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

17 sec (x1024) Camera Sony 960h Exview HAD CCD2

This topic has been archived. This means that you cannot reply to this topic.
494 replies to this topic

#351 GlennLeDrew

GlennLeDrew

    Hubble

  • *****
  • Posts: 15,850
  • Joined: 17 Jun 2008

Posted 01 December 2012 - 02:13 AM

Indeed, two bad pixels is still quite good, *especially* for the amount paid. Unless Elaine were to fire up a bunch of cameras and cherry pick one with one or no bad pixels, you're probably likely to fare worse on the replacement.

If the image posted is a positive, then are not the two light-toned pixels 'hot', or at least 'warm', instead if 'dead' (which is black)?

#352 yaesu77

yaesu77

    Lift Off

  • -----
  • Posts: 16
  • Joined: 04 Nov 2012

Posted 01 December 2012 - 04:22 AM

Thank you all.
Greetings.

#353 yaesu77

yaesu77

    Lift Off

  • -----
  • Posts: 16
  • Joined: 04 Nov 2012

Posted 01 December 2012 - 08:08 AM

Has anyone used this camera for planetary shooting?
Maybe with registax you can get good results.

#354 barbarosa

barbarosa

    Mercury-Atlas

  • *****
  • Posts: 2,649
  • Joined: 11 Apr 2010

Posted 01 December 2012 - 09:50 PM

Has anyone used this camera for planetary shooting?
Maybe with registax you can get good results.


Here is a single frame grab of an Io transit (not post processed) from a few nights back.

Attached Thumbnails

  • 5549445-JOVE IO.png


#355 yaesu77

yaesu77

    Lift Off

  • -----
  • Posts: 16
  • Joined: 04 Nov 2012

Posted 02 December 2012 - 04:37 AM

Wow Barbarosa :)
Setup? C5?

#356 dragonslayer1

dragonslayer1

    Apollo

  • -----
  • Posts: 1,290
  • Joined: 25 Feb 2012

Posted 02 December 2012 - 11:23 AM

Hey David, real nice pic. I have a stupid question, I purchased a VSS+ and have had it now for going on 2 weeks but with weather have not been able to try it. (Want bad weather, just purchase something for viewing LOL). Anyway, when people say of video "a single frame grab",are you talking a single pic of what is on your monitor? Like a cell phone, or DSLR picture of your monitor viewing screen? Thanks Kasey, PS again that's a real nice picture

#357 barbarosa

barbarosa

    Mercury-Atlas

  • *****
  • Posts: 2,649
  • Joined: 11 Apr 2010

Posted 02 December 2012 - 11:24 AM

CPC1100, LN-300-R (the small box), a old Meade 4000 2x Barlow. and some very good sky. :)

#358 barbarosa

barbarosa

    Mercury-Atlas

  • *****
  • Posts: 2,649
  • Joined: 11 Apr 2010

Posted 02 December 2012 - 01:03 PM

Hey David, real nice pic.

Thank you Kasey.

I have a stupid question

Never! At least not to me.

Anyway, when people say of video "a single frame grab",are you talking a single pic of what is on your monitor? Like a cell phone, or DSLR picture of your monitor viewing screen?


Let me try to clarify my careless use of frame grab. In this instance, the image is a SharpCap "Snapshot" made by using the "Snapshot" icon. However, I could not find any documentation for the feature, but I can speculate at little.

A Windows screen capture, screenshot, or screen grab (PrtScn or Alt-PrtScn keys in Windows) is a bitmap of the screen or the active window on the clipboard. Applications can then save or convert and save the bitmap to a file. A full or partial screen capture is not difficult to do in C++ or VB. That might be what the SharpCap Snapshot does.

What I did not do was to use a camera to capture an image of the display. Some people might use screenshot to mean a photo of the display. I do not know if there is any standard terminology.

This brings us back to the question of weather any of these are a frame grab in any useful sense. This gets me to thinking about interlaced video, analog to digital conversion, the video capture device, MPEG2 and my attention wanders. We need help from an expert.

#359 euphony

euphony

    Lift Off

  • -----
  • Posts: 12
  • Joined: 02 Dec 2012

Posted 02 December 2012 - 02:34 PM

That's a great pic Barbarosa,

What settings did you use for planetary with your CPC1100. I have the same telescope and will be hopefully receiving this camera in a week or two to have a play with.

Has anyone tried it on galaxies yet? I think I'll have a bit of experimentation with them when I get it :)
I use a mono Mintron at the moment but am a little underwhelmed with it.

