Cygnus Loop - Playing with narrowband filters and stock Olympus camera in moonlight

[thinker]

I’ve read that narrowband Ha filter is a “complete waste of money” with stock, unmodified DSLR cameras. But last Wednesday and Thursday (Nov 6 and 7) the moon was 64% and 80% (average) while I tried Baader Ha 3.5nm and Baader OIII 4.5nm on a 72ED teleskope and a stock unmodified Olympus E-M1 MarkII. 15 subs (Wednesday) and 20 subs (Thursday) 6 min from each filter, so 90 + 120 min each filter, 420min total (7 hours). ISO 6400, EQ6-R PPEC unguided. TS 0.79 + Metabones 0.71 causing severe vignetting requiring hard cropping. Only Rawtherapee, DSS and Gimp, manually stacking the two days together with equal ‘weight’ (but relative to integrated times).

You probably would argue that it is not a pretty image (understatement!), but the point with this exercise (and buying the filters) was to see if if I could get to use my gear, including stock Olympus cameras, during moonlight. And get more contrast with nebulae, by adding narrowband data to ordinary data. It appears to be (relatively speaking) a lot of cloudless nights around my habitat (60°N) when the moon grows bigger.

Comparing this image to other images from the Cygnus Loop, both mono and modified DSLRs, the image appears to me to have a lot of signal (?). My lack of PostProcessing skills will hopefully improve a little over time, but probably not fast or much, since I’m not too fond of PostProcessing (but I do find tweaking curves fascinating). [Since I only knew the Big Dipper when I started 15 months ago (really!) and was struggling to locate Polaris (!), and my PostProcessing ‘skill’ was basically cropping and scaling images, it’s been a steep climb to get anywhere!]

The Olympus E-M1 MarkII has PE of 81% according to Photons to Photos which is helpful, no doubt. And I did this image in -5°C to -8°C. This camera works well in single digit degrees (+) Celsius, preferably middle and lower end. Doing a test with long darks with this camera model in +20°C could possibly conclude that it is utter useless for AP. But in my climate I think it is great.

By the way, I did not try to make the background blue/green, it’s due to the OIII data (it is based solely on Ha and OIII data). Trying to ‘adjust’ the background with curves (without making all sorts of selections) appears to also reduce/remove the fainter blue wispy tendrils, which to me would be an even worse evil in this case. And is/’should’ an area ‘soaked’ with nebulae (both Ha and OIII) look ‘neutral’ if you don’t have this 'special desire' for 'neutral'? I took the usual shortcuts and ignored gradients and light falloff. This was after all just a try to see if there was a potential, or a ‘complete waste of time’. After all, I can reprocess it later on.

Anyway, I am quite excited by the test-result below. Does it not have ‘potential’ with perhaps more data and especially better PostProcessing skills/tools? I wouldn’t have given it a thought to try to image during 64% and 80% moon without these filters, so the fact that I got ‘anything’ and that it is ‘at least remotely recognizable’ (another phrase I’ve read) suggests it was not a ‘complete’ waste of money. I’m now considering buying APP when they support Olympus raw files (even though they now support DNG), especially because of their strong mosaic support (I think PI is total overkill for my PostProcessing interest). A cloudless moonlit night I now think is an opportunity with my particular Olympus model (below +10°C) when combined with narrowband filters, even though the OIII files from the night with 64% moon appear to maybe have a tiny little more contrast than from the night with 80% moon. The 4.5nm OIII filter is perhaps too wide for 99% moon (I’ve yet to find out), but I’m certainly not buying 3nm OIII Astrodon filters at three times the 4.5nm Baader price for my use! Use 'Ha-only' if OIII is suffering too much in that particular case.

Anyone care to fill me in on something that I’m missing in my wishful thinking? Please ignore all kinds of black-point errors, as it was just a try see if if I could get to use my gear, including stock Olympus cameras, during moonlight. I still find it hard (i.e. impossible) sometimes to notice any changes to an image when a member (dpreview, ‘Cloudy’, Astrobin) posts a new version of an image, with improvements. So there’s my ‘trained eye’ for you..

