Not too long ago, no one wanting to get into semi-serious, long exposure astronomical imaging would have considered a consumer digital camera as a viable option. Consumer-grade digital cameras were so far from being able to capture faint objects with low-noise long exposures that the notion was laughable.

Things have changed quite dramatically though. So much so that not only are people commonly using digital SLRs to capture objects in the night sky but some have even replaced their specialized CCD cameras in favor of 6-24 megapixel cameras from Canon and Nikon. And in many cases, astroimagers will own both types of cameras.

Yet, the questions often arise: Why have both? Should I get one versus the other? Which is better? As simple as those questions sound, the answers are not easy.

Below is a quick look at the general features and capabilities of each. When referencing DSLRs, I'm generally referring to the latest generation of mid- and high-level DSLRs (such as the latest Canon, Nikon, Pentax, Olympus, Sony, and Sigma cameras). When referencing scientific CCD cameras, I'm referring to those that have regulated thermo-electric cooling and monochrome sensors (like those from SBIG, Apogee, FLI, and QSI).

	Scientific CCDs	Digital SLRs
Imaging area1	medium (except at very high cost)	large
Cost	very expensive (for comparable chip size)	moderate to expensive
Sensitivity2	good to excellent	poor to good
Resolution	excellent	good
Weight/bulk	moderate to heavy	lightweight to moderate
Versatility3	almost none	excellent
Field use	complex/involved	simple to complex
Noise levels4	low to medium	low to medium
Temperature regulation4	excellent	non-existent
Software support5	very good	very good
advantage tie

It's immediately evident that there's no clear winner. The choice of which to buy depends almost entirely on your wants, needs, experience, and goals. That said, here are some things to remember:

A. Imaging with a DSLR can be much easier than with a scientific CCD camera. With a scientific CCD camera you must have a computer, a power supply, and the necessary cabling. When using a DSLR, all you really NEED is the camera (as well as the battery and memory card—but these are internal to the camera and you would have them already anyway). I emphasize "need" because it may be helpful to have the other items (a laptop, an external power supply, etc.) but they aren't required to start taking images... and great images, at that.

B. Thermal signal (noise) cannot be EASILY controlled in digital SLRs. The key terms here are "easily" and "controlled". The noise generated by thermal signal can be removed with dark frames but that isn't "controlling" it per se. It's correcting for it after the fact. There are modifications out there that involve attaching a heat sink to the camera but none of the modifications I've seen fall under the "easy" category. With those things in mind, remember that because of the lack of temperature regulation, you must be more diligent with dark frames. At cooler ambient temperatures, (colder than about 50°F) DSLRs produce quite reasonable levels of thermal signal. In the right conditions, you can often get away with no dark frames. But in warm climates, thermal signal can become a huge problem. You must take dark frames very near the time that you take the light images. Changes in ambient temperature between the time you take the light images and the time you take the dark images will negatively impact the resulting image. With regulated thermo-electrical (TE) cooling available in many of today's scientific CCD cameras, the precise time at which you take darks is irrelevant since a particular temperature can be maintained throughout your imaging session and even from night to night.

C. Scientific, monochrome CCD cameras have higher resolution than DSLRs. In order to achieve 1-shot color images, DSLRs have in place over the imaging sensor an RGB filter matrix (typically a "Bayer Matrix"). Incoming light contains red, green, and blue portions of the visible spectrum (as well as other colors in between and at the extremes, of course). But since each pixel in a DSLR is covered by either a red, green, or blue filter, it only collects one of those colors. The presence and levels of the other colors falling on that pixel must be approximated and calculated by the camera. Monochrome, scientific CCD cameras don't have this RGB matrix. Of course, with scientific CCD cameras you must use filters to achieve color images, which means more equipment, more cost, more complexity, and more overall time⁶—but the resulting image, all else being equal (optics, seeing conditions, etc.), will have considerably better resolution.

D. Scientific, monochrome CCD cameras have greater light sensitivity than DSLRs. This means you can take shorter exposures than with a DSLR. It also means that you are better prepared to delve into narrowband imaging (Ha, O-III, S-II, etc.) should you choose to do so. Of course, this doesn't mean that all DSLRs are less sensitive than all scientific CCD cameras. What it does mean is that if you are looking for the most sensitive, most quantum efficient camera, a dSLR is not the right choice.

E. If your goal is photometry/astrometry, scientific CCD cameras are the better choice. Determining accurate brightness and chemical composition requires accurate and consistent cameras. It requires chips with known and stable sensitivity and thermal signal characteristics. And in some cases it requires a very linear response to photons. These things are simply not possible with off-the-shelf digital cameras. That's not to say that these kinds of science are not possible with DSLRs. For the most accurate results, any findings should be confirmed with a linear CCD camera.

Conclusions
Several years ago the conclusion was simple. If you wanted to go digital and get great photos of deep sky objects, whether for "pretty pictures" or for more scientific purposes, the undeniable best choice was a monochrome CCD camera with regulated thermo-electric cooling.

Today, digital SLRs have come far enough that it's somewhat difficult to draw any definitive conclusions without knowing the user's intentions. Everyone's goals and situations are different. Ideally, you would have both types of cameras. But in the real world that isn't always possible nor is it always desirable. Both types of cameras can produce excellent results with the proper equipment, conditions, knowledge, and skills. A monochrome, scientific CCD camera is almost always going to yield greater sensitivity and greater control over thermal signal. Those facts alone are often enough for someone to make a decision. But if you want relative simplicity, versatility, and cost effectiveness without sacrificing much in the way of quality in the end result, today's digital SLRs cannot be beat.