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

ASI New Camera -- 294 Pro ! - beta testing

  • Please log in to reply
592 replies to this topic

#126 OleCuss

OleCuss

    Vanguard

  • *****
  • Posts: 2390
  • Joined: 22 Nov 2010

Posted 25 September 2017 - 04:21 PM

Well, I may be mis-interpreting this because I'm not an expert in all this.

 

But if you go back to the information released by Sony that I've found:  http://www.sony-semi...mx294cjk_e.html you'll note that Sony talks about having two high-speed interfaces which include the CSI2 and the SLVS-EC.

 

For the IMX294 sensor with the 4:3 aspect ratio the 14-bit frame rate is listed as being 24 FPS without specifying either CSI2 or SLVS-EC.

 

If you go to 12-bit it will give you a frame rate of 60 or 30 for SLVS-EC and CSI2 respectively.

 

If you go to 10-bit you get 30 and 15.

 

 

So maybe ZWO decided just to implement the 14-bit at 24 FPS and it is ending up actually being just a tad bit slower?

 

I don't know why the frame rate is actually slower at 10-bits but it would seem to be so?

 

The upshot is that you aren't going to get a truly high frame rate from this camera.  Personally, that is perfectly fine with me.  It would be awesome if ZWO or maybe QHY were to implement that 60 FPS at 12-bit but I have a feeling it would choke even an SSD?

 

I'm sort of assuming an ROI can be defined but I'm not sure that the SoC is going to pump out a faster frame rate even if that were the case.

 

I'd note that the PAL system happily uses a frame rate of 24 or 25 FPS and that the NTSC standard seems to be 30 FPS?  The designers of this sensor even state that this is intended for security systems which I don't think are often driven to high frame rates.

 

Not being an expert at the different standards I could be way off-base but it could be that the sensor designers set up the SoC to output frame rates which were compatible with existing commercial standards and have little to no interest in juicing up their SoC to output a frame rate which I don't think many security systems want to handle.

 

I know that the surveillance folk I talk to around here are generally in the NTSC camp for almost all their applications.  There are some specialized situations where they are using very high frame rates at a significant cost but that is by far the exception rather than the rule.

 

Anyway, I'm suspecting ZWO is happy with the 24 FPS since that should work just fine for what most of us would want to do with that camera.


Edited by OleCuss, 25 September 2017 - 06:52 PM.


#127 ccs_hello

ccs_hello

    Voyager 1

  • *****
  • Posts: 10303
  • Joined: 03 Jul 2004

Posted 25 September 2017 - 06:43 PM

There are few output methods and quite a few readout mode for these Exmor image sensors.

It is the camera system manufacturing firm's design requirements that will dictate what features are most useful and should come out

as a day-one camera capabilities.

 

In this new market, there are three collaborating forces:

1) basic imager output (what features are to be exposed/made available,

2) camera driver developers to understand (1)'s output and parse the data in a form that PC software can consume, and

3) PC software to utilize/take advantage of the camera offered image data

Still a long way to go, especially on less demanded features.

 

This shouldn't be any surprise.  Look at the real world camera systems outside the astro or EAA realms,

Have you found any camera system that will satisfy evryyone's demands?

 

Clear Skies!

 

ccs_hello



#128 A. Viegas

A. Viegas

    Mercury-Atlas

  • *****
  • topic starter
  • Posts: 2993
  • Joined: 05 Mar 2012
  • Loc: New York City/ CT

Posted 26 September 2017 - 09:05 AM

While I am traveling, I brought along my USB stick with my images from this past weekend.   So I will continue to leak these out as I get time...

 

So -- testing Binning

I used RAW8   and I show below  Bin1, Bin2, Bin3     all 15s exposures

Scope setup  CPC11  @ F6.3   - vignetting is obvious.

These are all taken directly from Sharpcap.  If you click on file you can see how many exposures.  I tweaked histogram a bit in each, but I was not totally consistent in exact number of frames and amount of histogram stretch...   yea yea...  my control and experimental variables are not consistent... I am not a good scientist... tongue2.gif

 

Al

 

Bin 1x1

M57 cpc11 294 bin1 15s Stack 16bits 13frames 195s
 
Bin 2x2
M57 cpc11 294 bin2 15s Stack 16bits 12frames 180s
 
Bin 3x3
M57 cpc11 294 bin3 15sraw8 Stack 16bits 11frames 165s

Edited by A. Viegas, 26 September 2017 - 09:08 AM.

