Help with AIP4Win tutorial

Hi Paul, I will work through the tutorial to see if I can reproduce the error statement you describe and let you know what I find. Meanwhile, I can explain the bright area in the image. These pictures were taken when a full moon was near Auriga and the glare creeps into the image. It is distracting, that's for sure. I'll try toreplace these test images witha settaken on a moonless night. Thanks for your interest. Sincerely, Tom

Thank you, Tom, for your prompt reply. I'm surprised that no one else hasn't picked up this processing problem. Perhaps there's some setting that needs to be made first (not mentioned in the tutorial) and most other users have long since passed over this hiccup. I'm just a naive newcomer. Well, it's said in a number of places that one can do reasonable (DSLR) photometry on moonlit nights. It looks like the caveat(s) need to be spelled out - I would suggest the moon should be 30+ deg. away from the field being photographed, but perhaps more experienced people could advise here. It's a pity that a moonlight-contaminated set of images was used for the tutorial, and a beginner one, too - it just makes the tutorial that less effective and off-putting to newcomers like myself. Have other users of the tutorial got reasonable results using this set of images (IMG_3147, 48 &49)? I'd certainly appreciate a replacement set of images. Finally, I'd like to suggest that the tutorial notes for the 2nd step of AIP4Win be amended: I found that to select the 2 alignment stars one needs first to click on the first star in the master frame then click the Star1 button on the Alignment tab, then click on the 2nd star and finally click on the Star2 button. That took a bit of trial and error to find out, again making the tutorial less than trouble-free to use. Perhaps these missing steps are a v2.3.1 innovation. Kind regards, Paul.

Hi again Paul, I was able to reproduce the Aip 32 message. Apparently the previous step in the tutorial "Calibration Setup" was either not completed correctly or the results were inadvertantly deleted. You can tell if the setup is complete by examining the "Calibrate Selected Image" button at the top of the page.If it's illuminated (look for a red circle around the letters "CAL"), you're ready to proceed. If the button is not illuminated, you'll need to redo the "Calibration Setup" section. I hope this helps. Tom

Hi Paul, Thanks very much for your feedback on the tutorial. I'll start working on replacing the images and revising the section on alignment stars. Concerning the moon, although it's a distraction, it should not affect your results unless it's really close (not sure how close) to your target star. This is because AIP4WIN subtracts the background illumination automatically. Check section 10.1.2 Aperture Photometry, p. 272, for an explanation of how this is done. All the best, Tom

Tom, Thanks for that. I missed clicking the Process Dark Frames button because in my quick reading of the tutorial instructions it's not mentioned specifically, just in passing with the phrase "On the “Flat” tab, select and process your flat and flat dark frames". Perhaps it might make things clearer (a.k.a. cater for poor readers like myself) if you specifically mention the button by name. If you don't click this button, then closing the Calibration>Setup dialogue window removes all the settings you have made. I also take it that you don't click the Process Flat Frames button - perhaps a comment to that effect might be worthwhile. And, does the "Apply Flatfield Correction" check-box relate only to the dark frames you have selected, or does it also apply to the images you are going to process in the next step? I would have thought that if it applied to the images, it should be checked ON - which implies that you have to Process Flat Frames. BTW, is there any point in the instruction "Open the image files to be stacked" in step 3 under the Multi-Image>Auto-process step? I have tried it with and without loading the image files and get what looks like the same result. Perhaps, after this, a few of the wrinkles in the tutorial instructions will be ironed out - then the next version of AIP4Win is likely to appear and change something... Ah, well, it's a good game. Kind regards, and thanks once again, Paul.

Tom, Thanks for your explanation, and your offer to revise the tutorial instructions. I'm looking forward to seeing your efforts. Wrt the Moon's influence/interference, I would have thought that the less of it you have, the better (a.k.a. the more accurate) the result you should get. I'm thinking here that the Moon's light will be a gradient across the field which might also include Air Mass effects, vignetting, and so on. If you can eliminate as many of these as you have control of, then you should be more confident of the results you get. Am I talking through my hat, here? Kind regards, Paul.

