793 replies to this topic

[Domdron]

Hi

iPhone's native iOS camera app averages in Night Mode only, but it does not align photos.

I don't think that's true -- the attached image (of my house) was a 2 sec exposure using the native iPhone Photos app, hand-held. No way that there's not some form of alignment going on there. 

 

Attached Thumbnails

• 

[astrshdr]

I've downloaded AS and still not had a chance to use it, but interesting topic above. I already commented a few things on the previous smartphone thread but a few more questions/observations, if I may.

 

1) It does seem a bit odd to let the iPhone stack 4 shots (4x1s) to get a sub. I assume that the iPhone does not align and stack as well as Astroshader?

 

So if AS stacks better, then would it not be better to only let the iPhone do its 1s shot then AS does all the stacking? Please explain (I understand there might be data size issues.)
Sorry, my previous answers may not have been 100% clear on this. AstroShader is doing all the stacking here; the entire image pipeline is build from the ground up starting with the sensor data. Both the sub-stack and rgb stack are done by AstroShader.
The whole 1s (or max exposure, e.g. 0.3s) thing is due to a hardware limitation on iPhone cameras where they have a very low max exposure for some reason. To partially get round this upper limit to allow say a 20s exposure I stack the raw sensor data, amongst other things.

 

2) Does AS rotate the image to align it? Maybe not an issue if on a short duration, but in 300s the image would show rotation with any Alt-Az and that seems to be the case with the stars in that orion image, some look stretched and that will blur the rest of the stack. 

Yes - the image will be rotated (and translated) to align.

(The star stretching in my recent image is actually from extreme field curvature due to me using a monocular rather than a telescope. The area outside the central 1/2 of the image is very warped. The iPhone camera lenses will also cause some field curvature towards the edge of images too.)

 

If AS stacked and rotated to align them (maybe a tick box option so EQ mounts don't get the processing overhead) it might improve things quite a bit.

This is a good idea, certainly for longer imaging runs.
The processing at present is fast—in the milliseconds range—as AstroShader uses hardware acceleration.

 

3) I found with NightCap that things improved greatly in terms of clarity when dropping the iso to 400 and even 200. Perhaps overdoing the ISO is leading to the stars not being more point-like.

I agree, there's definitely a tradeoff between stars and faint detail when choosing ISO, I think it's due to the native ISO range in the cameras making it easy to overexpose and causing the "blooming" around stars.

 

The app looks really great and hopefully I'll get a chance to give it an outing.

 

 

Hi Leafus,

Thanks for posting your questions... they seem like a good opportunity to see if I have understood what AS actually is capable of doing.  So I've answered some of your questions above in red (hoping my answers are correct).  I know NightCap quite well... just starting to understand AS.  I cannot answer your question #2 though, because I do not know how many registration points AS uses to align photos.  

 

I will begin to use AS very soon, using the iPhone's optics and with the phone attached to my telescope on my AZMP mount.  I know from experience with NightCap, that using the phone at magnifications higher than about 10x becomes exponentially difficult, without a tracking mount.  Without tracking, individual exposure times are necessarily kept so short that it limits photos to the brightest subjects.  Tracking permits longer exposures and the accumulation of multiple photos to average out electronic noise.  But as subjects become increasingly dim at higher magnifications, even with a tracking mount, they reach a threshold where exposure times and/or ISO cannot be pushed higher.  

 

I too, am looking forward to using AS.  I think AS has expanded the capabilities of phonetography, but testing is the only way to find out.  

 

I hope astrshdr can confirm or correct what I've written above.  I just want to make sure I understand correctly before I start using AS.  

Ray

Thanks Ray—you're pretty much on the on money with those answers there. I added a couple of comments for context.

 

You got me thinking about untracked a bit more...

At present I've using AstroShader untracked by taking a single image without re-aligning, e.g. letting the target drift across whilst taking short enough subs to avoid blur.

Maybe it would be useful to allow people to take a number of separate image runs, re-aligning the scope between, then select say 50 images to align and stack from their photos library?
 

[GeezerGazer]

Hi

I don't think that's true -- the attached image (of my house) was a 2 sec exposure using the native iPhone Photos app, hand-held. No way that there's not some form of alignment going on there. 

