Lodestar x2 color x Atik Infinity

[DonBoy]

One point that doesn't always come out in discussions involving the ICX828/829 sensor cameras like the LodeStar X2c is that it's color is not very pleasing to my eyes. It's not just this camera, the same color characteristics appear in the Xterminator and DSc cameras that use the ICX828/829 sensors.  This sensor appears to be biased toward the magenta with a much greater emphasis for red.   The images almost appear to be as if they were b&w images that were colorized using sepia and red.    The 825 based cameras have better color rendition and the Infinity software produces slightly better color than the UltaStar software, but I think that is or will be changing as the author of the UltraStarLive software is changing the way the color is decoded.  

Don

[Astrojedi]

I do prefer the color produced by the Atik Infinity vs. Starlight Live.

But note that the color issue is a function of the debayering algorithm used by the capture Software. The SX capture software supplied with the camera produces very accurate color with no hint of magenta - I can see little or no difference in color reproduction compared to an RGB matrix. Hopefully Starlight Live will be updated soon as well.

Also note that the Atik Infinity actually produces a yellowish color (especially with galaxies which are broad spectrum) without a LP filter due to the relatively high yellow sensitivity. Using a LP filter fixes this to quite an extent but slows down acquisition.

 

[DonBoy]

 Also note that the Atik Infinity actually produces a yellowish color (especially with galaxies which are broad spectrum) without a LP filter due to the relatively high yellow sensitivity. Using a LP filter fixes this to quite an extent but slows down acquisition.

I haven't noticed the yellow cast because I used the Idas P2 LPS filter with all my shots with the Atik 414EX color.

[Hibou]

15 hours ago, Astrojedi said:

Note that binning in SW has no impact on sensitivity. 

Stacking in SW "has no impact on sensitivity" either, but 2x2 SW binning, like x4 stacking, improves the S/N, which is the whole point. Glad you agree that it is physically possible after all :-)

My original "simple physics" argument was that a bigger chip, like a bigger telescope, necessarily collects more light. Yes, you may have to adjust the focal length and resolution to make use of that, and yes it may not be as easy for OSC astronomy as it is for my monochrome low-light imaging. But bigger chips do have a fundamental advantage that you should try to use. Otherwise "sensitivity" is an intrinsic chip property, and the 829 and 825 chips use the same maxed-out CCD technology. Bigger pixels do make a little better use of the area of an interline CCD, but the micro-lenses are supposed to compensate for that.

BTW you should be careful with Sony's claims of "x2 sensitivity" of the 829 over the 429, the supposed x2 advantage of the Lodestar X2 over the X1. They have achieved that in part by improving sensitivity at longer wavelengths, even into the infra-red, which may be good for their main industrial and surveillance market, but not so interesting for my application, nor for OSC astronomy. Perhaps that explains why the 829 colours look rather red.

These interesting comments from AstroJedi have confirmed my belief that you pay a high price for OSC astronomy. I would be tempted to stick with monochrome and just "Add a Splash of Color" from some archived source, as NASA's x-ray astronomy team so amusingly put it :-)

[Hibou]

6 hours ago, Hibou said:

...you pay a high price for OSC astronomy.

Having said that, I found an Ultrastar Mono and Ultrastar C comparison for M27 on the SX website, and I'm puzzled. The  stack of 14 x 20 second Ultrastar C frames appears as bright as the 40 x 10 second stack of Ultrastar frames. The scales and telescopes are the same, but I rotated and cropped the images below to make the comparison clearer. Surely there must be some mistake on the SX website ? The Ultrastar image looks just like the Ultrastar-C image reduced to grayscale, when it should be much brighter ??? Perhaps it's been re-scaled when converted to jpeg for the web ?

[Astrojedi]

7 hours ago, Hibou said:

Stacking in SW "has no impact on sensitivity" either, but 2x2 SW binning, like x4 stacking, improves the S/N, which is the whole point. Glad you agree that it is physically possible after all :-)

My original "simple physics" argument was that a bigger chip, like a bigger telescope, necessarily collects more light. Yes, you may have to adjust the focal length and resolution to make use of that, and yes it may not be as easy for OSC astronomy as it is for my monochrome low-light imaging. But bigger chips do have a fundamental advantage that you should try to use. Otherwise "sensitivity" is an intrinsic chip property, and the 829 and 825 chips use the same maxed-out CCD technology. Bigger pixels do make a little better use of the area of an interline CCD, but the micro-lenses are supposed to compensate for that.

BTW you should be careful with Sony's claims of "x2 sensitivity" of the 829 over the 429, the supposed x2 advantage of the Lodestar X2 over the X1. They have achieved that in part by improving sensitivity at longer wavelengths, even into the infra-red, which may be good for their main industrial and surveillance market, but not so interesting for my application, nor for OSC astronomy. Perhaps that explains why the 829 colours look rather red.

