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

ZWO ASI 294 MM PRO: First impressions and test images.

  • Please log in to reply
474 replies to this topic

#151 Coconuts

Coconuts

    Viking 1

  • *****
  • Posts: 585
  • Joined: 23 Sep 2012

Posted 20 September 2020 - 11:36 AM

John:  Please put up with my obtuseness (obstinacy?) just a wee bit longer.  I am well aware of the physical distinction between CCD charge transfer/summing and CMOS pixel reads.  I am also in full agree with your four bullets.  We're on the same page here.  We may not be able to call it "resampling", but I can certainly write a software process to accomplish your bullet #3.  In this case, the ADU value is four times greater, and read noise, adding in quadrature, is merely doubled.  So the signal to noise ratio is doubled.  That's not as good as the four-fold increase in S/N were I to use a CCD, but given that the latest CMOS sensors have far lower read noise than CCDs, at merely twice the read noise, you could still easily come out ahead.

 

My key goal is to shorten the total integration time required on any given object. A CCD will require four times less time on an object to achieve a given signal level if binned 2x2; imaging speed scales as inverse f-ratio squared, and as the square of the pixel size.  Fat binned CCD pixels speed up imaging, although we need to be careful to avoid undersampling in the process.  If we save a CMOS image at 1:1 resolution, and "bin" its pixel ADUs 2x2 per your bullet #3, the new image has one quarter the pixel count, and the ADUs of each new pixel are four times larger.  So far, just like a physically binned CCD.  The S/N improvement is only 2X, not 4X, but the read noise will be a lot lower with CMOS.

 

My point is that there is value in such software "binning" (even if we can't call it "resampling"), in that it speeds up imaging.  Would you agree?

 

All the best,

 

Kevin


Edited by Coconuts, 20 September 2020 - 11:39 AM.


#152 jdupton

jdupton

    Mercury-Atlas

  • *****
  • Posts: 2,809
  • Joined: 21 Nov 2010
  • Loc: Central Texas, USA

Posted 20 September 2020 - 12:38 PM

Kevin,

 

   I thought you were planning to be obstinate. I must have missed that part. smile.gif

 

   I agree with everything you wrote. In regards to bullet #3 in my prior post, you don't even have to really write your own code for software binning. If you really want a software binned image, you can resample and then scale by 4x after the fact. That will give you the same result. There are probably only a few reasons for doing a software bin rather than resample, so most folks will be OK with either if given a choice in their processing software.

 

   I agree with you also that there is a benefit to resampling (or binning) in software. You do get the advantage of the 2x SNR improvement. My main reason for pointing out the differences between binning and resampling earlier in the thread is that too many people may think that resampling in post-processing equals binning a CCD image. There are benefits to be had but they are not equal.

 

   Your point about read noise differences is well taken. While the read noise is doubled for CMOS binning, it starts out with a lower read noise advantage for most cameras. (I do recall, however, that several high-end CCD cameras had beat read noise down to the levels of modern CMOS chips through careful camera design. For such CCD cameras [eg QSI6120 with RN of 1.6e-], read noise was comparable to CMOS and binning would put them way over the top in any head to head comparison with CMOS.)

 

 

John


Edited by jdupton, 20 September 2020 - 12:39 PM.


#153 Coconuts

Coconuts

    Viking 1

  • *****
  • Posts: 585
  • Joined: 23 Sep 2012

Posted 20 September 2020 - 01:06 PM

Thanks, John!

 

"Fat pixels incoming!"

 

All the best,

 

Kevin



#154 yonafunu

yonafunu

    Sputnik

  • -----
  • Posts: 46
  • Joined: 03 Sep 2014

Posted 21 September 2020 - 03:33 AM

New post from QHYCCD :

 

New mono cam--QHY294M Pro, is about to be available--4/3 inch
format, with two modes of 11mp/47mp in one body!
Equipped with IMX492 chips, the QHY294 Pro series camera is capable of locking and unlocking the on-chip binning to provide two readout modes. The first mode reads the sensor "locked" mode to produce 11.6mp images with 4.63um pixel size and 14 bits per pixel. The second read mode unlocks the binning to produce 46.8mp images with 2.315um pixel size at 12 bits per pixel.