Thanks to Rutger who demonstrated that it may not be a 'complete waste of money' to use Ha filters on stock ‘OSC’ cameras with high PE (75+ %).




Imaged during moonlight, 64% and 80%, stock unmodified Olympus E-M1 MarkII and Ha/OIII filters

--
(Harvey) - Jane, I've been thinking...
(Jane) - Oh, do you want an aspirin?

 

[Rudy Pohl]

Very impressive thinker! It's nice to see some real data and actual results and not just an opinion. Thanks also for the link to the photons website so we can check out the potential of our own DSLRs.

Cheers,
Rudy

 

[thinker]

Very impressive thinker! It's nice to see some real data and actual results and not just an opinion. Thanks also for the link to the photons website so we can check out the potential of our own DSLRs.

Thank you for the kind words!

The PE (in isolation) may be a strong contributor, but high PE may (I’m just thinking loud here) correlate with a higher-quality sensor with low read noise and ‘what not’. Perhaps. So it all adds up, and e.g. the Olympus E-M1 Mark II with small 3.3 micron pixels compares relatively favourably to older modified FF cameras with comparatively (very) large (area) pixels? But I really don’t know.

However, maybe there is some/a lot of truth in ‘The proof of the pudding is the eating’, and ‘In theory there is no difference between theory and practice. In practice, there is.’ ...

 

[OutsideTheMatrix]

Nice image! What size filters were you using that caused the vignetting? Also how is your light pollution out there?

 

[Tomx72]

Very impressive work! Indeed, a narrowband filter is a no-nonsense solution with unmodified cameras too. It only takes more exposure time but you'll get there.

The sky is not exclusive to monochrome, high QE CCD systems!

 

[sharkmelley]

Great result! Well done for giving it a try!

Mark

 

[thinker]

Nice image! What size filters were you using that caused the vignetting? Also how is your light pollution out there?

Thank you!

No, I didn’t use filters that caused the vignetting. I wrote ‘TS 0.79 + Metabones 0.71 causing severe vignetting’, which means that I used a 0.79x reducer/flattener and then added a 0.71x Metabones Speedbooster which was really pushing the system. The telescope’s matching reducer/ flattener is only 0.85x, so I was really exceedingly pushing it, and I have absolutely no reason to complain! But I needed to crop more than I (unrealistically) was hoping for. Stacking two images that didn’t overlap too well (from different nights) didn’t help either.

Bortle class 4 is what lightpollutionmap.info suggests around my house. I live on a different continent than you do.

 

[thinker]

Very impressive work! Indeed, a narrowband filter is a no-nonsense solution with unmodified cameras too. It only takes more exposure time but you'll get there.

Thank you for the kind words!

Efficiency and more time required for imaging is indeed the central argument against using narrowband filters with DSLRs and other OSC cameras.

However, when you already have your gear, and you only need to buy filters to be able to use your mount, cameras, lenses and telescopes during clear moonlit nights (and perhaps more than double available clear nights), I find it to be very efficient to be able to do so, even though I may have to more than double the integrated time. After all, Ha and OIII need separate captures. But the result will be different, maybe better, even, and I found it fun and very rewarding.

But this it absolutely not (!) a thing for absolute beginners who are struggling to get ‘solid‘ (visible) images from their DSLRs without narrowband in nights without moonlight! (perhaps with the exception when they live in 'high-LP' 'red/white zones') What a frustrating experience that could be!

 

[thinker]

Great result! Well done for giving it a try!

Thank you! The potential gain of being able to image in moonlight was so much larger than the cost of a potential failure and selling the filters with a small loss. I got carried away when I began with AP 15 months ago, but frostbite in December ‘knocked me out’ for 10+ weeks, and the feet are still suffering and appear to be very vulnerable. Which kills my appetite for driving to other (and darker) places despite the local variations in weather and cloud covers (possibly because I’m surrounded by four mountains). The safety of my home to protect (and keep!) my feet is more important. Besides, I’ve got two of these Olympus cameras and the EQ6-R can easily carry two or more 3” refractors, so I can do a ‘Rey66 light’ OSN/one shot narrowband with just two cameras in the beginning. A potential massive gain with a ‘modest’ outlay and risk. And the filters could be used with a mono cam in the future, but I seriously doubt it.

