8 DX Sony DT 35mm F1.8 SAM +6 more
reygon Senior Member • Posts: 2,037
Re: So...

Great Bustard wrote:

MaxB1 wrote:

Having worked in the Semiconductor Industry (Eng'r Mgr) for 40+ years, I think that mfg. COST** of a larger sensor is the biggest problem followed by increased software issues. Also, the internal needs of the camera would be different.

...you think the cost of sensor 25% larger than APS-C would be the largest problem? I wonder what an APS-C sensor costs compared to an mFT sensor, as APS-C sensors are 65% larger.

** Mfg cost includes the decreased number of sensors on a wafer as well as the yield (which decreases dramatically with chip size.) After production testing also increases with chip size.

So... It's VERY expensive, complex and requires time from the initial set-up and then production costs. You are technically introducing a new standard as it will impact not only hardware but also software and peripherals (lens & etc.). In my previous work as product engineer for wiring harnesses for Ford and GM cars, it takes  2-3 years from design to prototype work before 'green light' for first production.

we haven't even discussed R&D and registering/approval of patent. How many companies worked and agreed together before micro 4/3rds became a reality?

Soooo.... your reasoning and justifications are very minute and not financially logical on the overall scheme. Personally I think 3D and better phones sensors are the commercially and viable investments. 3D lenses for camera are slowly coming our.

-- hide signature --

reygon
----------------------------------------------------------------
Take nothing but photos... Kill nothing but time... Leave nothing but footprints...

zxaar Veteran Member • Posts: 3,970
Re: Why not a 21.6mm x 21.6mm sensor?
1

Great Bustard wrote:

The entire image circle for mFT lenses could fit within such a sensor. While such a sensor would be 25% larger than an APS-C sensor (468 mm² vs 372 mm²), it would still be barely more than half the size of a FF sensor, so I don't imagine the cost of the sensor itself would be an issue.

The advantages of such a sensor, however, are substantial: you would not need to rotate the camera from landscape to portrait orientation (especially convenient if using on-camera bounced flash) and you could crop to any aspect ratio you like with no IQ penalty.

So, what are the downsides that negate the upsides?

To accomodate square of 17.3x17.3 you will need a circle of dia 25mm or so.  I believe the image circle for mft is 21.6 mm dia.

-- hide signature --

::> Knowledge is mother of efficiency.

ohmydentist Senior Member • Posts: 1,933
Re: Why not a 21.6mm x 21.6mm sensor?

4/3 is more suare than 3/2, why not go one step further and make it a square.  I'm all for it.  You can say goodbye to those vertical grip.

-- hide signature --

^__^
Just Shoot !

Detail Man
Detail Man Forum Pro • Posts: 16,666
Re: Why not a 21.6mm x 21.6mm sensor ?

zxaar wrote:

Great Bustard wrote:

The entire image circle for mFT lenses could fit within such a sensor. While such a sensor would be 25% larger than an APS-C sensor (468 mm² vs 372 mm²), it would still be barely more than half the size of a FF sensor, so I don't imagine the cost of the sensor itself would be an issue.

The advantages of such a sensor, however, are substantial: you would not need to rotate the camera from landscape to portrait orientation (especially convenient if using on-camera bounced flash) and you could crop to any aspect ratio you like with no IQ penalty.

So, what are the downsides that negate the upsides?

To accomodate square of 17.3x17.3 you will need a circle of dia 25mm or so. I believe the image circle for mft is 21.6 mm dia.

Bingo ! ...

Lights
Lights Veteran Member • Posts: 3,558
Re: Why not a 21.6mm x 21.6mm sensor?

It would certainly be nice for square format (which I kinda like) sort of like a modern Rolleiflex.

-- hide signature --

My Gallery is here -
http://www.pbase.com/madlights
Why so serious? :The Joker

Great Bustard OP Forum Pro • Posts: 35,667
Re: So...
1

reygon wrote:

Great Bustard wrote:

MaxB1 wrote:

Having worked in the Semiconductor Industry (Eng'r Mgr) for 40+ years, I think that mfg. COST** of a larger sensor is the biggest problem followed by increased software issues. Also, the internal needs of the camera would be different.

...you think the cost of sensor 25% larger than APS-C would be the largest problem? I wonder what an APS-C sensor costs compared to an mFT sensor, as APS-C sensors are 65% larger.