  • OleCuss, roelb and bdyer22 like this

#129 A. Viegas

A. Viegas

    Mercury-Atlas

  • *****
  • topic starter
  • Posts: 2993
  • Joined: 05 Mar 2012
  • Loc: New York City/ CT

Posted 26 September 2017 - 09:21 AM

Ok  here are two more examples of Bin1 vs Bin2  using Raw8 with the new ASI294

 

M56 - 10s exposures for each, stacked in Sharpcap Live.  Same setup as prior,  CPC11 at F6.3 ish.   NO adjustments whatsoever

 

Bin1

M56 cpc11 294 bin1 10s Stack 10frames 100s
 
Bin2
M56 cpc11 294 bin2 10s Stack 12frames 120s

 

 

Al

 

 

I think my science experiment skills are better here... you can see a slight improvement in SNR in bin2


Edited by A. Viegas, 26 September 2017 - 09:22 AM.


#130 jimthompson

jimthompson

    Apollo

  • *****
  • Posts: 1102
  • Joined: 06 Oct 2009
  • Loc: Ottawa, Canada

Posted 26 September 2017 - 09:27 AM

Hi Al,

 

I find it hard to compare the differences from the images posted within the CN webpage interface.  Are you able to crop out a little piece of the images and put them side-by-side so it is easy to see them at 100% zoom?

 

Thanks,

 

Jim T.



#131 A. Viegas

A. Viegas

    Mercury-Atlas

  • *****
  • topic starter
  • Posts: 2993
  • Joined: 05 Mar 2012
  • Loc: New York City/ CT

Posted 26 September 2017 - 09:36 AM

Here is another example of the sensitivity of the new ASI 294 Pro

Here is an Untouched Sharpcap stack of 17 frames at 15s each in Bin1  of the Crescent (NGC6888)  (CPC11 at F6.3)   actual field is 37' x 25'

 

not too shabby for 15s...   (the bright spot in the upper right is not amp glow, I think its my neighbor's dog "pee-at-night"  stadium light.   Fortunately, he knows I dislike it, so it only goes on 1 or 2x per night when his dog needs to go outside...    hehe

 

 

Al

 

NGC6888 cpc11 294 bin1 15s Stack 16bits 17frames 255s

 



#132 alphatripleplus

alphatripleplus

    ISS

  • *****
  • Moderators
  • Posts: 78977
  • Joined: 09 Mar 2012
  • Loc: New Jersey

Posted 26 September 2017 - 09:41 AM

 Fortunately, he knows I dislike it, so it only goes on 1 or 2x per night when his dog needs to go outside...    hehe

 

 

Al

 

Al,  this is a case where I would say post-processing has real value - unless you like to be reminded of that sort of thing.smile.gif



#133 A. Viegas

A. Viegas

    Mercury-Atlas

  • *****
  • topic starter
  • Posts: 2993
  • Joined: 05 Mar 2012
  • Loc: New York City/ CT

Posted 26 September 2017 - 09:46 AM

Hi Al,

 

I find it hard to compare the differences from the images posted within the CN webpage interface.  Are you able to crop out a little piece of the images and put them side-by-side so it is easy to see them at 100% zoom?

 

Thanks,

 

Jim T.

here you go Jim

 

Bin2

M56 cpc11 294 bin2 10s Stack 12frames 120s crop
 
Bin1
 
M56 cpc11 294 bin1 10s Stack 10frames 100s crop

 

 

Al


Edited by A. Viegas, 26 September 2017 - 09:47 AM.


#134 jimthompson

jimthompson

    Apollo

  • *****
  • Posts: 1102
  • Joined: 06 Oct 2009
  • Loc: Ottawa, Canada

Posted 26 September 2017 - 11:15 AM

Thanks Al, that is much better.  I am curious, why is the bin 2 image not noticeably brighter?  You used the same exposure of 10s per frame and roughly the same number of stacked frames, so shouldn't the bin 2 image be brighter?  Is the binning you have selected that provided by the Sharpcap software or is it on camera?

 

cheers,

 

Jim T.



