Jump to content

Banner.jpg.b89429c566825f6ab32bcafbada449c9.jpg

Atik 460 versus Atik 4000 shootout.


ollypenrice

Recommended Posts

I've had a lot of PMs and emails about this comparison and what follows is an imperfect shootout. It was made possible by our guest, Julian, who arrived with a new Atik 460 mono, EFW2 and a clone of one of our imaging rigs, a Tak FSQ85 on an EM200 mount. So, the same scopes, the same mount, the same evening, the same site and both cameras. To complicate matters Julian (lucky chap) has the Astrodon filters and we have the Baaders. Julian suffered a bit of focus drift which I spotted rather late on and his polar alignment was 'one night stand' rather than 'observatory' but, none the less, I think we now have a handle on Sony versus Kodak...

Firstly the 'real estate.' How much bigger is the Atik4000 chip? Here's M33 from the 460 dropped onto the same target from the 4000 with identical focal lengths. A clear win for the Kodak but how much does that matter to you? A personal decision...

CHIP-SIZE-copy-X2.jpg

Next up is an image intended to compare noise and sensitivity. We have applied just half the 460 image over the version from the 4000. The 4000 image has 9X10 mins and the 460 only 7X10 mins. The 4000 image was dark subtracted while the 460 merely had a bias subtracted instead of a dark. I tried to apply similar stretches, bringing both images about equally close to their own noise limits. A close look shows significantly less residual noise in the 460 image but there is no great difference in what the brighter signal is offering. I'd call this a clear win for the Sony despite its shorter total exposure. The 4000 will come quiet in the end but will take longer. You may need to look at this closely in the fullsize version because the difference is not enormous.

460-SEGMENT-copy-X2.jpg

Full size of the above: http://ollypenrice.s...Qd2SWf&lb=1&s=O

We had all been assuming that it was 'real estate' versus 'chip quality' and I think that this shows we were all right. I hope this gives a bit of an idea of how to quantify the two. Julian and I put this comparison together as an aside to processing a lot of data today so it is not a perfect objective test but it may be better than nothing and we hope you find it useful.

460 versus 8300? I don't have an 8300 chip and after using Julian's 460 it has dropped off my radar.

Both 4000 and 460 are great cameras. I expected to be partisan towards our much loved Atik 4000s but I have to say that the 460 is a mighty sweet CCD camera and serioulsy tempting...

Olly

Link to comment
Share on other sites

  • Replies 29
  • Created
  • Last Reply

Thank you Olly and Julian for putting this together. This is a really interesting comparison and I expect will give some people, including me, some food for thought.

I have to say that initially I was surprised at how the 4000 stands up to the 460 with regards to signal, even though the 460 has less exposure time, the galaxy shows equally good detail for both. Suggesting to me that if you have the hours to put into the targets, then either the 4000 will take longer to achieve the same result or the 460 may take longer if it means that a mosaic is required due to the smaller fov. It's also hugely useful to actually see the difference in chip size. It's hard to really appreciate that on CdC.

Once again Olly, a big thank you from me.

Link to comment
Share on other sites

Oh just a small PS; I'm often arguing that a perfect sampling rate in arcseconds per pixel is over rated as a priority and here I see no obvious benefit in terms of smoothness from the smaller pixels. At shorter focal lengths, such as with the Tak reducer, I do think it would show. If anyone would care to send me a second Baby Q and an Atik 460 I'll be most obliging, I promise!!

Ollly

Link to comment
Share on other sites

Perfect sapling rate in arcsecs per pixel is only really relevant for planetary AP. In that case you are gunning for fine detail, in DSO imaging, battling photon noise is the main issue. Very small electron wells in tiny pixels on the chip also mean low dynamic range, which can really be a problem in AP.

Link to comment
Share on other sites

I have had a look at the extra large image you linked to and the 4000 appears to have better detail.

I think I had better focus, Earl, that's all. Julian's setup is all new and unfamiliar so a bit of drfift went unspotted in the dead of night. The posted images have neither sharpening nor smoothing.

Olly

Link to comment
Share on other sites

Id be interested to see if i have this right...

the fsq85 at f5.3 with a 460 ex as a arcsec/pix of 2.08 with the 4000 its 3.38.

taking seeing into account seeing the 460 will show a 2 arcsec/pix star and it will take up 4 pixels as seeing will not be at 2.08 however the bigger pixels of the 4000 can deal with it better and only need 1 pixel or have i got this theory backwards?

Link to comment
Share on other sites

taking seeing into account seeing the 460 will show a 2 arcsec/pix star and it will take up 4 pixels as seeing will not be at 2.08 however the bigger pixels of the 4000 can deal with it better and only need 1 pixel or have i got this theory backwards?
The thing about seeing, is that an individual star has a distribution of brightness. So if you had a star perfectly centred on one single pixel, then there would be some "spill" onto adjacent pixels. How much would depend on the pixel size, the brightness of the star and the quality of the seeing at the time.

However, the chances are that the next star along, wouldn't be so accommodating as to align itself perfectly onto a single pixel :laugh: , so the whole situation becomes a bit more complicated.

