Embracing the New(er) EEVA Technology?

[Macavity]

After (quite a few!) years doing "Video Astronomy" -- Using my trusty Watec 120N+ 
(later Watec 910HX) 8" f/4 Newt, I am / was fairly happy with 8-bit hardware stacking
+ offline (stacking) processing.  But still hoping to improve previous DSO images? ?




I prefer to image a WIDE variety of DSO rather than emulate the Classical image(r)s. ?
(Ultimately I found myself inspired by "going a bit deeper" e.g. HG & ARP Galaxies):



BUT I SENSE I could now  *make life a bit easier* by moving to 16-bit Cameras
+ LIVE Stacking etc. At this stage of life, I might struggle to "invest" £1000+ in a
Camera... But think I might just stretch to a Lodestar X2 Mono (if still available)
AFAICS it uses the same CHIP as a Watec 910HX... AND 16bits > 8bits, right? ? 

I can anyway practice LIVE stacking with Watec Cams... ZWO ASI120MM/MCs? 
(I guess I still HOPE there will be some place for my "older technology" too!) ?

Edited April 2, 2019 by Macavity

[elpajare]

I agree with you on the type of objects to observe / photograph. This was one of the reasons to introduce me to the EEVA. I have self-limited, for the time being, objects of magnitude 12 or less.

If you prefer to add some color to your observations there are a few current chips that give very good results with short exposures and stacking.

[Macavity]

Colour certainly livens things up! And (with extra effort) I can do this via a
manual alignment of Narrow Band RGB on compact Planetary Nebulae:



N.B. I can NOW rid myself of (tri-coloured!) hot pixels by using Camera hardware. ?
And of course, get into such things as dark field subtraction... Flat field processing.

I suppose the only less inspiring experience is with diffuse nebulae. But I can still
image M42... And even use software masking for some "extended contrast" stuff? ?



I guess I will TRY pitting my Watec 910HX (using internal hardware stacking +
offline stacking / processing) versus my ASI120MM using (SharpCap) *LIVE* 
Software. I sense many modern cameras have similar Q.E's to Watec Cams! ?

16 versus 8 bits? But nothing is totally cut and dried until tried in practice?
I sense I will limit myself to non-cooled Cams. I like the simplicity of EEVA.
(Though I have no qualms re. using processing. "Whatever works" for me). ?

Edited April 2, 2019 by Macavity

[London_David]

I personally would not get a Lodestar now. While you can still get great observations out of it -- it's quite old technology -- so in terms of what you pay for what you get it's not good bang for buck.

I would look at the ZWO ASI cameras - I have been very pleased with mine. I've ordered directly from them and they've been helpful and responsive on the forums too. You can also buy from local vendors too if you want to have extra security for faster returns.

I have the ASI290 mono and the ASI294Pro colour. Rule of thumb on specs for these CMOS cameras is that you want low read noise and high QE. Pixel size can be small but you can bin at effectively no loss. There are a few options depending on sensor size which will determine your field of view (you should do a check with your equipment and proposed camera to see FOV on astronomy.tools). If your considering the Lodestar mono, I'd strongly recommend the ASI290 (or other IMX290 chip camera). For my money it's still much the best EAA mono sensor around in the amateur price range. The USB2 mini is about £300 and the USB3 version £350). There are other options but I've not tried them and tend to be more favorable for AP.

https://www.firstlightoptics.com/zwo-cameras/zwo-asi290mm-usb-3-mono-camera.html

[Martin Meredith]

If my Lodestar failed I would definitely consider getting another -- probably immediately! Or maybe an Ultrastar mono.

I'm not convinced that the Lodestar has been superseded for EAA use, at least for the kind of objects I like to observe. I presented some data on theoretical SNRs achieved after a fixed exposure (either with 30s subs or 10s subs) at the link below and the Lodestar wins out due to a combination of larger pixels and high QE. I hear what you say David re binning which can of course make up effective 'super pixels' to catch up with the Lodestar's performance, but the Lodestar is still a very competitive option for those of us that that like simplicity (no gain no pain) and sensitivity. And SX service is excellent, I can attest personally, even from over here in Spain.

Chris, I don't see them coming up second hand very often (which tells its own story) but they are well worth snapping up when they do.

