16.52
#...
·
·
1
like
|
|---|
Alan Brunelle: Hi Alan, Great to have you on board. This sounds like a perfect rig. All mosaicing will be done centrally and @James Tickner is working on that. At the moment, we aren't insisting on 2x drizzle [my elderly mac wouldn't cope anyway], but we are comfortable with whatever works for you best. After all, the principle behind the survey is inclusiveness. Re: field signup, it is fine to sign up at the last minute. One thing we have discovered in the 9months we have been running is that it is really hard to predict when those good conditions arrive, and many of us are signing up fields with <24hrs notice. Clear skies to us all Brian |
 #...
·
·
1
like
|
|---|
Brian Boyle: Thanks Brian! For the truncated processing that you guys are asking for, I have no need to push files of that much greater size onto yours or James' workload. And if the stars' fine structure are really irrelevant to the final product, then probably unnecessary. If I did send such data, I would not be insulted if you guys resampled down anyway. In any case, if a particular frame lends itself to a final workup with drizzle at 2X, I will just do that for my own purposes separately for posting. However, from the many frames I have been setting up in NINA today, it looks like a lot of stuff off of the galactic axis and the galaxies in the fields are often just too small for the frames to be all that exciting! Maybe there will be lots of IFN, but from my survey of the IR bands in Aladin, at least for the next several months, that seem scarce as well except for the far north. Fingers crossed on weather here! |
|
16.52
#...
·
|
|---|
Alan Brunelle:Brian Boyle: Hi Alan, All the best with the fields. It is great to have another northern observer - it really brings it home how challenging it is to find a dark sky in the north. So thank you so much for giving up your time to contribute to this project. Means a lot. Brian |
3.94
#...
·
·
1
like
|
|---|
|
Also a warm welcome from me, fellow Northener! A nice rig you have! At what f/stop are you planing to use your subs? I started at f2.8, switched to f2.4 but I am now down to wide open at f2.0, at least for my Samyangs quality of stars actually degrades with higher f/stops than 2.8, would like to hear of your experiences.... Michael |
#...
·
·
1
like
|
|---|
Michael Ring: Thank you Michael! Lets wait for my first contribution before anyone gives me thanks! And thanks for the compliments on my rig! As can be seen, the mount and the lens shade setup is probably a bit of overkill for the camera/lens for the 135, but it is meant to also work seamlessly with my FLT91. You can see the 135 swapped out for the FLT91 on the same mount, etc. here. Note, only the optical tube and camera (with its associated cables for focuser and heater) need be swapped. The computer, power distribution, guide scope and camera all stay with the mount. So I can swap on the fly in the field, with only a few cables to be plugged in and a simple change in NINA profile. And I am up and running. Also, no need to check polar alignment or guide camera calibration after a switch. The lens shade is critical for my home location. I have pretty dark skies for most areas I can see, but some neighbor lights cause issues, hence the longer shade. I have always strived to run my 135 wide open. I am not a fan of the many spiked bright stars. But I also want to get the best resolution and contrast I can wide open. That said, I have run into problems with stars when trying a 105mm lens and also my first Rokinon 135 that I got late last summer. Stopping it down did help, but defeated my objectives. So I returned it and bought a replacement from a different vendor. The new one is what I have now and it has checked out quite well, if not perfect. At least one test night gave me almost perfect stars to the corners, so that convinced me any problems I will run into in the future will be because of my fault! I do have two stop-down rings ready for an emergency though! One at 62 mm and one at 58 mm. Both relatively minimal, running at f/2.18 and f/2.33 respectively. With only 2 nights of imaging with the 135 since late last year, I have not used the stop down rings. The first attempt saw still a bit of tilt and backfocus issues, but nothing so bad that BXT couldn't fix it. After a back focus adjustment on my second night in early Feb, even better results with near perfect stars in two corners and a bit off in the other two, telling me I have a little tilt. Again BXT works wonders with one click. I would brag about the results, but I am only using a C-sized sensor and I am aware of others who get outstanding performance with full-frame sensors. So that humbles me! Also, my fields are 9.95 X 6.66 degrees, so cropping to the necessary 9X6 would likely eliminate any star defects, if James chooses to do so prior to mosaicing. Speaking of BXT, I know that @Brian Boyle has recently stated that there might be some changes to some of the processing requirements/allowances for panels submitted here on ABC. I would want to ask the specific question regarding the use of BXT, and even NXT on the data submitted here. After some experience with these tools with my typical processing practice, I have come to do star correction using the "Correct Only" and no longer using Stellar Adjustments. These I do early on, prior to any background correction or color correction. This because I have seen some color shifts in the process. So it is that I want to correct for. If I do any "deconvolution" I tend to do this later, but have been doing this later in processing. As well as BXT does its job in this area, I have noticed that it is getting very much overused (or at least to an excessive degree) in images I see commonly now. Anyway, getting down off my soapbox, I guess I would like to know what sort of actions in processing are expected for the data uploaded to ABC. |
|
3.94
#...
