0.00
#...
·
 |
|---|
|
Does anyone know why I should get the attached error message in PI when I did in fact have Darks/Bias/flats but it's not using the darks for calibration! Thanks Jonathan Screenshot 2021-12-09 at 21.48.26.png |
0.00
#...
·
|
|---|
|
Hi Jonathan, I haven't ever seen anything like that - so probably can't offer much help. Are you using one of the batch preprocessing scripts, or doing it the "manual" way? Is there anything "unusual" about the subframes and calibration frames you are using? Tim |
|
0.00
#...
·
|
|---|
|
Hi, I'm using WeightedBatchPreProcessing. There shouldn't be anything unusual about the darks, they are the correct exposure length, taken with the same camera. I'm wondering if there is something in the naming convention. |
|
0.00
#...
·
·
1
like
|
|---|
|
Hi Jonathan I think it’s talking about flat calibration in that message, rather than light calibration. It would try and look for dark flats first. So these should be the same duration as the actual flat frame. If WBPP hasn’t got the correct duration dark to match the flat, and dark optimisation isn’t ticked (i wouldn’t recommend ticking it) it will use bias frames to calibrate the flat and make the master flat. I think that’s how it works ! cheers Martin |
0.00
#...
·
·
1
like
|
|---|
|
That's right, Martin. Normally, the flats have much shorter durations than normal subs - so the standard dark frames can't be used. At that point you have two choices: Dark optimisation - that essentially scales the standard dark to try and match the exposure time of the flat. So if your dark is 200s and your flat is 2s, it would generate the dark for your flat as (dark - bias)*2/200 + bias. No dark optimisation - it just uses the bias instead. Which one is better depends on how your camera works - your camera dark current would need to be linear to use dark optimisation (DSLRs are notoriously non-linear - camera designers do all sorts of processing, even for raw frames. These make sense for standard photos, but not for astroiemagers). You can also try taking dark flats - darks which match the duration of your flats and see if that makes a difference. Personally, I would just stick with using the bias (no dark optimisation). You shouldn't get much more than one electron or so of dark current for your flats, and this will compare with ~ 20,000 electrons for your flat signal, so the noise is minimal. Colin |
0.00
#...
·
|
|---|
| worth having a read of this forum post https://pixinsight.com/forum/index.php?threads/wbpp-warning-message-%E2%80%9Cno-master-dark-will-be-used-to-calibrate-the-frames%E2%80%A6%E2%80%9D.16662/ |
|
0.00
#...
·
|
|---|
|
Hi, I've not got a lot further with this, I get the error message. I've got the selection set to Auto,. I notice that this is probably due to the log reference:Cosmetic Correction: disabled.Demosaicing with pattern: Auto!!! Error: No active light frames in the group.******************************************************************** Not sure why this should be as the light frames are loaded and look fine. The frames are OSC from my CDS Pro II (modded 600D). ********** BEST REFERENCE FRAME FOR REGISTRATION ***********!!! Error: Unable to detect the best reference frame.!!! Error: Missing reference frame, the process will stop.************************************************************ |
|
0.00
#...
·
|
|---|
|
Hi Jonathan, Sorry to hear it is proving so resistant to solution. If it's any consolation, I find myself repeatedly getting into tangles of various kinds with things here. And usually find myself on the wrong track for a prolonged period before I eventually figure out what the real cause of the problem is. That's probably just me though! I'm finding the PixInsight learning curve to be a very long incline, with no sign of the top of the hill yet. Good fun when it works though! In terms of trying to debug your problem, my inclination would be to try doing the calibration and preprocessing as individual operations rather than using the script. That might seem a bit daunting, especially if you haven't done it that way for a while. But it might provide a way of figuring out what is going wrong. Is it worth a go do you think? That's all I've got to offer. Now if you can help me to sort out my background modelisation issues...... Cheers! Tim |
|
0.00
#...
·
|
|---|
| Thanks, I’ll try that next, might have to wait for the weekend though when I finish work. |
|
0.00
#...
·
|
|---|
| So, with some further checking I find that all my darks are taken with ISO12600 and the lights with ISO1600. ISO 12600 is the default ISO in EKOS. I'll wait for the cooler weather at the weekend and then redo my dark library with the correct ISO!! |
|
0.00
#...
·
|
|---|
|
That sounds like it could be the answer! ISO 12600 seems a very odd figure to be the default. Is EKOS a bit of a weird beast? I don't really need an answer to that - I'm just acting curious. It looks like the weekend might be good for shooting darks (but not for anything else). Good luck. |
|
0.00
#...
·
|
|---|
| Yes, I’d never set the ISO so high. I think I may be able to reset the default in EKOS. That’s the highest ISO in the camera. |
2.61
#...
·
|
|---|
|
Jonathan, I used Indi/EKOS for a few years and it doesn't really have a default ISO, you always need to set one and that ISO needs to be consistent over all your Lights, Darks, Flats etc. Also Pixinsight WBPP will only match frames if the filter and other relevant settings are the same, so if you accidentally have your flats with a different filter set than your Lights you need to tell it in the calibration settings. Max |