** Mfg cost includes the decreased number of sensors on a wafer as well as the yield (which decreases dramatically with chip size.) After production testing also increases with chip size.

So... It's VERY expensive, complex and requires time from the initial set-up and then production costs. You are technically introducing a new standard as it will impact not only hardware but also software and peripherals (lens & etc.). In my previous work as product engineer for wiring harnesses for Ford and GM cars, it takes 2-3 years from design to prototype work before 'green light' for first production.

we haven't even discussed R&D and registering/approval of patent. How many companies worked and agreed together before micro 4/3rds became a reality?

Soooo.... your reasoning and justifications are very minute and not financially logical on the overall scheme. Personally I think 3D and better phones sensors are the commercially and viable investments. 3D lenses for camera are slowly coming our.

The way I was thinking is that there is a whole slew of different sized sensors for compacts, there are two different sizes for APS-C, the mFT (4/3) sensors, and now the new 1" (2.7x) sensors, so it sure doesn't seem like making a new sensor size has presented any difficulties.

I mean, what makes 21.6mm x 21.6mm (the sensor I am proposing) any more difficult to get going than the 1" sensor Nikon uses in their 1 series and Sony uses in the RX100?  Both those systems seem to be rather successful, no?

Great Bustard OP Forum Pro • Posts: 35,667
Re: Interesting angle.

Promit wrote:

Great Bustard wrote:

Then again, how hard could it be to redesign some hoods and sell them for $20 each?

Ask Olympus...or try that on the 7-14 or the 7.5, let me know how that works out.

One would think the wouldn't mind making $19 profit off of a $20 piece of plastic so that people could take full advantage of their new oversized sensor.  Then again...

At least none of the lenses seem to have internal/rear hoods.

Yes -- that would have been a bigger issue, for sure.

Great Bustard OP Forum Pro • Posts: 35,667
To be clear...
1

Lights wrote:

It would certainly be nice for square format (which I kinda like) sort of like a modern Rolleiflex.

...it's not a square format that I am proposing.  It's a square sensor that the entire image circle is inscribed in.  So, the camera records the whole of the image circle, which the photographer can crop (or not) to whatever aspect ratio they choose without any penalty.

micksh6
micksh6 Senior Member • Posts: 2,613
Re: So...

Great Bustard wrote:

MaxB1 wrote:

Having worked in the Semiconductor Industry (Eng'r Mgr) for 40+ years, I think that mfg. COST** of a larger sensor is the biggest problem followed by increased software issues. Also, the internal needs of the camera would be different.

...you think the cost of sensor 25% larger than APS-C would be the largest problem? I wonder what an APS-C sensor costs compared to an mFT sensor, as APS-C sensors are 65% larger.

Yes, cost. The thing is sensor cost grows exponentially with sensor size. Thom Hogan stated this, for example, and the reasons were exactly what MaxB1 provided above.

APS-C sensor won't cost twice more than m4/3 sensor, it will cost 5-10 times more, so economically this doesn't make any sense. It's better to make APS-C sensor for that money.

If there was any practical sense in doing it, this would have appeared on smaller sensor cameras, but it hasn't. It's more practical to rotate the camera.

Because of that, I think, it's safe to forget about square m4/3 sensor.

 micksh6's gear list:micksh6's gear list
Olympus PEN E-PL5 Olympus E-M1 Olympus M.Zuiko Digital ED 9-18mm F4.0-5.6 Panasonic Leica Summilux DG 25mm F1.4 Olympus M.Zuiko Digital ED 40-150mm 1:4-5.6 R +6 more
Lights
Lights Veteran Member • Posts: 3,558
Re: To be clear...
1

Right I realize that. I see the advantages, and all. But it still would give a large image circle (and a higher resolution at square in comparison with a rectangular sensor of the same design) but square would sort of be it's native format?...which I would like. Why not use the entire image circle in that case. It might cost slightly more. Existing OEM lens hoods might be problematical, as has been mentioned, but there are lots of round ones (Leica style, rubber, WA metal etc.) I would like the idea. No need for a vertical grip to add size and bulk (Like on a couple of my other cameras), whatever format a person wants, without holding at an odd angle..if the LCD screen was a simple tilt screen it would would work fine. Even the position of the viewfinder wouldn't have to be adjusted for verticals. When I have used vertical grips it de-contributes to the ergonomics, almost like having to learn to instinctively use two cameras, so would sure help in that regard.

