ZWO Seestar 50

[powerlord]

9 minutes ago, Giles_B said:

Totally agree, ZWO are certainly not alone, and in fairness, it doesn't seem to have an impact on the open software community - overall people still feel willing to contribute. And of course, there is an active community reverse engineering the asiair so it works with other hardware, if you feel tempted to go down that line.

I didn't think there had been any success though as of yet, or am I mistaken ?

 

[Giles_B]

I've not tried to use them, but there are custom kernels for the asiair at openastro https://www.openastro.net/

 

[powerlord]

Sure, I saw that stuff - but that's just using the asiair HW really - it's not added 3rd party devices to the asiair really. I mean, it's interesting and I may give it a try sometime, but originally I thought the plan was to try and 'hack' the asiair app/software to support 3rd party apps - that would have been more interesting to me at least.

stu

[LDW1]

11 hours ago, Giles_B said:

I love my Seestar and the ease with which a beginner can get into astro with the ZWO ecosystem. But we should all have open eyes about this relationship - in terms of open source, it is increasingly clear ZWO have been reprehensible in their abuse of open source development.

Cracked Asiair software code shows it is built on modifications of open source astronomy tools (the INDI server, astronomy.net, even parts of SIRIL). Where it supports other manufacturers, ZWO have modified the code to stop the software working with non-ZWO hardware. The code is open source under various LGPL licenses.  ZWO has refused to publish and distribute their modified source code in violation of the licence. They have been warned but still have not complied.

It's rotten behaviour but international copyright law is deep dark and expensive, so who's going to Sue? Not enthusiasts who have developed the original code for no pay in their spare time... https://disq.us/url?url=https%3A%2F%2Fwww.indilib.org%2Fforum%2Fdevelopment%2F10380-asiair-and-opensource-software-licences.html%3Fstart%3D12%2390031%3AXQ6jVceiroFaqGgGuH08bXO_qfk&cuid=7366295

The ZWO business model is a careful curated but closed ecosystem. It's very doubtful this will change 

incidentally, Apple pioneered this approach (albeit in a slightly different ways and grander scale). So I'm not saying that ZWO are especially villainous, more that this is becoming a successful model that other businesses are replicating.

So after all of this what do you want us, the average astronomer just looking to expand our horizons in this great hobby, to do about it ! Do we really care or am I just missing some important point ?

[Giles_B]

27 minutes ago, powerlord said:

Sure, I saw that stuff - but that's just using the asiair HW really - it's not added 3rd party devices to the asiair really. I mean, it's interesting and I may give it a try sometime, but originally I thought the plan was to try and 'hack' the asiair app/software to support 3rd party apps - that would have been more interesting to me at least.

stu

Okay - I'm pretty new to this, and wasn't aware that this sort of 3rd party support wasn't included - it sounds like its just implementing the open source stuff on the Asiair hardware, which I agree is much less interesting.

11 minutes ago, LDW1 said:

So after all of this what do you want us, the average astronomer just looking to expand our horizons in this great hobby, to do about it ! Do we really care or am I just missing some important point ?

"What should we do?" is the central question of moral philosophy, but a digression on these lines probably isn't warranted! I realise all this discussion is already quite off topic

It's kind of up to you whether you think it's important or something to care about. Personally, I like expanding my horizons in all directions.... including the backstory of ZWO's business model, the question of open sources, and all.

[LDW1]

7 hours ago, Giles_B said:

Okay - I'm pretty new to this, and wasn't aware that this sort of 3rd party support wasn't included - it sounds like its just implementing the open source stuff on the Asiair hardware, which I agree is much less interesting.

"What should we do?" is the central question of moral philosophy, but a digression on these lines probably isn't warranted! I realise all this discussion is already quite off topic

It's kind of up to you whether you think it's important or something to care about. Personally, I like expanding my horizons in all directions.... including the backstory of ZWO's business model, the question of open sources, and all.

You have got to be kidding, lol !  If you have any of their products I would get rid of them immediately, lol !  But maybe it is only you that cares about this trivial ....... !   PS:  Me I'm only going to see whats up there, lol !

