Orion Nebula with modified Sony A7S

[sharkmelley]

I took this last month using an H-alpha modified Sony A7S camera on a Takahashi Epsilon 180ED scope (500mm focal length at F/2.8).

3 hours of data in 30 second subs at ISO 2000 (yes, 360 exposures!) plus 10 min at ISO 640 and 10 min at ISO 200 for the bright core.

Traditional processing technique using darks, flats and bias, mainly in PixInsight.

Mark

 

[rnclark]

I took this last month using an H-alpha modified Sony A7S camera on a Takahashi Epsilon 180ED scope (500mm focal length at F/2.8).

3 hours of data in 30 second subs at ISO 2000 (yes, 360 exposures!) plus 10 min at ISO 640 and 10 min at ISO 200 for the bright core.

Traditional processing technique using darks, flats and bias, mainly in PixInsight.



Mark

Mark,

That is really beautiful. Well done. I do note there is quite a green cast at the low end.

Does the Sony record linear data in the raw file? I thought I read it is a compressed tone map. Did you record uncompressed or compressed raw?

Roger

 

[Jon Rista]

Now that is impressive! Well done! That is really a lot of signal there, too...500mm f/2.8 and 3 hours of data? That's like me getting about six hours of data with my 600mm f/4 (assuming similar skies). What a scope!

The A7s is seriously more sensitive to Ha than my 5D III, no question. You picked up a couple of nice Ha features below Orion Nebula that usually don't show up well unless you use narrow band imaging.

 

[sharkmelley]

Mark,

That is really beautiful. Well done. I do note there is quite a green cast at the low end.

Does the Sony record linear data in the raw file? I thought I read it is a compressed tone map. Did you record uncompressed or compressed raw?

Roger

Thanks!

I think you're right - there is still a slight residual green cast.

The Sony A7S always records compressed data, there is no choice. But the data coming from a raw converter like DCRaw or Adobe's DNG is definitely linear. However if you examine a histogram of the linear data there is a very obvious combing effect. For low data values, the histogram contains only even values and then the histogram steps become gradually bigger: 4, 8, 16 and finally reaching a step size of 32 at the high end. This is a direct result of the tone map encoding.

This encoding sounds quite destructive but I have yet to see any obvious artifacts caused by it.

Mark

 

[Michael S.]

Good Evening Mark!

Awesome widefield...maybe a bit to hard pushed/stretched to get out all those dust lanes for the high iso used, but in total a fine work.

thumbs up,

 

[BlackgumNate]

Wow, a work of art!

Nate

 

[Sir Canon]

WONDERFUL! i cannot believe that this is just 30 second exposures! truly impressed!

say, didnt i do a tiny bit of finishing post processing on this image a little while back?

 

[Tristimulus]

This is a good one, and a very interesting one too. Amazing what can be done with the digital tools at hand.

 

[sharkmelley]

WONDERFUL! i cannot believe that this is just 30 second exposures! truly impressed!

say, didnt i do a tiny bit of finishing post processing on this image a little while back?

--
Campaign to revamp Dpr's image compression!
I tend to overdo things

Yes you're right - your bit of editing encouraged me to go back and have a another go at processing it.

Thanks!

Mark

 

[sharkmelley]

Now that is impressive! Well done! That is really a lot of signal there, too...500mm f/2.8 and 3 hours of data? That's like me getting about six hours of data with my 600mm f/4 (assuming similar skies). What a scope!

The A7s is seriously more sensitive to Ha than my 5D III, no question. You picked up a couple of nice Ha features below Orion Nebula that usually don't show up well unless you use narrow band imaging.

Thanks Jon!

 

[Jon Rista]

Welcome. This is one of the most photographed objects in the sky, and phenomenal images of it don't come along all that often. I think I can count on two hands the ones I think rank at the top, and this is definitely one of them. Despite the bits of green noise and such.

I am quite impressed with your control of the core here. In my version, which sadly was only 2h20m with an unmodded 5D III, the core is still very bright, despite extensive HDR processing. I should probably give mine a go again, as I've learned a lot more about PixInsight than I did seven months ago the last time I reprocessed it. Still, I think you nailed the Ha, not overdone and overly red, not underdone and more brown. It's just right.