  • bobzeq25 likes this

#155 Coconuts

Coconuts

    Viking 1

  • *****
  • Posts: 585
  • Joined: 23 Sep 2012

Posted 21 September 2020 - 05:47 AM

This camera's native high resolution mode would be a better fit for wide field imaging with less undersampling.  Camera lenses would be a "prime" example.  1 arc-sec per pixel at 477 mm focal length; 2 arc-sec/pixel at 238 mm. 

 

Astromechanics already makes a thin (11.5 mm) ASCOM Canon/Sigma lens controller (aperture and focus) that satisfies Canon's 44 mm flange focal distance with the 20 mm. thick ZWO EFW and an ASI6200MM (17.5 mm sensor depth with the tilt plate left on):

https://astromechani...ller_6200mm.pdf

 

The QHY294M Pro has a sensor depth of 17 mm, which might require only a thin spacer, or a tilt plate gap.

 

All the best,

 

Kevin



#156 JerseyBoy

JerseyBoy

    Viking 1

  • *****
  • Posts: 537
  • Joined: 05 Aug 2014
  • Loc: Jersey, Channel Islands

Posted 21 September 2020 - 07:00 AM

This has just been posted on FB by QHYCCD:

 

New mono cam--QHY294M Pro, is about to be available--4/3 inch
format, with two modes of 11mp/46mp in one body!
Equipped with IMX492 chips, the QHY294 Pro series camera is capable of locking and unlocking the on-chip binning to provide two readout modes. The first mode reads the sensor "locked" mode to produce 11.6mp images with 4.63um pixel size and 14 bits per pixel. The second read mode unlocks the binning to produce 46.8mp images with 2.315um pixel size at 12 bits per pixel.



#157 AnakChan

AnakChan

    Apollo

  • -----
  • Posts: 1,136
  • Joined: 01 Sep 2014
  • Loc: Oz

Posted 21 September 2020 - 07:06 AM

Just a thought, the QHY294 should probably have its own thread.

#158 Peregrinatum

Peregrinatum

    Surveyor 1

  • *****
  • Posts: 1,506
  • Joined: 27 Dec 2018
  • Loc: South Central Valley, Ca

Posted 21 September 2020 - 12:10 PM

If the binning for the ASI294MM is software generated, why couldn't ZWO offer a firmware update to enable the native pixel size?



#159 Lead_Weight

Lead_Weight

    Surveyor 1

  • *****
  • Posts: 1,690
  • Joined: 21 Jul 2016
  • Loc: Houston

Posted 21 September 2020 - 12:24 PM

I would hazard a guess they are positioning the camera with larger pixels to fill a gap in their lineup which mostly consists of cameras with tiny pixels. But I don’t see why they wouldn’t just offer the option to do it both ways in the driver, possibly defaulting to the bin 2x2.


  • RossW likes this

#160 MikiSJ

MikiSJ

    Apollo

  • *****
  • Posts: 1,159
  • Joined: 24 Feb 2006
  • Loc: San Jose, CA

Posted 21 September 2020 - 12:41 PM

Just a reminder: the ASI294MC does not have larger pixels (4.63μm). ZWO, in firmware, bins the smaller native pixel 2.315μm in a 2x2 fashion for each of the RGB pixels to create an 'effective' pixel of 4.63μm.

post-15355-0-38956300-1599593385.jpg

The above is from the Sony datasheet for their IXM294 sensor. 


  • premk19 likes this

#161 Lead_Weight

Lead_Weight

    Surveyor 1

  • *****
  • Posts: 1,690
  • Joined: 21 Jul 2016
  • Loc: Houston

Posted 21 September 2020 - 01:43 PM

Just a reminder: the ASI294MC does not have larger pixels (4.63μm). ZWO, in firmware, bins the smaller native pixel 2.315μm in a 2x2 fashion for each of the RGB pixels to create an 'effective' pixel of 4.63μm.

post-15355-0-38956300-1599593385.jpg

The above is from the Sony datasheet for their IXM294 sensor. 

