Canon Vs Nikon

[rastis95]

Hi all,

Out of curiosity, with dslr's, why does it seem that the canon eos range is the first choice for people when doing astrophotography of this type. I haven't seen too many people using any of the nikon range. I would have thought that they would nearly be on par with each other. Ive been looking at getting a 450d myself but still doing my homework.
Any thoughts ?

Cheers
Scott:D

[Geoff45]

I gather that it's hard to get a true raw image out of a Nikon. It seems to do some processing of the "raw" image, which includes hot pixel removal, but unfortunately this also removes sharp star images as if they were hot pixels.
(Source: Michael Covington, Digital SLR Photography)

[luka]

As mentioned above, Nikon cameras do postprocessing of the "raw" images. This can be overcome by switching off the camera during postprocessing.

Until recently the main difference was that Canon used CMOS technology while Nikon was using CCDs (made by Sony). Canon generally had lower amp glow which made them more popular for astrophotography. However, some newer Nikon's are also CMOS based and I am not sure how they would compare.
And of course, some Nikon models are better than the others, for example quite a few people are using D40 for astrophotography.


For daytime photography, there is not much difference though.

[rastis95]

Thanks for the feedback,
With regard to cmos and ccd, i was led to believe that a camera with ccd was always better than cmos, don't know why :shrug:, something about cmos having to amplify the pixles ?. I will still probably go for the canon eos range, one day :whistle:.

Cheers
Scott:D

[iceman]

I don't own a Nikon so only going off 2nd hand information, but from those that have owned both, the Canon appears to have less noise and is therefore more suited to astrophotography.

There are some people doing brilliant things with Nikons though, so it can be done.

But i've seen owners of Nikons for terrestrial imaging, get a cooled modified Canon for astrophotography. So I guess that says something.

[rastis95]

You do have a point there Mike, I think my mind is made up, now if i can talk the wife into the 5D Mk II :whistle:.

Cheers
Scott:D

[Geoff45]

It is harder to get a uniform response from all pixels in a CMOS array than from a ccd array, that is, the sensitivity of the CMOS chip is more variable over the chip than is the case with a ccd. Probably not too important for pretty pics, but it does matter for things like photometric work.

[luka]

CCD technology was better several years ago but in the meantime CMOS has taken over (the difference is not that great though). Hence you have to check the date of the articles comparing the technologies. Most of manufacturers (including Nikon) are slowly migrating to CMOS.

Also I remember reading somewhere (a while ago) that the original Canon IR filter had higher transmission for H-alpha wavelengths and hence unmodded Canon camera was better suited for astrophotography. I don't remember seing a replacement IR filter for Nikon anywhere...

Basically, the choice is simple, if you don't have a DSLR, go with Canon. If you already have a complete system of lenses based on Nikon, stick with Nikon unless you have money to spare...

[rastis95]

Thanks all for the responses, has cleared things up now. All i need now is to get hold of one. Quick question though, do the cannon dslr's have a movie function and would that be able to be used for planetary work

Cheers
Scott :D

[Octane]

Scott,

The newer Nikons (such as the D90) have a video mode.

My 5D Mark II has full HD (1920x1080) video, and it is just astounding quality.

However, bear in mind that due to the size of your sensor, you'll need insane focal lengths to get a decent enough sized planet to grab video of. This is why the smaller chip sized web cams are more suited to planetary work. Have a look in the Solar System imaging section to see the kinds of camera and telescopes, people like Mike and Matt are using. To date, I don't know of anyone who's using the video modes in DSLRs for planetary work.

Regards,
Humayun

[rastis95]

Still looking into this one. I was speaking to the local photography guy about which is best. I was explaining that I would like the capability of astro work and that is the reason why i was looking into canon dslr's. Now this guy is a nikon man through and through, doesn't mind canon but reckons i should look a bit more closely at the nikon range. So ive been having a look at the d90, looks good but my limited knowledge on dslr's is pulling me up. Here is the site for the specs
http://www.nikon.com.au/productitem.php?pid=1281-86d7b52026

Now it was mentioned that nikon camera's did some post processing of the image to remove hot pixles ect, would the expeed system that the range now has, do something similar or not. Also it looks like they have changed over to cmos in line with the canon series, assuming this is a good thing.