Anyway, the key thing is to be happy with one's kit! We have a lot of choices and the 290 looks like a good option in the sensible number of pixels stakes ?

s

[SIDO]

Been thinking about the asi290mm mini myself, Qe at 80 and the fov seems perfect for an st80 or ed80 but being a eeva/eaa/astrophotography novice I don't want to be over sampling or anything else unpreceived, if someone could give me a heads up that would be great. 

Edited April 3, 2019 by SIDO

[JamesF]

I had the impression that binning on CMOS sensors wasn't as effective as on CCD sensors because the pixels are combined after they have been read rather than before.  I've not been following the debate intently however, so I may have the wrong end of the stick.

James

[HiloDon]

38 minutes ago, JamesF said:

I had the impression that binning on CMOS sensors wasn't as effective as on CCD sensors because the pixels are combined after they have been read rather than before.  I've not been following the debate intently however, so I may have the wrong end of the stick.

James

You are correct, James.  There was a lot of heated discussion on this in another forum.  Here is a good description from a reliable source that explains the difference.  One thing to also note is that binning a CCD color cam will lose the color, whereas the CMOS color is retained albeit less effective.

https://www.atik-cameras.com/news/binnning-the-differences-between-cmos-and-ccd/

Don

[noah4x4]

It is also worth viewing the Atik Horizon video available on the Atik site explaining how it cleverly applies <gain>. It has a full well of 20,000e at around 4,800 ADU, but amplification boosts that to 65,536 producing a stunningly good signal to noise ratio.

Everybody tends to focus on comparing the capabilities of "sensor x versus sensor y".  Frankly, I don't claim to fully understand all the technicalities, but it is clear to me that any appraisal of CMOS versus CCD needs to look beyond mere sensor performance and consider the additional capabilities of the electronics. Never mind the theory, what are the practical results?  Atik claim performance of its CMOS  is matching CCD at lower price and my experience of the Horizon has been awesome, albeit I won't ever win Astrophotographer of the year as no camera can ever compensate for a clunky mount and limited user skills.

[Macavity]

Great Stuff! It went ominously QUIET for a while? lol. It's been a while since I did
any "Video Astronomy" and I did need to catch up on the latest ideas / trends! ?

As we Astronomers know, it is often possible (unwise?) to stretch a budget 2X! ?
So maybe an "Ultrastar" is not completely impossible... Thanks @Martin Meredith

I also read this: (See section on EEVA) 
https://agenaastro.com/zwo-astronomy-cameras-buyers-guide.html
I find myself drawn (whatever the price) to ASI183 type cameras...

Having enjoyed the luxury of Watec "half inch" chips (Or "slightly dodgy" polar
alignment!) I am reluctant to go to smaller chips. My 8" f/4 Newt is OK for many
Messier, Caldwell DSOs etc., But l still aspire to (larger chip) "better framing"? ?

And my smaller(!) ZWO Cams give me some possibilities to TRY OUT this stuff...

Edited April 4, 2019 by Macavity

[Martin Meredith]

1 hour ago, noah4x4 said:

It is also worth viewing the Atik Horizon video available on the Atik site explaining how it cleverly applies <gain>. It has a full well of 20,000e at around 4,800 ADU, but amplification boosts that to 65,536 producing a stunningly good signal to noise ratio.

Everybody tends to focus on comparing the capabilities of "sensor x versus sensor y".  Frankly, I don't claim to fully understand all the technicalities, but it is clear to me that any appraisal of CMOS versus CCD needs to look beyond mere sensor performance and consider the additional capabilities of the electronics. Never mind the theory, what are the practical results?  Atik claim performance of its CMOS  is matching CCD at lower price and my experience of the Horizon has been awesome, albeit I won't ever win Astrophotographer of the year as no camera can ever compensate for a clunky mount and limited user skills.

SNR is theoretically independent of gain

In practice, if read noise is different at different gains then that will affect SNR -- though not by very much in the scheme of things. But SNR itself is independent of gain.

Some of the strategies employed on CMOS are ways to get around limitations of low full well capacities and should not be seen as boosting CMOS performance above the level of other types of sensors that don't have these limitations to start with. Of course, CCDs have their own disadvantages, but I think it is important not to oversell what gain actually is, which is just a simple multiplier dressed up to look like something else.

