[Xeteth (David)]

Wow there is a lot to digest here. Thanks all for your replies and info, it's much appreciated!

Originally Posted by Camelopardalis

You need to determine what better means to you before you can compare. There is no “one chip to rule them all”, they all suit different scopes differently and have different quirks and costs.

For example, you mention noise. There are a number of different sources of noise. Two of the most common are read noise and thermal noise. If you compare those contributions, you will see that the KAF8300 performs quite poorly in those respects compared with more recent CMOS sensors.

QE is quantum efficiency. It is the proportion of photons received that generate electrons that are subsequently measured as signal. Higher QE is always better.

I had mainly been focusing on thermal noise being the issue and had not really considered readout noise. I will definitely look into this now!

QE seems rather simple to understand... the QHY9 has a peak QE of 56%, would this likely mean I'd run into issues?

Originally Posted by RobF

I moved from a standard Canon 450D to a QHY9 years ago (2010) and its been a great camera. The DSLR was good on bright objects, but on more diffuse objects in summer (e.g. Rosette) it was a real struggle to pull out faint H2 structure.

Either camera would be an exciting improvement. The QHY9 8300 chip will give more noise, but that can be managed with proper bias/darks/flats.

I don't mind having to acquire a larger number of images to stack out any noise and am almost at the point where I'm confident with what I'm doing with darks/bias/flats, so it's great to hear that those issues can be overcome. Love the HA image you linked, pretty damn impressive!

Originally Posted by Atmos

This then moves into pixel sizes and how the ASI1600 with its smaller pixels stress' optics more. Let's just say that your refractor creates star sizes of 7 microns in red, 6 in green and 10 in blue. With the QHY9 (5.4 micron pixels) all of the stars will be contained within less than 2 pixels. As you move into the deeper blue (towards UV, where the spot sizes are more likely to reach 15 microns), the QHY9 becomes very insensitive so there is not much detectable bloat; stars are their true colours.

With the ASI1600 (3.8 micron pixels) the red and green are both below 2 pixels BUT the blue is at 2.6 pixels. It has a considerably higher QE towards the deep blue so there may even be a fainter blue halo reaching out to 3-3.5 pixels.
This means that you could have the centre 2 pixels being correct star colour but then a blue halo surrounding any non-dim star.

For narrowband none of this matters as you're only imaging a VERY thin slice of the spectrum. I've also just been picking random numbers purely for illustrative purposes. You're refractor, although not entirely cheap, is on the lower end of the triplet imaging refractors so it is built to a price point. Excellent value but it can have its limitations. Most of the halos and what I've been discussing can be processed out but my point in all of this is that if you're planning on LRGB images you will near certainly get better star colours out of the QHY9 than ASI1600. If you plan on doing narrowband then the ASI1600 is a definite winner over the QHY9 due to the higher QE and MUCH lower read noise.

Thanks for this write-up, certainly shed light on it all! One question - can the lower QE of the QHY9 (as compared to the ASI1600) be overcome simply by stacking more subs? Or will I still lose out?

Originally Posted by casstony

If you have the time and commitment to learn then the cooled mono cameras will give the best results; otherwise you can still do quite well with a much less noisy dslr.

I started with the 600D (almost identical to the 650D) which was very noisy. I now use the Nikon D5600 which is fine for most of the year in our southern climate.

The Helix was taken recently after I modified my d5600, 15x7 minutes. The Rosette image was taken last year before I modified my camera - I'm looking forward to imaging it again with the extra Ha signal and a few more processing skills.

On those rare nights when I'm tired and the things just won't work because I've done something wrong it's nice to have less complicated gear.

I'll certainly be keeping the 650D, it has it's perks for sure! Especially on those brighter targets. Love the rosette there, would be really interesting to see what you come up with given extra Ha signal.


Despite all this advice I am still somewhat unsure on how to proceed. I have the option of purchasing a QHY9 for about $2-2.5k at the moment. It seems like a great camera, but with the higher readout noise and lower QE than some of the CMOS cameras (namely the ASI1600) I am a bit concerned I'm not making the right purchase; which makes me think of getting the 1600 instead...

Given the two options, which would you guys likely choose?