Forgot Password
Pentax Camera Forums Home
 

Reply
Show Printable Version Search this Thread
04-21-2011, 10:57 PM   #1
Pentaxian
ogl's Avatar

Join Date: Dec 2007
Location: Siberia
Photos: Gallery | Albums
Posts: 7,294
New unusual sensor from SONY - 17.7 MP at 24.3 × 12.8mm

17.7 MP (8,192 × 2,160 pixels) and 120fps, 12bit CMOS. Or 60fps with 14bit.

Sensor size 24.3 × 12.8mm (diagonal 27.5mm). The pixel pitch of 4.2 × 4.2μm.

Sensor is for Digital Photo Cameras.




ƒ\ƒj�[�A1,770–œ‰æ‘f‚Å120–‡/•b‚Ì’´�‚‘¬CMOSƒZƒ“ƒT�[‚ðŠJ”* - ƒfƒWƒJƒ�Watch

04-21-2011, 11:01 PM   #2
ogl
Pentaxian
ogl's Avatar

Join Date: Dec 2007
Location: Siberia
Photos: Gallery | Albums
Posts: 7,294
Original Poster
~ 12.3 MP in 4:3 mode (crop 2!!!) and ~ 14 MP in 3:2 mode. IMO

Last edited by ogl; 04-21-2011 at 11:19 PM.
04-21-2011, 11:09 PM   #3
Administrator
Site Webmaster
Adam's Avatar

Join Date: Sep 2006
Location: Arizona
Photos: Gallery | Albums
Posts: 43,704
Perhaps this will be used in pro video cameras (w/ downsized resolution)?

Adam
PentaxForums.com Webmaster (Site Usage Guide | Site Help | My Photography)



PentaxForums.com's high server and development costs are user-supported. You can help cover those costs by donating. Or, buy your photo gear from our affiliates, Adorama, B&H Photo, or Topaz Labs, and get FREE Marketplace access - click here to see how! Trusted Pentax retailers:

04-21-2011, 11:12 PM   #4
ogl
Pentaxian
ogl's Avatar

Join Date: Dec 2007
Location: Siberia
Photos: Gallery | Albums
Posts: 7,294
Original Poster
QuoteOriginally posted by Adam Quote
Perhaps this will be used in pro video cameras (w/ downsized resolution)?
Said it's for digital photo cameras.
http://translate.google.ru/translate?u=http%3A%2F%2Fdc.watch.impress.co.jp%2...n&hl=&ie=UTF-8

maybe for future new mirrorless cameras from Nikon and Pentax?

"This sensor is about the same size as interchangeable lens type digital camera"

"The new sensor production start date has not been disclosed. The plan as it is a product newly developed sensor had not been cleared. However, plans to deploy this technology into a product internally so there is, he said that in recent years due in various products using this technology. "This next generation technology based on its ability to offer the entire Exmor. So far in the future technology is not at all have" (Mr. Nomura)."


I think that we will see first cameras with such sensor this year.


Last edited by ogl; 04-21-2011 at 11:19 PM.
04-22-2011, 05:35 AM   #5
Veteran Member




Join Date: Jan 2011
Location: Louisville, Kentucky
Posts: 443
If rumors are correct (a big if) this could be the sensor that Nikon will be using in their rumored pro-level MILC. Rumor has the sesnor size being slightly smaller than m-4/3.

For Pentax we get the supper smaller sensor NC-1 and maybe an 1.5 cropped MLIC camera late this year. Then again the camera coming out late this year might be the pro-level traditional camera. Rumor has it that Sony will have a new 1.5 cropped sensor with 24MP and that it will be used in the rumored Nikon D400. So just maybe Pentax will release a pro-level camera using that sensor and get a 3 tier tradional camera line-up.
04-22-2011, 05:53 PM   #6
Pentaxian
er1kksen's Avatar

Join Date: Oct 2008
Location: Forestville, NY
Photos: Gallery | Albums
Posts: 1,690
I'd be thrilled to see this in the upcoming A77.
04-22-2011, 07:26 PM - 1 Like   #7
Pentaxian
dosdan's Avatar

Join Date: Dec 2007
Location: Brisbane, Australia
Photos: Gallery | Albums
Posts: 1,706
That's a 3.8:1 AR. To crop to a 3:2 AR, which would give only 3240 x 2160 = 6.7 MP (using 2^20) or 7MP using (10^6), so they're using vertical interpolation to get the greater height for the quoted 14 MP for a 3:2 AR.

I think with such a high speed that they'll be offering the blending of multiple shots. Should be possible for static stuff to increase SNR or for super-resolution , but it would smear things where part of the scene had movement.

http://en.wikipedia.org/wiki/Super-resolution
http://www.photoacute.com/studio/examples/mac_hdd/index.html
http://www.photoacute.com/tech/superresolution_faq.html
https://www.pentaxforums.com/forums/digital-processing-software-printing/7470...king-test.html

Shows that really interesting things are coming out of Sony's sensor development department.

Dan.

Last edited by dosdan; 04-22-2011 at 08:55 PM.
04-22-2011, 10:11 PM   #8
ogl
Pentaxian
ogl's Avatar

Join Date: Dec 2007
Location: Siberia
Photos: Gallery | Albums
Posts: 7,294
Original Poster
QuoteOriginally posted by dosdan Quote
That's a 3.8:1 AR. To crop to a 3:2 AR, which would give only 3240 x 2160 = 6.7 MP (using 2^20) or 7MP using (10^6), so they're using vertical interpolation to get the greater height for the quoted 14 MP for a 3:2 AR.

3:2 - it means to mask sensor from 24.3 × 12.8mm (17.7 MP) to 19.2 х 12.8 mm -
it's 14 MP.

04-22-2011, 10:25 PM   #9
Pentaxian
dosdan's Avatar

Join Date: Dec 2007
Location: Brisbane, Australia
Photos: Gallery | Albums
Posts: 1,706
QuoteOriginally posted by ogl Quote
3:2 - it means to mask sensor from 24.3 × 12.8mm (17.7 MP) to 19.2 х 12.8 mm -
it's 14 MP.
The square pixel size (I don't think it's the pixel pitch) given doesn't add up:
4.2 μm * 8192 = 34.4 mm width
4.2 μm * 2160 = 9.1 mm height

And the total pixel AR of 8192:2160 = 3.8:1
doesn't match the sensor size AR of 24.3:12.8 = 1.9:1.

Dan

Last edited by dosdan; 04-22-2011 at 11:03 PM.
04-22-2011, 10:58 PM   #10
Pentaxian
dosdan's Avatar

Join Date: Dec 2007
Location: Brisbane, Australia
Photos: Gallery | Albums
Posts: 1,706
QuoteOriginally posted by dosdan Quote
The square pixel size (I don't think it's the pixel pitch) given doesn't add up: 4.2 μm * 8192 = 34.4 mm width 4.2 μm * 2160 = 9.1 mm height And the total pixel AR of 8192:2160 = 3.8:1 doesn't match the sensor size AR of 24.3:12.8 = 1.9:1. So it looks like the count of width pixels may have been doubled up from 4096.

The original Japanese article mentions a 45° pixel orientation. So the H x W dimensions would be:

4.2 μm * cos 45° * 8192 = 24.3 mm width
4.2 μm * sin 45° * 2160 = 6.4 mm height

So that explains the sensor width, but the height is only one-half what is quoted, so it's still an 3.8:1 AR vs. 1.9:1 AR.

Dan.

Last edited by dosdan; 04-23-2011 at 03:40 AM.
04-22-2011, 11:17 PM - 1 Like   #11
Pentaxian
dosdan's Avatar

Join Date: Dec 2007
Location: Brisbane, Australia
Photos: Gallery | Albums
Posts: 1,706
QuoteOriginally posted by dosdan Quote
4.2 μm * cos 45° * 8192 = 24.3 mm width
4.2 μm * sin 45° * 2160 = 6.4 mm height

So that explains the sensor width, but the height is only one-half what is quoted, so it's still an 3.8:1 AR vs. 1.9:1 AR.
What could be happening is that, in a vert. column, only every 2nd pixel is read in a pass (for speed?), maybe alternating between passes, so that would mean that out of 5120 pixels in a column, 2160 effective pixels are read at a time.

4.2 μm * sin 45° * 5120 = 12.8 mm height

Dan
04-22-2011, 11:27 PM   #12
Pentaxian
dosdan's Avatar

Join Date: Dec 2007
Location: Brisbane, Australia
Photos: Gallery | Albums
Posts: 1,706
Another thought. Here is a translation from part of the original article:

First "Dual Row Readout" The photodiode signals when the ADC transfer counter, it will read the two lines approach. Common Sensor was to scan the existing line by line. This image area is divided into upper and lower areas of the upper half of the ADC is placed on the columns, as well as the lower half of the area was placed under the column ADC signals simultaneously. Half the time was now transferred from the photodiode.


No sure what this means. Perhaps out of column of 5120 sensels either:

1. Every 2nd sensel is read at a time (as discussed in my previous message).

2. Or that two sensels (one above the other) are read at one time and combined, so giving a resultant 2:1 rectangular pixel out of 2 square sensels. (Actually, considering the 45° sensel orientation, that would be a double-diamond-shaped pixel from two diamond-shaped sensels.)