-- hide signature --

My Gallery is here -
http://www.pbase.com/madlights
Why so serious? :The Joker

MatsP
MatsP Senior Member • Posts: 1,664
Re: Why not a 21.6mm x 21.6mm sensor?

Lights wrote:

It would certainly be nice for square format (which I kinda like) sort of like a modern Rolleiflex.

-- hide signature --

My Gallery is here -
http://www.pbase.com/madlights
Why so serious? :The Joker

Rather a modern Hasselblad! (I'm a swede). I think a square sensor is a nice idea, and a square format. 4/3 is so close to 1/1 anyway. The sensor corners could be a little bit outside the image circle but not that big as 21,6x21,6. You could choose between square, portrait and landscape afterwards instead of turning the camera 45 degrees. Or choose format with a button on the camera. Like you could with the film APS format, if someone remember that.

 MatsP's gear list:MatsP's gear list
Olympus OM-D E-M5 Olympus E-M1 Olympus M.Zuiko Digital ED 40-150mm 1:4-5.6 R Samyang 7.5mm F3.5 UMC Fisheye MFT Olympus 12-40mm F2.8 +1 more
reygon Senior Member • Posts: 2,037
Re: So...

Great Bustard wrote:

reygon wrote:

Great Bustard wrote:

MaxB1 wrote:

Having worked in the Semiconductor Industry (Eng'r Mgr) for 40+ years, I think that mfg. COST** of a larger sensor is the biggest problem followed by increased software issues. Also, the internal needs of the camera would be different.

...you think the cost of sensor 25% larger than APS-C would be the largest problem? I wonder what an APS-C sensor costs compared to an mFT sensor, as APS-C sensors are 65% larger.

** Mfg cost includes the decreased number of sensors on a wafer as well as the yield (which decreases dramatically with chip size.) After production testing also increases with chip size.

So... It's VERY expensive, complex and requires time from the initial set-up and then production costs. You are technically introducing a new standard as it will impact not only hardware but also software and peripherals (lens & etc.). In my previous work as product engineer for wiring harnesses for Ford and GM cars, it takes 2-3 years from design to prototype work before 'green light' for first production.

we haven't even discussed R&D and registering/approval of patent. How many companies worked and agreed together before micro 4/3rds became a reality?

Soooo.... your reasoning and justifications are very minute and not financially logical on the overall scheme. Personally I think 3D and better phones sensors are the commercially and viable investments. 3D lenses for camera are slowly coming our.

The way I was thinking is that there is a whole slew of different sized sensors for compacts, there are two different sizes for APS-C, the mFT (4/3) sensors, and now the new 1" (2.7x) sensors, so it sure doesn't seem like making a new sensor size has presented any difficulties.

I mean, what makes 21.6mm x 21.6mm (the sensor I am proposing) any more difficult to get going than the 1" sensor Nikon uses in their 1 series and Sony uses in the RX100? Both those systems seem to be rather successful, no?

You said "so it sure doesn't seem like making a new sensor size has presented any difficulties" and even comparing other sensor sizes so kindly read again if I said "it's difficult". What I emphasized is it's not commercially logical (cost). As I said, you are introducing a new standard with odd size which is not too far from APS-C (Canon or Nikon size) except it's square. Hey, why not produce a circle sensor? it compliments the lens right but hey it will be more expensive to produce it (beside new production line there will be considerable wasted material from silicon wafers).

For the output and viewing, are you saying it's also better to view in square screens? The film academy, who also uses same camera and lenses principles, came up with 4:3 being the ideal ratio for projecting on a screen. It used most of the image circle a lens produced. It was known as the Academy Ratio. Ernst Leitz doubled the size for his 35mm (movie film) still camera. It has a ratio of 2:3 (still not square!!!) which are basis for APS chips so it still maximize the use of image circle a lens produced. Olympus, Panasonic and 4/3rds system is based on the Academy Ratio. Let's not even discuss printing square photos. So... it's not just engineering it's also complimenting the overall industry. I'm sure the 'beautiful and great minds' out there in the R&D considered and evaluated these so many times before as the camera and film industry progress to a million then to a billion dollar industry.

reygon
----------------------------------------------------------------
Take nothing but photos... Kill nothing but time... Leave nothing but footprints...

Keyboard shortcuts:
FForum MMy threads