Anyone who says PixInsight is unintuitive has never used IRIS.
IRIS was my tool of choice until I made the switch 6 years ago after seeing Rogelio's presentation at AAIC 2011.
If you ever miss IRIS, H, you can always do command line in PI
As someone who cut his teeth on TOPS-10 and early versions of UNIX I still appreciate the power of a command line and scripting. Not much good for the casual/non-technical user, of course.
By the way, SIRIL is meant to be IRIS for linux.
The name is actually IRIS spelled backwards + L for Linux. It is actively being developed unlike IRIS which seems to have reached the end few years ago
Dunno where "unintuitive" comes from, PI is just a workspace with a bunch of tools. It's like saying a garden shed is unintuitive.
If you want dialog boxes to hold your hand while you click "Next...next...next..." then it's clearly not for you
If I went into that garden she'd I fear I wouldn't make it out alive, or at least without chopping of a toe . The UI in PI could charitably be described as 'Novel'.
Having used a bunch of scientific software in my time, such powerful tools usually fall the wrong side of user-friendly...but...with such power comes great responsibility
They don't need to be user friendly when their users comprehend how to process data. Most "astrophotographer" don't and want/expect a one-click solution to their lack of ability to learn.
Ben, i have processed Canon RAW OSC files through DSS and i know it works for me. Have you configured the file bayer matrix settings correctly? There are a number of Utube videos on how to process OSC files in DSS.
There are threads on Cloudy Nights discussing ASI071 colour processing using DSS, here is one if them, with some clear screen prints from Jon Rista. https://www.cloudynights.com/topic/5...-color-issues/
I use DSS to stack colour FITS files from ASI071, no problems.
I think we've veered away from this threads subject. I havent processed yet with the latest PI update but its changelog is impressive and interesting to read. APP I'm still wrestling with and found some unannounced limits with large amounts of data. So it seems was written expecting people to work a particular way but unlike the author of PI they seem open to adjusting so people can be more adventerous or creative with how they work.
I still havent done stretching tweaks in APP but getting a clean integration file is still showing to be better than PI, way better than PIs BPP and since the steps (calibration, normalise, register etc ) I don't need to adjust for each set (i make master flat/dark/bias first then load in those plus lights) then i just need to press "integrate" and it runs through the batch steps with the settings I previously used so it can become a one step process for the mechanical stuff.
I still think today APP is not worth the money, was released as a full price product way too early. It does have the agorithms to do awesome stuff with your data, its not a dumbed down quick program, it can be faster than PI at some things but can also be slower, I havent done a direct comparison yet but dont think its really necessary either. But will see.
I think maybe for existing PI owners APP could replace the BPP script and I see that at least is how I'll be using it for a while. The best thing about APP though is it is pretty strong and a good competitor to PI, so its existence may encourage features and improvements in both products which will be good for all.
For people thinking to step up from DSS and particularly who just want a pretty picture APP will give you that easily right now. Its display does an autostretch that works well, better than PI usually does so you can save out a nice jpg pretty quick and its default processing settings do a good job so you can dump in light, flats, dark and jump to the integrate button and thats pretty much it.
unlike the author of PI they seem open to adjusting so people can be more adventerous or creative with how they work.
Juan doesn't suffer fools gladly but I've found him quite accommodating when approached with a cogent suggestion.
Quote:
Originally Posted by sil
I still havent done stretching tweaks in APP but getting a clean integration file is still showing to be better than PI, way better than PIs BPP and since the steps (calibration, normalise, register etc ) I don't need to adjust for each set (i make master flat/dark/bias first then load in those plus lights) then i just need to press "integrate" and it runs through the batch steps with the settings I previously used so it can become a one step process for the mechanical stuff.
Juan may be better in private but he's pretty aggressive with honest questions on the forum imho. Enough to avoid asking questions myself. And I dont suffer fools gladly either unfortunately, I don't mind someone who genuinely want to know but tire of people too lazy to even look for an answer themselves. I'm in no way bagging his product, PixInsight IS awesome, powerful and my program of choice, never regretted buying it.I'm nowhere close to your league either Rick btw! You do a lot of great stuff with PI and know it far better than I ever will and are able to help others better than I can too.
In what way(s) do you find it better?
This is entirely subjective on my part at this time. And as I said its comparing to the BPP script which most people use and itself says its not the optimal method of producing an integration frame.
Just to clarify, I do 95+% dslr on tripod untracked shooting. I consider three parts to a picture: 1) capturing data, and darks/flats/bias (but not essential, just helps) 2) produce an integration frame, a single fits file containing everything from step 1 included and applied as required, 3) is the cosmetic stretching, cropping, cleaning of the integration frame to get your pretty picture
So that's what I mean by an integration frame and I always keep my integration frames as these are the real starting point for reprocessing later on as i learn or in this case try out new software. So I've been throwing old data sets at APP and have looked at a few resulting integration frames in pixinsight along with my original PI integration file.