Oh, well, at 60°N the summers have four months without astronomical nights, another reason to try to use clear nights with moonlight whenever possible.

 

[OutsideTheMatrix]

Nice image! What size filters were you using that caused the vignetting? Also how is your light pollution out there?

Thank you!

No, I didn’t use filters that caused the vignetting. I wrote ‘TS 0.79 + Metabones 0.71 causing severe vignetting’, which means that I used a 0.79x reducer/flattener and then added a 0.71x Metabones Speedbooster which was really pushing the system. The telescope’s matching reducer/ flattener is only 0.85x, so I was really exceedingly pushing it, and I have absolutely no reason to complain! But I needed to crop more than I (unrealistically) was hoping for. Stacking two images that didn’t overlap too well (from different nights) didn’t help either.

Bortle class 4 is what lightpollutionmap.info suggests around my house. I live on a different continent than you do.

Thank you! I like that we have another astrophotographer who uses a mirrorless 4/3 camera! Can you find the QE and stats on my camera, the Olympus E-PL6? I could not find it on that site and wanted to see how it compares to yours. It is similar to the E-PL5.

 

[thinker]

Nice image! What size filters were you using that caused the vignetting? Also how is your light pollution out there?

Thank you!

No, I didn’t use filters that caused the vignetting. I wrote ‘TS 0.79 + Metabones 0.71 causing severe vignetting’, which means that I used a 0.79x reducer/flattener and then added a 0.71x Metabones Speedbooster which was really pushing the system. The telescope’s matching reducer/ flattener is only 0.85x, so I was really exceedingly pushing it, and I have absolutely no reason to complain! But I needed to crop more than I (unrealistically) was hoping for. Stacking two images that didn’t overlap too well (from different nights) didn’t help either.

Bortle class 4 is what lightpollutionmap.info suggests around my house. I live on a different continent than you do.

Thank you! I like that we have another astrophotographer who uses a mirrorless 4/3 camera! Can you find the QE and stats on my camera, the Olympus E-PL6? I could not find it on that site and wanted to see how it compares to yours. It is similar to the E-PL5.

The Olympus E-PL6 doesn’t seem to be there, but E-PL5 and E-PL7 are both there, and they are both very similar (both QE of 59%), so I don’t think it is unreasonable to assume that the E-PL6 is in the area of 59% (!). But you should note the ‘notes’. Those numbers are not necessarily very accurate ‘hard facts’, but good (?) indications (from my understanding).

However, Read Noise is also a factor here, and there are other factors. The E-PL6 is not in the Read Noise table either, but the E-PL7 is, and if you also look at other Olympus 16MP cameras (except the early original E-M1 Mark I) you will find that they are very similar, and you could say practically identical.

 

[OutsideTheMatrix]

Nice image! What size filters were you using that caused the vignetting? Also how is your light pollution out there?

Thank you!

No, I didn’t use filters that caused the vignetting. I wrote ‘TS 0.79 + Metabones 0.71 causing severe vignetting’, which means that I used a 0.79x reducer/flattener and then added a 0.71x Metabones Speedbooster which was really pushing the system. The telescope’s matching reducer/ flattener is only 0.85x, so I was really exceedingly pushing it, and I have absolutely no reason to complain! But I needed to crop more than I (unrealistically) was hoping for. Stacking two images that didn’t overlap too well (from different nights) didn’t help either.

Bortle class 4 is what lightpollutionmap.info suggests around my house. I live on a different continent than you do.

Thank you! I like that we have another astrophotographer who uses a mirrorless 4/3 camera! Can you find the QE and stats on my camera, the Olympus E-PL6? I could not find it on that site and wanted to see how it compares to yours. It is similar to the E-PL5.

The Olympus E-PL6 doesn’t seem to be there, but E-PL5 and E-PL7 are both there, and they are both very similar (both QE of 59%), so I don’t think it is unreasonable to assume that the E-PL6 is in the area of 59% (!). But you should note the ‘notes’. Those numbers are not necessarily very accurate ‘hard facts’, but good (?) indications (from my understanding).