Tom, or anyone else, The way the tutorial is written for the Calibration>Setup screens I find somewhat confusing as it seems to be too brief. If I understand Fig. 6.19 on Page 190 of HAIP the steps should be as follows: 1. Select Calibration>Setup. Select the Standard Calibration Protocol. 2. On the “Dark” tab, select and then process your dark frames (Process Dark Frames button) with Average Combine checked ON. Optionally you can save the resulting frame as a master dark frame. This processing will check ONthe Subtract Dark Frame option. 3. On the Flat tab, first select your flat frames, then check ON Subtract Flat-Dark and select your flat-dark frames. Process Dark Frame(s) with the Average Combine option and optionally save the result as a master flat frame. This step will check ON the Apply Flatfield Correction option. 4. Close the window. What these Process buttons do, I believe, is create temporary master dark and master flat frames, whether you save them or not. Without BOTH options checked ON (Subtract Dark Frame and Apply Flatfield Correction), you won't get both of the corrections shown in Fig. 6.19 applied, and it would be just a waste of time doing this processing if you have either or both these checked OFF. However, the screen shot in the tutorial does not show the Apply Flatfield Correction checked ON - leaving me confused as to why it should be OFF when you have put in the effort of processing the flat frames. Am I on the right track? I'm trying to learn this business by action and example from a position of true novice - I've never done this before. I hope my comments help make the tutorial that much better for everyone. Kind regards, Paul.

Paul, You're absolutely right about eliminating as many variables as possible to improve data quality, however, data quantity is also important. If you only go out on moonless nights when the weather is perfect and your target is directly overhead you'll have mighty few data points to work with. I guess this is where lab science and field science differ. In the lab you have the luxury of setting upyour experiment in a way that eliminates as many variables as possible. In the field, mother nature is in charge and you have to find a way to compensate for the variablesshe throws at you. That's why several members ofthe DSLRteam have worked very hard to develop the spreadsheets in the tutorial.These provide corrections for variations in air mass and star color. In the end, though, it'sa balancing act that's for sure. Best wishes, Tom

Hi Paul, I have rewritten the calibration setupsection of the tutorial. I hope I have addressed all the problems you encountered. I would really appreciate your giving it a try and letting me know how it works for you. Many thanks, Tom

Calibration Setup AIP4WIN Tutorial

1. Select "Calibrate" from the menu at the top of the screen.
2. Select "Setup…" from the sub-menu. This will open the "Calibration Setup" window.
3. At the top of the window, change "Basic" to "Standard".
4. On the "Dark" tab, click "Select Dark Frames".
5. Open the Dark images you wish to process. A green light will appear next to the "Select Dark Frames" button.
6. Click on the "Process Dark Frames" button. A green light will appear when processing is finished.
7. Now click the "Flat" tab.
8. Click on the "Select Flat Frames" button.
9. Open the flat frames you are using.
10. Check the "Subtract Flat – Dark" box.
11. Click on the "Select flat-darks" button.
12. Open the flat-dark images you are using.
13. Click on the "Process flat-frames" button. All green lights will appear when processing is complete.
14. Click on the "close" button at the bottom of the window.
15. The "Calibrate selected image" button (letters "CAL" in a red circle) at the top of the screen should now be illuminated.

Tom, Thanks for that. I must admit that my training and a large part of my working life has been in labortories as a professional chemist (not a pharmacist). In fact, I don't think I've ever really thought to make the distinction between laboratory and field work, but your comments make a lot of sense to me. I'm going to have to change my mindset. With that thought, it might be wise, then, to retain the existing tutorial images. What they need is an explanation as to the effects seen and that the data analysis method used by AIP4Win helps reduce these effects. That way, newcomers like myself are not put off by unexplained artifacts that one wonders whether a) you've done something wrong, b) why they are still there after applying dark and flat fields (which one would think should eliminate such instrumental artifacts), or c) whether they are going seriously to jeopardize the analysis. What say you? Kind regards, Paul.