Looking at the whole frame side by side using the same screen stretch they look identical. Which is what you'd expect and is good to see, shows they are both doing their jobs I think.Zooming in to the middle again both look to my eye identical but going to the corners I feel the APP one to be a bit better looking in that the stars are a little less distorted (I shoot camera lens remember so corners are all distorted). APPs integration also has several weighting options including "quality" so it may have used only the best 50% of frames for each star in the shot.
PI's BPP I find temperamental at times, its not always tolerant of the size of the dataset or not having bias frames, or flats whereas APP I can leave out flats, darks and bias and it happily runs the same process with the same setting each time reliably for me. It does choke on larger datasets and I'm still trying to find its limits there for me (the smaller batch sets i'm playing with atm have happily used 83x 36MP RAWs). Using Master calibration frames can be problematic too at times. Some programs make/want them at the largest display size and others at the largest RAW size (often a 50 pixel dimension difference), my Master Flats/darks wouldn't work in APP and I need to test the reverse since I'm unclear if I made those masters originally with PI or Photoshop.
I have NOT done a proper test of each program I just dont have time or drive space right now (I'm using an external hard drive as my working space for APP as it is). I'm sure someone will do a proper review and empirical comparison of APP and PI. Sorry I can't prove anything to you yet or show you a screenshot, I own both programs and its for my own interest I've done any comparisons at all, not to say which is better. PI certainly has the depth of tools and flexibility to cope as the user understands how and when to apply various techniques.
With all that in mind I'm confident in saying APP gives me a better integration file than PixInsight's Batch PreProcessing script. Certainly at least equal to.
Also for my way of shooting APP's registration and integration components allow me to mosaic all my shots in one go so i get a little strip instead of one frame, this make it easier to find a target, a comet for example, anywhere in my data if I wasn't certain I had the camera pointing at the right spot in the first place. In PI I have to go hunting through random frames to try to see if I might have even got my target for the session at all in any of them. APP gives my a single integration file with ALL the data in it so I can tell straight away or crop straight to the region and process to get my pretty picture knowing it already has all the data integrated I had to give it, so no need to re-register and integrate. So overall its a time saver in that regard.
The cosmetic side I havent really touched but it does an autostretch by default when you click on a file to display. This seems to be a less aggressive stretch compared top PixInsights Autostretch in STF which often stretches into a noisy image again for me. APPs one gives you a pleasing stretch for a newbie to save out as a jpg. DSS at least had no autstretch when I last used it so on that score APP would be a good step up for DSS users, it can be a minimal thought tool to get an image out of. The tools in APP seem to be the same as in PixInsight and I think Local Normalisation is coming in the future to PixInsight as well as the mosaic reg/integration. So my feeling is APP is using the same equations PI would be using, providing less ways to tweak them though so results will be close either way. its not a revolutionary program that's using all new algorithms for everything or the most optimised.
I don't see any problem with jumping between programs for getting the best end result. For me using APP to get my integration frame is preferable to BPP ( I honestly never came up with a better way to use darks and flats in PixInsight myself )and then loading into PixInsight for my stretching work which can be time consuming as I have a pretty complicated evolving workflow.
I don't think APP will grow to surplant PixInsight which (despite my dislike of Juan) is a damn awesome program with capabilities beyond my skill level to know how to use and possibly good enough for far smarter people than me to use too. The pros will see APP as maybe more simplistic and it sort of is but I dont think what it has is any less powerful as a result. It'll be interesting to see how pros with AstroArt and Nebulosity compare them with APP too. But its great to see a new strong program out there even if its launch was badly done.
Thanks, sil. I'm genuinely interested to know what you like about APP. I will get around to trying it myself one of these days but I have too much else on my plate right now.
I don't think that PI (or anything else) is the one true astro processing application. We all have different preferences and ways of working. Folks that value simplicity and ease of use are generally not going to be huge fans of PI I don't care so much about that, having started a career in computing during the late Jurassic, and I'm happy to climb steep learning curves if I think the results justify it.
APP shows promise and I think there is a niche for a processing package that combines ease of use with enough power to produce good results. As you said, a bit of competition could be good for all of us users.