Right, but the average person in the hobby probably doesn't understand this. They don't describe it at all on their website page. I'm just saying that positioning it this way helps to fill a gap in their lineup.

 

Screen Shot 2020-09-21 at 1.42.26 PM.jpg



#162 Peregrinatum

Peregrinatum

    Surveyor 1

  • *****
  • Posts: 1,506
  • Joined: 27 Dec 2018
  • Loc: South Central Valley, Ca

Posted 21 September 2020 - 02:13 PM

My 294MM comes thursday, hope to image the Bubble with it that night yay.gif



#163 MikiSJ

MikiSJ

    Apollo

  • *****
  • Posts: 1,159
  • Joined: 24 Feb 2006
  • Loc: San Jose, CA

Posted 21 September 2020 - 02:50 PM

Right, but the average person in the hobby probably doesn't understand this. They don't describe it at all on their website page. I'm just saying that positioning it this way helps to fill a gap in their lineup.

 

attachicon.gifScreen Shot 2020-09-21 at 1.42.26 PM.jpg

But the ASI294MM does not have 4.63µm pixels. It has 2.315µm pixels binned 2 x 2.



#164 dghent

dghent

    Viking 1

  • *****
  • Vendors
  • Posts: 719
  • Joined: 10 Jun 2007

Posted 21 September 2020 - 02:56 PM

Right, but the average person in the hobby probably doesn't understand this. They don't describe it at all on their website page. I'm just saying that positioning it this way helps to fill a gap in their lineup.

 

attachicon.gifScreen Shot 2020-09-21 at 1.42.26 PM.jpg

The way I understand it, the ZWO (or QHY firmware) doesn't do the binning in their own software or firmware. The sensor (IMX492) itself has multiple readout modes that the camera firmware can command the sensor to expose in: the locked 2x2 binning of the quad-bayer cluster to create 4x 4.63um pixels out of the 16x 2.513um pixels that comprise the quad-bayer cluster (thus producing 14bit data from 12bit pixels at 4x fewer pixel count), and the unlocked mode where no binning takes place and you get data from each 12bit 2.513um pixel, totaling 47mp of data. Any binning beyond this, such as classic user-controlled binning, would be done by the vendor firmware or host-side in the SDK.

 

These sensors can be driven in HDR mode as well, where the quad-bayer cluster is exposes 2 of the same color at one gain value and the other 2 pixels at a different gain value simultaneously, producing 2 different images which can then be combined to create a single high dynamic range image. This mode is for use on consumer cameras.

 

I believe the IMX294 also can be unlocked, but being that it is color only, some massaging the of the quad-bayer RRGGRRGGGGBBGGBB data would need to be done to create a RGGBRGGBRGGB pattern so that it can be properly debayered by existing software; at least most astro software.


  • JerseyBoy, Lead_Weight and Psittacula like this

#165 Lead_Weight

Lead_Weight

    Surveyor 1

  • *****
  • Posts: 1,690
  • Joined: 21 Jul 2016
  • Loc: Houston

Posted 21 September 2020 - 03:17 PM

But the ASI294MM does not have 4.63µm pixels. It has 2.315µm pixels binned 2 x 2.

I'm not in disagreement with you. I'm just speculating as to why they're advertising it like this.



#166 MikiSJ

MikiSJ

    Apollo

  • *****
  • Posts: 1,159
  • Joined: 24 Feb 2006
  • Loc: San Jose, CA

Posted 21 September 2020 - 03:58 PM

...

I believe the IMX294 also can be unlocked, but being that it is color only, some massaging the of the quad-bayer RRGGRRGGGGBBGGBB data would need to be done to create a RGGBRGGBRGGB pattern so that it can be properly debayered by existing software; at least most astro software.

If the description of how the individual pixels are binned to get 4.63µm effective pixels in the IMX294 is accurate on the sensor chip, then there would be a Bayer mask over the pixels as shown and simply removing the quad pixel arrangement in firmware would confuse the captured image. I don't see how the IMX294 used in the ASI294MC can be 'unlocked' to use the native 2.315µm pixels.