#135 A. Viegas

A. Viegas

    Mercury-Atlas

  • *****
  • topic starter
  • Posts: 2993
  • Joined: 05 Mar 2012
  • Loc: New York City/ CT

Posted 26 September 2017 - 11:50 AM

Hi Jim

I am using Sharpcap.   I had originally noticed (look back to earlier post in this thread) that I did not seem to get much benefit using Bin2 over Bin1 except for smaller file size.  Sam from ZWO said to use RAW8  vs. RAW16 for bin2,3,4   -  so that is what I was using.   It seems that Hiten has been more successful than me in getting binning to work with Sharpcap than I have.   I am not sure why.   But hey... that is why we are the guinea pigs testing this camera! shocked.gif



#136 XS_Man

XS_Man

    Vostok 1

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

Posted 26 September 2017 - 12:47 PM

Hi Al,
Hi Hiten,
and Hi All !

 

I still feel you are trying to fit a square peg in a round hole or said differently: I think the strength of this camera will be DSOs.

 

Have you found any camera system that will satisfy evryyone's demands?

 

I understand your reactions CCS and Hiten, but if it's possible to have more with a simple software upgrade or
easy electronic mod, why not asking for it ? Remember, ASI 294 price is more than 1000 dollars with cooling...
For 1000 dollars, I just want the best ! grin.gif

And if it's impossible, OK, but let me know why ?

 

OleCuss's explanations are very interesting and seems logical with Sony datasheet informations.
If you are right, our last hopes  are :

* "All-pixel scan (aspect ratio 17:9) with Horizontal/vertical 2/2-line binning" 
   2048 (H) × 1080 (V) with 120 fps at 12 bits,

* "All-pixel scan (aspect ratio 17:9)    4096 (H) × 2160 (V) with 120 fps at 10 bits,

I guess that, with classical USB 3.0,  4096 (H) × 2160 (V) with 120 fps at 10 bits is not possible.

 

But binned mode 2048 (H) × 1080 (V) with 120 fps at 12 bits is possible with USB 3.0.

 

Al, Astrojedi, is this mode accessible using Sharpcap or Firecapture ?
Is 120 fps possible in bin2 mode ? confused1.gif

 

I also don't agree with your comment about the FoV not being enough for imaging Jupiter. Damian Peach quite successfully does planetary imaging with a ASI290MM and a C14 (http://www.damianpeach.com/best.htm) and he seems to have some idea about how to go about planetary imaging. : )

 

 

Only Saturn or Jupiter are not enough in the field, it's better to have "the familly" with the moons. wink.gif 
For example my last picture with ASI 185 2.4 Mpix (Saturne is only 20° from horizon line !)  with 300 mm telescope :

 

http://www.webastro....ad.php?t=149983

 

 

 

Albéric


  • Ain Soph Aur and orionic like this

#137 A. Viegas

A. Viegas

    Mercury-Atlas

  • *****
  • topic starter
  • Posts: 2993
  • Joined: 05 Mar 2012
  • Loc: New York City/ CT

Posted 26 September 2017 - 01:16 PM

Very nice capture Alberic!!    with the amazing detail you got in that picture, I was hoping you may have captured Cassini as it plunging to its doom!   lol.gif lol.gif


  • Ain Soph Aur and paulymo like this

#138 Astrojedi

Astrojedi

    Soyuz

  • *****
  • Posts: 3964
  • Joined: 27 May 2015
  • Loc: SoCal

Posted 26 September 2017 - 02:47 PM

Jim, Al,

 

I ran some analysis from the camera output and it looks like in raw16 mode the ZWO drivers are averaging instead of binning when you select binning. I have reached out to Sam to understand this better.

 

If so, we will request Sam to fix this.

 

Hiten


  • Ain Soph Aur likes this

#139 DonBoy

DonBoy

    Apollo

  • *****
  • Posts: 1242
  • Joined: 06 Sep 2007

Posted 26 September 2017 - 03:15 PM

I'm speculating that Sam has set the driver for this camera the way he has been  setting drivers lately; in that, he has Raw 8 bin2 as hardware binned where there is a 4 times signal increase and a 2 times noise increase yielding a SNR of 2.  Raw16 bin2 as a software bin and should give a 2/1 noise improvement but no signal gain and thus SNR of 2. 