However, Read Noise is also a factor here, and there are other factors. The E-PL6 is not in the Read Noise table either, but the E-PL7 is, and if you also look at other Olympus 16MP cameras (except the early original E-M1 Mark I) you will find that they are very similar, and you could say practically identical.

They all seem to be very similar but your camera has taken a big jump in QE! I wonder how that will be seen in pictures?

 

[thinker]

[...]

The Olympus E-PL6 doesn’t seem to be there, but E-PL5 and E-PL7 are both there, and they are both very similar (both QE of 59%), so I don’t think it is unreasonable to assume that the E-PL6 is in the area of 59% (!). But you should note the ‘notes’. Those numbers are not necessarily very accurate ‘hard facts’, but good (?) indications (from my understanding).

However, Read Noise is also a factor here, and there are other factors. The E-PL6 is not in the Read Noise table either, but the E-PL7 is, and if you also look at other Olympus 16MP cameras (except the early original E-M1 Mark I) you will find that they are very similar, and you could say practically identical.

They all seem to be very similar but your camera has taken a big jump in QE! I wonder how that will be seen in pictures?

Yes, a big jump in QE, but even bigger dive in read noise at ISO 6400 which is now my favourite ISO! Wei-Hao Wang just posted his thoughts at the Cloudy place about Canon 5D2 versus Nikon 800 which has a quite similar gap when adding up QE and read noise. He wrote ‘The difference in image cleanliness is huge. Can I use 5D2 to create good images? Of course I can, but probably with 1.5x to 2x more integration comparing to using cameras with Sony sensors’. I don’t know what ‘image cleanliness’ means (see my OP about my ‘trained eye’), but there is certainly a difference in imaging time, both related to sub lengths and integrated time.

But there is another thing. I have bought/added Nikon D5300 for several reasons (price of body compared to a third E-M1.2 even used, native mount to gain access to e.g. Sigma Art 35/1.4 without adding expensive Speedboosters to avoid even narrower FOV, camera to give autofocus to three Sigma zoom lenses that I have only used as adapted manual focus lenses, and curiosity about how my Olympus compares to a ‘reference’ entry level AP camera). I imaged the Horsehead Nebula (early spring) with the Nikon and Olympus side-by-side and processed the D5300 raw files in Rawtherapee with exactly the same settings as with the E-M1.2 (including Rawtherapee’s ‘Daylight’ setting), and the result was what I’ve noticed so far: The nebula was more ‘pure’ red with the Olympus. Which may be related to the spectral response of the low-pass filter: Wei-Hao Wang said ‘Here is something’ interesting for everyone' and explained (and linked a chart to show) that Fuji has unusually high transmission of Ha, and ‘From this point of view, it is probably OK not to modify a Fuji camera’. So I wonder if this may be the case for Olympus as well. But we will probably never know since Olympus is not exactly ‘mainstream’ in DSO AP.

 

[OutsideTheMatrix]

[...]

The Olympus E-PL6 doesn’t seem to be there, but E-PL5 and E-PL7 are both there, and they are both very similar (both QE of 59%), so I don’t think it is unreasonable to assume that the E-PL6 is in the area of 59% (!). But you should note the ‘notes’. Those numbers are not necessarily very accurate ‘hard facts’, but good (?) indications (from my understanding).

However, Read Noise is also a factor here, and there are other factors. The E-PL6 is not in the Read Noise table either, but the E-PL7 is, and if you also look at other Olympus 16MP cameras (except the early original E-M1 Mark I) you will find that they are very similar, and you could say practically identical.

They all seem to be very similar but your camera has taken a big jump in QE! I wonder how that will be seen in pictures?

Yes, a big jump in QE, but even bigger dive in read noise at ISO 6400 which is now my favourite ISO! Wei-Hao Wang just posted his thoughts at the Cloudy place about Canon 5D2 versus Nikon 800 which has a quite similar gap when adding up QE and read noise. He wrote ‘The difference in image cleanliness is huge. Can I use 5D2 to create good images? Of course I can, but probably with 1.5x to 2x more integration comparing to using cameras with Sony sensors’. I don’t know what ‘image cleanliness’ means (see my OP about my ‘trained eye’), but there is certainly a difference in imaging time, both related to sub lengths and integrated time.

