| Would you be interested in contributing towards an AB all sky survey? | |
|---|---|
| No. I wouldn't find such a survey useful. | |
| No. Satisfactory data already exists for me elsewhere. | |
| No. I would find such a survey useful, but I don't have the time, location or equipment to contribute. | |
| Yes, I would be interested in taking part. One or two fields maximum. | |
| Yes, I would be interested in taking part. Prepared to do multiple fields. | |
| Login to vote and view results. |
4.64
...
·
 |
|---|
@Arny What's your camera/lens combination?
|
0.90
...
·
|
|---|
Thanks for all the suggestions for the processing guide. I approved them and made a few more changes. Quick summary: - The WBPP part is much easier to configure, since I utilized the “maximum quality” option in WBPP (completely forgot about that). Thanks for the suggestion @Michael Ring - The linear post processing part is completely re-written - The option for futher processing for publication was added It would be king of you to review this another time, I hope I approved all the changes but some more control can't be bad! CS Gerrit |
9.89
...
·
|
|---|
I shall interject to point out that we need a better defintion of what the image acceptance criteria are. When we say "SNR=30 or better" what does it mean? Is it expressed in decibel or just as a ratio (S/N)? And how do you compute it, robustly? Secondly, are we going for L-RGB or stick with RGB only? Note that LRGB isn't identical to L-RGB because the former requires stretching the data, so effectively the L in L-RGB is an alpha channel. And if so, how do you measure SNR in that case? I also feel we should be specifying the minimum PSF @ 10"/px across the frame to make it suitable for feathering at the blending zones.
|
4.29
...
·
|
|---|
andrea tasselli: I asked this in post #4 and via PM. This may seem trivial to some users (I suspect those with a background in Astronomy) but the calculation provided by Brian was definitely not something I've seen before on any of the amateur AP forums I regularly visit. Ideally, we follow the KISS principle here as well, but 2h @B2, 3h @B3 or 4h @B4 is probably too simplistic. |
|
0.90
...
·
|
|---|
I posted a table earlier to convert it to different f/ ratios and SQM values.
|
|
9.89
...
·
|
|---|
Not sure we sing from the same hymn sheet here. I want to know whether, given one master integration image, is it possible (and how) to measure the average SNR to see that it complies with the requirements or more integration time is needed. In fact, in my case, I'd need to test beforehand in order to book time in the remote system well ahead. f/ratio means nothing except in the context of a single lens system and how many users are likely to dark sky meter never mind SQM are wildly variable in time and pointing location.
|
|
4.64
...
·
|
|---|
Just started my first test with Panel 618, it is not yet completely dark but the moon will not rise until in 30 minutes. Using 2 minute exposures with my Samyang 135 f2.0@F2 and ASI071MCPro with gain 90. Nina Statistics tell me that I have 1800 blown out pixels, so I think I am on a good way, any thoughts on optimizing those settings, Stars acceptable? I think I see a little tilt on the left side, seems I will need to do another round of aberation inspector Guiding with my Staradventurer GTI is at 1.12", this is a really calm night, too bad that the moon will soon destroy all this. First statistics from Subframe Selector for the first 13 subs are now also available, FWHM is arround 1.42 Eccentricy is a little high with 0.74, SNR at 1.32 Michael   |
|
0.90
...
·
|
|---|
Hi Michael, that looks pretty good to me! Yes, the stars aren't perfect, but I guess we first can't expect that, and secondly they are better than many images shown online, so that is more than enough (at least in my humble opinion). To me, the settings seem good, but since I only use DSLRs, I can't tell that much about it. However, I guess we can't compare the SNR number you mentioned directly to the desired SNR of the survey. If I remember correctly, the SNR was specified for a star of a certain brightness, whereas the number mentioned in image processing tools is computed on the whole image and has another meaning. CS Gerrit |
|
4.64
...
·
·
1
like
|
|---|
The SNR number is for a single frame, nothing stacked yet. Wonder if those numbers will be compareable to numbers of other users in Bortle 4 skies..... I will definitely learn a lot in this project that will also make my astro pics in general better, glad that Brian started this. The number or stars is now sinking, the big bad moon is arround the corner to tell me it is time to soon go to bed....  |
17.56
...