Any advice appreciated

Cheers
Scott:D

[bojan]

Have a look at Christian Buil's webpage here and elsewhere on his site:
http://www.astrosurf.com/buil/d70v10d/eval.htm

Also, Canon has generally better support re astro work.

[rastis95]

Now that was a read and a half :lol:, thanks for the link. Reading that the ir cut on the nikon seems to be a big problem. That article refers to a ccd chip on the d70, with some terrible thermal noise and amp glow, any one herd how the new cmos on the d90 is going, and this pre processing is another problem. So far the cons are outweighing the pro's :shrug:

Cheers
Scott:D

[Omaroo]

That's a really OLD article guys. The D300/700/90 are all VERY different to the D70.

[rastis95]

Chris,
Have you ever used your d40 for astro work, would be a far cry from the SBIG ST-8i (by the way looks like your getting the hang of that little puppy), and compared it with any shots from the 350d ?

Cheers
Scott:D

[Octane]

Scott,

It seems like your heart is set on the Nikon. May I ask, why that is?

If you feel that you'll be using it primarily for terrestrial work, and the occasional outing on the back of a telescope, then, it won't really matter too much which way you go. However, just take a look in the deep space imaging section at anything that isn't taken with dedicated astrophotographic CCDs, and you'll soon see that it is all Canon. There's a reason for it -- it's tried and tested, the accessories and software are all available (in some cases for free), and, it's a relatively cheap (300D/350D) and easy way of getting into DSLR imaging.

Also, remember the median filtering crap that's applied to Nikon RAW files...

Regards,
Humayun

[luka]

Scott, see the post by Chriss in this thread (http://www.iceinspace.com.au/forum/showthread.php?t=15806) and also see this page (http://www.astropix.com/HTML/I_ASTROP/NIK_CAN.HTM) for more comparison which shows several quite important points.



Most of people use Canon because Canon used to be the clear winner in the low noise area and the choice was simple. However, we are talking about newer Nikon models and Nikon changed a lot recently. People should not use several years old articles comparing some old model (D70) - technology changes at amazing speed nowadays.
For example, Nikon D40 has been proven to be very good for astrophotography with noise levels comparable to Canon. The IR filter can be replaced as well. And speaking of cheap, you can pick up a brand new body for ~$350 on ebay.
I also read somewhere that D90 has excellent response for H-alpha with standard IR filter.

[Octane]

luka,

My points still stand.

Furthermore, you'll note that I haven't referenced, and, for that matter, I have not read Christian Buil's exhaustive comparative research and am not using his research to vindicate my points. Just my own experience. Though, I do use his awesomesauce software.

I went down to the lake here in Canberra on Australia Day with a colleague of mine who has a Nikon D90 to do some long exposure night time landscape work. I remember the noise that was prevalent on his LCD screen. I'll see if he still has the shots so he can send them on over to me so that I can compare them with my old 350D. I haven't had a chance to do long exposure work with my 5D Mark II, just yet.

Horses for courses.

Regards,
Humayun

[luka]

Humayun, my apologies if I made it sound like you referenced Christian Buil's article.
Your points still stand and I agree with all of them but what I tried to say is that one of the possible reasons people use mainly Canon in the forums is that until recently there was no real competition in the field of astrophotography (AP). Nikon was based on CCD technology which was not as good as CMOS for AP, but things changed.

Now I am not trying to say that Nikon is better nowadays but the technology changed a lot recently and old comparisons are useless. I have seen some excellent results for D40 but could not find no direct comparisons for newer Nikon models. Just people's experiences which range from great to bad. This guy (http://www.sfu.ca/%7Edmunro/DSLRnotes.html) even praised D50 over comparable Canons (at the time of writing).
However, any conclusions based on subjective tests are useless if you ask me. I would like to see more tests like Christian Buil's articles and until then I agree (as I already posted above) that Canon is a better choice.