These interesting comments from AstroJedi have confirmed my belief that you pay a high price for OSC astronomy. I would be tempted to stick with monochrome and just "Add a Splash of Color" from some archived source, as NASA's x-ray astronomy team so amusingly put it :-)

Binning in SW does not improve sensitivity or SNR. 

[Hibou]

2 hours ago, Astrojedi said:

Binning in SW does not improve sensitivity or SNR. 

You claimed earlier that I would not find software to bin OSC cameras. Well I found PixelInsight and ImageJ for starters. Now apparently you accept that it can be done, but claim that it's not worth doing ?

I never implied that sensitivity could be improved by software; In fact I explicitly said that sensitivity was an intrinsic chip property.

I also said that binning is similar to stacking. Repeated independent measurements of a constant signal superimposed on random noise does improve S/N. Stacking uses the same pixels read at different times, while binning uses different pixels read at the same time. They are both largely independent measurements.

Yes, I know that some people don't believe in binning, even in hardware binning when for short exposures much of the noise is read noise. If you are one of them, just say so, and we can at least agree to disagree. Perhaps you don't even believe in stacking.... (Gallic shrug :-)

[aparker]

5 hours ago, Hibou said:

Having said that, I found an Ultrastar Mono and Ultrastar C comparison for M27 on the SX website, and I'm puzzled. The  stack of 14 x 20 second Ultrastar C frames appears as bright as the 40 x 10 second stack of Ultrastar frames. The scales and telescopes are the same, but I rotated and cropped the images below to make the comparison clearer. Surely there must be some mistake on the SX website ? The Ultrastar image looks just like the Ultrastar-C image reduced to grayscale, when it should be much brighter ??? Perhaps it's been re-scaled when converted to jpeg for the web ?

I think it's very hard to judge "brightness" without the original FITS files in hand

[Astrojedi]

8 hours ago, Hibou said:

You claimed earlier that I would not find software to bin OSC cameras. Well I found PixelInsight and ImageJ for starters. Now apparently you accept that it can be done, but claim that it's not worth doing ?

I never implied that sensitivity could be improved by software; In fact I explicitly said that sensitivity was an intrinsic chip property.

I also said that binning is similar to stacking. Repeated independent measurements of a constant signal superimposed on random noise does improve S/N. Stacking uses the same pixels read at different times, while binning uses different pixels read at the same time. They are both largely independent measurements.

Yes, I know that some people don't believe in binning, even in hardware binning when for short exposures much of the noise is read noise. If you are one of them, just say so, and we can at least agree to disagree. Perhaps you don't even believe in stacking.... (Gallic shrug :-)

Of course you implied sensitivity could be improved by SW binning otherwise why else would we be having this discussion.

You claimed that the Op should go buy the Infinity color as you can bin it and achieve/exceed the sensitivity of other sensors. My whole point was that HW binning does increase sensitivity but the Infinity does not preserve color if you bin.

Then once you realized that HW binning does not retain color you started claiming binning can be done in software to improve SNR and that there is no difference.

Also hardware binning is completely unlike stacking (software binning is mathematically equivalent to scaling). The process by which SNR is improved and the math involved are completely different for the two. Like comparing apples to oranges. Just because one works does not mean the other works.

And when did I even mention stacking? And since when are these approaches subject to a "belief system"? If the math and science behind them are sound then I will use them (and yes stacking and hardware binning both work and I use them). 

 

 

[Hibou]

10 hours ago, Astrojedi said:

Of course you implied sensitivity could be improved by SW binning otherwise why else would we be having this discussion.

You claimed that the Op should go buy the Infinity color as you can bin it and achieve/exceed the sensitivity of other sensors.

I "implied..." ? It was you who introduced "sensitivity" and clearly you have your own definition. If you read the CCD specs you will see that sensitivity is an intrinsic property of the chip and I never "implied" anything else.

And I didn't claim that he should "go buy the Infinity", but rather that he should "go for the cheaper solution".

But there's no point in continuing "this discussion", which you started by telling me to "Please stop giving people bad guidance and inventing new Physics concepts". 

You're simply disappointed that your own ambiguous guidance was ignored and the guy bought the Infinity, which is a great camera manufactured in his home country, but which you dismiss as suitable only for "casual astrophotography" whatever that is.

Adieu.

[stash_old]

36 minutes ago, Hibou said:

I "implied..." ? It was you who introduced "sensitivity" and clearly you have your own definition. If you read the CCD specs you will see that sensitivity is an intrinsic property of the chip and I never "implied" anything else.

And I didn't claim that he should "go buy the Infinity", but rather that he should "go for the cheaper solution".

But there's no point in continuing "this discussion", which you started by telling me to "Please stop giving people bad guidance and inventing new Physics concepts". 