Edited November 19, 2023 by LDW1

[cimh]

Had a brief spell on the bubble nebula last night - the image was coming on nicely but after 15 mins a couple of bad frames were added so I aborted. This gave me the impetus to collect individual frames and try stacking in Siril. I only managed 8 mins before the clouds came. Siril stacked 8 mins but after I removed the poor frames I ended up stacking 6.5 mins. I think it might be worth doing to avoid losing a session because of a few poor frames but  I am new to Siril and I am struggling to improve on Seestars final stack. I can often improve the clarity of the DSO but at the expense of background noise (I am in light polluted sky). I know 6 or 8 minutes isnt long enough especially where I live,  although given the short time I had it is amazing what Seestar can do.

Are others doing this? Is the master dark any use? I tried to include it in the stack but was told that it had the wrong number of channels. Any suggestions welcome

 

 

 

[Giles_B]

I'm stacking in sirilic which gives easy control over the stacking parameters. The master dark is the wrong orientation and needs to be rotated 90 degrees or it won't stack. I didn't find the master dark made a difference anyway.

I'm finding making improvements with sirilic and siril a fine art, and what works best varies with the image. I'm not at the computer but will post later giving a worked example.

[cimh]

1 hour ago, Giles_B said:

I'm stacking in sirilic which gives easy control over the stacking parameters. The master dark is the wrong orientation and needs to be rotated 90 degrees or it won't stack. I didn't find the master dark made a difference anyway. I'm finding making improvements with sirilic and siril a fine art . . . . 

Thank you. I was following this video. https://youtu.be/8iL2bzewBrQ Two things I noted. When he converts the FITS, he selects De-bayer at 1'.47". When I tried that it said my FITS were in the wrong format. I wonder if that was why it said my dark had different channels to my lights?   But as you say perhaps not worth worrying about. I also tried out the free GraXpert on an image with a lot of background gradient, following a review by Cuiv. the Lazy Geek, - with very good results

 

 

[Cosmic Geoff]

23 hours ago, Giles_B said:

I'm stacking in sirilic which gives easy control over the stacking parameters. The master dark is the wrong orientation and needs to be rotated 90 degrees or it won't stack. I didn't find the master dark made a difference anyway.

Since the Seestar makes both the pre-stacked FITS and the separate FITS files available, has anyone made a direct comparison to see if stacking the files oneself confers any advantage? I am aware that one is supposed to supply a dark, maybe a flat, and preview all the images to weed out any duds caused by clouds, satellites or whatever, but it is commented above that the master dark didn't work anyway, and the Seestar lights are typically so dark with their 10 sec exposure that one can't see anything in them, which would make screening 200 or so of them extremely tedious.

So far, despite having had the Seestar for two months, I have not mastered the post-processing to the point where I can improve on the instantly delivered Iphone image. ☹️

There's not much point in generating a FITS collection of several hundred MB and moving it around, if the 10Mb stacked FITS is equally useful.

[Elp]

35 minutes ago, Cosmic Geoff said:

Preview all the images to weed out any duds caused by clouds, satellites or whatever

Stacking software provides feedback as to good subs or not, so you don't need to do this. DSS provides a score to each registered image so you can remove those with a low score, Siril provides a plot so you can see the amount of lower quality subs and you can set a FWHM rejection parameter for the software to reject. Kappa sigma rejection removes or averages out satellite trails.

[cimh]

33 minutes ago, Cosmic Geoff said:

Since the Seestar makes both the pre-stacked FITS and the separate FITS files available, has anyone made a direct comparison to see if stacking the files oneself confers any advantage?

I have just started staving individual fits. In my short time with the Seestar I have taken about 20 images of DSOs. Twice the Seestar stack has introduced rogue shots. 1 of the triangulum galaxy where 2 satellite trails were included (I am ok with those they are interesting) and once with the bubble nebula (when the seeing was bad) see my image a few posts above. That made me think how would I feel if I was 30 minutes into a great image when Seestar stacks a rogue. That would be frustrating. So by saving the original fits, if that happens, it will be easy to fix.

However, while stacking is a piece of cake in Siril. Due to bad weather I have only been able to stack twice and no my finished product is not better than Seestar's as on both occasions Seestar did not stack any rogues. So the main reason to save then for me is as a precaution (and I am nerdy and enjoy playing with software like siril)

[powerlord]

5 hours ago, Cosmic Geoff said:

Since the Seestar makes both the pre-stacked FITS and the separate FITS files available, has anyone made a direct comparison to see if stacking the files oneself confers any advantage? I am aware that one is supposed to supply a dark, maybe a flat, and preview all the images to weed out any duds caused by clouds, satellites or whatever, but it is commented above that the master dark didn't work anyway, and the Seestar lights are typically so dark with their 10 sec exposure that one can't see anything in them, which would make screening 200 or so of them extremely tedious.

So far, despite having had the Seestar for two months, I have not mastered the post-processing to the point where I can improve on the instantly delivered Iphone image. ☹️

There's not much point in generating a FITS collection of several hundred MB and moving it around, if the 10Mb stacked FITS is equally useful.

The  main difference is going to be selectivity. Stacking live, it's gonna stack a frame if it thinks it's decent. then it's on to the next one.

When you stack afterwards you have the benefit that ALL the frames can be analysed and the software can make decisions about which is the best 'master', and depending on software, weighting, ones to not bother stacking, etc.

screening is pretty straight forward - at least on a mac - I'd imagine something similar is possible on windoze.

On a mac space bar 'quick views'. i.e. pops up an instant picture of the file. For fits, once you install a quickview viewer for fits (I use the free Quickfits), I can just go to the directory, press the space bar to quickview the first file, and then use cursor keys to move down the list. takes no time. cmd+delete to delete any rubbish ones as I go.

then into Astro Pixel Processor, where it will also make decisions about which to stack if I want based on quality.

However, if beginner, I'd start with the seestar stacked fit, and learn to make that better first. Here's a free process I'd suggest based on Siril, and installing starnet2++ and setting it up with Siril.

- open it in Siril. set siril to autostretch (you will now see it)

- go to plate solve, and enter target, set FL to 250, and pixel size to 2.9nm and plate solve.

- go to colour correction and do photometric colour correction

- save that if you like as file-platesolved-colourcorrected.fit

- histrogram stretch - use auto.

- change view to linear.

- use siril to do a star removal, and create star mask

[you are now looking at a starless version].

- save this as a tiff.

- open the stars only one it created in siril and save that as a tiff.

Now, go into you favourite photo editor that supports layers and load both the tiffs.

you will use a blend layer on the stars - depending on the software it might be called 'lighten' or 'screen' - experiment with them

you can now turn that layer off, and concentrate on the nebulosity - play with curves, contrast, saturation - try to bring out the nebulosity while keeping the background darker, but not pitch black.

add some denoising.

pop the stars layer back on, maybe add some saturation to that too, and probably curves to pull the brightness of the stars down a bit.

- save it as a jpg.

The above may sound like a mammoth task, but with a bit of practice it really only takes 10-15 mins.

btw - nothing about specific to S50 - that is a general colour workflow which will work for ANY astrophotography.

stu

[Giles_B]

I've been restacking the Fits from scratch. I'm a total beginner, but I think I'm getting some improvements in the image with a bit of playing in Sirilic, Siril and Photoshop. I worked on 'Ghost of Cassiopeia' IC63 tonight. I had about four long sessions over the past month. All the images were quite rotated and I wanted to see if I could get something a bit better with reprocessing.

Here's the original and the reprocessed images:

 

Obviously it's cropped a bit, but other than that I took the following steps:

I began by looking at the Fits - four nights and 768 subs
I divided these into 7 x roughly 30 minute sessions in Sirillic (my thinking is that I should get less rotation this way - not sure if that's correct)
Dragged and dropped files into Sirillic - added the Dark Library dark that had been rotated 90 degrees in ASTAP to give the correct orientation
Used identical stacking parameters for all layers.

Under the "properties" tab in SIrilic I selected:

Stack - mean
Rejection type - linear fit
Weighting - WFWHM
Rejecting filter - WFWHM 40%

Once stacking complete the stacked image was too badly rotated with banding in a horrendous diamond shape around the nebula. So I Retried in Sirilic - removed first 3 sessions, and this gave a better stacked image.

Once this was done I did a few quick steps in Siril:
1. Background extraction - removed any samples that went over the nebula or bright stars

2. Saved then Platesolved the stacked Fit in ASTAP

3. Reloaded into Siril and did a photometric colour callibration

4. Removed green noise

5. Did a starnet removal

All that took about 5 minutes maximum

 

I then went into the Generalized Hyperbolic Stretch Transformation - this is the step to spend time on. First set the type of stretch to "generalised hyperbolic transform" (which is the default)

1. Select a point in the middle of the curve to set a symmetry point

2. Set the shadow protection point just below the symmetry point

3. Set the local stretch intensity to something quite aggressive - I started with 5

4. Move the stretch factor slider until just before the curve starts to break up into spikes

5. "Apply" and repeat about a dozen times, gradually backing off on the 'local stretch intensity'

6. After about a dozen times the curve will be stretched out and the background will be getting a bit light - select 'Linear Stretch (BP Shift)' in the type of stretch and gradually use the slider to drag the curve back toward the black point (to the left) - don't put your curve below the black point as this will lose your data

7. Keep repeating these steps. Turn of different colour channels and stretch channels separately if you have the time. Try to stop the curve from getting spikey. If you go to far, you can always unstretch by using the "inverse hyperbolic transform" type of stretch

Once you've got the most detail you can from the nebula, save the image as a "Tif" file and open in photoshop

1. Use the camera raw filter to darken the image a bit, by playing with contrast and shadow

2. Denoise the image (I used topaz denoise)

3. Save the image and reopen in Siril and resave as a Fit file

Now just use the star recomposition tool to add the starnet layer back onto the starless image. Play with the stretch of the background and the star layer, changing the black point if you want the sky darker.

Check the image and crop if necessary!

All in all this can be done quickly, but time invested on the stretching pays off in my opinion.

The only thing I'd add is that sometimes the stretching works better if the stars are left in the image and removed at a later time. In this case I save the image, do a starnet removal, then go back to the original RGB image, later recombining with the unstretch stars. I find this works well for very faint objects.

 

 

 

[DaveL59]

If its not already been posted here's one that shows the S50 under dark skies

 

[DaveL59]

and a head to head of S50 and Dwarf II

 

[Ande]

Is anybody else having issues with their Seestar? I seem to be having a running battle with mine….

Firstly, I usually get it to point at one of the brighter stars, Vega or similar, in order to get a decent focus. It will struggle to locate it and will try to plate solve. Then it will moan about not seeing enough stars. It’ll keep taking pictures, then take the full 30 seconds to try to get it’s bearings.  It will invariably fail, and keep repeating the process several times before reporting it’s inability to complete the task. Eventually, I’ll get the star centred, then it will fail to focus, so I have to do it manually with the aid of a Bhatinov mask.  Once it’s finally in focus, and pointing where it should be, I’ll choose a target. It’ll eventually find it, and then start imaging. However, it is discarding more frames than it is keeping. Far more…. It can take an hour to give me 5-6 minutes of data.  Every 6 or 7 passes it will inform me of star trails, but most of the interim pictures are ignored anyway, so it’s a moot point.  And every now and then it’ll moan that it can’t see any stars during imaging. It’s done this when looking at the Pleiades so I’m not at all sympathetic. Especially when I can see it clearly with my 61 year old eyes. 
 

Another odd thing it will do is suddenly present me with a picture where it looks like the brightest stars have melted downwards, as if the telescope has been knocked. But all of the lesser stars are still sharp, so it makes no sense. Although, somewhat infuriatingly, it’ll happily add the frame to the stack. 
 

With all of the above, I am trying to give the Seestar the best possible chance. I systematically level it before starting anything, regardless of whether it asks for it or not. And if the moons up and bright, and say sat in the east, I’ll pick a target in the west.  I’ll try and place it where it is sheltered from the wind. 
 

I’m getting exhausted keep running outside, because every time I realise it’s not adding frames I go dashing out expecting to see clouds and, possibly, rain. And it’s most frustrating when the sky is crystal clear giving the Seestar no excuse. Right now I feel like sticking my boot under it 🤨

 

Oh, lastly, when I carefully pick the Seestar up, by gripping the tripod, I can feel a slight wobble in the ‘scope. I just assumed that this was a little play in the bearings, or gearing. Just wondering if this is par for the course, or if it should be totally free of any play?  I do screw it fairly firmly to the tripod, so it cannot be attributed to the way it is mounted. 
 

Does any of this ring a bell, or have I got a Friday afternoon special?
 

 

[LDW1]

2 hours ago, Ande said:

Is anybody else having issues with their Seestar? I seem to be having a running battle with mine….

Firstly, I usually get it to point at one of the brighter stars, Vega or similar, in order to get a decent focus. It will struggle to locate it and will try to plate solve. Then it will moan about not seeing enough stars. It’ll keep taking pictures, then take the full 30 seconds to try to get it’s bearings.  It will invariably fail, and keep repeating the process several times before reporting it’s inability to complete the task. Eventually, I’ll get the star centred, then it will fail to focus, so I have to do it manually with the aid of a Bhatinov mask.  Once it’s finally in focus, and pointing where it should be, I’ll choose a target. It’ll eventually find it, and then start imaging. However, it is discarding more frames than it is keeping. Far more…. It can take an hour to give me 5-6 minutes of data.  Every 6 or 7 passes it will inform me of star trails, but most of the interim pictures are ignored anyway, so it’s a moot point.  And every now and then it’ll moan that it can’t see any stars during imaging. It’s done this when looking at the Pleiades so I’m not at all sympathetic. Especially when I can see it clearly with my 61 year old eyes. 
 

Another odd thing it will do is suddenly present me with a picture where it looks like the brightest stars have melted downwards, as if the telescope has been knocked. But all of the lesser stars are still sharp, so it makes no sense. Although, somewhat infuriatingly, it’ll happily add the frame to the stack. 
 

With all of the above, I am trying to give the Seestar the best possible chance. I systematically level it before starting anything, regardless of whether it asks for it or not. And if the moons up and bright, and say sat in the east, I’ll pick a target in the west.  I’ll try and place it where it is sheltered from the wind. 
 

I’m getting exhausted keep running outside, because every time I realise it’s not adding frames I go dashing out expecting to see clouds and, possibly, rain. And it’s most frustrating when the sky is crystal clear giving the Seestar no excuse. Right now I feel like sticking my boot under it 🤨

 

Oh, lastly, when I carefully pick the Seestar up, by gripping the tripod, I can feel a slight wobble in the ‘scope. I just assumed that this was a little play in the bearings, or gearing. Just wondering if this is par for the course, or if it should be totally free of any play?  I do screw it fairly firmly to the tripod, so it cannot be attributed to the way it is mounted. 
 

Does any of this ring a bell, or have I got a Friday afternoon special?
 

 

You are definitely having a lot more issues just to perform its basic operations than most fellow astronomers. I really don't know why, what Bortle class skies did you say you had ? Mine operates, 95% of the time, with minimal effort !

Edited November 28, 2023 by LDW1

[Ande]

Last night’s disastrous session was from a Bortle Class 5. 

[powerlord]

one thing is, do make sure its focused BEFORE you try to goto anywhere - it needs to be focuses to plate solve. You can always focus again later. Note sure about other issues.

[Ande]

I use the focus figures from the previous session as a starting point. However, the Seestar, if asked to then autofocus, will blow the stars up to blurry snowballs, have a half-hearted attempt to focus, and then declare that it can’t see the stars. Then it’s Bhatinov mask time as I manually get the stars sharp again. 

[Stu1smartcookie]

9 hours ago, Ande said:

Is anybody else having issues with their Seestar? I seem to be having a running battle with mine….

