Jump to content

  •  

CNers have asked about a donation box for Cloudy Nights over the years, so here you go. Donation is not required by any means, so please enjoy your stay.

Photo

Sony IMX294 STARVIS 8MP 4/3 sensor

  • Please log in to reply
67 replies to this topic

#1 Sigerson

Sigerson

    Lift Off

  • -----
  • topic starter
  • Posts: 19
  • Joined: 28 Mar 2017

Posted 08 April 2017 - 02:13 PM

It might be worth keeping an eye out for the new Sony IMX294 STARVIS 8MP 4/3 sensor if it comes to fruition.  I have yet to find a data sheet but it is mentioned as a new sensor in this Sony technology brochure (page 2).  

 

http://www.sony-semi...tail/170301.pdf

 

https://www.ephotozi...rom-sony--30796

 

This mid-sized CMOS sensor will offer larger pixels in a moderate resolution with BSI technology boasting high dynamic range (HDR).  It might hit a sweet spot for fast optical systems with larger image circles.  Maybe astro camera manufacturers will give this sensor consideration for new EAA cameras.

 


#2 dwkdnvr

dwkdnvr

    Vostok 1

  • *****
  • Posts: 104
  • Joined: 09 Jul 2012
  • Loc: SE of Denver

Posted 08 April 2017 - 07:57 PM

this could be interesting. I always thought that a 4/3 version of the sensor in the A7S would be an interesting idea, but that would be a 3MP sensor and that's probably not gonna fly in a 4k world. It'll definitely be interesting to see whether ZWO picks this one up. I'm not entirely sure that the market is big enough to support this as an alternative to the 1600, but if it's more sensitive it might be an interesting tradeoff. (assuming a mono version, that is - article wasn't clear on that point)



#3 Relativist

Relativist

    Cosmos

  • *****
  • Posts: 8151
  • Joined: 11 Oct 2003
  • Loc: OC, CA, USA

Posted 08 April 2017 - 09:17 PM

So what is the pixel size?



#4 ccs_hello

ccs_hello

    Voyager 1

  • *****
  • Posts: 10237
  • Joined: 03 Jul 2004

Posted 08 April 2017 - 09:38 PM

Should be around 5.4um by 5.4um or so.

With BSI, it's going to be very good.

(And SONY will keep us waiting for 18 months for real production truly happening, and it's going to be on allocation for volume buyers only...

hint, hint: mirrorless camera mfgs.)



#5 Relativist

Relativist

    Cosmos

  • *****
  • Posts: 8151
  • Joined: 11 Oct 2003
  • Loc: OC, CA, USA

Posted 08 April 2017 - 09:56 PM

Should be around 5.4um by 5.4um or so.

With BSI, it's going to be very good.

(And SONY will keep us waiting for 18 months for real production truly happening, and it's going to be on allocation for volume buyers only...

hint, hint: mirrorless camera mfgs.)

Then it should be comparable to the A7r II in sensitivity. Yes it would be worth exploring for an EAA camera, especially if we can bin. BSI only buys so much, it won't reach A7s levels on just based on that (higher QE might help though if it has it). Something to look forward to next year. 



#6 ccs_hello

ccs_hello

    Voyager 1

  • *****
  • Posts: 10237
  • Joined: 03 Jul 2004

Posted 09 April 2017 - 07:17 AM

Think this way:

 

Popular IMX224 has 3.75um pixel pitch, 1.23 M pixel, and it's not even BSI.

 

A7R II sensor has 4.5um approx. pitch.

 

A7S sensor has 8.4 um pitch.

 

Real estate property is at a premium.


Edited by ccs_hello, 09 April 2017 - 07:19 AM.


#7 Relativist

Relativist

    Cosmos

  • *****
  • Posts: 8151
  • Joined: 11 Oct 2003
  • Loc: OC, CA, USA

Posted 09 April 2017 - 01:32 PM

Right, the A7r II is also BSI, so hence we could reasonably speculate close performance given the close pixel pitch size. Other aspects of the sensors will make some difference as well.

#8 Astrojedi

Astrojedi

    Soyuz

  • *****
  • Posts: 3785
  • Joined: 27 May 2015
  • Loc: SoCal

Posted 10 April 2017 - 12:08 AM

Based on what I know this is going to be a very good sensor for Astro/EAA...