Hi Dom, 

Nice home and photo!  The recent iPhones do have a robust system to compensate for handheld variations in long images with either Optical Image Stabilization (OIS) on older models, or the improved In Body Image Stabilization (IBIS) which is more effective in adjusting the sensor (5 axis adjustment) to compensate for movement.  Also, if the accelerometers detect movement, as in hand held imaging, the phone will compensate in Night Mode, by limiting exposures to 2, 3 or 5 seconds, instead of 30.  It's like magic.  But I've found no evidence that Night Mode actually aligns photos using a single registration point in multiple photos.  In a way, you are correct, there is alignment going on, but it is with the sensor on iPhone's starting with the 12 ProMax... with prior iPhone models, it's the lens assembly that moved for image alignment to compensate for handheld movements.  

 

Here's a good description of what OIS can do in Night Mode from the Apple website: https://support.appl.../en-us/HT212803

 

This link describes what IBIS does in the iPhone 12 ProMax and more recent iPhones:

https://www.macrumor...hone-13-models/

 

In each link, you have to scroll down a bit to get to the pertinent info. 

 

You should not be able to take a Night Mode, 30s exposure, hand held because the phone will detect movement and limit/restrict the averaging time.  If you place your phone on a steady rest or tripod, where it is held perfectly still, it can raise the total averaging to 30s, if that length of time is needed for best exposure.  

Ray

Edited by GeezerGazer, 02 February 2023 - 04:38 PM.

[GeezerGazer]

You got me thinking about untracked a bit more...

At present I've using AstroShader untracked by taking a single image without re-aligning, e.g. letting the target drift across whilst taking short enough subs to avoid blur.

Maybe it would be useful to allow people to take a number of separate image runs, re-aligning the scope between, then select say 50 images to align and stack from their photos library?

I use an alt/az tracking mount for photos with a total exposure time limitation to prevent field rotation,  which is good enough for me to complete decent phone images.  But if AS permitted the importation and alignment of ~50 images from the Photos library, that could change the dynamics for those who use stationary, non-tracking optical systems.  

 

What you suggest would permit iPhone users, for the first time, to do what Android phones have been doing for several years for planetary and DSO photos... aligning and averaging multiple individual images... but you would be accomplishing this all in one app!  Android systems rely on 2-5 different apps to do the same thing.  That would be an impressive feat!  

Ray

[GeezerGazer]

iPhones have been used to take multiple images that have been aligned and processed previously, but not in one app!  In the following link, CNer Loren Ball produced an excellent iPhone image of Luna taken with only 90mm of aperture.  But he used four different apps either on his iPhone SE or his computer for the process.  At the bottom of his narrative in post #12, just before the image, he explained what he used:  https://www.cloudyni...artphone-image/

 

AstroShader is designed to complete that same process without leaving the app.  THAT'S why I am excited about AS!  

 

We have complete cloud cover today, followed by 3 days of rain... but Mon-Wed next week looks to be clear.  I'll be outdoors!  

Ray

[astrshdr]

I use an alt/az tracking mount for photos with a total exposure time limitation to prevent field rotation,  which is good enough for me to complete decent phone images.  But if AS permitted the importation and alignment of ~50 images from the Photos library, that could change the dynamics for those who use stationary, non-tracking optical systems.  

 

What you suggest would permit iPhone users, for the first time, to do what Android phones have been doing for several years for planetary and DSO photos... aligning and averaging multiple individual images... but you would be accomplishing this all in one app!  Android systems rely on 2-5 different apps to do the same thing.  That would be an impressive feat!  

Ray

At present AstroShader should be able to deal with field rotation when using an AZ mount.

The image below was from some recent testing with an AZ-GTi, I think it was 150x 10s exposures (excuse the B9 and cloud 😆). I've left it uncropped so you can see where the alignment process has rotated the image during the capture.

 

I'm also going to look into the idea of adding images from the library to stack. I'm imagining maybe taking a series of images over several nights with the same settings, so you could stack the stacks.

 

[GeezerGazer]

Last night was steady and transparent enough for a few images... albeit, with a full moon rising!  I started about 18:30 hrs, and had to close up at 20:45 hrs, because Luna was getting too close to the main subject.  So I switched to Luna for a few awful shots.  A full moon has never been a favorite of mine.  