To give an audio analogy, gain is recording level. If you use a cheap sound card (say 8-bits per sample), you need to worry about setting the recording level high enough to capture faint sounds (plus noise!) but low enough to avoid clipping. If on the other hand you use a 24-bit sound card you can probably leave recording level fixed and still capture faint sounds and loud sounds without clipping. 

Martin

Edited April 4, 2019 by Martin Meredith

[SIDO]

Vacuum tube eeva?

[London_David]

From my reading the Atik post doesn’t explain that the comparison to make is signal to noise ratio per micron, rather than just per pixel, since CCD pixels tend to be bigger. While read noise increases with signal because it is done after, it does so significantly slower than the signal increases. This is particularly relevant on CMOS since read noise gets down to very low numbers compared to CCD.

This SharpCap post has a longer explanation with the maths on it (the talk I posted in this forum on goes into a lot of detail on imaging including this which is interesting and has less maths...)

https://forums.sharpcap.co.uk/viewtopic.php?t=262

Atik until recently only made CCD cameras and would have an interest in putting CCD technology in the best light. What’s in the post is not wrong, as I understand it, it leaves out part of the story. Robin Glover  who developed SharpCap is neutral to CCD or CMOS technology since the software processes both sensor types and was a non profit hobby project until recently.

Specs don’t tell the whole story of course but in QE the imx290 is rated at greater than 80% QE and the Lodestar at 77% so very similar. But the pixel size is 8.2 on the Lodestar microns compared to 2.9 microns. So you can bin (2x2, or 3x3) and get more signal over the same size of sensor area with the imx290 chip than the Lodestar. Which, given everything else being equal, makes it more likely you will pick up fainter detail, and the option to sample at a higher rate.

All the same, my original point is that for a newcomer something like the asi290 mini is £150 cheaper than the Lodestar. So even if you think both cameras can do roughly the same thing, the asi290 is probably a better bang for your buck.

I’ve been happy with mine and I’ve found it very easy to use with SharpCap which sets everything automatically depending on your light pollution and individual sensor readings.

The Lodestar is a good camera and will give you good results (reading in the other forum you can see excellent interesting observation reports using it). The SLL software is clearly excellent (and one of the very few options if you want to use a Mac), however there are newer cheaper options that are at least as good.

The sad truth is sensor capabilities are probably going to be a small difference relative to seeing and light pollution which will be a more significant restriction on what you can see. I think any camera talked about here can give you good observing results (and you can see those results in the forums). As Martin says the important thing is to enjoy using the kit. 

@SIDO If you check out the astronomy tools site it’ll give you a sense of over and undersanpling with your equipment and seeing but I would not worry too much about oversampling and be careful about undersampling. As mentioned you can always bin if you are oversampled. 

 

 

 

Edited April 4, 2019 by London_David

[Macavity]

On 04/04/2019 at 14:33, London_David said:

The sad truth is sensor capabilities are probably going to be a small difference relative to seeing and light pollution which will be a more significant restriction on what you can see. I think any camera talked about here can give you good observing results (and you can see those results in the forums). As Martin says the important thing is to enjoy using the kit. [My Emphasis]

 

True (all) Dat! ? For the moment I'm leaving my Camera choice up "in the air"!
Indeed some excellent input (grist for the mill!) here and elsewhere though. ?

I am now tempted by a Mono Ultrastar, but worry about the "Black Streak"? ?

After a long layoff (mostly solar imaging) I forget stuff... Notably the virtue of
smaller (1/2") chips: They are within "Coma limit" of my budget 8" f/4 Newt! ?

I may "struggle on" with my *eight bit* Watec 910HX for a while. I've no idea
how to factor in 8-bit hardware pre-stacking to the general equation of EEVA! 
It may not be ALL that bad? As indeed some of my past (Wat 120N+ images).

THIS may have been taken with an ST102... I processed the life out of it. (sic)!



For those who don't mind "post-processing", it may illustrate my problem?
ONE mere click within GIMP (Photoshop?) can make or break the image! ?
Notice, in the corners, I am fighting (what I call) the "Generic HEAT Haze"?!?