Tom, That's a whole lot better and easy to follow for a newcomer. I would add that both the Subtract Dark Frame and Apply Flatfield Correction checkboxes should be ON before you click the Close button (step 14). And to reinforce this I would also recommend that you replace the screen shot of the Calibrate>Setup with one that shows all processing completed just before you click the Close button. Just one other thing you might mention: the calibration will be using the Average Combine method to process the darks and flats - it just assures a newcomer that everything is as it should be. It's starting to come together. Paul.

Tom, or anyone else, Having now got my calibrated image I find there are a number of points in the tutorial for the instrument magnitudes step that I didn't find helpful. 1) The instructions advise you adjust the radius of the aperture to "slightly larger than the star image" and the adjust the annulus size appropriately without giving some help. Is there an easy/interactive way of doing this without resorting to the Help system? For example, does one first inspect all the stars that you are going to select, choose what appears to be the largest (in my image that appears to be the variable star), then click on it. To inspect the bounding circle and annulus one needs to enlarge the image. If you are not satisfied with the results then I have been doing the following: i) Click the selected star again - this de-selects it. ii) Adjust the radii and click Save, then click the star again. iii) Repeat i &ii until satisfied. iv) If the star you have chosen is not the variable star, then de-select it and then select the variable star. The major problem with this method is that the enlarged image jumps to mid field each time one clicks on a star, or clicks on any radius field. This results in a tedious processing step. Are there better/quicker ways of doing this? Another concern is that the setting of the radii does for all stars, so that smaller diameter stars are going to have a not insubstantial amount of background around them. Is this an issue? 2) The instructions also say "enter a “zero point” that gives a reasonable value for the instrument magnitude". Are there any guidelines for this? For a newcomer where does one start: 1, 10, 100, 1000...? I've left it at the default 25.0 as I have no criterion to judge it against. And, what does "a reasonable value for the instrument magnitude" mean? What do you regard as "reasonable"? 10, 100, 1000...? Again, I have no criterion against which to judge the instrument measurements I get (typically 15 to 17 with Zero Point = 25.0, -0.1 to 2 with Zero Point = 10.0). I can see that all this does is shift the measurements up or down, but what is a "reasonable magnitude" to aim for? 3) Finally, the Result tab appears to show only the last selected star's Raw Instrumental Magnitude. Is there a way of going through the list of stars selected to check on the magnitudes you have written down? If not, perhaps the developers might like to consider this facility in a future upgrade. I briefly looked at the Report tab but couldn't make sense of it and the Help system doesn't cover it. Instead, it talks about a Details tab that my version doesn't have. I'm still in there plugging away, Paul.

Hi Paul, It's very interesting that our backgrounds are both similar (science) and different. I'm a retired Navy oceanographer. My experience is primarily in the field (actually, at sea), working from the tropics to the arctic and often collecting data under some pretty arduous conditions. It's interesting how ourindividual backgrounds give us a different perspective and approach toa problem. I really appreciate your careful analysis of the tutorial. I've decided to give it a complete scrub so I look forward to any additional recommendations you might have. Warm regards, Tom

Hi Paul, Thanks again for your feedback on the tutorial. Concerning the radii settings, I think the best thing I can do here is to add an example image of a star with the aperture and annulus rings. I also will include the following three cautionary points, 1) make sure the aperture ring is slightly larger than the star so that no photons are lost, but 2) not so large that an adjacent starcontaminates the dataand 3) adjust the size of the annulus rings so that no stars are included between the inner and outer ring. Starting with the brightest star is a good idea, too. That will ensure that the aperture is large enough for all your stars. Yourchoice of the zero point is truly arbitrary because it's only the difference in Instrument magnitudes between target and comparison starsthat's important. So I think I'll just eliminate that instruction. Going with the default value should be fine. I'll check it out first, though. Concerningthe Raw Instrument Magnitude, I'm not aware of a record that's maintained in the software. It would be nice, though. I have acouple of notebooks filled with all the numbers I've recorded by hand over the last couple of years. Warm regards, Tom