·
|
|---|
Thanks to everyone for their contribution and hard work over the past day. @Astrogerdt the pipeline is looking good. I made a couple of small proposed edits. 1) I thought it might be helpful to state up top what the aims of the pipeline were: Quality, homogeneity (as far as possible) and recoverability [i.e. minimise destructive processes] 2) I also include a min requirement for 10arcsec/px and no resampling from WBPP (with drizzle integration if required). @andrea tasselli I would say just RGB (not LRGB) based on the recoverability principle. @Victor Van Puyenbroeck the calculation I used was largely to motivate the 2hours @ Bortle 2 etc - backed by others. I reproduce it below for completeness and transparency. Many may not understand it, but I don't think you necessarily need to have an astronomy PhD. Mind you, I feel similarly befuddled when I am presented with lines of code in order to explain the AB index. It's just not a language I understand. Do we need to more specific on depth? My view is this is an area where there is a dilemma between "trusting the observer" and "specifying the pipeline". My view is that the proposed depth is well specified SN=30 at 22.5mag/arcsec per 10arsec spatial resolution element Getting to that point is harder, since it will depend on a number of imponderables; transparency, night sky brightness (presence of auroral activity) and the photographer on the ground is clearly best placed to make an assessment re: how far they are from their own ideal. As it is currently scoped, I think the pipeline strikes a good balance between prescribing the depth and providing guidance of what is needed in terms of exposure time to get there. In truth, those times are also driven by what it is reasonable to expect an observer to give up to do a community program. 2-4 hours seems a not unreasonable proposition. In a similar vein to "let's not over-specify field to field flatness" since we don't know how to resource it or achieve it, I suggest we also adopt "let's not over-specify field to field depth" since we don't realy know what we will get. I suspect we will learn a lot from the first test lunation, and "learning by doing" is likely to be the best way forward. Even if it means re-doing a little. For that reason, I would want anyone to embark on a huge survey campaign in the first "test" lunation. To a large extent the Northern summer [where most AB members are located] helps us in this regard, but this would also apply to the Southern ABC participants. CS Brian ----------------------------------------- My SNR calculations [abandon hope all who enter here?] Per sub-frame SNR = Obj/(Obj+Sky+RN**2+DN)**0.5 Obj = ADU from Obj Sky = ADU from Sky RN = Readnoise in ADU DN = Dark current in ADU And SNR [Total] = SQRT(Number of sub-frames) x SNR (single sub) Vega (mag=0) is 1010 photons/s/A/cm^2 at the top of the atmosphere. For any mag (m) the photons arriving at your telescope are 1010 x 10^(-0.4*[m+0.15*sec(zd)]) per second per Angstrom per cm^2 [The sec(zd) term reflect atmospheric extinction] So 22.5 mag/arcsec at a zd of 30 is around 8.7e-7 photons/s/Angstrom. Multiplying this by
This should get you to a number for Obj in the formula above. Further if you know your sky brightness in mag/arcsec [from your Bortle class] , the same calculation [ignoring the atmospheric extinction] will give Sky. RN and DN as per sensors spec sheet for gain used. |
|
0.90
...
·
·
1
like
|
|---|
I go with @Brian Boyle, I think overcomplicating the SNR matter is non-beneficial for us. We can't achieve a perfect, uniform SNR in the integrations. We made it as good as it can be without significantly more effort. Just a little bit of thinking about the processing we will do: maybe we can measure the SNR of every panel and determine denoise settings based on that to achieve a more uniform SNR in the final result. Since every SNR will be at least somewhat similar, this could work. I approved a few more comments on the processing guide. @Michael Ring suggested a procedure to verify the required field coverage, you can find that at the bottom of the document. As I lack the knowledge about the used scripts, it is up to the community to evaluate that. One question about the processing guide is left for me: I opted for a pedestal of 512 during image calibration. As I have a DSLR, I don't have to worry about that as much as most others. Is that a reasonable value of should I change that? Apart from that, I consider the acquisition and processing guide to be complete as of now. CS Gerrit |
|
17.56
...
·
|
|---|
I have now submitted the survey announcement. Once moderated, the survey - at least the test lunation - has started. Many thanks for everyone's contributions to date. CS Brian |
|
17.56
...
·
·
1
like
|
|---|
@Michael Ring Can you give everyone permission to edit the field list - so they can book out the fields on the sheet itself? Or would you prefer to do that via the forum. Some bookings might get missed this way - but also leaving people to edit the field spreadsheet is not without its risks. I will have a crack at Field 39 tonight - around the meridian for the 1.5hours (30 x 180sec) I will have on it before the moon rises. |
|
4.64
...
·
·
1
like
|
|---|
@James Tickner created the sheet, he is the one who can make it available in editing mode.
|
1.20
...
·
·
1
like
|
|---|
You should have comment (but not edit) access at the moment. My thinking was that people could express interest in a sheet by adding a comment with their Astrobin ID in the Field ID comment. The Google Sheet commenting mechanism allows people to see others comments and respond to them, so if people really want to they can conduct a 'chat' that way, pass on notes that they want to relinquish a field etc. Anyone can see that a field is marked as a little coloured triangle appears. If a field ID cell is empty (no comment marker) and the field isn't marked as allocated or completed then that means the field is free to pick. Periodically (eg once per day) I'll transfer the field requests into the 'Allocated To' and 'Allocated Date' columns. I'm more than happy to give edit access to anyone involved in coordinating the project (I'll need a Google ID to do this - PM me if you want to). I didn't open up edit permissions to everyone mainly to avoid accidental (or God forbid malicious) stuff-ups. The document would be the primary reference for what fields need doing and if it gets messed up it's going to lead to all sorts of confusion. Happy to get everyone's input into this approach though. @Brian Boyle I'll mark field 39 as allocated to you now, but if you get a chance please test the 'comment booking' mechanism as well - good to see if it works as expected. |
|
1.20
...
·
|
|---|
Brian Boyle: I think it only fitting that the honour of the first field goes to the originator of the idea  My Samyang 135 mm lens arrived a couple of days ago. I've only managed to grab one 20 s sub between clouds and rain, but was impressed with the star quality right to the corners - seems like it lives up to its reputation. I managed to pick up a Nikon D5000 for $125 from Ebay and once it gets here I'll have my first go at an astro-modification. Fingers crossed! Hopefully I'll be up-and-running by the time the weather clears up. |
|
17.56
...
·
|
|---|
James Tickner: Thanks @James Tickner and sorry for the mix up. I will add an explanatory note to the announcement to that effect. As luck would have it, fog olled it about half an hour into astronomical night, so I have try again tomorrow. Annoyingly, it is clear at a friends place about 100m (in elevation) further up the hill. Brian |
|
1.20
...
·
|
|---|
Brian Boyle: Bummer! I'll leave the field allocated to you though - hopefully clear skies will find their way over the ditch shortly. |
|
9.89
...
·
|
|---|
James Tickner: Be careful there with using an IR-UV blocking filter replacement as there is no way you can add it on the camera. At least inexpensively. |
|
4.64
...
·
·
1
like
|
|---|
@James Tickner : Welcome to the Samyang club, mine arrived last week, so far my stars in the corners are a bit meeehhhh but I managed to dial in backfocus with Nina's aberration inspector last night so hopefully next clear skies things will improve. I will post 3d printable lens-rings in the lens forum for the Samyang in the near future in case you are interested. Michael |
|
1.20
...
·
|
|---|
andrea tasselli:James Tickner: @andrea tasselli Can you explain in more detail - I'm not quite sure I understand what you're referring to. I'm planning to pull the camera apart (there's some good Youtube videos out there for this model), remove the glass filter in front of the sensor and replace with a new piece that has a transmission band of approximately 410 - 680 nm (95% at H-alpha and 92% at S2 according to the manufacturer's datasheet). Please let me know if I'm missing something important - this is my first attempt at this. Thanks! |
|
9.89
...
·
|
|---|
James Tickner:andrea tasselli:James Tickner: Usually there are two choices of filter replacements: full bandwidth or UV/IR cut. I see that you chose the latter and that's the way of doing it as you can't put a clip-on filter in front of the sensor (or maybe only very expensively). Losses a bit in flexibility but it covers 99% of the possibilities out there. I guess you might need to reposition the sensor if you want to retain autofocusing with most lenses out there. Be careful when you remove the filters, piezo or otherwise as you risk wreaking the thing. I know I did. |
|
1.20
...
·
|
|---|
Thanks for the tips @andrea tasselli From what I've read, going full bandwidth tends to lead to bloated stars due to poor chromatic correction at the extreme wavelengths, so definitely something to avoid. As far as I can tell the thickness of the new filter is the same as the one I'm removing, so I hope focus shouldn't be affected too much. I deliberately bought a cheap 2nd hand camera to practice with first time! If I destroyed our family camera I'd be in big trouble  |
|
9.89
...
·
|
|---|
I booked field 58 through the form leaving a comment there but no feedback just yet, that's the one just above@Brian Boyle field (in Dec) and let's see what happens.
|
|
4.64
...
·
|
|---|
@andrea tasselli, I saw your booking in the comments of the google page, so I guess James will reserve the filed for you when he is online again. You can doublecheck the comments by clicking on the comment icon in the top right edge of the google page
|
