M8 in 8 seconds

[codemonkey]

Ok, more like 696x8s, but all the same.

Because I wasn't guiding I set up SGP with several targets, though they were all M8. I do this because I can then set it to center the target at the beginning of each "target" this keeping M8 fairly well positioned throughout the night, even though I wasn't guiding. Unfortunately for me, I stuffed up in all but two of the targets, and set it to finish entire events, rather than cycling through the different events, resulting in getting a crapload of luminance and red subs, but bugger all blue or green.

I've basically spent all day since about 8:30am weighting, registering and stacking these subs. Unless you've got a super-computer, I do not recommend it. For those wondering why I did it, I had a choice between guiding and using my flattener, so I used the flattener and did very short subs.

Anyway, I think it's an ok result, especially for 8s subs. Sub count was:

L = 388
R = 262
G = 24
B = 22

Image was cropped because I preferred the composition like this. Full version (same crop) on astrobin (http://www.astrobin.com/254826/).

[Mighty_oz]

Looking pretty good to me for 8 secs subs :) Lots of detail there.

[Atmos]

For a moment there I was expecting to see a 2s LRGB image (8s total) :P There is quite a nice amount of detail in there, although a bit noisy towards the background which is to be expected with 8s images :P
Surprisingly, the core hasn't been blown out :lol: :D

For only about 3 minutes of GB data they don't seem to have been impacted much on the image overall :) There is a part of me that wouldn't mind getting one of these cameras for LRGB imaging as it gives a wider field than my 694 and has significantly lower blue response :P

[glend]

Lee, Obviously taken with the ASI1600. If your mount is reasonably aligned, unguided 20" subs would be easier - less subs required. Yes that many of thise subs is going to mean alot of storage and memory used. What camera settings did you use?

[codemonkey]

Cheers :-)



Cheers Colin. haha, no, the core certainly didn't blow out on this one :D

I think the majority of the noise comes from (a) me being lazy and using unsharp mask on it because I gave up trying to get decon to play nice and (b) the noisy B & G data.

I've attached a small, 100% crop of one of the darker areas show in the image, in luminance, without any sharpening... it's a fair bit less noisy than the LRGB image.



Cheers Glen. The mount is reasonably well aligned, but the Linear Fast Reverse sometimes has significant RA drift that you can only battle with fast (every second) guiding. It doesn't happen all the time, and is basically a product of the PE from multiple belts coinciding, according to Avalon. When it's having one of those days, I doubt I could get anything of use with 20" unguided subs at this kind of resolution. I'm also particularly anal about the quality of subs I use.

I expect I'll be back to guiding the week after next (or whenever the skies clear after that) so it's not something I plan on trying to optimise.

As for camera settings, I used unity gain, with whatever ASI default the offset to for unity. I was using the 1.0.2.8 ASCOM driver, USB 3, speed set to 45. It dropped one sub all night.

[codemonkey]

And here's the full image (http://www.astrobin.com/254839/) (ok, I cropped off the ragged edges), with much less contrast and no sharpening at all. Noise wise it looks pretty good to me to be honest.

[SimmoW]

Hmm, thanks for posting Lee, I am currently tempted to get this mono version over any 8300 alternative, with the price being so good. What are your thoughts on the cam? Especially if I team it with my 430mm FL tak? Keep posting pics!

[gregbradley]

A nice dreamy looking Lagoon. Nicely done and worth the effort.

Greg.

[atalas]

Nice work Lee!so,8 sec subs eh.....so the mount wouldn't allow you longer?I would of thought with good polar alignment 2 min would be ok.

Anyway I like what you've been doing with the little cam...very cool.

[codemonkey]

Cheers Simmo. I haven't used the cam enough to have a firm opinion on it, so my advice is based mostly off specs and the experiments done by others.

I wouldn't buy an 8300-based camera--I was never prepared to take the hit of higher read noise, dark current (and thus noise) and low QE. You have to trade all of that away to get a bigger FOV than you get with sony-based cameras and I was never willing to do it. If I had more light polluted skies and if my main targets were very large, maybe I'd have considered it, but neither are true for me.

First off, the ASI 1600 is a bit of a risk. There's not been much work done with it yet, and it may have some annoying issues (detailed below). It's up to you whether those (possible) issues are deal-breakers or not.