I also recall reading somewhere (though I may have confused myself) that since seeing is essentially a ransom process - turbulence changes during a single image and between one image and the next, that even without dithering some of the randomness gets removed by stacking - just as it does with noise. So the result of stacking many subs has the effect of improving the resolution of the final image.

BTW: great piece of work on the 4000/460 comparisons.

Link to comment
Share on other sites

As a sequel to this post, here's Julian's first full image with his new setup. The processing is mine. I'm not familiar with colour processing from the Astrodon filters but I'd like the chance to try them some more.

This had an hour per colour, an hour and ten minutes luminance and about two hours of Ha. SInce the Ha added only a little this is really just over four hours' worth in the Atik 460. I don't know about you but I'm impressed.

Olly

M33-byJulian-Shaw-XL.jpg

Link to comment
Share on other sites

Very Nice image there, I have been debating the 460 as my upgrade path from the 383, but for my taste i think im going to have to get something bigger for the Wider FoV, which there are only a few options at present really.

Link to comment
Share on other sites

Very Nice image there, I have been debating the 460 as my upgrade path from the 383, but for my taste i think im going to have to get something bigger for the Wider FoV, which there are only a few options at present really.

That's how I feel. The only small pixel full frame I know of is the H36 and I can't say I'm too keen. I'd have the Atik 11000 but that would be too undersampled with my beloved focal reducer!

Olly

Link to comment
Share on other sites

That's how I feel. The only small pixel full frame I know of is the H36 and I can't say I'm too keen. I'd have the Atik 11000 but that would be too undersampled with my beloved focal reducer!

Olly

Have you considered an Apogee Alta® F16000?

Link to comment
Share on other sites

Thanks very much for writing up this comparison Olly - It's very useful indeed. I must admit I'd been looking at the 4000 as my "dream CCD" for some time (long before the 460 came out) but then the waters got seriously muddied with the release of this new Sony chip... As you noted, FoV vs sensitivity seemed to be the main consideration between the two and as Sara noted, it really helps to see two actual images overlaid to compare these on screen.

I guess one also has to sometimes consider the cost too (well, in order to make a business case to SWMBO!), and on that score the 460 would seem to have another plus point (ie I'm not spending £2,100, I'm saving £650 :rolleyes:), but putting that aside, with the atrocious weather we've had in the UK this year (and subsequently much less opportunity to get out) for me I think the sensitivity is now beginning to get a higher weighting. However, I suspect that by the time I have the cash and or have made a decision, Kodak will have brought out another chip to upgrade the 4000 and I'll have to start all over again!

Link to comment
Share on other sites

Heh, it's a killer. Around the table here we've even considered two 460s in parallel with two scopes as a serious widefield alternative to one setup with a larger Kodak chip. It's taken Sony ten years to make a bigger chip so holding of breath probably insn't an option...

Olly

Link to comment
Share on other sites

I've got a colour 460 and have taken M33 with it too as part of my testing, so I thought it might be interesting to post it here too. Its 11 x 5min subs taken with a WO FLT123 with the 0.8 reducer, so f4.8, with an IDAS filter (remembering I live in the light polluted South Wales Valleys rather than deepest, darkest France!), Flats but no darks or bias. My processing skills need some work too.

I am really enjoying the camera though, and love being able to gather enough data in one night for a reasonable image. Although I'm also looking forward to seeing what sort of images I can get with long capture runs during the winter :smiley:

Helen

post-374-0-13387200-1348575283_thumb.jpg

Link to comment
Share on other sites

Great to see the OSC version joint the party, Helen. You can never tell how the OSC version of a chip will work. The Atik 4000OSC I know to be good but the SBIG 8300 OSC has not made many friends. Your result looks very good indeed for 55 minutes. Remarkable, even.

Olly

Link to comment
Share on other sites

Thanks Olly. I've been REALLY pleased with the camera/scope combo. I've got a few more processed images which I need to post up, all of which have relatively short integration time but good results. Watch this space....

Helen

Link to comment
Share on other sites

Thanks for posting an OSC version too Helen - I have to say that for "only" 55 mins, that really is very good indeed. Some while back I all but decided that to get any kind of a decent image I probably needed at least 10 hrs on an object, whether with a DSLR or a mono CCD, so <1 hour is fantastic

Link to comment
Share on other sites

Hmm

I am wondering if it's worth trading in my trusty but venerable 4000 for the 460? I would probably get about £2k for the 4000 meaning the new camera and electronics of the 460 would cost me a couple of hundred net.

In my rig I could do with a bit more resolution to get better images of the smaller galaxies etc and would only be sacrificing a small bit of FOV to get bette signal to noise - hmm any opinions?

Regards

Rob

Link to comment
Share on other sites

Archived

This topic is now archived and is closed to further replies.

  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...

Important Information

We have placed cookies on your device to help make this website better. You can adjust your cookie settings, otherwise we'll assume you're okay to continue. By using this site, you agree to our Terms of Use.