You're simply disappointed that your own ambiguous guidance was ignored and the guy bought the Infinity, which is a great camera manufactured in his home country, but which you dismiss as suitable only for "casual astrophotography" whatever that is.

Adieu.

I think I am right in saying both Starlight and Atik are UK companies even if they use Chinese components (doesn't everyone!) .

No point falling out over it - just agree to disagree. For me the pictures tell most of what I need to know (looked at many different people's pics with both camera's) the rest just goes above my head - "nit picking" as we say in the UK. For me the over riding plus was the Starlive software development was seems to be on going and frequent.

In the end the "man/woman" behind the hardware and the skies create the final picture/video - that reminds me I need one hell of a lot more practice.

[Hibou]

1 hour ago, stash_old said:

I think I am right in saying both Starlight and Atik are UK companies..

Yes, you are right Stash_old, they are both UK companies. Atik was launched 10 years ago by Steve Chambers (UK) and Rui Tripa (Portugal). Atik is based in Norwich, where the cameras are designed and developed, but most of the equipment is made in their factory in Lisbon, Portugal (Mystyco's home town :-). Both companies use Sony Japanese CCDs, or ON-semi (previously Kodak) for larger formats to produce excellent cost-effective cameras. I'm not sure if there is much Chinese content. And yes, both Starlight Live and Infinity software are excellent UK products too. Nothing to disagree with there :-)

I was surprised by how "frank" AstroJedi's response was to my initial email, and actually apologised if I had offended anyone. Apparently "tell it like it is" is popular now, but I belong to an older generation on a different continent.

[Astrojedi]

On 2/22/2016 at 08:59, Hibou said:

All these pretty pictures do not impress me as much as simple physics :-) The Atik Infinity and the SX-Ultrastar use the same CCD and cost the same i.e. over €1000 in Portugal; they will give similar performance. The SX-lodestar costs half as much (€500) for the same CCD technology but with half the resolution and half the chip area.  The smaller chip will give a smaller FOV for the same focal length, and its lower resolution (larger pixels) will give a brighter image. But the larger chip will still win, even for intensity, if you can adjust your focal length and binning to give the same FOV and resolution.

 

Let's recap... Here is your original  post... See text in bold... You are implying that the Atik Infinity Color will 'win for intensity' due to its larger size as you can bin.

Lets get the facts straight...

1. You cannot bin the Infinity in hardware/charge domain and keep color

2. Binning in software does nothing for sensitivity or SNR yet you brought it up in defending your point.

3. Unbinned the Color Infinity and Ultrastars require long total exposures at slower f ratios. In my option these are best used at F2.

Looks like we are in agreement now.

 

[2STAR]

On ‎22‎/‎02‎/‎2016 at 00:28, HiloDon said:

SX also makes an 825 sensor based camera in competition with the Atik Infinity called the Ultrastar.  It has the same housing as the Lodestar and uses Paul's Starlight Live software.  The software has been recognized as one of the best for EAA and is not only a free download thanks to Paul, but is compatible with both Mac and Windows.  I have both the Ultrastar C and the Lodestar X2c.  Both have their advantages as Hiten pointed out.  Here are some comparisons.  LSX2c first then Ultrastar C.

I like both.  They both do a great job for EAA.  I think the Ultrastar works best with Hyperstar and must be stacked to reduce random noise more noticeable in the 825 sensor.  The lodestar is very easy to use and I always recommend it for those just starting out in EAA.

Hope this helps.

Don

f 1.9 is ultra fast , nice pics 

[stash_old]

Will the Admins please close this thread down its going nowhere - whats bin(joke) said has been said and people sensitivity(joke) is getting that the longer the exposure(joke) to this thread we are not helping the original question  

[Carl Reade]

Without going into tech details the Atik Infinity has opened up my sky and scope with quality images. It may not be as fast as some but I can see my target in a single 30 sec exposure and the software does the rest. I am using a SW 150p F5 with LP filter (prob F6 with it) and three sodium street lights each less than 10 meters from the scope and it still gives me what Im after. These pics were 30 sec and prob let stack less than or around 10 apart from M42 which just needs a second of exposure. some have levels adjusted live to bring more out. Works for me.

The colour cams work fine unless you need to see something in less than 30 secs for dim stuff. You can cover a lot of objects in a single session easily.

Carl

 

 

[HiloDon]

Thanks, 2Star.  I'm really sold on Hyperstar for EAA/near real time viewing and imaging.

I agree with Stash.  The OP had his questions answered and made his dicision.