Chris

#360 barbarosa

barbarosa

    Mercury-Atlas

  • *****
  • Posts: 2,649
  • Joined: 11 Apr 2010

Posted 02 December 2012 - 03:59 PM

[

What settings did you use for planetary with your CPC1100. I have the same telescope and will be hopefully receiving this camera in a week or two to have a play with.

Has anyone tried it on galaxies yet? I think I'll have a bit of experimentation with them when I get it :)
Chris


AGC - Off
Shutter - Electric
Exposure - 1/250(?)

BLC and similar set to OFF or a minimum value.
Experiment a little with others such as Sharpness.

The camera has several, seemingly interactive menu settings. Jupiter is bright so I just try various speeds until I get reasonable image. Then I adjust it using the the video proc settings in whatever software I'm using (usually SharpCap). Sometimes after adjusting contrast or brightness I go back and try faster or slower speed, and then readjust in the software.

One quirk that I fail to remember is that any changes that you make revert to the last saved settings after a short time. To keep a setting you must exit and save; the image goes dark and the camera restarts with your saved settings.

For DSOs, this camera's strong point is the long integration, but again the scope, the sky the target and any reducer/filters are variables that interact with the menu settings and the software settings that you set. Very small changes in the camera and software can have a large affect on the image.

I've only imaged two galaxies; not good but better than the EP view. You should like what you see though with clusters, and nebulae (Orion, Bubble, Veil, Saturn, Snowball, Crab, etc.)should look much better than with an EP.

I should mention that while you can use this cam at f/10 for bright DSOs, there a clear gain if you use a .5x reducer. I often use a /6.3 reducer on the scope, a 1.25 diagonal and a .5x reducer on the c-mount adapter.

The Meade f/3.3 reducer also works well with these small chip cams.

If you catch me on NSN and ask, I will be glad to show you the current settings or try different settings.

#361 slipfit

slipfit

    Sputnik

  • -----
  • Posts: 26
  • Joined: 02 Apr 2012

Posted 02 December 2012 - 07:08 PM

Thanks Barbarosa, I have the large version of this camera. It does well but have not got the settings right. I will give these a try as soon as the skies clear. I have not tried both focal reducers like you said, but will try that too. Thanks again.

#362 euphony

euphony

    Lift Off

  • -----
  • Posts: 12
  • Joined: 02 Dec 2012

Posted 03 December 2012 - 02:20 AM

Thanks Barbarosa, that's a great help.

I have got both an f6 and f3 reducer, so I'll experiment to find what works best with what objects. I've got a Sky Wathcer light pollution filter too (visually I can't see any difference, but I might see one when using the camera).
I've also invested in an IR Cut-Off Filter. So I'm all set and ready to go once the camera arrives.
The main obstacle of course is the temperamental weather over here!
Many thanks again for your incredibly helpful and detailed response. I've not been on nsn before, but will certainly try and get on there.
All the best

Chris

:)

#363 euphony

euphony

    Lift Off

  • -----
  • Posts: 12
  • Joined: 02 Dec 2012

Posted 03 December 2012 - 02:26 AM

That should be 'Sky Watcher'!

#364 euphony

euphony

    Lift Off

  • -----
  • Posts: 12
  • Joined: 02 Dec 2012

Posted 03 December 2012 - 08:59 AM

Did I read somewhere in this thread that you don't need to take darks with this particular camera? Or did I imagine that? :question:

If you don't, what is the reason for that? I understood that in astro imaging you always needed darks, flats etc. :hmmmm:

#365 nytecam

nytecam

    Voyager 1

  • *****
  • In Memoriam
  • Posts: 12,692
  • Joined: 20 Aug 2005

Posted 04 December 2012 - 05:55 PM

Did I read somewhere in this thread that you don't need to take darks with this particular camera? Or did I imagine that? :question:If you don't, what is the reason for that? I understood that in astro imaging you always needed darks, flats etc. :hmmmm:

This is a remarkably low-cost CCTV security cam aimed at the sky :grin: The option to apply darks and flats is limited and negates the prime reason in using these cams eg instant viewable results after a 17s/20s exp with minimum fuss that are better/deeper than an EP view ;)

#366 euphony

euphony

    Lift Off

  • -----
  • Posts: 12
  • Joined: 02 Dec 2012

Posted 05 December 2012 - 02:42 AM

Magic, thank you :)

I have been using my Mintron camera for the same thing, a live view, and it does indeed allow you to see things that you can't see through the EP. Although I'm a little disappointed with the quality, as it's extremely grainy and pixellated no matter what settings I try.
I've stacked a lot of individual frames and got better results, but still looking for somethign less pixellated. I'm going to experiment with this before I take the leap for an ATIK camera or something similar. It gives me lots of time to save up for that too! :lol:

