Anyone Using The Atik 4000 camera ?

[jcm]

I am trying to work out whether its worth upgrading my 16HR to the Atik 4000. Is anyone using the 4000 ? I love ( probably the wrong word to use ) my 16HR and I would only sell it if the 4000 produced similar quality images.

John

[RobH]

Hello John.

I have the kodak 4021 chip in my SXVF H16, which has the same pixel size etc. The 4000 series chips are the Kodak 4022. I don't know what the difference is but I expect it is slight.

The big difference between this and the 16HR is the FOV. With the larger chip it is very much bigger, but the resolution is almost (but not quite) as good. I doubt if one could tell the difference in resolution with our skies.

This means that you can hit a wider field but still be able to crop down to the 16HR FOV and get a good image.

The 16 has very little dark noise, but the kodak chip has more, so with long exposures darks are more important, although this can be alleviated somewhat by dithering between subs, the using SD mask, Sigma clip, or median combine modes whan stacking.

I don't have to dither....I let the guiding take care of thet

1 1/4 inch filter work, but when you start to use focal reducers, you get vignetting, but this can be dealt with by using flats.

I don'yt know about the Atiks, but the SVX is very light, certainly lighter than my 16HR, which is worth considering.

Cheers

Rob

[jcm]

Hi Rob,

Thanks for all the info.

A difficult decision this one. I found some images taken with the 4000 on an Atik web site. They looked pretty good.

Will have to think about this one for a while.

John

[NickH]

FOV is impressive, I have both now, the 4021 is part of an obsy project in Spain, and I have the 314L for my everyday use. The 314l/16hr is vastly more sensitive, and delivers amazingly low noise. The 4021, huge FOV, good sensitivity, slightly noisier, but you have set point cooling, which really helps. Be prepared to do mroe worek in dark frames, 1.25 inch filters do work (just), and it's all round a great cam the 4000/4021... I would ideally were I you, keep both...you'll see why when you start imaging smaller DSO objects...

[jcm]

Thanks Nick for the extra info.

Keeping both sounds a good idea but it will damage the bank balance more

John

[ollypenrice]

John, I'm glad I found this post because I'm considering a 4000 as well. It's the field of view I'm after. Like you I have a 16HR but that will certainly NOT be going up for sale. It's too good, and in my situation having more than one camera is a good idea. I'm interested in the 'just' when NickH says 1.25 filters do just work. I take it they really DO work 100%?

Olly.

[NickH]

They do, but the JUST comes from some imaging done last year on the horsehead where a minor amount of clip at the extreme edges was seen..

[jcm]

John, I'm glad I found this post because I'm considering a 4000 as well. It's the field of view I'm after. Like you I have a 16HR but that will certainly NOT be going up for sale. It's too good, and in my situation having more than one camera is a good idea. I'm interested in the 'just' when NickH says 1.25 filters do just work. I take it they really DO work 100%?

Olly.

I do not want to sell my 16HR either.

The Starlight Xpress SXV H16 is priced very close to the Atik and is smaller and lighter ( better made ?) - this is probably the one I would actually buy.

John

[NickH]

be prepared to also factor in a copy of Maxim then...

[narrowbandpaul]

i have been doing a report on the performance of that sensor you are interested in.

it will be done hopefully in a few days.

we have it in the H16 from SXV.

no set point cooling, and I agree with nickH about needing Maxim.

The SXV control prgramme is not particulary well written.

paul

[RobH]

As Nick says, 1 1/4 " filters do work, but only just, and, as I mentioned earlier, when you start to use FR's and barlows, expect some vignetting.

I don't bother with the Starlight Xpress software, I use maxim as my imaging software and the H16 works fine with that.

You need to leave it running for half an hour or so, to be certain you get it down to the lowest temperature, or your darks won't match.

That said, I have to leave my 16HR running for a good while before using it otherwise the computer crashes to blue screen!!

I'm certainly not planning on selling my 16HR. The H16 is in addition to it, and will get a lot of use on widefield targets in the summer.

Cheers

Rob

[NickH]

The Atik has set point cooling, and comes with staggeringly good software (in terms of ease of use, plus loads of new stuff coming from them which will be probably free for Atik users)

[jcm]

I do have Maxim DL.

I suppose the Atik does have the feature of set temp control. There is also an LE version with less accessories and only one stage temperature reduction. As £2500 is a big lump of cash - I am taking a lot of time to consider this purchase.

John

[ngc2403]

I think that some long hard though is need over this, i think that the comparison between the H16 and the 16hr is not very fair, as the too are different types of chips and the software and manufactures are different too.

The non-set point cooling is a nightmare, trust me, you can't tell what darks are good to use and there is no way too know... this means that unless the camera is used in the as temperature environment all the time the darks are all different and new one must be taken very night.

I have been told that the SONY chip in the 16Hr is lower noise than the kodak one so the non-set point cause even makes problems

[ollypenrice]

I think the comparisons were between the Atik 4000 and the H16 and this is a worthwhile comparison. The presence of the 16HR in this discussion was that it was the camera that those of us thinking of going 'Big Chip' were already using.

As a software numbskull I love the idea of sticking with the familiar Atik control stuff. For people of normal intelligence a switch to Maxim would be less of a challenge!!