[rastis95]

Thanks all for the feedback,

Humayun im not overly stuck on the Nikon dslr's, actually in the back of my mind im still looking at the 450d or higher. Its just with the advancements Nikon seem to be making lately across there range they were worth a look, just doing some research how these newer models have been fairing, especially in the area of AP.

Software seems to be a bit of a problem with a trend toward supporting canon. Another big thing is that the Nikon's still seem to do this post processing altering the RAW files but can't confirm this. And the apparent bad transmission of H-alpha through the IR Filter, something Canon can boast about. Still a few unknown's

Luka you said you read somewhere that D90 has excellent response for H-alpha, so maybe things are changing a bit :shrug:. Problem is there wouldn't be to many newer Nikon's being used for AP to report back on. Its like in the old days, Beta was always better but we ended up with VHS ( Im not that old by the way but I do remember it ). Maybe the playing field is leveling out a bit.

Cheers and thanks for the feedback, its much appreciated
Scott:D

[Zuts]

All the major AP camera manufacturers, SBIG, FLI, QSI use CCD sensors and always have so I certainly dont understand this statement?

Cheers
Paul

[bojan]

CCD techology was always better performer than CMOS in terms of sensor noise, and it was purely process problem, there were no fundamental theoretical problems that would favour CCD over CMOS.
Canon (and Foveon, I think) sensors changed that, so today both tehnologies give similar results.

However, the definite advantage of CMOS over CCD is power consumption: CCD requires relatively high voltages to operate, and CMOS does not. Also, CMOS allows direct addressing of any pixel in the array, and CCD does not (architectural reasons).
Also, CMOS process is used for digital circuit manufacturing, while CCD uses totally different process so it is not possible to mix digital circuitry on the same chip with CCD, one of things that bring the manufacturing costs down.
That is why affordable commercial cameras are much more likely to use CMOS sensors. If someone still uses CCD (in mass production) there must be a trade-off somewhere to offset the manufacturing cost.

While CCD is still better in terms of noise (one of the reasons is, there is no digital circuitry on the same chip so readout noise is easier to suppress), it is much more expensive to manufacture the cameras using it.

[bojan]

BTW, since I posted the link to Christian Buil's (old) article, let me clarify in more detail why I did it:
Firstly, it was because (as someone pointed out earlier) only this kind of comparisons, with enough support of facts and images and measurement data), are really useful.
Also, the models he was comparing they are still around (second hand), and for people with tighter budgets (and some people on this forum tend to forget the fact that the majority of prospective astro-photographers belong to this category) they are still a viable option.
Finally, I deeply respect the work of people like Christian, who share their knowledge, experience and expertise with the rest of us without thinking about their own material benefit.
In my opinion it is a good thing to keep an eye on such websites, there are always updates, and, by visiting them perhaps they will let be known that we appreciate their work :thumbsup:

[rastis95]

Sorry to revive the great debate here but a couple more questions.
Through all the info from here and elsewhere Ive decided on the canon series. Question is, the new models usually come with say a twin lens kit. One would assume that these lenses are of a good quality, but how good ?.
Would I be better off buying the body only and buying a really good all round lens separate. I would like to use it for everyday use with a decent amount of zoom say 200 - 250 mm ? Would a lens of this type also be suited for wide field astrophotography.
The 450d seems very popular at the moment with proven results, should i go with this one or be looking at a higher model. There is a 40d second hand body only in the classifieds that looks good.
Your Thoughts

Cheers
Scott :D

[Geoff45]

I used a Canon 350 D with a 90-300mm zoom lens. Stars all had flares like coma (which it wasn't because they were the same all over the field). The standard 18-35 mm zoom lens worked fine. If I were doing it again, I'd go for the body only plus a good quality non-zoom lens.
Geoff

[bojan]

Kit lenses are not top quality, but good enough for normal everyday photography.
This is obvious on starfield photos where stars away from dead centre of the frame are not round and small at maximum aperture, so you have to close it down to f4 - f5.6 or even lower to achieve good result.. however that depends on individual criteria for "good" and of the purpose of taking images... Totally distorted star images are still good for photometry - which is quite serous work and could provide results of high scientific value.....