Rick, at the moment I fear you'd not find much benefit from time with APP over PI. I am really disappointed that it was released at all, its a beta program really. You'd comprehend the tools which are familiar but less controllable than PI but then any of those controls you want that aren't there might put you off. If you are dealing with hundreds of Lights its not so reliable yet, and its use of Java is itself a negative point. But it IS strong and shows great promise, but we've all been let down by promises before and Mabula (lead dev) might decide to pack it in today, or worse microsoft buy it no capture side to it will put some off too as will the lack of documentation (a large reason I dislike Juan All I can say is its not Snake Oil and I will use it for the immediate future for part of my workflow but yeah if you're time poor don't worry you're not missing out on wonders right now. I've got AstroArt and Nebulosity etc too but found them too rigid when I bought them to work with how I wanted to work. So i'm several years behind with those programs and not vaguely able to comment on comparing for those users. I hope it bodes well for astronomy in general, however people come into it, and more options, more growth, more word getting around and more cool pics getting into people brains and sparking curiosity and interest. It's good for us all.
I agree I don't think PI is the "one true astro processing application" nor is anything else at this time. We have some strong products and some amazing work being produced with them and as long as they continue to be developed and improved we have more flexibility in how we want to use them for scientific or creative purposes.
I'm interested as trying to figure which software (capture & processing) to go with. So far have only used sharpcap & APT (capture) - DSS (stacking) - startools (processing and occasionally photoshop as I have it for other stuff).
I'm interested as trying to figure which software (capture & processing) to go with.
Neither APP nor PI will capture so those are out on your criteria. Otherwise have you tried FireCapture (free)?
I was hoping to have a pic to show PI vs APP and just to complicate things PI has proven better as a one click solution in this test so far. But I havent an integration frame to compare yet. The other morning I went out and shot Orion, full set of darks, flats, bias (mucked up my flats settings so they arent ideal) threw them all into PI's BPP script, chose first file as registration target and and output folder. Pressed go and went to bed. 200 frames later when I woke up it had ran through and I have an integration frame. No errors or crashes (PI is updated to the release from a week or so ago). Trying the sam with APP it stopped partway through waiting for user input SO I'm stepping through the steps individually which is going to take longer but I hope to have an integration frame to show and compare this week. Maybe, if telstra fix my damn line (I have no internets at home for a week and its playing havok with my home automation stuff and sleep and this machine is one I rarely touch and APP is a fresh install (I've been playing from a laptop downstairs instead.)
Anyway unsure what to expect at this point and I'll post when I can however it turns out. As i think i mentioned earlier a new install of APP will need some user info if you just give it frames and press integrate, it runs through the numeric tabs in sequence on its own util it needs input but after some use it should be storing the changes you make to the steps to tweak them and soon enough it does become a one click step but this machine I'm testing APP and PI on I havent been using for APP so its not there yet. I am still to test how the mast calibration frames from one perform in the other too. But in my laptop testing I'm evolving how I store my master calibrations to make future reprocessing of old data easier to my needs. So its forcing me to rethink things and keep notes better on all my source frames
Or am I just weird wanting to go slow and learn a new program and understand what its features do in order to try to get the best out of it?
Ok as promised I have a screenshot comparing APP and PI using same data as quick and dirty as a newbie would do. This does not involve tweaking of setting to get the best from either program. In PI I used the BPP script as is, and in APP I had to run the numeric tabs in order manually touching as little as possible. I tried to make it as fair as possible and not the best that either program should be capable of.
Annoying for me it contradicts my earlier quick tests and i still stand by those comments and can't account for the deviation I got for this test where i was trying to give each program a fair chance.
Shots were Nikon D800E with f2.8 200mm lens on tripod. Sets of lights (200 frames), darks(35), flats(77) and bias(38) were taken. The flats were over exposed so not ideal. In body light pollution filter in place as always.
PI I loaded everything into the BPP script set reference frame (first light) and save directory, hit the go button and went to bed. In the morning I had my finished integration frame with rejection maps attached. It chugged through it all, no crashes or user interaction needed.
APP however took days to go through it all and I had to step each tab in its process myself as it required some user input during it. So I would set steps off before bed or work as needed to go through each step. My quick laptop tests didnt require as much interaction.
My feeling is it took much longer on APP by default than PI just accounting for actual processing time. The screenshot attached shows the integration frame result from both programs left is PI, right is APP.
The flatter windows underneath sho closeups of the center of each integration frame.
Observations:
1 - application of calibration data has improved overall vignetting of the right side, the APP image. PI clearly exhibits diagonal lines in the vignetting and less smoothness overall.
2 - worryingly the APP stars all show a vertical streak artifact I have yet to see in any of my testing. PI is much better and shows my lens was out of focus, no hint of those streaks (they are not in my lights either). I think this is an integrationartifact of the default settings of APP, again this is the first time I've seen this so it doesnt confirm my earlier observations of better strs than PI.