But there is another thing. I have bought/added Nikon D5300 for several reasons (price of body compared to a third E-M1.2 even used, native mount to gain access to e.g. Sigma Art 35/1.4 without adding expensive Speedboosters to avoid even narrower FOV, camera to give autofocus to three Sigma zoom lenses that I have only used as adapted manual focus lenses, and curiosity about how my Olympus compares to a ‘reference’ entry level AP camera). I imaged the Horsehead Nebula (early spring) with the Nikon and Olympus side-by-side and processed the D5300 raw files in Rawtherapee with exactly the same settings as with the E-M1.2 (including Rawtherapee’s ‘Daylight’ setting), and the result was what I’ve noticed so far: The nebula was more ‘pure’ red with the Olympus. Which may be related to the spectral response of the low-pass filter: Wei-Hao Wang said ‘Here is something’ interesting for everyone' and explained (and linked a chart to show) that Fuji has unusually high transmission of Ha, and ‘From this point of view, it is probably OK not to modify a Fuji camera’. So I wonder if this may be the case for Olympus as well. But we will probably never know since Olympus is not exactly ‘mainstream’ in DSO AP.

That reminds me! In your camera, does ISO 6400 show up in a different color? In previous cameras, anything above ISO 5000 was boosted- that is, artificially attained. Looks like the site you linked to allows for that, as for the E-PL7 anything above ISO 5000 has open data points, and for your camera it's anything above ISO 12,800.

and you are correct for Ha transmission rate for both Fuji and Olympus cameras. I saw an spectrum response curve for the E-PL5 on Cloudy Nights and the only ILC that was better at transmitting Ha was the Fuji. The E-PL5 transmitted about 35% of Ha and the Fuji camera they tested transmitted 42%. All the other ILCs tested were under 30%.

I have a Fuji camera also and that is so good with IR that I can easily do handheld IR at ISO 100 even at f/5.6

--
In the councils of government, we must guard against the acquisition of unwarranted influence, whether sought or unsought, by the military industrial complex. The potential for the disastrous rise of misplaced power exists and will persist. We must never let the weight of this combination endanger our liberties or democratic processes. We should take nothing for granted. Only an alert and knowledgeable citizenry can compel the proper meshing of the huge industrial and military machinery of defense with our peaceful methods and goals, so that security and liberty may prosper together.
-Dwight D. Eisenhower, 1961

 

[thinker]

[...]

Olympus. Which may be related to the spectral response of the low-pass filter: Wei-Hao Wang said ‘Here is something’ interesting for everyone' and explained (and linked a chart to show) that Fuji has unusually high transmission of Ha, and ‘From this point of view, it is probably OK not to modify a Fuji camera’. So I wonder if this may be the case for Olympus as well. But we will probably never know since Olympus is not exactly ‘mainstream’ in DSO AP.

That reminds me! In your camera, does ISO 6400 show up in a different color? In previous cameras, anything above ISO 5000 was boosted- that is, artificially attained. Looks like the site you linked to allows for that, as for the E-PL7 anything above ISO 5000 has open data points, and for your camera it's anything above ISO 12,800.

At ISO 8000 a text pops up saying ‘Extension’ (but no different colour that I can notice), and the read noise curve becomes completely flat at 1.6 electrons after ISO 6400. Which (combined) suggests that above ISO 6400 there is only synthetic/imitation gain (multiplication), no improved precision (smaller differences) in electron count that could/would have produced finer levels of brightness/shades. The open data points starting above ISO 12800 is misleading and should have started at ISO 8000, I think (I really don’t know, but since there is no improvement in read noise, I wouldn’t go above ISO 6400 for this camera in any case).

and you are correct for Ha transmission rate for both Fuji and Olympus cameras. I saw an spectrum response curve for the E-PL5 on Cloudy Nights and the only ILC that was better at transmitting Ha was the Fuji. The E-PL5 transmitted about 35% of Ha and the Fuji camera they tested transmitted 42%. All the other ILCs tested were under 30%.