Edited by DonBoy, 26 September 2017 - 03:15 PM.


#140 Astrojedi

Astrojedi

    Soyuz

  • *****
  • Posts: 3964
  • Joined: 27 May 2015
  • Loc: SoCal

Posted 26 September 2017 - 04:18 PM

Don,

 

The RAW8 mode uses software binning (i.e. summing adjacent pixels). The RAW16 I don't believe does software binning but rather averages adjacent pixels in software which is why you don't see an increase in brightness.

 

Also note software binning provides almost all the benefits to shot noise reduction / brightness / "SNR" as hardware binning even in CCDs. But because the read noise is so much higher in CCDs, hardware binning also improves SNR a little bit by bringing more signal over the read noise threshold.

 

The already very low read noise of CMOS sensors makes hardware binning irrelevant. Even without hardware binning CMOS sensors produce much better SNR (all else being equal).

 

Hiten


  • mikefulb likes this

#141 DonBoy

DonBoy

    Apollo

  • *****
  • Posts: 1242
  • Joined: 06 Sep 2007

Posted 26 September 2017 - 06:16 PM

Hiten, thanks for the clarification.



#142 Kaikul

Kaikul

    Surveyor 1

  • *****
  • Posts: 1718
  • Joined: 23 Jan 2014

Posted 26 September 2017 - 08:16 PM

Hi Al,

 

One of the things that I noticed in many images from this camera is that the stars are considerably tight. Did you use a UV/IR filter or any filter (like the Astronomik one that Hiten used) that would prevent star bloat? Or is it something in the camera that does this? 

 

Kaikul


Edited by Kaikul, 26 September 2017 - 08:18 PM.


#143 A. Viegas

A. Viegas

    Mercury-Atlas

  • *****
  • topic starter
  • Posts: 2993
  • Joined: 05 Mar 2012
  • Loc: New York City/ CT

Posted 26 September 2017 - 10:06 PM

Hi Kaikul

No filters used. 3 telescopes so far used. AT65 apo, F8 250mm refractor and C11 at F6.3. I did use a manual filter wheel but I removed the filter threads so the opening is 31mm and this way I can do darks easier.

Al
  • Kaikul likes this

#144 XS_Man

XS_Man

    Vostok 1

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

Posted 27 September 2017 - 11:59 AM

Hi,

 

Very nice capture Alberic!!    with the amazing detail you got in that picture, I was hoping you may have captured Cassini as it plunging to its doom!   lol.gif lol.gif

It would have been cool !  grin.gif

 

For this picture I use an ASI 185 with frame rate between 40 to 100 fps (transparence of sky decrease

greatly during the night !). In France (45° latitude), Saturne is very low, only 20° above Horizon ! So high

frame rate is the only way to fight against bad seeing. And next years, planetary imaging will be real

nightmare...

 

That's why I'm searching for the maximum possible performances for astro cameras that will use

IMX294. And I'm sure that on forums, multipurpose imaging fans will have the same requirement if they

plan to use it for planetary/lunar/deep sky imaging.

 

Please Al, when it will be possible for you, could you make the test in bin 2 mode 2048 (H) × 1080 (V)

to know if 120 fps is possible. Both in RAW8 and RAW16.

If it works, it would not be a perfect solution but 120 fps is a good frame rate.

 

It's really important request, not only for me.

 

Many thanks Al !

 

Albéric


Edited by XS_Man, 27 September 2017 - 12:01 PM.

  • Ain Soph Aur likes this

#145 Astrojedi

Astrojedi

    Soyuz

  • *****
  • Posts: 3964
  • Joined: 27 May 2015
  • Loc: SoCal

Posted 27 September 2017 - 12:37 PM

That is a fantastic capture Albéric. I see what you are trying to do. I never bothered with the moons : )

 

I am still not sure if this camera or any camera can deliver the frames rates you are looking for at the high resolutions you want. Assuming USB 3 + the camera + sensor silicon can output at that frame rate, you will need a very fine tuned very high performance system to get to 150+ fps at the resolutions you are talking about.

 

Also note that $1100 is relatively cheap for a 4/3 sensor cooled camera with such exceptional sensitivity. Just a few years ago you would have easily paid $3000+ for the privilege and if you were lucky gotten 1 frame in 5 seconds. : )


Edited by Astrojedi, 27 September 2017 - 12:38 PM.