[rastis95]

Thanks for the replies.
Geoff, with your 90-300mm lens trouble, was that a canon lens or a different brand. Ive been doing a bit of research on the cameras itself but when it comes to the lenses its a bit of a Grey area.
General question, what would the main difference be between the 450d and the 40d. I know the 40d has a lower megapixel but the same size sensor (im assuming lower noise across the chip) besides this most functions seem the same.

Cheers
Scott:D

[Octane]

Scott,

Better build quality, and 14-bit over 12-bit imaging, which improves dynamic range.

I'd get the one in the classifieds. I'm seriously considering buying that and using it as a third backup camera.

If I were you, I'd go for either the 450D or a 40D ($800 to $1300), and buy the 200mm f/2.8L II USM for about $1,000.

Regards,
Humayun

[Geoff45]

It was a canon lens.

[robin]

I always get a smile when the canon v. nikon debate arises whether its which is better for astro photoraphy or general/professional work. From a pro's point of view, it would clearly seem that Canon has cornered the market & once one person gets good results & builds on it, word soon gets around that its the way to go & that brands popularity soars. Ive been using Nikon for 25 years & shoot professionally with 4 bodies in continuous rotation. I was eyeing off a Canon 20Da a while back for astrophotography & it didnt even concern me one iota that it wasnt a Nikon.Didnt make the purchase as my $$$ went on a trip to Europe but that wouldnt stop me from considering a canon body if it was the best choice for astro photography. For my professional work, wouldnt touch anything else but Nikon.

[rastis95]

Hi all again,

Thought someone might be interested in this. I was about to ask the question what the difference was between a full frame sensor and a APS-C sensor found in the canon series. Then i found this article:

http://www.robgalbraith.com/public_files/Canon_Full-Frame_CMOS_White_Paper.pdf

Bias towards canon of course, its a couple of years old but i think it contains tonnes of information for people considering buying a dslr, I still haven't gotten around to getting one myself yet but all in good time:whistle:
Hope it helps someone, filled in a few blanks for me

Cheers
Scott :D

[AlexN]

As a current user of both Nikon and Canon cameras, I will say that for terrestrial uses, they are very much on par, They each have their strengths and weaknesses for certain tasks, I find the Canon to be the strong winner for "action" photography, where as the Nikon excels in landscape/portrait still life photography. The Nikon has better colour reproduction straight out of the camera, although the canon images can be made to look identical with a little tweak here and there in photoshop..

Lenses (for astro and terrestrial use):
Canon lenses in my opinion are nicer. They are more solidly built, they generally focus just that weeee bit faster than the Nikon equivalent, and also, are usually more accurate when continiously autofocusing on moving objects.. The one exception would be the Nikkor 70-200 F/2.8 VR, which I think outdoes the Canon 70-200 F/2.8L IS by the slightest of slight margins..

Astro work.
Canons are supported a lot more by astro imaging software, and seem to be easier to integrate into an astro imaging setup without having techincal difficulties.. Programs like DSLR Focus/DSLR Shutter etc do support Nikon cameras (some models) however there is near on total support for the Canon models.

In the older models, the Canon's were a lot less noisy than the Nikon's. This is going back to the days of the 20D and the D70s, however nowadays, comparing, say the Canon 5D Mk2 and the Nikon D3 or the Canon 1D vs the D3x, I think you would be very hard pressed to find a winner...

The choice then falls back on "what are you most likey to use it for the most?" If its purely for astrophotography, Buy a Canon...

There are plenty of reviews and comparison tests to look at online, but dont take one reviews word as gospel, definitely read everything you can, as some reviews can be biased etc. Also, try to look past the older comparisons (ie 20D vs D70/D70s) whilst they are good articles for historical purposes, the results are no longer true in the Canon vs Nikon great debate..