#9 Sigerson

Sigerson

    Lift Off

  • -----
  • topic starter
  • Posts: 19
  • Joined: 28 Mar 2017

Posted 10 April 2017 - 09:34 AM

Based on what I know this is going to be a very good sensor for Astro/EAA...

 

 

Is there anything you can add without violating your non-disclosure agreement?   wink.gif

 

How about aspect ratio – 4:3 or 16:9?  (I’m betting it is 16:9 like a larger IMX274 because people just have to have their 4K video these days)

 

Image diagonal close to 21.6mm?

 

Pixel size so we can start running the numbers?



#10 Relativist

Relativist

    Cosmos

  • *****
  • Posts: 8151
  • Joined: 11 Oct 2003
  • Loc: OC, CA, USA

Posted 10 April 2017 - 10:25 AM

The pixel size can be approximated well simply knowing the sensor size format and resolution.

#11 Astrojedi

Astrojedi

    Soyuz

  • *****
  • Posts: 3785
  • Joined: 27 May 2015
  • Loc: SoCal

Posted 11 April 2017 - 11:20 PM

 

Based on what I know this is going to be a very good sensor for Astro/EAA...

 

 

Is there anything you can add without violating your non-disclosure agreement?   wink.gif

 

How about aspect ratio – 4:3 or 16:9?  (I’m betting it is 16:9 like a larger IMX274 because people just have to have their 4K video these days)

 

Image diagonal close to 21.6mm?

 

Pixel size so we can start running the numbers?

 

Unfortunately I cannot provide any more detail at this time.



#12 XS_Man

XS_Man

    Vostok 1

  • -----
  • Posts: 179
  • Joined: 20 Jul 2016
  • Loc: FRANCE

Posted 13 April 2017 - 11:19 AM

Hi all ! wink.gif

 

As I you guys, I have great expectations for this sensor.

Both for planetary/lunar imaging and short exposure deep sky imaging.

 

I don't know if someone is aware and able to reply me ? I wonder if Sony has planed

only color version or if a monochrome will be proposed ?

 

Sony announced an other Sensor, IMX 390, but I don't understand why they made a so stupid

choice, it's a joke !

 

http://image-sensors...ensor-with.html

 

To summarize, it's an IMX 290 a little bit larger (2.45 vs 2.1 Mpix) with anti-flickering system.

No interest for astronomy use...

In fact there's no real differences between these 2 sensors ! A kind of clone, not more.

 

I'm desperatly  waiting for a big brother of IMX 290...

 

Albéric



#13 Sigerson

Sigerson

    Lift Off

  • -----
  • topic starter
  • Posts: 19
  • Joined: 28 Mar 2017

Posted 13 April 2017 - 12:22 PM

I think I may have found some specifications on the Sony IMX294 from what I would consider a fairly reliable source - FRAMOS. 

 

I expected there were three main possibilities as to how this image sensor would be designed.  I had previously mentioned two of the three.  I did not mention the third as I thought it was the least likely and the least desirable.  But strangely, the design doesn’t even match what I thought was the third possibility.  I really hope this information is not correct, but it sure looks like the real deal.

 

Quick summary of the three possibilities I envisioned:

 

1. 4:3 aspect ratio and 8MP resolution giving the largest pixel size around 5.3um to 5.4um.  This is what I think most of us would like to see.

 

2. 16:9 aspect ratio with 8.3MP resolution to support the UHD-1 4K television video standard of 3840 x 2160 pixels.  This has precedence with the IMX274 sensor shown in the same STARVIS chart.  The IMX294 would basically have been a scaled-up version of the IMX274.  The pixel size would have been around 4.9um which is 9.26% smaller than 5.4um.

 

3. A combination of 1 and 2 above - using the 4:3 aspect ratio AND supporting 4K video.  This is done by significantly increasing the number of pixels to at least 11.06MP so the sensor can be cropped to 3840 x 2160 pixels for 4K video.  This reduces the all-important pixel size to 4.5um which is 16.67% smaller than the 5.4um it could have been.

 

 

But the apparent specifications of the IMX294 sensor show something else.  It looks like they stopped a little short of supporting the 4K video standard.  For me, the question is why?  Maybe it has something to do with a limitation or efficiency of the STARVIS process.  Maybe someone who has experience in image sensor design and/or has knowledge of the design requirements for this particular sensor can shed light on this decision.