 

The night had high humidity, 85-90%, the camera was my iPhone 12 ProMax, the initial test used a stationary tripod with a solid connection holding the phone.  Where possible, I tried to keep settings as near as possible to equal.  All images are unprocessed, un-edited.  

 

My first test was to see how AS compared to other apps for its ability to average images.  The comparison included AS, iOS Night Mode, Nocturne and NightCap.  Nocturne is an automatic, single button start-to-finish app that I included which does align and stack multiple images from a stationary mount.  In the recent past, it has been my go-to app for 1x images with the iPhone (no supplemental optics).  Recognize that Nocturne has an unfair advantage in this first test since the other three apps could not or  were not permitted to align and stack subs.  

 

In this photo, AstroShader (AS) is on the L, and the iOS Night Mode is on the R.  As usual, Night Mode wants to brighten the entire FoV, as if it's taking a photo of a dimly lit terrestrial scene.  Night Mode seems to be designed for that purpose, having no manual controls except for the number of seconds used for exposure.  I set exposure to 30s in Night Mode.  For the AS image, I set ISO to 5004 with a 1s exposure that was averaged for 15s (no alignment/stacking).  

 

Click on the photo to see better resolution:

 

At ISO 5000, AS did show more noise than any other app in this test.  But it did provide a better rendition of the sky than Night Mode.  

 

The following comparison was between Nocturne-L and NightCap-R.  The Nocturne image took 2m 25s to complete (with unknown settings).  NightCap was set to ISO 5000, with 1s exp, averaged for 15s.  The NightCap image was very dark compared to the other photos; NightCap can provide increased brightness in two ways, but I all of that turned off; Nocturne provided the best star images with the most stars in any of the photos and is the only app that aligned and stacked images in this first comparison.  

 

I'll post about other parts of the tests from last night... later today. 

Edited by GeezerGazer, 07 February 2023 - 04:10 PM.

[GeezerGazer]

The second comparison last night used the same apps with the iPhone mounted on my AZMP for tracking.  Again, I only permitted averaging without any aligning/stacking.  Results were slightly better because of the tracking.  But for this test I extended averaging from 15s to 30s.  In AS, this is basically taking a single sub.  For brevity, I'll post just the comparison between AS and NightCap, both at ISO 5000, 1s exp, 30s average: 

 

The most obvious difference is the brightness of the images.  Let me explain... NightCap permits brightening an image in two ways, but I had both of them turned completely off.  I probably should have had the screen brightness in NightCap turned up at least 50%.  On the other hand, the AS photo is over exposed.  What I found was that making adjustments to the ISO in AS does not effectively change the brightness of the image that is on the picture taking screen.  This is troublesome because it means that a series of photos needs to be taken to determine the best setting for a perfect exposure.  In NightCap Long Exposure mode, the brightness of the image is seen on the picture taking screen before hitting the shutter release.  Averaging in NightCap does not change the image brightness... it only averages the images to eliminate electronic noise.  

 

The next comparison image was using an AT 125 EDL refractor with a 17.3mm Delos with the phone attached.  On the L is a single sub at ISO 5000, 1s exp averaged for 20s.  On the R, I used 20 subs at the same exposure to see how accurately alignment was made in the app.  Alignment and stacking looks very good.  In the single sub on L, there are some artifacts resulting from the way AS deals with bright stars, I marked them.  In the aligned and stacked image on the R, much of the artifact issue was eliminated... not all, but this was using only 20 subs.  Two things come to mind.  Using averaging in NightCap, I have never found artifacts like those seen in the AS sub exposure.  It is obvious to me that NightCap and AS are using different mechanisms to average multiple frames.  In settings within NightCap, if I set exposure at 1/10th sec, and averaging to 1 sec, the averaging of 10 images occurs in 1 sec.  In AS, if I set exposure to 1/10th sec it takes 10 individual exposures, each taking about 1/2 sec.  Fast exposures for AP are only needed for the moon; even 1/100 sec is not sufficient as the phase moves closer to full.  I have use exposures from 1/100 up to 1/750 sec exposures on Luna in the past with good success.  Last night, one day past full, 1/100 was not fast enough, even with the ISO at its very lowest setting.   