·
|
|---|
|
For uploading the best is to do no processing after stacking. When the decision is taken that we use BXT we willl anyway have to apply it to all images and not every contributor is the happy owner of a BXT license. In the beginning I think at least Brian and I did SPCC color calibration, but along the process we saw some color casts in my images so I restacked everything (also because of better flats I aquired) and send over the 1x drizzled data fresh from stacking. I guess your 2x drizzled data will also be fine, but @James Tickner is the expert on this topic.... We also saw the color cast you mentioned, we have a lengthy thread about that and James was also in contact with Russell about that topic. And yes, I think we all are kind of plagued with tilt issues with our lenses, fortunately we can crop out a bit with our lens/camera combination but this will happen in post process by James when the mosaic is created. Michael |
1.20
#...
·
|
|---|
|
Coming in a bit late to the conversation, but I'd like to echo Michael's comments. Please upload any contributions in as 'raw' a state as possible - stacked and flat/dark field corrected of course, but without any other corrections. We're putting together a programmatic pipeline that tries to treat every field in the same fashion to produce a consistent final mosaic, and starting from the same point makes this easier. The BXT colour cast is an interesting one. It seems that when correcting optical aberrations it doesn't accurately preserve flux, with the result that star brightness and colour can vary significantly across the field. The differences between original and final flux can be a factor of 2 or more. This is turn interferes with matching star colours and brightness when neighbouring fields are stitched together. This is a real shame because the impact of BXT on star shape is fantastic - it does a particularly good job of removing corner aberrations from tilt. We're trying some corrections for this but they're a bit experimental at the moment. |
#...
·
|
|---|
James Tickner: James, I know that you are away at this time and I will await comments and any testing you do on the field that I uploaded. It is raw. This is field 1015. From my assessment, I am getting more star aberation deeper into my field than I did a while ago when I had my rig out. Since then I added not only a new (and better) UV/IR cut filter, but also a new dovetail plate for my camera. No doubt, I have reintroduced some issues. Another concern is I may have overtightened my mount rings for the lens. And I also may not have had the f-stop set completely open since I am seeing a diffraction spike pattern on some of the moderately bright stars as well that mimic what is seen with the f-stop in use. I now wrap the f-stop adjustment ring and bayonet joint in black electrical tape because I noticed a significant light leak through that part of the lens when doing darks during the day. That problem is solved, but may have accidentally turned the f-stop a bit! Most of my aberations still result in fairly small, tight stars, so may not matter when the final sampling is employed on the images. As far as my experience with BXT with my AstroBin submissions, I mostly use just the star-corrections-only feature. I never use the star reduction feature of BXT, the shift is too great and it can create other artifacts. Bill Blanshan's approach is much more gentle. Occasionally I sometimes some BXT sharpening, but only after I see how the stars-only corrections work on the deconvolution of non-stellar features. Early on with the introductin of BXT, the community often mentioned color shifts in stars, etc. with the use of BXT. Because of that, if I employ BXT is is always one of the first processes during my processing routine. Then background correction. And only after that, SPCC. In this way, I figure that any color shifts from the prior processes are resolved by SPCC. Because SPCC is deficient in only coming up with a singe, simple set of correction factors for the whole field, I do not think it will be able to solve color inconsistencies between panels in a very large mosaic from different contributors. When makers of PixInsight complete their project (I can't remember the name) to come up with a CC function that can work in much smaller tiles using the GAIA data better, then I think that would be the solution. |
|
16.52
#...
·
|
|---|
|
A update from me following April lunation... Nearly 2/3 of Southern Sky done with a first pass of fields. Still have a backlog of fields to process and an even bigger backlog to process to correct doughnout problem. However, I think our most pressing issue is now mosaicing and QC of existing fields. I say that because I will shortly run out of fields to observe. I have about 18 fields between 10h and 20h in the Dec strips -15 and -10 I could go to more northernly decs, but then I will be around 45deg from the zenith at best and, to date, I have tried to capture fields with zds of 45 or less. The other option would be to do a quick survey of the southern sky with my Sigma Art 40mm lenss [thanks to @Michael Ring for the suggestion] It is fast and gives good images, here are 2 x 1hour fields I did two nights ago in almost full moon. The images had moonlight gradients removed using GraXpert and mosaiced using PI. If we can get this in a hours in 98% moon, just think what we can do when the moon isn't up. Some time ago @James Tickner produced a field list for the Sigma Art 40mm +FF (20% overlap). With everyone's agreement I plan to spend next lunation to see how much of the Southern Sky I can cover. There are 29 fields in the Southern sky for the Sigma Art field of view. In the next lunation, I should be able to see at least 2/3rds of these fields. If I spend on hour on each, I ccould complete in 3 May nights. Do this sound like a plan? Brian  |
|
3.94
#...
·
|
|---|
|
Sounds like a really good plan! @James Tickner has already started taking images in the area of 10° north, perhaps it is worth the effort that you go below your 45° above horizon limit to have at least some coverage of the area around 0° I currently run 2 cameras and will soon run a 3rd one but the weather gods are really not that kind to Switzerland, most moonless nights are wasted because of the bad weather and to dust from the sahara desert traveling over Europe. Michael |