 

[Sir Canon]

youre welcome it looks great

 

[Rutgerbus]

Very impressive image....why using 30sec exposure and ISO 2000 and not for instance ISO1600 and 1min exposure or 45sec...??? Normally one shoots in the range of 400-800-1600-3200 etc...not the inbetween ISO's due to better dynamic range and S/N ratio (Linear exposures) I believe (correct me if I'm wrong)

 

[RustierOne]

Very impressive image....why using 30sec exposure and ISO 2000 and not for instance ISO1600 and 1min exposure or 45sec...??? Normally one shoots in the range of 400-800-1600-3200 etc...not the inbetween ISO's due to better dynamic range and S/N ratio (Linear exposures) I believe (correct me if I'm wrong)

I believe the 30-second non bulb mode exposures are necessary to avoid Sony's "star-eater" problem. But even with that 30-second limit, he has produced an excellent image.

 

[RustierOne]

I took this last month using an H-alpha modified Sony A7S camera on a Takahashi Epsilon 180ED scope (500mm focal length at F/2.8).

3 hours of data in 30 second subs at ISO 2000 (yes, 360 exposures!) plus 10 min at ISO 640 and 10 min at ISO 200 for the bright core.

Traditional processing technique using darks, flats and bias, mainly in PixInsight.

View attachment 1213022

Mark

Mark that is truly an amazing image! When viewed original size, I take my glasses off (I'm quite myopic) and closely examine the image. I could get lost in the view. That Takahashi reflector is an awesome optical device.

So with the 30-second limit to avoid the "star-eater" issue are you losing much performance? I would expect it takes a lot of computer horsepower to process 360 images, even at 12 MP.

You must have been willing to live with the shortcomings of that camera to invest in having it modified. What a combo - Sony & Takahashi!

--
Best Regards,
Russ

 

[dethis2]

Very impressive image....why using 30sec exposure and ISO 2000 and not for instance ISO1600 and 1min exposure or 45sec...??? Normally one shoots in the range of 400-800-1600-3200 etc...not the inbetween ISO's due to better dynamic range and S/N ratio (Linear exposures) I believe (correct me if I'm wrong)

I believe the 30-second non bulb mode exposures are necessary to avoid Sony's "star-eater" problem. But even with that 30-second limit, he has produced an excellent image.

 

[Jon Rista]

Very impressive image....why using 30sec exposure and ISO 2000 and not for instance ISO1600 and 1min exposure or 45sec...??? Normally one shoots in the range of 400-800-1600-3200 etc...not the inbetween ISO's due to better dynamic range and S/N ratio (Linear exposures) I believe (correct me if I'm wrong)

 

[sharkmelley]

Very impressive image....why using 30sec exposure and ISO 2000 and not for instance ISO1600 and 1min exposure or 45sec...??? Normally one shoots in the range of 400-800-1600-3200 etc...not the inbetween ISO's due to better dynamic range and S/N ratio (Linear exposures) I believe (correct me if I'm wrong)

--
www.rutgerbus.nl
Photographic Moments
A photon only stops "existing" when it is captured by your sensor.

The Sony A7s uses a dual gain readout system. Up through ISO 1600 it uses a low gain mode. From ISO 2000 and up it uses a high gain mode. The high gain mode actually improves dynamic range at ISO 2000 over ISO 1600, so you actually get better results using that on the A7s.

The 30 second exposures avoids the "star eater" issue that results from Sony's forced use of certain kinds of noise reduction in Bulb mode (required for exposures longer than 30s). This NR tends to nuke the bright center pixel(s) of stars, as it mistakenly thinks they are hot pixels.

Exactly. You saved me the explanation

Mark

 

[sharkmelley]

Mark that is truly an amazing image! When viewed original size, I take my glasses off (I'm quite myopic) and closely examine the image. I could get lost in the view. That Takahashi reflector is an awesome optical device.

So with the 30-second limit to avoid the "star-eater" issue are you losing much performance? I would expect it takes a lot of computer horsepower to process 360 images, even at 12 MP.

You must have been willing to live with the shortcomings of that camera to invest in having it modified. What a combo - Sony & Takahashi!

 

[RickColman]

The camera is self modified. It was a bit scary ripping apart a brand new camera!

Mark

Hmmm. An interesting idea ...

*** would you be willing to share the details of how yoy modified the camera? ***

If it is not too scary, I might take a whack at it.

-- Rick