#367 euphony

euphony

    Lift Off

  • -----
  • Posts: 12
  • Joined: 02 Dec 2012

Posted 05 December 2012 - 03:06 AM

I'm just enjoying your pictures on your website with your Lodestar :D

That could be an option for me in the future :hmmmm:

#368 yaesu77

yaesu77

    Lift Off

  • -----
  • Posts: 16
  • Joined: 04 Nov 2012

Posted 05 December 2012 - 01:09 PM

I'm just enjoying your pictures on your website with your Lodestar :D

That could be an option for me in the future :hmmmm:


Hi my friend, you've tried to compile a video with registax? should greatly improve the final result.

#369 mattflastro

mattflastro

    Vendor - Astrovideo Systems

  • -----
  • Posts: 622
  • Joined: 31 Jul 2009

Posted 06 December 2012 - 12:10 AM

I've been playing for the past couple of weeks with both the small and the large versions of the LNTECH cameras.
I have been following this thread and there are a few things that haven't been said (or I haven't noticed them) about this cam :

- in manual frame integration mode (sense-up off, lens manual, x2 to x1024 ) , the 3D noise reduction set to ON does a frame stacking or several frames interpolation of sorts . It stacks several consecutive long integration frames .
This brings the benefits of stacking such as better signal to noise but one big problem. Because the stacking is blind, without a frame alignment algorithm , the mount must track well for a much longer period of time . Instead of good tracking for the up to 17sec (x1024) , now it is several times the 17seconds. That means a good or at least decent polar mount .

- both cameras use an analog front end chip which includes a programmable gain amplifier from 6dB to 40 dB and CDS 10 bit A/D converter model AD9943 from Analog Devices. This is an excellent chip in itself .
- however there is a problem in the fact that there is no manual gain adjustment . There is only AGC with the option of OFF/Low/Mid/High . Based on what I noticed, (I haven't actually measured the gain in low agc vs high agc) it appears that the AGC OFF selects a low fixed value for the programmable gain amplifier. Obviously, this would lower the signal to noise at the A/D input and even using the maximum integration time will not compensate for the low input signal due to this too low gain. The only instance when the programmable gain amplifier is set to highest gain is when AGC is set to High (and after the camera integrates several frames) , but this does also many other undesirable things to the image and is unusable for astro video .
- I am investigating a possible hack of some sort in order to be able to set the menu to AGC OFF while being able to manually select the 40dB gain for the prog. gaim amp . This would maximize the input signal while not having any of the bad effects on the image that would be produced by the AGC set to ON.
- if this hack succeeds, the camera becomes several times more sensitive . Integration times could then be reduced accordingly , with all the advantages of not needing a good tracking polar mount. Or much more sensitivity can be had while using the 3D noise reduction/frame stacking with the shorter integration times.

-another interesting fact is that while the older production cameras use the icx672 ccd , the newest ones use the icx810 . The difference is that the icx810 is a SuperHAd II ccd , NOT an Exview ccd . You can see the ccd model # etched on the chip if you look thru the optical window at the chip itself. Somewhere along the chip edges, next to some of the bonding pads there is written sony icx672 on the silicon . The specifications show very minimal differences between the 672 and 810 . The same geometry , same electrical performance with the 2 exceptions of slightly lower sensitivity of typ 2350 mV for the icx810 instead of 2450 mV for the icx672 and about 5% less sensitivity in IR . The benefit is fewer pixels going bad for the superhad vs exview .

#370 euphony

euphony

    Lift Off

  • -----
  • Posts: 12
  • Joined: 02 Dec 2012

Posted 06 December 2012 - 02:32 AM

Hi my friend, you've tried to compile a video with registax? should greatly improve the final result.


I've been using deepsky stacker. I have tried registax but for some reason it doesn't seem to produce as good results as deep sky stacker, perhaps it's the settings I have used. I tend to find the stacked images still grainy and pixellated in either application, no matter what I do :(

#371 ccs_hello

ccs_hello

    Voyager 1

  • *****
  • Posts: 10,623
  • Joined: 03 Jul 2004

Posted 06 December 2012 - 07:26 AM

Thanks Matt!

If the mfg changed the image sensor (to ICX810/811AKA which is a 960H SuperHAD II, not ExviewHAD II ICX672AKA), they should change the camera's model number.