  • Ain Soph Aur likes this

#146 XS_Man

XS_Man

    Vostok 1

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

Posted 27 September 2017 - 03:59 PM

That is a fantastic capture Albéric. I see what you are trying to do. I never bothered with the moons : )

 

Hi Hiten,

 

Many thanks. For me both Saturn's moons and Jupiter's moon's are interesting to consider.

 

Now, to show you all my expectations for this sensor...

One day I will choose Saturn and its moons.

 

Next day, bad seeing for planets, I will choose DSO target. For exemple NGC7635 with

short exposures, here with my T300/1600 on EQ-6 mount (NO AUTOGUIDING) :

 

http://www.webastro....ad.php?t=151008

 

First image is the color layer with QHY5III 178 color plus 0.65x focal reducer.
Exposures are 7 seconds.

 

Second image is the monochrome Layer with a QHY174 M cooled.
Exposures are a mix of 2 seconds and 5 seconds.

 

Third image is the merged image monochrome + color.

 

IMX 178 have only 2.4 microns pixels, it's not enough even with focal reducer.
With 4.63 microns pixels, I expect a much better SNR, more signal for color layer.

 

Albéric


  • MGD and 42itous1 like this

#147 ccs_hello

ccs_hello

    Voyager 1

  • *****
  • Posts: 10303
  • Joined: 03 Jul 2004

Posted 27 September 2017 - 06:34 PM

re: post 144

 

Albéric

 

Re: if you'd like to get the high frame rate with lots of pixels, I'd suggest Point Grey (now owned by FLIR) 

https://www.ptgrey.com/Camera-selector

 

They know their biz (especially in high frame rate department.)  Unfortunately no IMX294 yet, but check their offering on

other 2M and 3M pixel image sensors based imagers, it should shine some light on how camera systems can be optimized.

 

Clear Skies!

 

ccs_hello



#148 A. Viegas

A. Viegas

    Mercury-Atlas

  • *****
  • topic starter
  • Posts: 2993
  • Joined: 05 Mar 2012
  • Loc: New York City/ CT

Posted 27 September 2017 - 08:54 PM

Alberic maybe Hiten can run the usb3 speed test for you. I am currently traveling and will not be back to run anymore tests until the second weekend in October.

#149 XS_Man

XS_Man

    Vostok 1

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

Posted 28 September 2017 - 11:32 AM

re: post 144

 

Albéric

 

Re: if you'd like to get the high frame rate with lots of pixels, I'd suggest Point Grey (now owned by FLIR) 

https://www.ptgrey.com/Camera-selector

 

ccs_hello

 

Hi CCS,

 

Yes, I have a Point Grey Blackfly with IMX104 few years ago. A good camera for planetary
imaging but now IMX224 is far better.

 

The main problem is that Point Grey are designing cameras for Industrial vision and surveillance,
not Astronomy. So no cooling, limited maximum exposure time and no anti-amp glow system.
For FSI sensors it's a big problem when used for DSO short exposures !

 

But, I agree, Point Grey is exploring interestings ways :
- USB 3.1 Gen1,
- 10 GigE,
- IMX 420 is planned to be integrated !

 

https://www.ptgrey.c...cmos-technology

 

As no monochrome version of IMX294 is avalaible, only other solution is global
shutter IMX 420 7 Mpix sensor with 4.5 microns pixels. I hope that ZWo or QHYCCD
will plan a camera with this sensor.  For sure price will not be cheap !

 

 

Alberic maybe Hiten can run the usb3 speed test for you. I am currently traveling and will not be back to run anymore tests until the second weekend in October.

Hi Al,

 

OK, I understand.
Hope Hiten could try.

 

Albéric


Edited by XS_Man, 28 September 2017 - 11:42 AM.


#150 iam1ru12

iam1ru12

    Messenger

  • *****
  • Posts: 453
  • Joined: 25 Feb 2006
  • Loc: Raleigh/Durham, NC

Posted 29 September 2017 - 02:46 PM

Alberic,

Those are wonderful images you provide links to.  Unfortunately my French is quite poor, mi español es mucho mejor.  I believe you images are processed, correct?  They are not screen shots or saves of stacked images as view when you captured the view, correct?

 

Thanks!

-Mike




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