Thank you for the confirmation! It’s good with feedbacks that can serve as ‘reality check’. After all, that was the main purpose for this thread (i.e. ‘reality check’).

 

[OutsideTheMatrix]

[...]

Olympus. Which may be related to the spectral response of the low-pass filter: Wei-Hao Wang said ‘Here is something’ interesting for everyone' and explained (and linked a chart to show) that Fuji has unusually high transmission of Ha, and ‘From this point of view, it is probably OK not to modify a Fuji camera’. So I wonder if this may be the case for Olympus as well. But we will probably never know since Olympus is not exactly ‘mainstream’ in DSO AP.

That reminds me! In your camera, does ISO 6400 show up in a different color? In previous cameras, anything above ISO 5000 was boosted- that is, artificially attained. Looks like the site you linked to allows for that, as for the E-PL7 anything above ISO 5000 has open data points, and for your camera it's anything above ISO 12,800.

At ISO 8000 a text pops up saying ‘Extension’ (but no different colour that I can notice), and the read noise curve becomes completely flat at 1.6 electrons after ISO 6400. Which (combined) suggests that above ISO 6400 there is only synthetic/imitation gain (multiplication), no improved precision (smaller differences) in electron count that could/would have produced finer levels of brightness/shades. The open data points starting above ISO 12800 is misleading and should have started at ISO 8000, I think (I really don’t know, but since there is no improvement in read noise, I wouldn’t go above ISO 6400 for this camera in any case).

and you are correct for Ha transmission rate for both Fuji and Olympus cameras. I saw an spectrum response curve for the E-PL5 on Cloudy Nights and the only ILC that was better at transmitting Ha was the Fuji. The E-PL5 transmitted about 35% of Ha and the Fuji camera they tested transmitted 42%. All the other ILCs tested were under 30%.

Thank you for the confirmation! It’s good with feedbacks that can serve as ‘reality check’. After all, that was the main purpose for this thread (i.e. ‘reality check’).

I'm going to try and link you to the transmission curve because it's very useful. If I can't find it, the thread is titled something like "electronic camera and M43 camera captures of deep space objects." It was originally about Panasonic cameras but users later posted photos taken with M43 cameras also.

One thing I am wondering about with read noise is that could noise reduction also be factored in there, or is that done with NR off? Also, with higher ISO it is important to preserve dynamic range, so it would be interesting if we could find DR stops at each ISO data point; I know Sensorgen had this info, but that site went defunct. Based on E-PL5 info that was listed on it I remember that I found ISO 800 and 3200 to be the ones that retained the best combo of read noise and dynamic range. This was borne out to me in my captures, as anything above ISO 3200 made my stars appear white (blown highlights) no matter the shutter speed (because the DR was too narrow.)

edit- graph is at the bottom of this page, click to enlarge!

https://www.cloudynights.com/topic/451681-a-night-with-the-micro-four-thirds-cameras/page-3

It's also here:

https://www.cloudynights.com/topic/447365-modified-olympus-e-pm2/

Some DSO pics here:

https://www.cloudynights.com/topic/400283-panasonic-gh3-for-near-real-time-ap/#entry5127685

 

[thinker]

[...]

One thing I am wondering about with read noise is that could noise reduction also be factored in there, or is that done with NR off?

Read noise is a sensor property, so NR definitely off.

Also, with higher ISO it is important to preserve dynamic range, so it would be interesting if we could find DR stops at each ISO data point; I know Sensorgen had this info, but that site went defunct. Based on E-PL5 info that was listed on it I remember that I found ISO 800 and 3200 to be the ones that retained the best combo of read noise and dynamic range. This was borne out to me in my captures, as anything above ISO 3200 made my stars appear white (blown highlights) no matter the shutter speed (because the DR was too narrow.)