Olly.

[RobH]

Regarding darks, it really isn't that critical, certainly with the H16, as long as there is some slight movement between subs.

If there is a little movement, stacking using SD mask, sigma clip, or median combine will deal with most of the noise.

I have a few darks at different temperatures, and I use the closest one to the temperature the light was taken at.

That said, set point cooling would be a bonus, as you can do one set of darks and that's it, and applying matched darks is undoubtedly better than not applying them.

With my 16HR, I don't bother with darks though, they aren't needed until you get to 15 minute subs and above.

Cheers

Rob

[ngc2403]

how did you get the temperature darks for the H16? it has no temperature control or any sensor to tell the temp?? unless you were measuring the outside temp...

I don't think that the small variations in the temperature make too much difference on a seasonal basics but winter darks will not work on summer lights and seen as almost ever other CCD maker has some sort of temp control i think Starlight Express are missing a trick by not having it.

Things like photometry need very accurate noise control and the SXV H16 fails in some respects to deal with dark current.

[RobH]

how did you get the temperature darks for the H16? it has no temperature control or any sensor to tell the temp?? unless you were measuring the outside temp...

I measure the outside temperature and record this in my file names. I got hold of an old fridge and put the camera inside, together with a thermometer that had a remote sensor, and adjusted the temperature by having the door open a crack. I then shot a series of darks at different temperatures. Saves loads of time and you can do it in the day.

I don't think that the small variations in the temperature make too much difference on a seasonal basics but winter darks will not work on summer lights and seen as almost ever other CCD maker has some sort of temp control i think Starlight Express are missing a trick by not having it.

Agreed. Although dithering helps a lot. Yes, Starlight should have set point cooling.

Things like photometry need very accurate noise control and the SXV H16 fails in some respects to deal with dark current.

I wouldn't debate this, I just take (hopefully) pretty pictures but photometry is a whole different ball game.

Cheers

Rob

[narrowbandpaul]

dark thermal noise isnt a big deal. you just need to remove it.

how do you remove it?

darks at the same temp as your lights, and lots of em

no set point cooling is a zeroeth order flaw with kodak sensors

a single dark subtraction may add noise to the image, just like applying a single flat

[ngc2403]

rob that is some long lengths you have gone to, but it is the only way to get the right stuff for your images

[roundycat]

I think I only just picked up on this thread and from what I have seen one point is being laboured too much. I use an Artemis 11002 which has poor temp regulation to the extent that I just run the cooler flat out and never worry about what temp the chip is at. As Rob said, dithering and the correct type of combining algorithm will usually take care of noise.

I always use a hot pixel filter set to 5% in Maxim and then rely on dithering and SD Mask to take out what is left. The resultant frames are incredibly clean and not a dark frame in sight.

The moral of this is simple, to me anyway. Hard earnt empirical results trump theory.

Dennis

[ngc2403]

Hey Dennis,

Your simple theory about image processing is not a 'Hard Earnt Empirical result'. it is shortcut to good quality images.

Just because the hot pixels out side the 5% are remove don't men they are all gone, the ones inside the 5% are still there it's just that you can't tell they are noisy.

i think this is the key difference here in your opinions, if you like the look of it then that is good enough however if you can imaging well under the seeing limit say wide field then this method might remove the stars from the image all be it only the ones confined to a single pixel and so they look just ike hot pixels.

Quasars might get removed through this method though it would be unlikely.

basically if you imaged a sooth object them without FPN removal the object would look noisy, and would look nothing like the real thing. more likely is the effect on an uneven object will cover the real details with the object roughness and the FPN combine which your methods will never remove.

[narrowbandpaul]

not having a temp controller is a big issue especially with the noisier kodak chips. Having to obtain dark at the end of the imaging session or having to cut the imaging session short to capture darks is a major PITA.

this is never truer when one considers narrowband imaging when typical exposures are of the order of 20mins. Having to capture an hour and a half of darks at like 4am is not nice. and that was using the less noisy sony chip, it would be ultra required if it were a kodak sensor.

having temperature regulation means as long as you note the sensor temp during acquisition you can capture them the next day. which maximises the dark sky photon capturing time.

paul

[roundycat]

Ally, please give me some credit! The 5% figure is chosen by me not hoisted on me. I can make it 1% if I want. The hot pixels that are left are taken care of by dithering. I have got quite a few wide field images under my belt and none of them as far as I know show stars that are represented by a single pixel. As I have said before I am in the business of making pretty pictures which closely represent the nebulosity or galaxy brightness that is generally accepted for a given object. If a star or quasar or anything else that is one pixel or less goes missing I couldn't care less. If you cannot see it you cannot miss it.

I also always use flats which get rid of all the other unwanted stuff and I have never knowingly presented a finished picture which is noisier than any of the original subs.

Maybe we should settle this at twenty paces. You pick an object and we'll both image it and process it. You with your theoretically perfect approach and me with my sloppy shortcuts. Meantime here is a Rosette close up with no dark subtraction, just to get back on topic a bit. Is this smooth enough?

Dennis

[roundycat]

Paul, just a quick reply. See above picture to show you how unimportant darks can be. They were not used here. Also remember that if you take temp controlled darks they can be taken on a cloudy night, no lost imaging time.

Dennis