 

All of the images presented thus far have been unprocessed.  So I thought I'd show a comparison of two photos taken just a few minutes apart last night that have been processed.  On the L is AS, taken at ISO 3000, 10x5s subs processed in AS to decrease light and sharpen and then processed in Photos on Mac to adjust (reduce the blue curve) and reduce saturation.  On the R is NightCap, taken at ISO 1000, 1s exp, averaged 10s, processed in Photos on Mac to increase brightness, shadows and black point (no sharpening).  That was about the best I could do with each image.  

 

Regarding AS, I am still excited about this app.  I think it holds great potential.  The aligning/stacking mechanism seems to work very well.  Where I saw issues was in the averaging process to produce subs, using supplemental optics (telescope).  I don't know if that can be changed in any way, but the artifacts I saw from bright stars need to be eliminated.  AS compares very well to Nocturne for 1x phone nightscapes.  If given the same processing time as Nocturne for longer exposures and/or more subs, results might be just as good.  Although, the longer the averaging in AS for subs, the brighter the image seems to be... that would need to be changed so that stars can be contrasted against a darker sky.  I'd also like to see short exposures permitted down to about 1/400s for images of the moon.  The only option to that would be to use an ND or polarizing filter when photographing a bright moon.  

 

My biggest immediate concern is the image on the picture taking screen does not show the brightness of the image that is about to be taken.  Changing the ISO did not change the image brightness on the screen... which means I had to take a series of photos to find the best exposure before taking a lot of subs... unless I was doing something wrong!  

 

Finding best phone focus in AS was easily accomplished on bright stars.  In use, there is little reason to change the best infinity focus, unless some other part of the optical chain is altered.  Once best focus is attained from the telescope, AS can be focused and it holds that focus setting until it is intentionally changed or the app is closed.  That's a very good feature.

 

The aligning/stacking feature worked so well in AS, I wondered if images taken in other apps like NightCap or from the native iOS camera app, might be imported and stacked by AS... or maybe even a new, separate app designed for just that purpose.  I also appreciated the excellent sharpening tool in AS... it works so well, I had to be judicious about using it. 

 

I intend to do more experimenting with AS.  If I find anything of importance, I'll post.  

 

 

[GeezerGazer]

After some feedback from Astrshdr about the potential causes of the artifacts, I re-opened the above images, expanded them to their maximum, and found that AS, NightCap, iOS Night Mode and even Nocturne, ALL show the same artifacts, even in the 1x images, where no other optical system was used .  I simply noticed them first in AS!  The source of these artifacts may be internal to the iPhone's lens assembly and/or sensor.

 

If you want to see the artifacts, click on the 1x images in post 32, expand them to their maximum and look at the brightest stars (mainly at the top) in each image.  Then move to dimmer stars in the same photo where you should be able to see a dark lip around one or two sides.  They are present in the NightCap photo which is very dark.  Using the original photo in that comparison, I lightened it and could clearly see the artifacts.  

 

Different settings might avoid or minimize the artifacts.  

Ray

[astrshdr]

Firstly, thank you for the comprehensive review and the feedback. Some really useful points I can improve, I'll try to take a few of the key ones in turn.

 

Firstly the black artefacts—I'll continue to investigate and test this, but have a suspicion it is optical in nature (microlensing, Airy pattern or other lens induced diffraction) however the processing (probably lightening) in AstroShader may exacerbate this. ...Hopefully will find a resolution to this.

 

The ISO not affecting the preview and then over lightening the final image—so this was actually a "fix" for an issue some others reported where beyond the native max exposure (1s), increasing the exposure didn't result in a lighter image. This is a kind of "exposure spoofing" but I think has caused more problems than it solved. This is useful feedback, I might remove this and simply have a min-to-max (max usually 1s) exposure range if I can speed up some of the processing—so aligning everything. With respect to the ISO not affecting the preview, this might be a bug, it should definitely although can be slow as the frame rate is 1/s. I'll look into this...

 

For the min exposure going to 1/400, I'll get this sorted in the future release.

 

I will also look at importing images from the library to align and stack in a future release.