#167 sebastien79

sebastien79

    Lift Off

  • -----
  • Posts: 12
  • Joined: 26 Jan 2008
  • Loc: France , Vendée

Posted 22 September 2020 - 11:46 AM

Just arrived ! 

 

First Sensor Analysis with Sharpcap (sensor temp : -10°C) :

 

Annotation 2020-09-22 175355.jpg

 

 


  • lakerunr and Peregrinatum like this

#168 khursh

khursh

    Viking 1

  • ****-
  • Posts: 621
  • Joined: 23 Jan 2018

Posted 22 September 2020 - 11:54 AM

Hi Sebastien, 

 

What was ambient temp if you recall?



#169 sebastien79

sebastien79

    Lift Off

  • -----
  • Posts: 12
  • Joined: 26 Jan 2008
  • Loc: France , Vendée

Posted 22 September 2020 - 12:03 PM

Hi Sebastien, 

 

What was ambient temp if you recall?

Hi Kevin, 

 

Temp ambient 20°C, power cooler approximately at 67% 



#170 Peregrinatum

Peregrinatum

    Surveyor 1

  • *****
  • Posts: 1,506
  • Joined: 27 Dec 2018
  • Loc: South Central Valley, Ca

Posted 23 September 2020 - 07:28 AM

Just arrived ! 

 

First Sensor Analysis with Sharpcap (sensor temp : -10°C) :

 

attachicon.gifAnnotation 2020-09-22 175355.jpg

Looks great, better than the published curves!



#171 Dean J.

Dean J.

    Viking 1

  • -----
  • Posts: 652
  • Joined: 13 Aug 2011
  • Loc: Vista, CA

Posted 23 September 2020 - 05:11 PM

I was just reading about the QHY294M Pro and the “expanded pixel mode” that is available on that camera. Has anyone heard anything about whether ZWO will ever offer the same feature on the ASI294MM?

#172 suvowner

suvowner

    Viking 1

  • *****
  • Posts: 593
  • Joined: 22 May 2016
  • Loc: Arkansas

Posted 23 September 2020 - 07:54 PM

I was just reading about the QHY294M Pro and the “expanded pixel mode” that is available on that camera. Has anyone heard anything about whether ZWO will ever offer the same feature on the ASI294MM?

I asked this question in the q/a section on the zwo website for this camera, and there response wa no mad.gif

https://astronomy-im...ct/asi294mm-pro



#173 Dean J.

Dean J.

    Viking 1

  • -----
  • Posts: 652
  • Joined: 13 Aug 2011
  • Loc: Vista, CA

Posted 23 September 2020 - 08:55 PM

Well, that is interesting. I had been waiting for the mono IMX571 cameras but the QHY294M might be worth looking into.

Edit: On second thought, maybe I don’t want a 46.8MP camera. :-)

Edited by Dean J., 23 September 2020 - 10:12 PM.

  • Stelios likes this

#174 andysea

andysea

    Skylab

  • *****
  • topic starter
  • Posts: 4,073
  • Joined: 03 Sep 2010
  • Loc: Seattle, WA

Posted 23 September 2020 - 11:06 PM

I think that would be a good option to have actually. However the full resolution will only give you 12 bit depth. It would be interesting to see what the image quality of the sensor unbinned is.



#175 Dean J.

Dean J.

    Viking 1

  • -----
  • Posts: 652
  • Joined: 13 Aug 2011
  • Loc: Vista, CA

Posted 23 September 2020 - 11:20 PM

I have been getting along just fine with the ASI1600MM so the 12 bit thing really doesn’t concern me. It’s the file size that concerns me. My processing slowed down quite a bit with the new ASI2600MC vs. the ASI1600 so I don’t want to get files much bigger than the 26MP IMX571 mono.

Edit: So, I guess I just talked myself out of the QHY294M and the pixels of the ASI294MM are too big for my use situation (imaging at 380mm fl).

Edited by Dean J., 23 September 2020 - 11:33 PM.

  • leviathan likes this


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