Who knows? Couple of years back I built a cooling case for the WAT910HX.
(Contemplated adding a small cooling FAN... A Peltier Chip chip even! lol



I sense there are two types of heat-effect? On-Chip heating AND also the
surrounding electronics... Ambient temperature? A lot of air cooled EEVA
Cams have small cases... get VERY hot... At least on Summer Nights now? ?

Aside: EEVA'ers can be forgiven if... "Their Brain Hurts"? [Monty Python!]? ?

</wibble>!

Edited April 6, 2019 by Macavity

[elpajare]

The lighter parts of the edges can be eliminated by applying Dark Fields. The software that I use ( and many others) captures and applies it automatically.
You can also choose a color chip and you will save triple ( or quadruple !) RGB-L exposure.
I have used the IMX224 which is very cheap with a Newton 8 "f4 and gives very good results in small and weak objects such as galaxies, globular clusters and planetary nebulae.

With this combination the exposures are 15 seconds for this type of objects with maximum gain.

I wish you a good choice

[Macavity]

Thank You! Ideed M31 & M33 etc. are not (exceptionally so) the "easiest" objects
at this image scale. But I have a few more scopes / lenses to try on larger and/or
low surface brightness objects. Not all of my "historical" data has dark frames, but
there are other "recovery" possibilities too. The Watec 910HX has quite a few hot
pixels (now)... I mostly used the darks to remove those. But I find they can also be
treated (better?) by the camera processor.... So still a lot of options available! ?

Lest we be downhearted I located one of the plots of M31 surface brightness...
I "know" I can just about get to Mag +20+ with my "very best" skies.
But these closer Galaxies can go on... and on... ?


 

Edited April 6, 2019 by Macavity

[NickK]

Even non-aligned 30 second exposures can give a decent result with a CDD; I don't bother with noise here (squiggly pixels):

[Martin Meredith]

Plenty of signal in that M51Nick. It is hard to get the faint outer plumes coming off the three horns. 

Just to take it down a notch, here's a stack of 15s exposures that I just dug up from 2015 and live-stacked (offline, as it were, but with nothing that couldn't be done live), completely uncalibrated (since I couldn't find any suitable frames). I've not managed to get as much of the faint stuff as you, but I think it shows that there's still life in what is a CCD-based guide camera with monster pixels in spite of relative high read noise compared to CMOS, and operating in alt-az to boot.

BTW I think 15s is too short on this object for CCDs, but I'd like to repeat the experiment with proper calibration to see how much of the plume is possible and to check out if it is possible to reduce what I suspect are in part read-noise related striations in the quadrant at around 4 o'clock re M51's core (although some of these I see in your image so presumably real).

 

 

Martin

Edited April 7, 2019 by Martin Meredith

[Mike JW]

Chaps - just gone through this post twice now and followed links. So helpful (and confusing at times due to my limited knowledge),  many thanks to you all.

Ultrastar Mono - would buy another one. I have tried the Lodestar x2, a couple of early GPCAMs and also a Watec 120.  I would go for the Ultrastar anytime, partly because I find the software so simple to use. I sort of copped with Sharpcap but  the Altair software defeated me - after all, I simply wish to stick a camera in the focuser and get views similar or deeper than a big dob and in a short time as possible, with minimal tweaking at the laptop.

Cheers,

Mike

[Martin Meredith]

Mike, I must admit the Ultrastar mono is tempting.

I just want to add that although the kind of sensor I use is remarkably small (in pixel count too) by modern standards (it has 50 times fewer pixels than a 20MP sensor), for most objects not only is it more than adequate (e.g. 6784 members of the 6940-strong NGC catalogue i.e. nearly 98% are under 20' in diameter and therefore fit comfortably in the FOV of this camera on a 800mm FL scope), but the kind of detail one can get with these huge pixels is consistently surprising to me. E.g. here's a recent shot of NGC 3294  (ignore the name):

 

 

This shows tons of detail in this many-armed galaxy. But this represents only a small part (maybe 1/30th) of the field of this tiny sensor (hence the name: NGC 3294 was an unexpected bonus). Here's the full image

 

Yes, there are objects where a 20MP sensor is going to be worthwhile, but they are relatively few in number and one must ask if it is worth the extra compute, memory and long-term storage costs most of the time, esp. for galaxy lovers. Sure, long-term storage is cheap, but compute and memory costs provide practical limits on what one is able to accomplish in terms of flexible processing in near real-time.  