 

 

One final point—which applies to all apps—the main camera usually has the best sensor in terms of light sensitivity. In addition the ultra-wide camera on the pro models has some significant distortion from the lens at the edges of the image, this does get digitally corrected but I suspect will still cause problems when stacking uncropped.

[astrshdr]

Oh and I'll also fix the default white balance and add an auto-white balance option.

(It can be manually set right now in AS but the default is far too cool so making the images blue if left unchanged).

[GeezerGazer]

Astrshdr suggested in a PM that I check focus, among other things.  So tonight I mounted the phone on a tripod, used 3x readers, then I took a series of images making 1 stop adjustments near what I previously use as a focus point at 1x, using the wide (main) camera lens on my iPhone.  Sure enough, focus was off by 4/100.  Two nights ago I use 94 as focused.  Tonight I used 90 as the focus point.  

 

Next I took a series of 15 photos moving the White Balance ~100 for each photo.  I liked it best at around 4500K to 4800K.  

 

Then I took another set of 15 images to find a suitable ISO and Exposure for a single 1sec photo, which turned out to be quite good at ISO ~700.  I found artifacts around stars by pixel peeping, which are probably from the phone's camera lenses or the very small sensor.  But they are not bothersome like the artifacts in the images from two nights ago.  I really can only see them at the most excessive magnification, where star images are completely pixelated.  

 

I wanted to lower the ISO to ~400 for less noise, so raised the exposure to 1sec averaged for 2sec, with an aligned stack of 20.  On the left is the original photo without post processing.  On the right is the same image with modest processing in the AS app and in Photos on my Mac. Click on the photo to open it for better resolution.  

 

These images are much improved, and provide a basis for better expectations using AS with my scope.  I would say that like other forms of AP, even at 1x, focus is critical!  

The images also look better and respond to post processing better when the White Balance is more neutral.  Many thanks to Astrshdr for his suggestions.  

Ray

Edited by GeezerGazer, 09 February 2023 - 02:56 AM.

[GeezerGazer]

I captured M36 last night from my red zone home using AS in my iPhone 12 Pro Max, through an f:4 8" Newt, using an Ackermann .73x reducer/corrector, with Delos 17.3, all mounted on a AZMP alt/az tracking mount.  This image was a stack of 20, 3second exposures at ISO 2408.  This is a cropped image and it was slightly lightened in Photos editing tools and compressed to fit here.  Stars appear round across the FoV.  Less need for precision stacking because of the tracking mount, but, just the same, no star trails in this image.  Good job AS!  Click on photo to see details: 

 

 

 

 

 

 

[Pinac]

My first experiment with Astroshader, just trying a simple configuration without putting too much thought into it:

20 exposures, 0.8 secs per exposure, ISO 1200.

Holding the iPhone 11 on top of a tripod, no further equipment. Conditions were mediocre (lots of humidity in the air, the sky bright from lights of the small town nearby).

The result is of course as good as an image with a "proper" camera (Leica Q2, see next post), but still interesting enough so that I will continue using Astroshader to see what can be achieved.

 

Pinac

Attached Thumbnails

• 

Edited by Pinac, 12 February 2023 - 10:44 AM.

[Pinac]

A simple Leica sample for comparison (single shot, all settings on "automatic")

Attached Thumbnails

• 

Edited by Pinac, 12 February 2023 - 10:44 AM.

[Pinac]

Another trial with the iPhone 11 behind a Celestron Hummingbird 56 at mag 12x; 5 exposures at 1.2 secs each; Bortle 6 sky (SQM 19.55).

Horribly unsharp (possibly not properly focused),  but the the shape of M42 is clearly visible, so the light gathering capacity of the app is proven.

More practice needed.

Attached Thumbnails

• 

[astrshdr]

Another trial with the iPhone 11 behind a Celestron Hummingbird 56 at mag 12x; 5 exposures at 1.2 secs each; Bortle 6 sky (SQM 19.55).

Horribly unsharp (possibly not properly focused),  but the the shape of M42 is clearly visible, so the light gathering capacity of the app is proven.

More practice needed.

Thanks for sharing the results. Also, Leica Q2... nice.

 

Yes, that last one looks like it might be out of focus. My (very much best guess) technique for focus with telescope/binos is...