 

Don’t get me wrong, this is still a cool sensor and it might be fantastic for EAA cameras.  I just think it could have been better for EAA and I wonder why it was designed like this.  Now that I have the array and pixel size I am going to play with various optical systems using this sensor in CCDCalc.

 

If I haven’t bored you to death and you are still with me, here is the link to the specifications I found for the color version of the IMX294.

 

 

https://www.framos.c...jk-c-19482.html


  • mclewis1 likes this

#14 Relativist

Relativist

    Cosmos

  • *****
  • Posts: 8151
  • Joined: 11 Oct 2003
  • Loc: OC, CA, USA

Posted 13 April 2017 - 12:34 PM

Sony does not make any sensor for EAA, our use case is a niche and secondary market.
  • OleCuss likes this

#15 Sigerson

Sigerson

    Lift Off

  • -----
  • topic starter
  • Posts: 19
  • Joined: 28 Mar 2017

Posted 13 April 2017 - 01:25 PM

Sony does not make any sensor for EAA, our use case is a niche and secondary market.

 

 

I am not sure if that remark was aimed at me or at XS Man, but at no point did I say that Sony has made any sensors for the tiny EAA market. 



#16 mclewis1

mclewis1

    Thread Killer

  • *****
  • Posts: 18674
  • Joined: 25 Feb 2006
  • Loc: New Brunswick, Canada

Posted 13 April 2017 - 02:27 PM

4:3 ratio with square pixels will be a benefit for us. This means the sensor won't be highly elongated (as a 16:9 aspect ratio layout would) and that means less pressure on focal reducers ... although you certainly aren't going to be able to use the less expensive small diameter ones with this sensor.


  • OleCuss likes this

#17 XS_Man

XS_Man

    Vostok 1

  • -----
  • Posts: 179
  • Joined: 20 Jul 2016
  • Loc: FRANCE

Posted 13 April 2017 - 03:37 PM

I agree with Curtis, astronomy market doesn't exist for Sony, main production are sensors
for smartphones ! After, far away comes surveillance/industrial vision markets.
And flickering of road signs is a so important problem in life that a sensor is designed
to solve it ! It makes me laugh (and cry...).

 

Thanks for Framos link, 4.63 microns pixels size remains an interesting size. I guess that
it's a reliable information, Framos is a well known professionnal supplier for sensors.
What would be the interest for them to give wrong information to customers ?

 

So in conclusion, we have to wait for the official datasheet of IMX294 from Sony...

Albéric


  • Sigerson likes this

#18 Relativist

Relativist

    Cosmos

  • *****
  • Posts: 8151
  • Joined: 11 Oct 2003
  • Loc: OC, CA, USA

Posted 13 April 2017 - 05:38 PM


Sony does not make any sensor for EAA, our use case is a niche and secondary market.



I am not sure if that remark was aimed at me or at XS Man, but at no point did I say that Sony has made any sensors for the tiny EAA market.

No worries, It's a general comment/reminder, if I were to aim it at anyone I would say their name.
  • will w and Sigerson like this

#19 ccs_hello

ccs_hello

    Voyager 1

  • *****
  • Posts: 10237
  • Joined: 03 Jul 2004

Posted 13 April 2017 - 06:17 PM

Just to copy down on what's known now

 

IMX294CKJ

10.71 MP 

3792 x 2824 pixel pitch 4.63um x 4.63um

effective image area  17.56mm x 13.08mm (aspect ratio: 4:3)

BSI type and meets the Starvis grade.  It is ERS (electronic rolling shutter), not the "global shutter" type.

 

4K UHD (16:9 aspect ratio <-- fits most of the modern monitor) is 3840x2160.

IMX294's horizontal pixel count is slightly shy of 3840.  Usually that is not a problem, just two thin black vertical bars on left and right edge. You'll not notice them easily.

 

Clear Skies!

 

ccs_hello 


  • Sigerson likes this

#20 ccs_hello

ccs_hello

    Voyager 1

  • *****
  • Posts: 10237
  • Joined: 03 Jul 2004

Posted 13 April 2017 - 06:25 PM

Hi all ! wink.gif

 

As I you guys, I have great expectations for this sensor.

Both for planetary/lunar imaging and short exposure deep sky imaging.

 

I don't know if someone is aware and able to reply me ? I wonder if Sony has planed

only color version or if a monochrome will be proposed ?