RE: Gain control issue, I had stated here and here in CN. I.e., if using Electric-Lens and no "Asomething setting", it's long exposure. In this mode, gain is stuck in "Gain=Low (fixed)" mode and not manually settable.
In the sidebar, I had also asked the seller Elaine to contact factory (Korean company) to see if they can change the firmware on RJ11 DSP. The answer back from the factory was "no".
(I understood factory point f view: long exposure + high gain == more hot pixels will show, then people complain...)
We the astro guys know that effect. TEC will "cure" it any way. Better to have choices than no choice. :) :).

In other modes, the digital summing of multiple medium-length exposures will kick in.

Clear Skies!

ccs_hello

#372 mattflastro

mattflastro

    Vendor - Astrovideo Systems

  • -----
  • Posts: 622
  • Joined: 31 Jul 2009

Posted 06 December 2012 - 09:18 AM

ccs_hello,

The programmable gain amplifier is set to LOW by the RJ11 DSP when AGC is set to OFF. I also contacted the factory and asked them about the firmware with about the same success as you . However, the camera board itself gives some clues and hope . There is a connection on the board for in circuit programming of the DSP . There's an app note from RJE describing how to customize various parameters using a programming tool, and there's a datasheet describing the DSP to some extent . So it might be doable at the DSP level.
There's another brute force hardware hack that might be possible , which does not involve doing anything with the DSP , but I'll only know if this is a viable option sometime next week after I hook up a logic analyzer to the AD9943 and see exactly when and how does its gain get programmed .
The problem with the other AGC modes except OFF is that the camera tries to adjust the image brightness by closing a feedback loop over several frame integration cycle and after each successive frame in which the software decides the brightness is too low, it boosts the gain AND the image brightness. After a few iterations the whole image is completely washed out.
If the front end gain could be hacked to be set manually instead of being in this closed feedback loop, the problem would disappear and the camera would gain a number of dB compared to what it is now . Considering that it's already a good cam, the boost would push it into "great" territory.
Intercepting the gain setting command from the DSP to the front end might be easy enough because it's done over a serial 3 wire interface and the front end has only 4 programmable registers. The gain is set by only one of them, and this register only controls gain , nothing else, by writing to this 011h address a 10 bit value in the range 000h to 3FFh for gain from 6 to 40 dB . So the plan is to hook up a logic analyzer to the 3 wire interface and monitor all traffic on it, then just traffic to address o11h which is the gain setting register. After I have that information , it will be clear if I can piggy back a small processor like a PIC or AVR to intercept the gain setting from the DSP . Sounds complicated but in the end it might be easy to implement.

#373 mattflastro

mattflastro

    Vendor - Astrovideo Systems

  • -----
  • Posts: 622
  • Joined: 31 Jul 2009

Posted 06 December 2012 - 03:03 PM

Couldn't resist and probed the serial data interface from the DSP to the AD9943 . I tested what the DSP is sending in various menu combinations and the conclusion is rock solid: I will be able to hack the board to manipulate the FRONT END gain manually, regardless of the DSP menu AGC settings .
In other words, with the menu AGC set to OFF (which right now I verified sets a gain of 5.3dB ) I will be able to manually adjust the front end gain to maximum.
By the way, I just verified that in its factory form, when the menu AGC is set to HIGH , the DSP doesn't program the front end gain to the maximum value of 40dB . It only writes a value of 857 to the gain register which sets it to 35.3dB. I assume this is to avoid hot pixel complaints .
I am ordering some parts and will have a prototype modification working next week.
I attached an image showing the DSP writing the highest gain value to the front end . Top waveform is the serial clock signal and bottom is the address/data signal .

Attached Thumbnails

  • 5557694-AD9943 HIGH AGC sm.JPG


#374 GlennLeDrew

GlennLeDrew

    Hubble

  • *****
  • Posts: 15,850
  • Joined: 17 Jun 2008

Posted 06 December 2012 - 10:34 PM

So, can this camera perform out of the box, or is it a tweaker's project?

The number of posts made about this cheap but seemingly limited device is astonishing.

#375 mattflastro

mattflastro

    Vendor - Astrovideo Systems

  • -----
  • Posts: 622
  • Joined: 31 Jul 2009

Posted 06 December 2012 - 11:52 PM

So, can this camera perform out of the box, or is it a tweaker's project?


Glenn,
It depends on your definition of performance out of the box .
I'd be interested in what would you like this camera to do , or better yet , what would you like your ideal camera to do for that matter ?
If we were to start the conversation with a clean slate, what would you like your astro video cam to be like? As far as sensitivity , functionality, setup, etc. What do you find good/bad about other existing cameras? I noticed you own at least a Mallincam , is there anything you could think of as an improvement for your camera?


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