Hi Tom, Yes, I've struck this difference of perspective when I have rubbed shoulders with engineers and biologists in my work. I guess each technology imbues its adherents with its own view of the world. I'm glad to help with the tutorial. While I can usually figure out what's going on, although there were a few points that stumped me, I'm thinking here of a newcomer who might be easily put off this interesting and rewarding field just through trying to struggle with unclear advice. It's the same with computer software manuals - it's quite a skill to be able to write effectively for a broad audience without talking down to the more tech-savvy ones. Idon't believe that rewrite isn't warranted as, in my opinion, you are at least 80-90% there already. I'm about to post some questions re the Finishing Analysis step - I find there are some questions regarding the spreadsheet your team has so kindly put together for us. Best wishes for your revamp, Paul.

Hello again, Tom, and thanks for replying. I realize that the tutorial can't be a user manual (the software distributors would need to pay you for it, if it were), but there are some points that probably need to be somewhat spelled out to help the newcomer get results. What you suggest regarding setting the radii seems to fill this bill, especially if it is backed up with a screen shot or two. Now, on this Settings panel I notice that the default values for the radii put them at 6.0, 9.0 and 15.0. My question is: having decided on your inner radius, let's say it's 10.0, should one then keep the step distances the same, i.e., 10.0+3.0 and 10.0+9.0? If there are practical minima or maxima to these steps, perhaps they should be mentioned. For what it's worth, I chose to keep the step sizes the same and then inspected each star selected at higher magnification (Ctrl-I) to ensure that the stars were inside the inner circle, and that the annulus was free of other stars, then adjusted accordingly. The zero point is a bit of a tricky one to explain to the newcomer. Perhaps it should be left at the default of 25.0 to begin with and optionally adjusted to give all instrumental magnitudes >0 (simply because, for the non-technical person, positive numbers are just easier to deal with and to understand). Their absolute values have no effect on the Transformation Coefficient value determined (the slope of the least-squares line of fit). I take it that the other parameters on the Single Image Photometry>Settings panel can be left at the default - if so, a comment to this effect will allay any concerns. Finally, reviewing results: This part of the AIP4Win program could definitely be improved (Richard Berry, are you reading this?). Here you have selected a suite of stars but there is nowhere you can see the results of these selections listed. A good program would automatically provide you with this facility, even if it has to be displayed in an entirely new window/panel, and also provide the facilities to print it out. I suspect that the Report tab is eventually going to do something like this, but as it doesn't seem to be working for me, I could be wrong here. Kind regards and good wishes, Paul.

Hi Paul, I agree completely with your point about how an unclear tutorial can be off-putting, especiallyfor a newcomer. I've been there too many times myself. I plan todevote some hunker down time this weekend to the AIP4WIN section. Hopefully, I'll be able topost results early next week. Tom

Hi Paul, I like your expanded procedure and will incorprate it into my revisions. You are right, too, about both boxes being checked. I'll revise that image, too. Thanks again for your input. Tom

Hi Paul,

Let’s see if I can address your specific questions above.

Concerning the annulus radii, I’ve taken another look at section 10.1.2 in HAIP and don’t see a formula for determining what values to use. I would say as long as the annulus is large enough to be representative of the background brightness and located so that stars are not inside, it should be ok.

I agree that the best way to handle the zero point is to simply leave it at the default value. As far as I know, any changes are purely esthetic. I also agree that adding a comment to leave the other settings at default would be helpful.

If you would like to leave comments for Richard Berry, he moderates two Yahoo groups. One called AIP4Win and the other AIP4Win Photometry.