 

Sony announced an other Sensor, IMX 390, but I don't understand why they made a so stupid

choice, it's a joke !

 

http://image-sensors...ensor-with.html

 

To summarize, it's an IMX 290 a little bit larger (2.45 vs 2.1 Mpix) with anti-flickering system.

No interest for astronomy use...

In fact there's no real differences between these 2 sensors ! A kind of clone, not more.

 

I'm desperatly  waiting for a big brother of IMX 290...

 

Albéric

Wrong thread though.  There are other thread om IMX290.

 

IMX390 has the built-in HDR and there is a special internal circuit to optionally control the start of the frame sync to coincide with the florescent or LED lighting fixture's firing cycle (match AC frequency to avoid flickering.)

IMX390's pixel pitch (3.0um) is slightly larger than IMX290's 2.9um.

 

Clear Skies!

 

ccs_hello


Edited by ccs_hello, 14 April 2017 - 07:37 AM.


#21 premk19

premk19

    Viking 1

  • *****
  • Posts: 663
  • Joined: 19 Apr 2015

Posted 14 April 2017 - 12:02 AM

I agree with Curtis, astronomy market doesn't exist for Sony, main production are sensors
for smartphones ! After, far away comes surveillance/industrial vision markets.
And flickering of road signs is a so important problem in life that a sensor is designed
to solve it ! It makes me laugh (and cry...).

 

Thanks for Framos link, 4.63 microns pixels size remains an interesting size. I guess that
it's a reliable information, Framos is a well known professionnal supplier for sensors.
What would be the interest for them to give wrong information to customers ?

 

So in conclusion, we have to wait for the official datasheet of IMX294 from Sony...

Albéric

Not sure why you find it so amusing. It makes a lot of sense to have an anti-flicker sensor for reading LED signs and stop lights for surveillance and traffic cameras (primary market for this sensor), cameras in self driving cars etc. 



#22 ccs_hello

ccs_hello

    Voyager 1

  • *****
  • Posts: 10237
  • Joined: 03 Jul 2004

Posted 14 April 2017 - 07:39 AM

Yes, anti-flickering is very important for high-end augmented drive safety (lane departure/front collision)

as well as for self-driving cars.



#23 XS_Man

XS_Man

    Vostok 1

  • -----
  • Posts: 179
  • Joined: 20 Jul 2016
  • Loc: FRANCE

Posted 17 April 2017 - 06:52 AM

 


Not sure why you find it so amusing. It makes a lot of sense to have an anti-flicker sensor for reading LED signs and stop lights for surveillance and traffic cameras (primary market for this sensor), cameras in self driving cars etc. 

 

 

Yes, anti-flickering is very important for high-end augmented drive safety (lane departure/front collision)

as well as for self-driving cars.

 

It's a 'nervous laught' when I saw the annoncement !  I was expecting the big brother of IMX 290 and what a disappointment !

 

And now anti-flickering for On Semi :

 

http://image-sensors...-reduction.html

 

 

BSI type and meets the Starvis grade.  It is ERS (electronic rolling shutter), not the "global shutter" type.

2 good news for us. A global shutter sensor is more expensive and less effective.

 

Now the question is to know if a monochrome version is planned.

 

Albéric



#24 OleCuss

OleCuss

    Vanguard

  • *****
  • Posts: 2279
  • Joined: 22 Nov 2010

Posted 17 April 2017 - 09:53 AM

I believe there are some applications for which we might really prefer the global shutter.  Most of us aren't doing that kind of thing, however, so I'm not looking for a global shutter myself.

 

But several members who are truly very knowledgeable/expert on this stuff are not saying evil things about this sensor for NRTV so I'm going to maintain hope that 1-2 years from now someone could be selling a wonderful camera with this sensor.  I hope they will consider including GPS timing and calibration so that it could potentially also be used for occultation timing.



#25 gdd

gdd

    Mercury-Atlas

  • -----
  • Posts: 2536
  • Joined: 23 Nov 2005
  • Loc: Lynnwood, WA (N/O Seattle)

Posted 17 April 2017 - 11:02 AM

Wonder how the anti-flicker feature will affect AP. Perhaps bad for meteors/satellites and good for DSO's?

 

Gale




CNers have asked about a donation box for Cloudy Nights over the years, so here you go. Donation is not required by any means, so please enjoy your stay.


Recent Topics






Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics