Really tried to use KStars and EKOS
Jasem Mutlaq
mutlaqja at ikarustech.com
Mon Jun 16 05:24:56 BST 2025
Hello Mark,
Can you elaborate on the Manual Rotator bit? Right now in the Align module,
you can "Load & Slew" an image, and if you don't have a rotator, a manual
dialog will popup to ask you to manually rotate until the desired Position
Angle detected in the FITS file is achieved.
--
Best Regards,
Jasem Mutlaq
On Sun, Jun 15, 2025 at 11:48 PM Hy Murveit <murveit at gmail.com> wrote:
> Mark,
>
> I haven't ever messed with the rotator code. That's more Jasem's and Toni
> Schriber's department.
>
> Re the mouse-command rotation, if I had to take a wild guess I'd say we
> just have to figure out what are the
> right keys. Perhaps experimenting with which key is the alt key, assuming
> there are other possibilities, and also making sure shift is depressed
> before the other keys and before the left mouse press-and-hold.
>
> If you can compile and debug/test, just put a printout at the start of
> this method to see which keys are pressed:
>
> https://invent.kde.org/education/kstars/-/blob/master/kstars/skymapevents.cpp?ref_type=heads#L797
> and to see what needs to be done to get inside this if statement
>
> https://invent.kde.org/education/kstars/-/blob/master/kstars/skymapevents.cpp?ref_type=heads#L802
>
> Hy
>
> On Sun, Jun 15, 2025 at 2:31 PM Mark Casazza <markcasazza at gmail.com>
> wrote:
>
>> Hy,
>>
>> Shift-Ctrl-Alt mosue drag does not work on Intel hardware running Kubuntu
>> and a standard "Windows" keyboard.
>>
>> I'm not worried about the screen grab & solve feature. I now know it is
>> there and will probably use it some day, but I don't see it as a common
>> need.
>>
>> Any word on the *manual rotator*? That seems to be my only remaining
>> blocker (assuming we can figure out the FOV rotation issue).
>>
>> Mark Casazza
>> http://casazza.net
>> Home of the Clear Sky Alarm Clock and Tonight's Sky
>>
>>
>> On Fri, Jun 13, 2025 at 10:20 AM Hy Murveit <murveit at gmail.com> wrote:
>>
>>> Mark,
>>>
>>> Re the rotation, please note it isn't shift and most drag. Try this:
>>> hold down shift, then hold down control & alt, then (with all three
>>> still pressed) drag the mouse and you should see the FOV rotate (as opposed
>>> to the sky).
>>> I should make this simpler, but please verify this works for you. I just
>>> tried it and it worked for me. (On my Mac running an Ubuntu VM, the alt is
>>> the "Option" button.)
>>>
>>> Re the 1 monitor, what I do is initiate the screen grab, then use the
>>> keyboard to pop-up different windows until the one with the image I want to
>>> grab comes to the foreground.
>>> On my Mac/VM/Ubuntu setup, that is using Option-Tab to get the various
>>> windows to cycle to the top. I do this on my 16" laptop successfully.
>>> Please give it a try and let me know.
>>>
>>> Hy
>>>
>>> On Thu, Jun 12, 2025 at 7:44 PM Mark Casazza <markcasazza at gmail.com>
>>> wrote:
>>>
>>>> Robert,
>>>>
>>>> I assumed there was some feature to turn on the overlay. I found it.
>>>> Admittedly when in the field with very slow internet it really won't be
>>>> that helpful. An offline full sky image would be very helpful.
>>>>
>>>> Hy,
>>>>
>>>> I watched the video and my shift mouse button rotates the sky, not the
>>>> FOV. And I must say doing the screen capture plate solve on 1 monitor is
>>>> infuriating. The wrong app kept coming up. I'm no fan of the screen shot
>>>> plate solve, but I also don't see myself using it. I do like the rotate
>>>> with shift and the mouse. I was hoping there was some feature like that.
>>>> When I click the star icon to pull the location into the scheduler, I get
>>>> what seems like a nearby location, but not the exact one. I can explore
>>>> this a bit more. *Any idea why my sky rotates and in the video the FOV
>>>> rotates?*
>>>>
>>>> I would be happy to share my spreadsheet of targets. Even at its size I
>>>> find myself adding more all the time. It is basically a subset of multiple
>>>> catalogs with objects big enough that they frame reasonably well with the
>>>> longest focal length all the way up to mosaics with my shortest focal
>>>> length. I'll DM you the spreadsheet. I agree many objects don't show off
>>>> well, but I still enjoy the challenge of those types of targets. (i.e. LBN
>>>> 543).
>>>>
>>>> The remaining miss is the *manual rotator*. I am so looking forward to
>>>> hearing that my search was insufficient and there is a manual rotator.
>>>>
>>>> I am greatly encouraged so far. EKOS is looking better with each reply.
>>>>
>>>>
>>>> Mark Casazza
>>>> http://casazza.net
>>>> Home of the Clear Sky Alarm Clock and Tonight's Sky
>>>>
>>>>
>>>> On Wed, Jun 11, 2025 at 9:04 PM Hy Murveit <murveit at gmail.com> wrote:
>>>>
>>>>> Mark,
>>>>>
>>>>> First off, thanks again for your feedback. We can't learn if we don't
>>>>> hear back from our users, so appreciate it. Also, I want to mention out
>>>>> front that you're welcome to make very specific suggestions or, better, to
>>>>> actually send in merge requests. I'd be happy to work with you to help you
>>>>> with that process.
>>>>>
>>>>> Now to address your specific concerns, which seem to me to be:
>>>>>
>>>>> - You're looking for an efficient way to compose images including
>>>>> rotation,
>>>>> - You're having issues with catalogs
>>>>>
>>>>> *Composition*: I agree that composition needed to be improved, and in
>>>>> fact I made some steps to improve what was there in the past year, but
>>>>> likely the job is not done. However, please take a look at this video
>>>>> https://www.youtube.com/watch?v=r-_xM8gZpM0 and in particular between
>>>>> 2:30 and 4:00 I discuss how you might compose images and adjust rotation
>>>>> and add it to the scheduler. I think it's a reasonably quick process once
>>>>> you've created your field of view--just search for the object, drag the FOV
>>>>> to compose, and rotate with a keyboard shortcut and mouse drag. The same
>>>>> process would work if you weren't using the Imaging Planner at all, but
>>>>> rather just looking at HIPS overlays or other imagery on the skymap. Once
>>>>> you're happy with a composition, you can use a button on the scheduler tab
>>>>> to copy the skymap's center coordinates to the scheduler. I don't think
>>>>> there's any way to copy the rotation angle to the scheduler, but you could
>>>>> just type that in (the necessary rotation value is drawn on the skymap at
>>>>> the tip of the arrow) and, sure, we can add that. Note--it seems like
>>>>> there may be an issue with the keyboard command I mention in the video--it
>>>>> might have changed from the time I recorded that video. I just tried it and
>>>>> holding shift-control-option and then left clicking-and-dragging the mouse
>>>>> worked for me. If you have issues with that, try to shift first, then the
>>>>> other keys, I've seen that come up. We can remap that shortcut if need be.
>>>>> Anyway, please let me know where that technique falls short.
>>>>>
>>>>> *The ImagingPlanner's catalog*: You should be able to download an
>>>>> updated catalog from Data -> DownloadNewData. My original catalog had 555
>>>>> objects, but I updated it in the Spring to 770. Still, of course, that's
>>>>> not the 2000 you want. Here are things to consider. The full Messier, NGC
>>>>> and IC catalogs are available to you in KStars -- "Find" looks them up
>>>>> online. Many other designations as well. That won't work with the Imaging
>>>>> Planner, but it would allow you to center the skymap and compose as I've
>>>>> described above. This should get you the coordinate you desire, though
>>>>> perhaps not much imagery. However, as Rob mentioned, if you use the HIPS
>>>>> imagery (see the View menu) then you'd get the full DSS imagery. You can
>>>>> even download a local copy of DSS and get much quicker rendering. However,
>>>>> this DSS imagery is nothing like a pretty Astrobin picture of a target,
>>>>> which is one of the reasons I put together the Imaging Planner.
>>>>>
>>>>> In my design of the Imaging Planner, I chose to support only objects
>>>>> where I could display a pretty picture. I got those nice images from
>>>>> Astrobin, using only images the authors had given appropriate permission
>>>>> (Attribution Creative Commons, either just that, or including
>>>>> Non-Commercial and/or Share-Alike). The reason I was limited to ~770 was
>>>>> that there were many objects I wanted to show but at the time I didn't find
>>>>> images that had appropriate permissions. Of course, there are probably more
>>>>> images out there now, and that's just searching work that needs to be done.
>>>>> If you have a source for object images that have that kind of sharing
>>>>> license (i.e. either yours, or others' that you know of) please let me
>>>>> know, or I can show you how to put together a supplemental Imaging Planner
>>>>> catalog.
>>>>>
>>>>> Also, even if nice images were available for all NGC/IC objects, I
>>>>> think that it is probably best not to include (at least by default) every
>>>>> NGC/IC item as that would probably bury the "more desirable" objects. So,
>>>>> when I was looking, I certainly ordered my search by popular objects and
>>>>> objects from successful imagers. Still, I'm sure there's room for many more
>>>>> objects--why not your curated list of 2000 objects if we can find images
>>>>> for them?
>>>>>
>>>>> *Scheduler template* I think we effectively have this already, it's
>>>>> the Sequence (.esq) file you can load from a menu. The other things are
>>>>> target dependent (name, coordinates, rotation). What am I missing?
>>>>>
>>>>> Hopefully that'll get the conversation started. Please let me know
>>>>> what you think,
>>>>> Hy
>>>>>
>>>>>
>>>>> On Wed, Jun 11, 2025 at 11:53 AM Robert Lancaster <rlancaste at gmail.com>
>>>>> wrote:
>>>>>
>>>>>> Hi Mark,
>>>>>>
>>>>>> Great feedback, I note one item on your list might be easily
>>>>>> resolved. You said this:
>>>>>>
>>>>>> The area of the sky is displayed, but minimal nebulosity; even the
>>>>>> Veil Nebula is only shown as a quadrilateral.
>>>>>>
>>>>>> Did you try the hips overlay on the skymap? If you turn that on and
>>>>>> select “DSS Colored” you will get a great deal of nebulosity for the veil
>>>>>> and many other targets. This feature is great for framing and looking at
>>>>>> how targets will look on your sensor. Pairing this with the fov rectangle
>>>>>> for your sensor is a game changer. I use this feature constantly myself.
>>>>>>
>>>>>> Thanks,
>>>>>>
>>>>>> Rob
>>>>>>
>>>>>> On Jun 11, 2025, at 2:35 PM, Mark Casazza <markcasazza at gmail.com>
>>>>>> wrote:
>>>>>>
>>>>>>
>>>>>> Jasem,
>>>>>>
>>>>>> Let me start with EKOS as I spent the most time trying to make it
>>>>>> work for me. And before I get into application details, let me describe my
>>>>>> use cases.
>>>>>>
>>>>>> I generally image 1, or at most 2, objects a night from dusk till
>>>>>> dawn at a borderline Bortle 2-1 site. As a result, I do not bother with
>>>>>> filters. Camera orientation is very important to me, and because I only
>>>>>> image 1, or sometimes 2, objects in a given night I can provide the camera
>>>>>> rotation. When I do image from my home, a bortle 5 location, I use a strong
>>>>>> dual band filter and image with and without the filter. Due to my
>>>>>> simplistic filter use I provide the filter changes.
>>>>>>
>>>>>> I did not dig deep into filters in my exploration of EKOS, because
>>>>>> rotation alone was a deal killer. I was unable to find "manual rotator"
>>>>>> where the UI prompts the user to rotate the camera a number of degrees in a
>>>>>> given direction. NINA provides this and during the plate solving process
>>>>>> first establishes proper rotation, then sky location. I am able to provide
>>>>>> a tolerance (I use 5 degrees) and a number of iterations before a failed
>>>>>> rotation (I use 5 again). This process only occurs once per night and once
>>>>>> set all is good.
>>>>>>
>>>>>> But before I can go out and image an object I need to frame it up
>>>>>> myself. While EKOS has a way to set framing in KStars and an Imaging
>>>>>> Planner tool both fall short in this area. The process to set rotation is
>>>>>> so click heavy that I gave up on my first attempt after 15 minutes. In NINA
>>>>>> I can set the rotation in a few seconds. This speed becomes devastating
>>>>>> when I want to image 2 objects in the same night and I need to find the
>>>>>> optimal camera rotation for both objects. This requires quite a bit back
>>>>>> and forth to try various orientations on each object. If each orientation
>>>>>> took minutes, not seconds, this could take many hours. As with all my
>>>>>> feedback I woudl love to hear that there is a feature or method that I have
>>>>>> yet to discover. This is how I tried the framing process in EKOS:
>>>>>>
>>>>>> 1. Open Kstars
>>>>>> 2. Select Search icon
>>>>>> 3. Type in target object name or designation. (Note "eleph" does
>>>>>> not find the Elephant Trunk Nebula, but instead only offers the comet
>>>>>> Telephus 1991 KC)
>>>>>> 4. Select the object from the list
>>>>>> 5. The area of the sky is displayed, but minimal nebulosity; even
>>>>>> the Veil Nebula is only shown as a quadrilateral.
>>>>>> 1. This is a show stopper, but I assume there is some way to
>>>>>> supplement the data to get better images displayed.
>>>>>> 6. I have already set my camera(s) up as FOVs so I can see how
>>>>>> the object fits each potential focal length. This is better than NINA!
>>>>>> 7. To rotate the frame I need to:
>>>>>> 1. Settings
>>>>>> 2. FOV
>>>>>> 3. Edit FOV
>>>>>> 4. Pick camera
>>>>>> 5. Edit
>>>>>> 6. Change rotation value but providing a new number.
>>>>>> 7. Okay
>>>>>> 8. Okay
>>>>>> 9. See results and repeat as needed.
>>>>>> 10. Yikes!
>>>>>> 8. Once I get a good framing I need to note the RA, Dec, &
>>>>>> rotation to set up the exposure sequence. It woudl be very helpful to
>>>>>> quickly jump into EKOS with a framed object and add it to a schedule.
>>>>>>
>>>>>> I tried to use the Imaging Planner as well. This is probably the
>>>>>> place to develop the features I'm seeking. This tool has the necessary
>>>>>> nebulosity visible but lacks:
>>>>>>
>>>>>> 1. The ability to see the image in my camera's FOV (with various
>>>>>> focal lengths) unless I select "Center on SkyMap", but then I really need 2
>>>>>> monitors...
>>>>>> 2. A robust target list. I wish 555 targets was a good list, but
>>>>>> reality is I work with a list of 1,780 possible targets in my primary list
>>>>>> with the full NGC/IC and other catalogs a tab away in a highly customized
>>>>>> spreadsheet. Only occasionally does NINA not have my target by designation
>>>>>> and I have to enter RA & Dec.
>>>>>> 3. Any obvious way to quickly get an object into a schedule.
>>>>>>
>>>>>> In contrast with NINA the process is:
>>>>>>
>>>>>> 1. Open NINA
>>>>>> 2. Select profile (location, telescope, camera)
>>>>>> 3. Select "Framing"
>>>>>> 4. Type in the name or designation of the desired target and pick
>>>>>> it from an auto-populated list. (i.e., "eleph" to find the elephant trunk
>>>>>> nebula)
>>>>>> 5. Select Load Image.
>>>>>> 1. This could download the sky from various Internet resources
>>>>>> 2. I prefer to use the offline Sky Map for quick response.
>>>>>> 6. I can now use the mouse wheel to zoom in and out and drag the
>>>>>> sky around to properly frame up the shot.
>>>>>> 7. If I find the object is too big or small for the focal length
>>>>>> I initially selected I can adjust the focal length value on the page and
>>>>>> the camera's field of view will be updated. This allows quick views with
>>>>>> various compressors and extenders.
>>>>>> 8. There is a "Rotation" slider that rotates the camera field of
>>>>>> view. I can rotate as needed. (This feature is what EKOS really needs.)
>>>>>> 9. With all the above changes the RA, Dec, and rotation are
>>>>>> updated so creating a session is reduced to:
>>>>>> 1. Selecting "Add Target to Sequency"
>>>>>> 2. Select "Legacy Sequencer" or "Sequencer"
>>>>>> 3. Selecting the prebuilt template to apply.
>>>>>> 10. At this point I can go into the sequence and add all the
>>>>>> details like # images, duration of images, start and stop time, etc.
>>>>>>
>>>>>> Regarding filters, as I stated I have not gone deep with EKOS, but I
>>>>>> know even NINA falls short of what I think are essentials. The biggest
>>>>>> thing that I have discovered when using strong narrow band filters is the
>>>>>> need for a very different exposure setting for both auto focus and plate
>>>>>> solving. I ask NINA to plate solve every 3rd image to make sure nothing
>>>>>> traumatic has happened to greatly move the mount away from the object. I
>>>>>> have not dug deep enough into EKOS to know if you have a simular
>>>>>> capability. But I did see the ability to auto focus throughout the night
>>>>>> and that will drive those longer duration and higher gain exposures. Also,
>>>>>> the polar alignment process uses plate solving and can become a time killer
>>>>>> if you cannot use gain to allow for short exposures and quick feedback as
>>>>>> you make adjustments. Again, I have not had EKOS out under dark skies
>>>>>> because I got stopped with the rotation issues.
>>>>>>
>>>>>> All this said, the product is impressive and I see great potential. I
>>>>>> look forward to the day I can get off Windows altogether.
>>>>>>
>>>>>> Mark Casazza
>>>>>> http://casazza.net
>>>>>> Home of the Clear Sky Alarm Clock and Tonight's Sky
>>>>>>
>>>>>>
>>>>>> On Tue, Jun 10, 2025 at 12:31 AM Jasem Mutlaq <
>>>>>> mutlaqja at ikarustech.com> wrote:
>>>>>>
>>>>>>> Hello Mark,
>>>>>>>
>>>>>>> Thank you for your email and we'd love to hear more from you. Please
>>>>>>> share your experience and what it is that you miss in KStars/Ekos.
>>>>>>>
>>>>>>> --
>>>>>>> Best Regards,
>>>>>>> Jasem Mutlaq
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>> On Tue, Jun 10, 2025 at 7:23 AM Mark Casazza <markcasazza at gmail.com>
>>>>>>> wrote:
>>>>>>>
>>>>>>>> As a fellow software developer I want to offer feedback for KStars
>>>>>>>> and EKOS as replacements for my Windows tools (SkyTools & NINA). I really
>>>>>>>> want to get off Windows and I've managed to convert every device except the
>>>>>>>> two that control telescopes. That is six of eight computers now running
>>>>>>>> Kubuntu with no regrets.
>>>>>>>>
>>>>>>>> The EKOS / NINA feature gap is too wide to even consider making the
>>>>>>>> jump for imaging. KStars with Observation Manager seems closer to the
>>>>>>>> feature set I need for visual observing with a 20" truss tube scope, but it
>>>>>>>> would be a major investment in catalog population for objects like quasars
>>>>>>>> and globulars in other galaxies.
>>>>>>>>
>>>>>>>> I would be happy to share the details of my A/B comparisons and
>>>>>>>> suggest a few "easy" wins that could really help along with bigger items
>>>>>>>> that might be worth including in your roadmap. I fully appreciate that dev
>>>>>>>> teams have their plans and priorities so I will wait to hear back from you.
>>>>>>>>
>>>>>>>> I fully respect the great work you have done and I have several
>>>>>>>> friends very happy with their Linux telescope control and image acquisition
>>>>>>>> solution. I even recommend it whenever I find someone open to a Linux
>>>>>>>> approach. I admit that I am spoiled and a power user. There are features I
>>>>>>>> would be lost without that seem to be missing.
>>>>>>>>
>>>>>>>> Clear skies,
>>>>>>>>
>>>>>>>> Mark Casazza
>>>>>>>> http://casazza.net
>>>>>>>> Home of the Clear Sky Alarm Clock and Tonight's Sky
>>>>>>>>
>>>>>>>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.kde.org/pipermail/kstars-devel/attachments/20250616/6062620a/attachment-0001.htm>
More information about the Kstars-devel
mailing list