Warm regards, Tom

Hi Paul, I've gone through the tutorial and tried to incorporate the changes we discussed plus others I thought would help. Before I send it in to replace what we have now, I'd really appreciate your taking a look at it. I apologize for the awkward way I've had to attach the images. Many thanks for all your help. Warm regards, Tom

Attachment	Size
AIP4WIN Tutorial - 2011.doc	34 KB
Image 1.png	2.1 KB
Image 2.png	14.9 KB
Image 3.png	3.8 KB
Image 4.png	1.79 KB
Image 5.png	10.36 KB
Image 6.png	11.53 KB
Image 7.png	3.02 KB
Image 8.png	11.62 KB
Image 9.png	13.77 KB
Image 10.png	84.22 KB
Image 11.png	4.65 KB
Image 12.png	218.26 KB
Image 13.png	912.96 KB

Hi all! There's a brand new version of AIP4Win out, and I wonder whether it would make sense to also brush up the AIP4Win tutorial to make sure it's still in-line with the new version. I think this could be done in the DSLRphotometry group, for example, which is open to new members anyway. I know Richard Berry actually got some feedback from Citizen Sky volunteers and used it to improve AIP4Win (see here: http://www.willbell.com/aip4win/aip.htm ). One might also consider covering the Magnitude Measuring Tool (MMT) of AIP4Win which is the recommended (by the author) method to extract photometry from a series of images. CS Heinz-Bernd

Hi again Paul, Well, it looks like my revamp of the tutorial may be for nought - ugh! I'll download version 2.4.0 and see what changes I need to make to the changes. Tom

Tom, Thanks for giving me the opportunity to comment on your editorial changes. I've downloaded them and will peruse them tonight. However, I am about to go to Newcastle, a large regional centre 230 km south of here, for a few days as my wife has a specialist appointment. This means I will be off line until Friday (my time) at the earliest. Concerning vers. 2.4.0: In the words of a famous Prime Minister's wife (equivalent to your First Lady): "Wouldn't it rot your socks?!" It's a pain in the posterior to keep track of version changes and deciding whether to keep notes for two versions or just the latest. I don't envy you. I've installed 2.4.0 simply because 2.3.1 didn't support my DSLR's files (2.4.0 doesn't list my camera as supported - Canon EOS60D - but its predecessor is listed and it appears to read the files OK). I notice that it simplifies the green-channel greyscale conversion routine, but that's all I've had a look at so far. However, I'd like to ask Berry and Burnell why they don't offer a batch processing option here - it's a pain handling 25 and more files each of images, darks, flats and flat-darks. I'm sure this would be a boon for CCD imagers as well as DSLR imagers. Even adding short-cut keys for Open Image (Ctrl-O) and Save as FITS (Ctrl-S) would be a small step in that direction. Are you monitoring this discussion thread, Richard and James? Cheers for now and best wishes, Paul.

Hi Paul,

Thanks very much for looking over the changes. Hopefully, by the time you’re back from Newcastle (not carrying coals there I hope) I’ll have the 2.4.0 issues sorted out.

You are quite correct about the potential for batch processing. It wouldsure reduce the repetitive drudgery. The short-cut keys you mention are, in fact, already there. Looking at the row of buttons across the top of the screen, the "open" button is at the far left. The "save as FITS" button is the third from the left.