1. Set exposure to ~0.1s and ISO to max.
2. Zoom in on a bright star.
3. Then set the focus slider in AstroShader roughly in the centre (0.5) and focus with the telescope as best I can.
4. Following this trying tiny adjustments using the AstroShader focus +/- buttons.

If anyone else is trying anything different to focus, or has any ideas to improve this, I'm all ears

[Pinac]

Thank you, will try that.

[Pinac]

Tried again today M42 and M45, with, it seems, a bit more success.

 

Sky @ Bortle 6 (SQM reading 19.5).

 

iPhone 11 mounted behind Celestron Hummingbird 56 @ 12x magnification.

 

22 exposures of 0.125 secs each, ISO max.

 

Minimally edited (need to further study the editing process)

 

Starts to be fun!

 

 

Edited by Pinac, 15 February 2023 - 04:09 PM.

[Pinac]

[GeezerGazer]

The last two nights, we had only a few very high, thin clouds.  At times, none were present, yea!  So I was in the alley next to my red zone (Bortle 7-8) home for some pictures.  The first night, I used AS, iPhone 12 ProMax, AT 125 EDL and either a 17.3 Delos or an APM Super Zoom 15.4-7.7mm, mounted on AZMP.  Humidity is still very high, 85+%.  I took about 25 images, deleting most.  AS images are automatically saved as TIFF files which eats up memory.  I do wish that I could choose a different format such as a HQ JPEG for most photos, reserving TIFF when needed for difficult or extreme processing.  In most cases, these phone photos don't require much time to take, and results are known immediately.  So reserving TIFF for the occasional image would be a better choice for me personally.

 

Conditions were not great for imaging, so take this unprocessed image with a grain of salt... it is not a keeper.  But it shows that AS does provide a pretty robust system of averaging and stacking images.  Total time was only 24 sec, with 3 short stacks of 8 one second images.  The bright EoF is from the distance being incorrect between the phone and eyepiece, using the APM Zoom.  This image does show the bright star artifacts that I've previously described.  

 

Since I previously wondered if incorrect focus was contributing to the artifacts, I took 12 short images, with Sirius centered.  I placed focus on 90, pinched to zoom in on Sirius, and then focused the scope, zoomed out.  Each of the 12 images were at intervals of 2 on the focus preview screen.  Through the entire sequence from 80 to 100 on the focus scale, the TIFF image showed the artifacts not much different than this image: 

 

The next night, I left the telescope in the house, taking just the phone out for imaging.  This time, I used the AZMP for tracking to get the longest exposure necessary to grab stars.  But I started with some shorter exposures.   I have now taken about 100 images using AS, and with all those images, I saw artifacts, except in 3 images!  I don't know what differences occurred between the images because AS does not allow EXIF data to collect settings, but 3 images do not show artifacts.  Here is one of them.  

 

I needed to re-read some of the information in this thread, and noted that Domdron's AS image of Pleiades in post #17 also showed bright star artifacts.  I have also taken comparative images using NightCap, NightMode and Nocturne, to see if they also produce these artifacts.  The other camera apps do show a dark lip or edge around bright stars, but they are very different than the artifacts associated with my AS photos.

 

I feel confident that my focus test eliminated focus issues as a source of the artifacts.  

 

I have used AS with an f:4 8" Newt, 125 EDL refractor, and the native phone camera mostly at 1x with a few images at 2.5x (telephoto lens).  Last night I also applied a 1.7x supplemental lens over the native camera lenses.  This lens is a Beastgrip telephoto lens that allows the native iPhone camera to maintain its native focal ratio.  

 

NOTE:  I found that when using the native 2.5x lens, ISO parameters (min-max) changed significantly, to lower values, permitting a maximum ISO of 2016 instead of the normal 7616 maximum. It is possible that the numerical scale changed without effecting actually exposure/ISO value.  But I cannot say for certain.  The 2.5x lens is an f:2.2 whereas the 1x (wide) lens is an f:1.6, so the telephoto lens needs all of the ISO available to it.  

 

NOTE:  Twice during use a pop-up has appeared on the previewer screen where focusing is done.  The pop-up says:  "Focus:  Fixed for this camera type."  When this appears, the focus is locked and on the settings page, the exposure slider and ISO slider are also locked.  I could not clear the pop-up without closing the app.  Last night, closing the app did not clear the pop-up, so I powered the phone off and re-booted.  That cleared the pop-up message, and I continued to take photos with manual control.  