Firstly, I usually get it to point at one of the brighter stars, Vega or similar, in order to get a decent focus. It will struggle to locate it and will try to plate solve. Then it will moan about not seeing enough stars. It’ll keep taking pictures, then take the full 30 seconds to try to get it’s bearings.  It will invariably fail, and keep repeating the process several times before reporting it’s inability to complete the task. Eventually, I’ll get the star centred, then it will fail to focus, so I have to do it manually with the aid of a Bhatinov mask.  Once it’s finally in focus, and pointing where it should be, I’ll choose a target. It’ll eventually find it, and then start imaging. However, it is discarding more frames than it is keeping. Far more…. It can take an hour to give me 5-6 minutes of data.  Every 6 or 7 passes it will inform me of star trails, but most of the interim pictures are ignored anyway, so it’s a moot point.  And every now and then it’ll moan that it can’t see any stars during imaging. It’s done this when looking at the Pleiades so I’m not at all sympathetic. Especially when I can see it clearly with my 61 year old eyes. 
 

Another odd thing it will do is suddenly present me with a picture where it looks like the brightest stars have melted downwards, as if the telescope has been knocked. But all of the lesser stars are still sharp, so it makes no sense. Although, somewhat infuriatingly, it’ll happily add the frame to the stack. 
 

With all of the above, I am trying to give the Seestar the best possible chance. I systematically level it before starting anything, regardless of whether it asks for it or not. And if the moons up and bright, and say sat in the east, I’ll pick a target in the west.  I’ll try and place it where it is sheltered from the wind. 
 

I’m getting exhausted keep running outside, because every time I realise it’s not adding frames I go dashing out expecting to see clouds and, possibly, rain. And it’s most frustrating when the sky is crystal clear giving the Seestar no excuse. Right now I feel like sticking my boot under it 🤨

 

Oh, lastly, when I carefully pick the Seestar up, by gripping the tripod, I can feel a slight wobble in the ‘scope. I just assumed that this was a little play in the bearings, or gearing. Just wondering if this is par for the course, or if it should be totally free of any play?  I do screw it fairly firmly to the tripod, so it cannot be attributed to the way it is mounted. 
 

Does any of this ring a bell, or have I got a Friday afternoon special?
 

 

The only time i have the problem you first described is if i have mischieviously tried to goto Jupiter ... i then get the same massage about not having enough stars . I am 100% sure you do this , but please make sure to callibrate the mount (Compass ) and level the mount as accurately as you can . 

[Ande]

33 minutes ago, Stu1smartcookie said:

The only time i have the problem you first described is if i have mischieviously tried to goto Jupiter ... i then get the same massage about not having enough stars . I am 100% sure you do this , but please make sure to callibrate the mount (Compass ) and level the mount as accurately as you can . 

I’m no stranger to the compass calibration dance 😂. And I do level before each session, regardless of whether the Seestar asks for it or not. In fact, I bought a tripod leveller to better do the job.  What concerns me more than the preliminaries, is the huge amount of dropped frames. Over an hour’s imaging last night, on the Crescent Nebula, earned me 9 minutes of data. See pic.

[Giles_B]

Hi Ande,

It sounds to me like there is a problem with your SeeStar and I'd be tempted to send it back - hopefully you did not order it direct from ZWO

My own experience using it in Bortle 5-6, and sometimes under pretty cloudy skies - is that it gets surprisingly good images even in those conditions within a few minutes to half and hour. Autofocus is always needed but takes less than a minute, with no need for additional intervention. They time I have seen star trails is when it is pointing to objects that are nearly at its maximum altitude, and then it discards the images. I think it does drop a reasonable amount of frames, however nowhere near as many as you describe - perhaps four times less than in your example.

The only part of your experience that sounds "normal" is the slight play in the scope, which I agree seems like normal play in the gearing when the motor is not engaged.

[Ande]

Thanks @Giles_B  I’m pretty sure it’s faulty too. After all, the first couple of sessions it performed as it should. Something has definitely gone awry since. Fortunately, I purchased it from FLO, so hopefully they’ll do me proud when I bring it to their attention.