Martin

 

Edited April 8, 2019 by Martin Meredith

[elpajare]

Large sensors and many megapixels are fashionable but I am also of the opinion that most objects can be seen in great detail with these small ICX 829 or IMX 224 sensors.

As an example, in the northern hemisphere there are 1.472 objects of magnitude equal to or less than 12. Objects considered "large and bright" ( more than 20 arc. min), appropriate for these mega-sensors are only 154 ( 10.6%). If we go to higher magnitudes ALL objects are small and weak in this way small and sensitive sensors well combined with a telescope is much more useful for fans who dont want to limit themselves to the "big and bright".
The most coveted objectives of astrophotographers and suitable for this type of mega sensors are the bright nebulae and these only represent 7% of all the observable objects in the northern hemisphere.

[SIDO]

12 minutes ago, elpajare said:

Large sensors and many megapixels are fashionable but I am also of the opinion that most objects can be seen in great detail with these small ICX 829 or IMX 224 sensors.

As an example, in the northern hemisphere there are 1.472 objects of magnitude equal to or less than 12. Objects considered "large and bright" ( more than 20 arc. min), appropriate for these mega-sensors are only 154 ( 10.6%). If we go to higher magnitudes ALL objects are small and weak in this way small and sensitive sensors well combined with a telescope is much more useful for fans who dont want to limit themselves to the "big and bright".
The most coveted objectives of astrophotographers and suitable for this type of mega sensors are the bright nebulae and these only represent 7% of all the observable objects in the northern hemisphere.

Just bought the asi385mc observe in the northern hemisphere and am liking your post alot, I went with a smaller 1/2 sensor for those same reasons you sumed up quite well.

For use on my 150/750 with reducer at f2.5 so 1.1 x .63° and my 80mm at about 2 x 1° f2.5

Thinking of buying a large sensor camera for my finderscope though.

 

[Macavity]

I suppose my "ideal" would have been a *16-bit* (hardware-stacking!) 1" Watec-thing,
pumping out higher ADC resolution (than 8-bit!) images on standard coax at 25 fps? ?

Not yet sure how I will approximate this. lol. EEVA has moved to (RT) software stacking.
I *imagine* a "fair frame rate" (different images) at ~25+ fps is going to retain options!
With an 8" f/4 Newt (wobbly HEQ5) a slightly larger than 1/2" chip would be handy. ?

There is so much FUN stuff you can do with less-obvious targets using EEVA-ish Cams:

Hubble Deep Field... Uranian Satellites... Meteors... (Mercury Transits... Lunar Eclipses...):

EEVA doesn't get me embroiled with "classic" auto-guiding, cooling... YET! [teasing] ?
I am moved to try out a (neglected!) standard DSLR / Lenses / Scopes... for Wide Fields.
A casual wide field image got me far more "likes" than anything else I've produced! lol

 

Edited April 8, 2019 by Macavity

[SIDO]

It is interesting the many reasonings for perusing eeva from battling light pollution to eliminating the dark adaptation requirement for visual observation and as well those more interested in quick and dirty imaging for personal observation or to share those images with other like observer's or just general like sharing as dedicated imagers do, for me with moderately dark skies it's an ongoing physical disability that moves me toward eeva wereas continued visual observations have become too taxing and eeva may be my only comfortable means to continue.

The dangerous part of all this is my already accumulating digital eyepiece collection an my wife's newly evolving opinion of it ?

Eventually I intend to replace visual observations entirely with eeva like alternatives and develope hardware and software work flows around my physical limitations, this new forum on eeva and it's now many discussions have helped greatly the realization that this is the best option for my particular situation and  continued observation of the deep sky.

Whatever your reasoning and expectations, Best of Luck and Clear Skies Everyone ?

                          Freddie...

[SIDO]

2018 Interview with Robin Glover developer of Sharpcap, discussion touches on the evolution of astrophotography with modified webcams to modern-day (2018) AP and EAA. The discussion also includes Robins take on Cmos technology and how it has drastically improved over time, an interesting >30 minutes.