 

(Continued in a short while in the next post.)  

 

 

 

 

 

[GeezerGazer]

Continued from post #46:

 

In use, I find the Exposure slider in settings to be quite fiddly for sub-second exposures, because the sub-second range of 1/100 to 1sec encompasses such a small section of the Exposure slider.  IOS limits iPhone exposures to 1sec.  My thought was that the Exposure slider should also be limited to 1sec, where ISO adjustments would show up within that 1/100 to 1sec range.  In a PM discussion with Astrshdr, I learned that ISO brightness adjustments have purposely been extended to show in multi-second subs that are made by averaging, even though averaging photos plays no part in exposure brightness.  This may change, by allowing sub-second exposures to take up more space on the Exposure slider, OR by creating a new slider to select the number of photos for subs and leaving the Exposure slider for exposures between 1/100 and 1sec.  

 

The Exposure slider is effectively carrying out two functions.  The first function is to set the exposure value between 1/100 and 1sec in duration.  The second function is to set the number of one second photos that is intended to be averaged as subs.  With this dual function, only exposures of 1sec can be averaged before being sent as subs for alignment.  Sub-second exposures can only be taken as a single image.  This could be an important issue for users with a non-tracking mount where nearly all exposures will be less than 1sec.  Averaging is a tool to reduce electronic noise; the more averaging that's completed, the more noise that can be eliminated with higher ISO settings.  But averaging does not affect exposure brightness.  That's why I think ISO changes which DO affect image brightness should be limited to sub-second exposures.  

 

Multiple sub-second, single images can be taken, aligned and stacked, and I have learned that when they are aligned and stacked, averaging is also taking place.  So that does help create a better final image.  

 

AS is attempting to do a LOT, in terms of accommodating tracked vs. non-tracked imaging and  native phone images vs. images taken through a telescope.  But I think it is succeeding!  Previously, I guessed that if AS was given enough time to process images, it could produce images as well as Nocturne.  I think last night, it actually surpassed Nocturne with this image of Orion, taken with the 2.5x lens with a 1.7x supplemental lens (about 4x), mounted on my AZMP.  Except for the artifacts around Rigel, the belt stars, Betelgeuse and a few others, the photo shows more stars with better resolution than I have previously seen from my phone.  Keep in mind that the original TIFF file was 97.5 MB, converted to a HEIF file of about 9 MB, then converted to a JPEG of 4.5 MB, and then compressed to less than 1/2 MB to fit here.  

 

Click on this image to see detail; it includes a lot more stars than I could see visually from my red-zone home (Bortle 7-8), although I did see the jet fly through: 

Edited by GeezerGazer, 21 February 2023 - 07:38 PM.

[Pinac]

Good stuff, GeezerGazer.

 

Just for comparison, single shot at exposure „max“, other parameters standard, with iPhone 11 Pro.

Attached Thumbnails

• 

Edited by Pinac, 22 February 2023 - 02:04 PM.

[astrshdr]

The last two nights, we had only a few very high, thin clouds.  At times, none were present, yea!  So I was in the alley next to my red zone (Bortle 7-8) home for some pictures.  The first night, I used AS, iPhone 12 ProMax, AT 125 EDL and either a 17.3 Delos or an APM Super Zoom 15.4-7.7mm, mounted on AZMP.  Humidity is still very high, 85+%.  I took about 25 images, deleting most.  AS images are automatically saved as TIFF files which eats up memory.  I do wish that I could choose a different format such as a HQ JPEG for most photos, reserving TIFF when needed for difficult or extreme processing.  In most cases, these phone photos don't require much time to take, and results are known immediately.  So reserving TIFF for the occasional image would be a better choice for me personally.

 

Conditions were not great for imaging, so take this unprocessed image with a grain of salt... it is not a keeper.  But it shows that AS does provide a pretty robust system of averaging and stacking images.  Total time was only 24 sec, with 3 short stacks of 8 one second images.  The bright EoF is from the distance being incorrect between the phone and eyepiece, using the APM Zoom.  This image does show the bright star artifacts that I've previously described.  