Warm regards, Tom

Hi Paul, Here is a recent post by Richard Berry on his AIP4Win-Photometry Yahoo Group. Hi Tom-- The most valuable feature for Citizen Sky users is the ability to open a raw color DSLR image directly as a green-channel monochrome image. You set this in the Preferences > DSLR Conversion settings. Choose DeBayer, Convert Color to Grayscale, and set red=0.0, green=1.0, and blue=0.0. Then click "Save." All DSLR images will now be opened as green-channel monochrome images. To enable dark frame subtraction and flat-fielding, use the calibration menu to set up a master dark and master flat. The bias, dark, and flat frames must be opened as green-channel monochrome images. My personal opinion is that using your DSLR's built-in "noise reduction" feature (which is just dark frame subtraction) gives better results than shooting separate bias and dark frames, but AIP4Win allows you to do so if you want to. Flat fields are tricky to shoot with 50mm lenses on DSLRs, but if you do them carefully, they work well. I place a small sheet of opal glass over the lens.and use afternoon sky light to make my flats. One nice feature of photometry with DSLRs and 50mm lenses is the optics geometry doesn't change, so flat-frames are good for days or even weeks. Ghis group identified and we corrected a bug in reading the date/time stamp from DSLR images. For most people, the date/time is set to local civil time, so on the Observer tab, you must remember to set the Time Zone setting to you time zone. Observers should always -- a matter of scientific due diligence -- check the Julian Day that appears in the output, just in case. In the Magnitude Measurement Tool, on the Images tab, check the Calibrate Files when Opening checkbox. When the MMT opens your DSLR images, they are calibrated before they are displayed. The MMT allows you to queue up any number of images, so for my testing, I would usually shoot a dozen eight-second exposures at ISO 100. I used a *.STAR file with eight comparison stars. You need to use a fairly large aperture to accommodate trailed star images. Capella makes an excellent guide star in the "Guide star offset, re-center, follow" mode. Because DSLR images load rather slowly, the actual photometry run took three or four seconds per image, but the group of a dozen images was done in less than a minute. I think the output format which is most appropriate for Citizen Sky is the Instrument Magnitudes Report. This gives you instrumental magnitudes for the target and for each comp star. If epsilon Aurigae is low in the sky, select the Each Star option for air-mass reporting. I've done several photometry runs on eps Aur with my Canon 350D and a 50mm f/2.8 macro lens, and was pleasantly surprised to find the image-to-image standard deviation in the magnitude for epsilon was around 0.005 mags. --Richard

Hi Paul, I've installed 2.4.0 on my computer and used it to process an imaging session. I'm attaching a file of the updated tutorial and one image to replace the old "DSLR Conversion Settings" dialog box. I would really appreciate your feedback on the new tutorial. Many thanks, Tom

Attachment	Size
AIP4WIN Tutorial - 2011b.doc	34.5 KB
Image 2b.png	15.98 KB

Tom, Firstly, I apologize for not getting back sooner - I've been sidetracked for a few days. I've attached my suggestions (made in red) to your initial tutorial revamp. See what you think, but what you did reads a good bit clearer for a newcomer, in my opinion. I'll have a look at the v2.4.0 version (at least I presume it is) and get back to you more promptly. Best wishes, Paul.

Attachment	Size
AIP4WIN Tutorial - 2011 - ed CPL.doc	38 KB

Hi Tom, I've had a read through of this version and again it reads well. I think you have incorporated the v2.4.0 changes quite nicely. I also think the comments/suggestions I made to your earlier revamp mostly still stand, so you might like to review these in the light of the v2.4.0 differences. Just a few minor comments: 1. I notice that you haven't opted to use the MMT process - I presume that this represents a reasonable departure from the current tutorial and probably needs to be tested a bit before you endorse it. 2. I, personally, am a bit suspicious of the camera's noise reduction algroithm for photometry work. If I remember a test reported by Christian Buil (http://astrosurf.com/buil/d70v10d/eval.htm) he came to the conclusion that there were some problems with it - I'll have to re-read this article. Cheers for now, and well done with the reworking of the tutorial, Paul.

Hi Paul, Thanks very much for your excellent recommendations and corrections. I'll start working to incorporate them right away.

My feeling about the MMT is that it is too complex for someone makinga first attempt at photometry. My goal has been to get a beginner through the simplest procedure possible to obtain a reasonable result. The MMT is a powerful and sophisticated tool and would be a good candidate for an intermediate level tutorial - probably best written by Richard Berry if he would be willing. Concerning the automatic noise reduction, I've heard arguments both ways. On the plus side some say it's superior because it is performedunder exactly the sameconditions (temperature)as the imaging. It's certainly a lot easier, too. Okay, back to work. Warm regards, Tom

Hi Paul,

I like all your suggestions and believe I have them all incorporated now. The recommended image is also attached.

I think were getting close.

Thanks again.

Tom

Attachment	Size
AIP4WIN Tutorial - 2011b.doc	39 KB
Image 13b.png	615.55 KB