I think the ASI 1600 would pair well with your 430 scope and better than the 8300 due to the pixel size. You'd get about 1.8"/px with the ASI 1600, and 2.6"/px with an 8300. The former is undersampled, the latter much more so.

The ASI has significantly lower read noise, dark current and about the same QE (we think) and the price is great, so while the pixels are smaller, you're probably not losing much at the pixel level in terms of SNR, whilst still getting better sampling.

There's a few down sides to the ASI though. It has amp glow, and if you want to do super long subs, it's going to be a problem. It's fine with short subs. At this stage I'm not sure I'd want to be using it for subs over 10mins in length. Having said that, due to the reasonable QE and the low read noise (< 2e @ unity), you shouldn't need to do long subs anyway as, for broadband images at least, you should quickly become sky limited.

The next downside is, based on others' reports, there seems to be differences in calibration frames between sessions. What I mean by that is if you shoot calibration frames one night, then power down the camera, you may not be able to use those calibration frames the next night. I've not been using any calibration subs, and when I do I hope to keep the subs fairly short and avoid doing anything besides flats & bias, which I can reasonably capture each night.

Final downside (imo) is there's been a lot of issues with the drivers and compatibility with SGP. Seems to have been working well with other software. I expect these issues to be transitory, that's the great thing about software eh? Still, if reliability is more your thing, that's a tick against the ASI 1600 at the moment.



Thanks Greg!

[codemonkey]

Cheers Louie :-)

Eh, the mount might have been able to do better, it depends on the day, as I mentioned in my post to Glen. This is a mount that really need to be guided. It can (and usually does) perform well guided, but definitely not something you want to try serious unguided work with.

I've seen the mount drift 250" in RA over the space of an hour. That's about 0.5" in 8 seconds if my math is correct (it's frequently not!). I'm sampling at 0.93"/px and I consider 0.5" drift in the sub length to be acceptable, but I wouldn't want to go much over. Keep in mind that's just average drift, and doesn't consider any of the little bumps you get when tracking.

By keeping subs at 8secs I was able to get FWHM of < 3" for 696/734 subs. FWHM was as low as 1.7" in some subs.

I hate having blurry data, and I'd rather have more noise, thus the sub length.

[dylan_odonnell]

Thanks for this. Really fantastic result. The more I see from this camera, the more tempting it is to jump over from my OSC RGB CCD to this relatively cheap mono CMOS.

I did a similar image using my f2 Hyperstar setup and the ZWO 174mm and 1 second subs on Orion .. my whole image run was over in a couple of minutes! It doesn't have the chip size or resolution of the 1600 though. Lots of potential for getting great results in a single night rather than spending a month waiting for the conditions and time to get your data.

Although CCDs have slightly more dynamic range it matters little at that price point and is barely noticed after processing anyway.

[Placidus]

That's a very good Hourglass in there.

[rustigsmed]

nice work lee, great hourglass as mentioned below.

definitely got me thinking of getting one of these cams to run on my goto dob (for colour) while I capture L with my 694 on the eq8.

[Rod771]

Pretty good Lee!

I'm keeping an eye on that camera.

[E_ri_k]

Pretty good for 8s subs! Decent detail, well done!
Erik

[codemonkey]

Thanks Dylan :-) I'd personally take this over any OSC CCD, but that's me. Its dynamic range is actually quite good. At zero gain you have about 12.5 stops. The QSI 583 has 70db of dynamic range, which I believe is equivalent to ~11.67 stops. Of course the ASI has adjustable gain so you can tweak that, and its nice to have that flexibility.



Thanks Mike!



Sounds like a plan, Russell!



Thanks Rod :-)

[codemonkey]

Thanks Erik!

[RickS]

A good result for such short subs, Lee, but it's a fairly bright FOV which helps.

[codemonkey]

Thanks Rick. It is indeed a bright FOV, thus why I selected it, knowing that I was restricted to short subs.

[Stevec35]

A very good result for such short subs!

Cheers

Steve

[Shiraz]

that's a great illustration of how lots of short subs = extreme dynamic range. You have managed to to get fine detail in the hourglass region and acceptable low noise elsewhere. And apart from all that it is a very attractive image as well.

[codemonkey]

Thanks Steve!



Thanks Ray :-) Could have done with more data, as is always the case for me... I get distracted easily.