3- more worrying APP if you look close produced an integration frame that is larger than the registration target in both width and height. You can see the ugly blank portion in the top right corner where lights didnt cover that
4- the colour cast I can only attribute to the differing default settings. APP takes you through a calibration step and a normalise one. Again with newbies in mind this gets you closer to a nice final image to show off with as little tinkering as possible.
Absolutely NO cosmetic processing has been done, both files loaded into PI were black with a few point stars visible, I used STF to adjust the levels for display, dragged this to the workbench and from ther dragged to each image so they both have the same visual adjustment applied for this comparison.
In recent months I've seen a number of people asking about moving beyond DSS and photoshop and I've tried to post with them in mind, experts in any package have nothing to fear from APP today. I still think it will improve and stabilise fast and it shows some promise to me, unfortunately my first level field test doesnt show that to you all. Until this I was exclusively doing full mosaic integrations which are of worth to me rather than to a registration target. But I cant account for the difference in integration sizes, both programs appear to have used the full raw dimensions but APP did a litlle more, perhaps its a stretch of the images, that may also account for the star deformation too.
I'll leave it there, need sleep and got python code to work on this weekend but I might point the author of APP to this and he can investigate if he wants.
2 - worryingly the APP stars all show a vertical streak artifact I have yet to see in any of my testing. PI is much better and shows my lens was out of focus, no hint of those streaks (they are not in my lights either). I think this is an integrationartifact of the default settings of APP, again this is the first time I've seen this so it doesnt confirm my earlier observations of better strs than PI.
Ok, hands up here. I reprocessed this data with PixInsight and those streaks came back. This set of data is not one I took quickly just for this test, insomnia gave the chance to get outside early for some data collection. After writing the above I'm done comparing the two for now and so processed the data more seriously my usual way. Whereas as mentioned above I was trying to keep default settings in both programs for the comparison, not optimal settings (which may not exist, I am forever tweaking and testing mine to matchmy gear and data for the way I want to work. So the streaks came back. Last time I stepped through viewing the jpegs from my camera and the streaks arent visible (still not this time either), but this time I threw ALL the RAWS into pixinsight and zoomed in, a time consuming process and found three frames where there was a small streak from vibration, it was in three positions and added to make a long streak. So i deleted those frames and reprocessed my preferred way and all good again.
So my bad blaming the programs and making false assumptions when it was in my data after all . Sorry. So I dont know if i can conclude APP was better after all at recognising the signal was an outlyer to be removed or if it removes more signal out of the box (remember the context of my comparison is how it works by default, since little documentation or use I can't try it at its best potential since I just dont know how yet. In PI I go for quality and I err on the side of caution when choosing values that remove signal so more tweaking on my part there... the constant search for perfection
New APP out last night too, tons of changes that could effect all my negatives above too. I think I'll wait for releases to settle and I feel its at last a "final" before comparing again.
Thanks sil for your efforts in comparison. Could you please update this thread when you have further experience with APP? Whilst I bought PI, my trial of APP showed promise, unfortunately, whilst I liked it, I had a few issues with it that were not fixed prior to my trial running out. Still think it may be worth adding to the arsenal in a couple of months as further development occurs. At this stage I'm reluctant to spend 125 euros - if at was 50 euros, I'd buy it to encourage development.
cheers
Gary
Something I observed recently. I havent done anything much with APP recently, been working on projects with PI. I shoot nikon DSLR with an in body filter I reviewed elsewhere here. So my colours are out of whack in my subs as a result. Running a batch through PI's BPP script to produce a single integrated frame and stretching the colour cast is ugly and obvious. Can of course been removed and I do.
Same batch ran through APP to produce an integrated frame i chuck that into PI along side PI's and the APP one is clearly a good colour, no cast and stars etc look about right.
NOW, fire up PI's new Phometric Colour Calibration and hit the PI integrated frame and rest/autostretch it and now both integration frames are VERY VERY close. Damn this new PCC rocks!
More importantly APP is doing a more accurate colour calibration process by default without the plate solving etc. So again simplisticly I guess its 98+% colour accurate without any fuss while there are tons of options in PCC which may be better for certain data and targets (I left its default ). In PI I would say its slightly more saturated than APP, that might give APP the edge to be artistic with stretching and saturation but seriously it seems to be almost nothing between them. Again this is observational, not true testing. I just tried to make the best integration file from both programs before deciding which to use for the next part of the project. So I won't say any more but its great to see the integrated data has such accurate colour without the cast (light gradients are still present) I'll have to see if the result is the same with light polluted data, I'll try to keep it in mind.
. The UI in PI could charitably be described as 'Novel'. 