Don

[RobertI]

1 hour ago, Carl Reade said:

Without going into tech details the Atik Infinity has opened up my sky and scope with quality images. It may not be as fast as some but I can see my target in a single 30 sec exposure and the software does the rest. I am using a SW 150p F5 with LP filter (prob F6 with it) and three sodium street lights each less than 10 meters from the scope and it still gives me what Im after. These pics were 30 sec and prob let stack less than or around 10 apart from M42 which just needs a second of exposure. some have levels adjusted live to bring more out. Works for me.

The colour cams work fine unless you need to see something in less than 30 secs for dim stuff. You can cover a lot of objects in a single session easily.

Carl

 

 

Carl, I'd love to see more of your captures with this setup, can you start another topic and show us some more if/when you have them. Thanks.

[Carl Reade]

33 minutes ago, RobertI said:

Carl, I'd love to see more of your captures with this setup, can you start another topic and show us some more if/when you have them. Thanks.

Will do Robert this is only my second outing with the cam since October thanks to the weather. I will be broadcasting on VAL as well when I get a chance. Glad they are of interest?

[mystyco]

Got my camera today. Sadly there's a terrible weather around here and they even say it could snow. Luckly the wind pushed some of the clouds away and I quickly aligned the telescope, placed the camera in the focuser, and got my first image in 5 minutes or so. Very straightforward and intuitive. But I'm not 100% sure if my focus is perfect. I didn't had much time because of the clouds.

With that said, here's my first image. All I did in the image below was noise reduction, and nothing else. This is 5'' x 10.

The camera is definitely very sensitive. But is it better for me than my current DSLR setup? When I get clear skies, I will know. I enjoyed the "live" experience very much.

 

[mystyco]

M42 in 40 seconds. So far, the camera seems very good for bright and big objects. However for the faint objects it has disappointed me: too noisy, very hard to get a clear picture. This is my personal opinion, and at least for now. But I need a couple more days to fully test all it's potential. This image of Orion has impressed me.

 

[mystyco]

 

[Hibou]

8 hours ago, mystyco said:

M42 in 40 seconds. So far, the camera seems very good for bright and big objects. However for the faint objects it has disappointed me: too noisy, very hard to get a clear picture. 

Thanks for the update Mystyco. It would be interesting to see photos of faint objects under comparable conditions with your SLR camera when you have had a little more time with the Infinity. (With details of the exposure, number of frames etc).

You probably already found another photo of M42 with the Infinity on the Atik web site (a mosaic with longer exposures so not directly comparable). And of course HiloDon, DoctorD and others have also posted M42 with the Lodestar and Ultrastar for comparison (it's one of the examples of the use of non-linear scaling in Don's LL manual).

[stash_old]

11 hours ago, mystyco said:

M42 in 40 seconds. So far, the camera seems very good for bright and big objects. However for the faint objects it has disappointed me: too noisy, very hard to get a clear picture. This is my personal opinion, and at least for now. But I need a couple more days to fully test all it's potential. This image of Orion has impressed me.

 

I am going to get a lot of flack for this but  IMHO you perhaps need to think why you want to do EAA (sorry Video Astro) - Do you want very photographic images as seen in mags - which as you know takes lots of processing and lots of data and playing in PS and the like - what I like to call Photographers doing Astronomy or do you want near real time viewing of objects in space(Astronomers using Photography). Nothing wrong ,just personal choice, doing either !!

We cant all live in Hawaii ( Am I jealous = yep) but Carl Read's pictures,just one example, are great considering his problems as he stated. I also believe you will improve  as you get used to doing things differently and more familiar with the software - I did using just a LN300 camera and Sharpcap.

But in the end its what ticks your box's - your choice.

[mystyco]

37 minutes ago, stash_old said:

I am going to get a lot of flack for this but  IMHO you perhaps need to think why you want to do EAA (sorry Video Astro) - Do you want very photographic images as seen in mags - which as you know takes lots of processing and lots of data and playing in PS and the like - what I like to call Photographers doing Astronomy or do you want near real time viewing of objects in space(Astronomers using Photography). Nothing wrong ,just personal choice, doing either !!

We cant all live in Hawaii ( Am I jealous = yep) but Carl Read's pictures,just one example, are great considering his problems as he stated. I also believe you will improve  as you get used to doing things differently and more familiar with the software - I did using just a LN300 camera and Sharpcap.

But in the end its what ticks your box's - your choice.

The images that I posted, to me look absolutely amazing. My jaw literally dropped when I saw M42 on my screen. My concern are the faint objects, even M81 which is quite large looked like a grainy tv screen, no color, no detail, nothing just static and noise. But what I really didn't liked at all was the chip size. Super hard to find and center the objects in my FOV. But like I said I need more time to fully test it. I'll take it out again tonight, and post some images again. Like I always say: astrophotography is hard, and the trick is to find something that works for you. My DSLR has been working flawlessly. Will the Atik work better for me? I hope, becauE I really love the idea of getting a finished  color image in 30 seconds or less. Time will tell.