While I’m still a beginner (but perhaps not ‘complete’ beginner), my approach (for now) is that blown-out stars is a ‘feature’. There are compromises and priorities. I want more nuances in nebulae (in my brain that gives more 'depth') and think about adding stars as a separate operation when/if I learn that. I’ve seen some nice images with RGB stars done in connection with narrowband, so it appears to be possible (!). So for me, DR is a non-issue. I want to compromise on efficiency (again) and prepare for a little extra work. But ask me later..

edit- graph is at the bottom of this page, click to enlarge!

https://www.cloudynights.com/topic/451681-a-night-with-the-micro-four-thirds-cameras/page-3

It's also here:

https://www.cloudynights.com/topic/447365-modified-olympus-e-pm2/

Some DSO pics here:

https://www.cloudynights.com/topic/400283-panasonic-gh3-for-near-real-time-ap/#entry5127685

Thank you, again! I’ve only more-or-less just ‘scrolled through’ the pages so far, but you helped me to a ‘moment of insight’! I was considering buying SII filter for mapped-colour nebulae (its red colour should to well in moonlight), but my not-that-passionate relationship to PP has stopped me. Perhaps a quarter (¼) of that of the Ha-transmission in addition to usually being a much weaker signal than Ha, how miserable that would have been. Now that’s something I will not do without a modded cam! Maybe Nikon D5300 for a DIY mod in the future. But I’m not tampering with my Olympuses.

I also stumbled over ‘It is pointless to use an Ha filter with an unmodded cam’, but I already knew that..

 

[thinker]

With Stellarium I just noticed that the moon illuminations were 72% (not 64%) and 80%. After processing the stacks for each day, there appeared to be a minor difference in contrast, but the OIII ooc jpg files from 80% illumination were surprisingly brighter than the 72% day. I think it does well for the price, and I’m still not considering 3nm Astrodon.

Just in case someone else should be tempted to try this 4.5nm OIII filter.

--
(Harvey) - Jane, I've been thinking...
(Jane) - Oh, do you want an aspirin?

 

[OutsideTheMatrix]

[...]

One thing I am wondering about with read noise is that could noise reduction also be factored in there, or is that done with NR off?

Read noise is a sensor property, so NR definitely off.

Also, with higher ISO it is important to preserve dynamic range, so it would be interesting if we could find DR stops at each ISO data point; I know Sensorgen had this info, but that site went defunct. Based on E-PL5 info that was listed on it I remember that I found ISO 800 and 3200 to be the ones that retained the best combo of read noise and dynamic range. This was borne out to me in my captures, as anything above ISO 3200 made my stars appear white (blown highlights) no matter the shutter speed (because the DR was too narrow.)

While I’m still a beginner (but perhaps not ‘complete’ beginner), my approach (for now) is that blown-out stars is a ‘feature’. There are compromises and priorities. I want more nuances in nebulae (in my brain that gives more 'depth') and think about adding stars as a separate operation when/if I learn that. I’ve seen some nice images with RGB stars done in connection with narrowband, so it appears to be possible (!). So for me, DR is a non-issue. I want to compromise on efficiency (again) and prepare for a little extra work. But ask me later..

edit- graph is at the bottom of this page, click to enlarge!

https://www.cloudynights.com/topic/451681-a-night-with-the-micro-four-thirds-cameras/page-3

It's also here:

https://www.cloudynights.com/topic/447365-modified-olympus-e-pm2/

Some DSO pics here:

https://www.cloudynights.com/topic/400283-panasonic-gh3-for-near-real-time-ap/#entry5127685

Thank you, again! I’ve only more-or-less just ‘scrolled through’ the pages so far, but you helped me to a ‘moment of insight’! I was considering buying SII filter for mapped-colour nebulae (its red colour should to well in moonlight), but my not-that-passionate relationship to PP has stopped me. Perhaps a quarter (¼) of that of the Ha-transmission in addition to usually being a much weaker signal than Ha, how miserable that would have been. Now that’s something I will not do without a modded cam! Maybe Nikon D5300 for a DIY mod in the future. But I’m not tampering with my Olympuses.

I also stumbled over ‘It is pointless to use an Ha filter with an unmodded cam’, but I already knew that..

That's a good point. I too am looking around for a second hand camera to modify! About the star colors, you can always add that in later when you layer images!