View Single Post
  #1  
Old 21-02-2013, 10:54 AM
garywiz
Registered User

garywiz is offline
 
Join Date: Jan 2013
Location: Balnarring
Posts: 15
CCD/CMOS with Digital Output vs. Video (SVID, Comp)

I have not yet embarked on Astrophotography as I have just started getting into astronomy with my first real kit. But, Astrophotography will not be far off, as it is a great interest of mine, so I have been "planning my attack" as it were.

I am primarily interested in cameras which can do video, as well as frame-integrated image capture (for various reasons).

However I am an electronics engineer and video expert, and I have been looking at two different types of cameras:

- CCD/CMOS cams like the Stellacam and Mallincam, designed specifically for astrophotography
- CCD/CMOS cams like those from Basler, designed for industrial low-light applications.

Though marketing literature for the Stellacam boasts "Best Astrophotography camera ever created!!", I actually find this quite hard to believe. Why? Because those cameras convert their CCD/CMOS output to composite video (or S-video, which has essentially the same D2A conversion problems).

Essentially video in the form of component, composite or S-Video is *always* inferior to direct bit-by-bit transfers from the sensor itself. I do not own one of these cameras, but assume that they are PAL/NTSC compliant, meaning that they have all the requisite quality loss problems that any D2A converter will have when converting raw sensor data to what is now an archaic video representation format which is not only analogue, but is prone to noise, cable problems and so forth. Worse, these video-based cameras require that you have a A2D capture converter on a PC, meaning that the camera converts D2A (one potential loss area), the cable transmits analog data (another potential loss area), and the capture card (the most significant potential loss area because of the spotty quality of these devices) converts it yet again back into the digital signal that is captured by video or imaging software.

By contrast, a camera like those from Basler (such as the ACA640-100GM), allow direct capture of exact sensor data pixel by pixel, meaning that there is no possible loss, distortion, or conversion between the camera sensor and image capture.

Now, the decision is still not easy. Each of these cameras has its problems. For example, the BA 100GM above has known artifacts which may interfere with deep-sky photography. But, solving that is a matter of choosing the right industrial camera, the correct image sensor, and the right interface (GigaE or USB3) among the many many industrial manufacturers, which allows for a wide range and set of choices instead of the limited choices among "astrophotography webcam vendors".

I am posting to see if my above analysis makes sense, and whether there may be actual advantages I can't imagine that would justify the D2A-A2D losses incurred with video formats.

In fact, I can't even imagine why camera vendors like StellaCam and Mallincam would NOT use digital transfer, as their original sensors are digital. Am I missing something? It wouldn't be the first time!
Reply With Quote