I definitely love my Nikon, for what I use it for its brilliant, I definitely love my Canons also..

[Omaroo]

Excellent synopsis Alex - spot on. I, too, love both my Nikons (all four of them) and my Canon. As you say - I prefer the Nikon colour straight out of the camera for landscape and portraiture - the 350D comes no where near it - and I've tried all manner of white balance tweaks. For deep-space astro, the 350D wins for me because of the software available to drive it. I can use the Nikon for reasonably successful single-frame planetary but cannot get the same result on my Canon. Go figure.

[gregbradley]

Thanks Alex for that informative writeup.

My first astro camera was a Nikon D70 and it was very good. Just as a point of interest - the 6.3mp chip in that camera is a Sony and is the same as used in the Starlight express MXx25C and the QHY8. In those days the Nikon had bad amp glow and also did not have true RAW files (unprocessed saved data).

As far as I know (please correct me if I am wrong) that is still true today.
Nikon has its own file format called a NEF file and this has some smoothing run on the image before it is saved. This is done to reduce noise (astroimagers know you sometimes need to smooth areas of noise in the dim areas of an image to get a pleasing result).

This has the effect of reducing smaller. delicate, fine detail in an astro image. Exactly how much is debatable - perhaps not much at all but it is present.

This was and still is one of the major arguments against using a Nikon.

I know up to the Nikon 200D (I think that was its designation) the Canons were clearly superior. However later models I have no knowledge of and Alex does.

Having said all that I believe a QHY8 would be far superior to any DSLR and a smaller SBIG 2nd hand would also be far superior.

Dedicated CCD cameras have the following advantages/disadvantages:

1. High cost - disadvantage although QHY have brought the costs down a lot and are closer to a DSLR with a couple of lenses.
2. Cooled - far lower noise, a major advantage
3. 16bit processing - this means a larger dynamic range although the new DSLRS are 14bit and it could be argued that 16bit is not fully utilised in CCD cameras as the image data is only in a small part of the range it can differentiate in.
4. SBIGs have internal self guiding which means autoguiding is all in one package and you have to add on the cost of a good autoguiding system to a DSLR to compare costs (as a guide I'd say $1000 minimum for a decent autoguiding/guide scope/rings/mounting gear setup and easily could go to $2500 for this alone + the problems of flexure can enter where the guide cam and imaging camera are moving very slightly from each other and you get eggy stars).
5. filtered imaging means each pixel counts when doing LRGB imaging versus a DSLR where it has a coloured grid overlay of microfilters on the chip so 4 pixels are required to create one colour pixel in other words your 10 megapixel chip is only really 2.5 megapixels of colour.
6. Unmodified DSLRs are a lot less sensitive than dedicated mono astro cameras and modified cameras are also although less so. Sensitivity is typically
3 times greater for a cooled dedicated astro camera than a DSLR.
7. If you want to do narrowband imaging (Ha filter for example) it is ideally done with a dedicated astrocamera. You can use Ha with a DSLR but it is less sensitive to this and unmodified it may be a waste of time as the standard Canon/Nikon filter blocks that light. Modified cameras it is not.
8. Dedicated astro CCD needs a laptop and a battery or power supply. DSLR is more portable and does not absolutely require a laptop so this could be a big advantage.
9. DSLRs can be used for other photography - this is often appealling as well. However point and shoot cameras these days do an awesome job and an excellent one costs less than $100 now and they are a lot more portable and fit in your pocket.

So if it comes down to money a one shot colour QHY8 would actually not be much more at all.
A 2nd hand SBIG camera would also not be much more when you consider not needing a separate autoguiding setup (which you will need to do serious imaging).

Results will be much better with the dedicated astro-CCD.

I guess it comes down to how hard are you going to get into this hobby - mildly interested or full on? If full-on you will eventually want the dedicated astro camera so you could just start at that level and save yourself some cost of changing over later.