Both would also explain the quoted figures.

Dan.

Last edited by dosdan; 04-23-2011 at 03:55 AM.
04-23-2011, 12:42 AM   #13
Pentaxian
RonHendriks1966's Avatar

Join Date: Jan 2009
Photos: Gallery
Posts: 8,714
That is a strange thing. One new sensorsize that is to be expected is new movie-wide-screen size. This should give HD-wide movie at 2560 x 1080 pixels wich is 2,37 : 1 size.
04-23-2011, 01:18 AM   #14
ogl
Pentaxian
ogl's Avatar

Join Date: Dec 2007
Location: Siberia
Photos: Gallery | Albums
Posts: 7,294
Original Poster
QuoteOriginally posted by dosdan Quote
What could be happening is that, in a vert. column, only every 2nd pixel is read in a pass (for speed?), maybe alternating between passes, so that would mean that out of 5120 pixels in a column, 2160 effective pixels are read at a time.

4.2 μm * sin 45° * 5120 = 12.8 mm height

Dan
Insane....But your calculation is logical.
04-23-2011, 01:21 AM   #15
Pentaxian
dosdan's Avatar

Join Date: Dec 2007
Location: Brisbane, Australia
Photos: Gallery | Albums
Posts: 1,706
The illustration in the original article shows a very wide AR, but that may be an artist's impression, rather than from Sony.

The definitive question is: "Is this sensor's AR 3:8.1 or 1.9:1?" I suspect it's 1.9:1. That way, a crop to 3:2 wouldn't lose much MP, but that would require non-square pixels.

If it's using non-square pixels, this is unusual in DSLRs. The Nikon D1 was a camera where the number of sensels was much greater (4x) than the pixel count in the output image. See the Feb 2006 addendum to http://www.lonestardigital.com/techcorner.htm. The Nixon D1x had a very strange translation of its sensels to image pixels: 4024 x 1324 -> 3008 x 1960. http://www.robgalbraith.com/bins/content_page.asp?cid=7-4833-4834

If it's reading every 2nd sensel in a vert. column, this would halve the possible QE because only half of the usable sensor area is being used to capture photons at a time.

Edit: concerning method 1, it wouldn't halve the QE at all when reading every 2nd vert. sensel, if the new value for the 2nd sensel was combined with the held value of the 1st sensel to update the contents of the composite pixel.

Next pass, the 1st sensel's new value would be combined with the 2nd sensel's held value to refresh the composite pixel again. So the composite pixel could be refreshed 120 times a sec, but with temporal interpolation, the "rise-time" is slower. It's a bit like interleaving-with-interpolation in video.


Both method 1 & 2 would satisfy the 24.3mm × 12.8mm sensor size with 3.8:1 total pixel layout AR, but with an 1.9:1 image AR due to non-square pixels (1:2 AR). (Some video formats use non-square pixels.)

Dan

Last edited by dosdan; 04-23-2011 at 01:14 PM.
Reply

Bookmarks
  • Submit Thread to Facebook Facebook
  • Submit Thread to Twitter Twitter
  • Submit Thread to Digg Digg
Tags - Make this thread easier to find by adding keywords to it!
mp, pentax news, pentax rumors, sensor, sensor from sony
Thread Tools Search this Thread
Search this Thread:

Advanced Search


Similar Threads
Thread Thread Starter Forum Replies Last Post
K-5 uses Aptina sensor rather than Sony?? kales Pentax News and Rumors 20 11-24-2010 08:58 AM
Sony outsources some sensor production to Fujitsu deadwolfbones Pentax News and Rumors 10 10-02-2010 03:24 PM
K-x Sony Sensor vs K-7 Samsung Sensor karl79 Video and Pentax HDSLRs 9 09-23-2010 09:35 AM
New Sony sensor SNR on par with D3 FF! JohnBee Pentax News and Rumors 38 09-19-2010 05:19 PM
Will Sony supply the Sensor for future bodies? Reportage Pentax News and Rumors 10 06-22-2010 06:47 PM



All times are GMT -7. The time now is 08:03 AM. | See also: NikonForums.com, CanonForums.com part of our network of photo forums!
  • Red (Default)
  • Green
  • Gray
  • Dark
  • Dark Yellow
  • Dark Blue
  • Old Red
  • Old Green
  • Old Gray
  • Dial-Up Style
Hello! It's great to see you back on the forum! Have you considered joining the community?
register
Creating a FREE ACCOUNT takes under a minute, removes ads, and lets you post! [Dismiss]
Top