IMG_8425.jpeg

 

Since I previously wondered if incorrect focus was contributing to the artifacts, I took 12 short images, with Sirius centered.  I placed focus on 90, pinched to zoom in on Sirius, and then focused the scope, zoomed out.  Each of the 12 images were at intervals of 2 on the focus preview screen.  Through the entire sequence from 80 to 100 on the focus scale, the TIFF image showed the artifacts not much different than this image: 

IMG_8424.jpeg

 

The next night, I left the telescope in the house, taking just the phone out for imaging.  This time, I used the AZMP for tracking to get the longest exposure necessary to grab stars.  But I started with some shorter exposures.   I have now taken about 100 images using AS, and with all those images, I saw artifacts, except in 3 images!  I don't know what differences occurred between the images because AS does not allow EXIF data to collect settings, but 3 images do not show artifacts.  Here is one of them.  

IMG_8426.jpeg

 

I needed to re-read some of the information in this thread, and noted that Domdron's AS image of Pleiades in post #17 also showed bright star artifacts.  I have also taken comparative images using NightCap, NightMode and Nocturne, to see if they also produce these artifacts.  The other camera apps do show a dark lip or edge around bright stars, but they are very different than the artifacts associated with my AS photos.

 

I feel confident that my focus test eliminated focus issues as a source of the artifacts.  

 

I have used AS with an f:4 8" Newt, 125 EDL refractor, and the native phone camera mostly at 1x with a few images at 2.5x (telephoto lens).  Last night I also applied a 1.7x supplemental lens over the native camera lenses.  This lens is a Beastgrip telephoto lens that allows the native iPhone camera to maintain its native focal ratio.  

 

NOTE:  I found that when using the native 2.5x lens, ISO parameters (min-max) changed significantly, to lower values, permitting a maximum ISO of 2016 instead of the normal 7616 maximum. It is possible that the numerical scale changed without effecting actually exposure/ISO value.  But I cannot say for certain.  The 2.5x lens is an f:2.2 whereas the 1x (wide) lens is an f:1.6, so the telephoto lens needs all of the ISO available to it.  

 

NOTE:  Twice during use a pop-up has appeared on the previewer screen where focusing is done.  The pop-up says:  "Focus:  Fixed for this camera type."  When this appears, the focus is locked and on the settings page, the exposure slider and ISO slider are also locked.  I could not clear the pop-up without closing the app.  Last night, closing the app did not clear the pop-up, so I powered the phone off and re-booted.  That cleared the pop-up message, and I continued to take photos with manual control.  

 

(Continued in a short while in the next post.)  

Good stuff, GeezerGazer.

 

Just for comparison, single shot at exposure „max“, other parameters standard, with iPhone 11 Pro.

 

 

Thanks for sharing the images GeezerGazer and Pinac.

 

On the weird bright star artefacts, GeezerGazer and me are trying to figure this out. Updates to come in the future...

 

1. In the next release I'll change the default image type to either JPEG (minimal compression) or HEIF. Unless do say 10mins total exposure where I'll save all the uncompressed data.
2. On the note on ISO: Different iPhone cameras have different ISO ranges. In AS I just allow the full range to be selected—so unfortunately what's there right now is all that is possible.
3. The fixed focus pop-up appears when selected cameras on certain iPhone models that have fixed focus e.g. ultra-wide angle camera on iPhone12. This definitely sounds like a bug as it should just warn and let you carry on. I'll get this sorted out...
4. On the exposure range slider, this will change slowly over the coming months. First will be making the usable range more sensitive. Then later will redo it completely when I finish the new faster processing pipeline comes online.

[Domdron]

Venus + Jupiter through my telescope this evening with the AstroShader app on my phone, 4 0.066s exposures at ISO ~400 (I think). 

 

There’s a reflection in the middle which is repeated 4 times, because the app aligned the planets.

 

This was taken with the eye guard extended. If I collapse the eye guard and put the phone closer, I get a big reflection with a “hole” in the middle. See other attached pictures. Looks like the primary mirror shape with s dark reflection of secondary in the middle? Any idea why this appears and how it could be avoided?

 

 

Attached Thumbnails

• 
• 
•