Greg.

[AlexN]

Well said Greg..

You are very correct, up until Nikon's D200, the Canon cameras were far superior.. These days its really a matter of personal taste more than a clear superiority...

I also could not agree more, that if someone was going to buy a Canon 40D/50D and get it modified, they may as well save that little bit extra and buy the QHY8. When used to its full potential, there really is no comparison between a DSLR and CCD for astro imaging. Yes, DSLR images can be great, and can have low noise. but how often do you see DSLR images where the photographer chose not to use Darks simply because they didnt deem it necessary?

I just got through posting in a thread on Cloudy Nights about this exact same topic, and the general gist of my post was that for astro photography, All roads eventually lead to a dedicated cooled CCD imager. be it mono or OSC, large format or otherwise, multi megapixel or small resolution... One way or another, every serious astro imager will eventually want a cooled CCD, and majority will convince whoever it is they have to convince in order to buy one..

Im currently looking for another CCD.... Horses for courses sort of thing.. :)

[jkrah]

for whats its worth.. if your into hacking / modding at all..

It seems to me like are more firmware hacks and mods available for the Canon.. I have a low end Nikon (D40) and have not been able to find any (firmware) hacks at all :(

.. but I'm pretty sure I even saw a SDK mentioned somewhere for the Canon.. I certainly got the impression Canon might be more 'hacker' friendly..

[Chippy]

Excellent info, and I agree with pretty much everything that's been said. Just a small point of clarification: for high ISO the noise levels are clearly in favour of Nikon over Canon at present (at the respective price points) and have been for some time. It's not a huge lead, but is quite measurable and definable, and to be honest its not really up for debate. Canon have definitely narrowed the gap with the 5DMarkII, when you take into account its higher resolution (not so with the 50D though).

Overall the Nikon D3 is the clear leader, with the D700 and D300 also VERY good, and better than their Canon counterparts - albeit with currently lower pixel counts. The D3X isn't so good, but at the price they have set, it can almost be ignored anyway ;-)

This of course doesn't change all the other various arguments for Canon being the better/more popular AP platform. But for high ISO noise Nikon wins.

[bojan]

Guys,
do you have some numbers to support your comments?
Because words like "good", "better" etc do not tell the real story, it is more individual impression one has about various product.
Also, we have to take into account price/performance ratio.
Only when data are presented in such a way, the decision about what is better (for the purpose) and by how much is clear cut.
Christian Buil's form of presentation is the way to go.

[Paul Haese]

This debate is like Holden and Ford or Ford and Holden. Both systems produce quality shots.

I am on my 8th or so Nikon, had them for years. I love the glass they make, bodies are another matter but one thing stands out in my mind in that regard. Nikon have the same mount as when they first started making DSLR's. I have old lenses that I still use from time to time.

In terms of terrestrial v Astro, I am one of the people that Mike mentioned who uses a Nikon for Terrestrial and a Canon for Astro. Why? Well that is simple. The Canon has been modified and cooled. No one seems to be doing this with a Nikon CMOS camera. If they were I would have gone with the Nikon.

[Omaroo]

Here's one from the other night using the D40 Scott. The image has been compressed lots, so it's a bit nasty here at 1260px.

Compared to the 350D it has a much better rear LCD and the menus are far better. Battery life is double. Noise? Nah - not really. A little tiny bit of amp noise past 600secs, but otherwise fine. The only annoyance is the lack of a cable remote - so I use an IR one with a timer on my watch.

[TrevorW]

In a lot of instances it's subjective what one photographer can acheive with a Canon may differ to another. Personally I've seen some really stunning stuff done with DSLR's in comparison too CCD camera's even on this site and in fact it's only when these CCD camera's are matched to and equally expensive scope with an equally experience photographer that IMO excelling results are acheived.

Even if we all like to aspire too obtaining a top of the range CCD cooled camera for a lot it's in the realm of Lotto win and we make do with what we have and try too maximise the final result.

IMO most of the people presenting arguments here do excellent work that is world class.