Clipped/noisy subs - Possible FITS pedestal/black-level regression: Page 8

Clipped/noisy subs - Possible FITS pedestal/black-level regression

Nasa-HSIA

mbits Is this test app for coming new version?

mbits

Nasa-HSIA Yeah probably. It's a higher version. Obviously can't say for sure what ZWO's plans are though. Fingers crossed it fixes this issue.

Nasa-HSIA

mbits Please share your experiences to us after using this test app to observe DSOs.

Umbra

I also just got the new version, V3.1.0 Beta01
I will let you guys know what I see in this version.

hlc

WARNING! I've updated to the trial app/firmware version provided by ZWO via email and my S50 is now completely unusable - it's not saving any subs AT ALL - IRCUT or LP - just totally black image! I'm just about the contact ZWO directly.

Umbra

hlc
You have stretched the images, correct? Because when loading they appear black for me too, as all these fit images do (ergo: that's normal), but they view perfectly fine after stretching or viewing through autostretch.

basically, fit file subs transfer over perfectly for me onto computer. I am sick, so I haven't checked the clipping yet though.

Edit: I have an S30 though. but the images load into siril fine and I can view and stack them normally. I will check for clipping soon.

hlc

Umbra nothing to stretch as it didn't save a single sub, in either of the capture modes (live stacking or continuous). Just kept saying 'not enough stars'. Perfect alignment in EQ mode, pointing at M45, so there were plenty of stars ;-) (and the S30Pro did a fantastic job using older firmware). I couldn't spend too much time on it as it was about to cloud over, so I'll have another go when I next get clear skies.

Umbra

(Good?) Update:

I just checked a single 10 second sub on the histogram in siril and compared it to an older, pre-beta sub. I zoomed in to 10x zoom, applied no stretch and did not debayer either. The subs after the update look cleaner and there appears to be no clipping anymore. problem solved?

Guys, you who know siril better and got access to the beta, please provide input as well so we know I am not just an isolated case. I am very new to this hobby and am still learning, so someone more skilled than I would be appreciated to give input.

Nasa-HSIA

Umbra Hoping that these known issues can be solved in the coming new APP.

Michael875

Umbra Can you try overstretching the image by dragging the middle slider in Siril all the way to the left? You will get a white picture and it would be interesting to compare the number of black pixels before and after the update.

I also got an invite to the beta, but no clear skies within the next week in sight.

hlc

Umbra certainly looks promising! Don't know why mine failed completely (S50 vs S30 perhaps?) - I'll have another go. (Edited to add I used the Android beta app, not iOS - no idea if that would make any difference though. I'm not inclined to try the iOS beta app yet as I'm using that for my S30pro)

shamrock

Thanks for the update Umbra – looks promising.
But hlc's finding that "... S50 is now completely unusable - it's not saving any subs AT ALL ..." is concerning.
I have the s50!!

t5KI0Kra

I have the beta version. It has a firmware update with it.
It's several hours till dark here and we've only had one clear night in the last 3 or 4 wks, but tonight may give a few hrs.
I've tried everything in daylight and scenery mode and it is saving images.
Whether that means it will save astro subs who knows.
But at least it hasn't bricked anything.
Just thought I would post this in case hic's post is scaring folks off.
It did make me think twice, but I only had a paper weight anyway.
Dark skies folks

Lilo

Tim1343
Stargazing
Skip skip everything
Continuous capture
And it will take single .fit frames

Umbra

Michael875
Just did that. the new images are pretty much perfect white, while the old subs are... grainy, to say the least. So the clipping issue is resolved, but now we straight up have people with S50's where they don't save subs at all. I mean, sure, the beta is a little buggy, but wow, thats an issue.

You know what to do, guys. report it to zwo. email them. give them feedback. that's the purpose of Beta, after all.

by the way, here's 2 overexposed images. the old sub is the grainy one, the pure one is the white one.

t5KI0Kra

A fail from me as well. EQ mode. It did everything right. Set up properly, went to the target. I could see it on the screen. Took the first shot and the screen went a total blank. it stayed that way for each shot. I tried 30 and 10 sec. Just the message couldn't stack, not enough stars. After one of those messages it even suggested I try a slower shutter speed !!! No images saved to the S50, I checked that I did have save all images checked.

hlc

t5KI0Kra out of interest, was that with the Android or iOS app? Wondering if we can narrow it down to either the device (30/50) or app os. (my fail was S50 with Android - haven't tried another combination)

t5KI0Kra

hlc Android on a samsung tablet. S50

Nasa-HSIA

Umbra Are these images taken using LP or UVIR cut filter? I do not find the issue by using UVIR cut filter.

Michael875

Umbra That does look like they resolved the issue - at least for the S30.

hlc

Umbra Yep, have emailed ZWO support.
Also, I checked earlier and it looks like it did actually save one .fit file after all (when it was in 'continuous' mode rather than live stacking), but it's completely black/empty. No data in the histogram. Took a photo in scenery mode no problem (that I can see).

Umbra

Nasa-HSIA
I am using the LP filter (both images taken in "v3.0.2" and "v3.1.0beta01" are with LP filter).
Regardless, the issue has been resolved in the beta firmware i received, but now zwo should just fix the other issue where some users get blank subframes..

Nasa-HSIA

Umbra Thanks for your sharing. Some SeeStar users have reported a high rejection rate during shooting, regardless of exposure. Did you encounter a similar issue when using app version 3.0.2? If so, was this issue fixed in app version 3.1.0 beta01?

Umbra

Nasa-HSIA
So, I do feel like I got a lower number of rejections, yeah. whereas with V3.0.2 I got about, say, 100 rejections for every 450 good frames, I got about 40 rejections for 450 good frames last night (not exact numbers, I'm just remembering vague numbers. I will take more careful note tonight.)

I have a theory that the bad, clipped data in the subframes causing excessive noise messed with the firmware. It might have seen the noise and that triggered some form of threshold that caused the higher rejection, and that caused a higher number of image drops in v3.0.x which seems to no longer be the case in v3.1.0 b01... but I could have just gotten lucky. Clouds-willing, I will be able to do some more shoots tonight to see for sure. I only had about an hour last night...

I am really hoping I am not just an isolated case.

« Previous Page Next Page »