Canon v Nikon, astro imaging

[salmja79]

Hi,

I've been into astronomy for about 12 months and i'm looking to take my first tentative steps into astro-imaging. Recently I converted a webcam and have started to get some good results but i'm looking for something with a little more clout! I've been looking at a number of astro images taken with DSLR cameras in magazines such as the Sky at Night and find that most seem to be taken with a Canon camera and very few images are taken using a Nikon. Is there any reason for this?

I ask as i'm looking to get a DSLR and was thinking of of purchasing the Canon 450D.

Regards

Sal

[Beyond_Vision]

There is more support in astro software and hardware for the Canon range of camera's. They also tend to be less noisey than thier Nikon counterparts but the gaps is closing.

[blinky]

Does the Nikon not have some kind of auto noise reduction technique that applies even to RAW frames - thus removing some of the faint stuff??

[Beyond_Vision]

Yes Craig the Raw images are not actually 'raw' as you say they have noise reduction applied which can mean loss of detail. I think this can be overcome by switching the camera off then on during the noise reduction process, which is a bit inconvenient.

[salmja79]

Thanks for this info.

Sal

[ollypenrice]

Hi Sal,

You know that a standard DSLR is largely blind to H alpha? This limits the use of a standard camera to things other than emission nebulae. Many of the images you see are taken with de-filtered or re-filtered DSLRs. Just checking. You can find out about this if you Google 'Astronomiser.' They do it quite cheaply and expertly.

You will always see plenty of DSLR images in magaines becasue the large chip suits their print requirements... and because they are usually good images, of course.

Olly

[brianb]

Does the Nikon not have some kind of auto noise reduction technique that applies even to RAW frames

Same applies to the newer Canons. Also the super high pixel counts of the newer models are more of a nuisance than a help .... bigger files, slower to process, no more resolution in the final image which is limited by the lens anyway.

IMO the Canons with DIGIC III processors are the best for astro work.

Nikon aren't learning so much as Canon are going backwards.... the latest fad is superhigh ISO settings, meaning very aggressive in-camera noise reduction ... fact of the matter is, the sensors work best at their native speed of around ISO 200. This hasn't changed, and won't do so, the detection of photons is already as efficient as it can get - if you want a smoother image, you either capture more photons (longer exposure) or process the detail out of the image.

[dph1nm]

Same applies to the newer Canons

Do you have any evidence/reference for this? I have not seen any suggestions that Canon clip their stars in the way Nikon do (did?). In fact does anyone know if the very latest Nikons still do this?

NigelM

[zork]

since the subject is here... is is possible to use a nikon d40x for astroimaging? my wife got one and i want to use that one for taking pics when i get a telescope.

[jgs001]

I can't comment on the Nikon's but the 450d has two types of NR... ICNR (which is full dark subtraction) and a High ISO Noise Reduction. I think it's this later one that is being referred to. It is thankfully, a switchable option in the menu.

[dark knight]

The Nikon D80 does indeed have NR, which quite rightly as John says can be switched off and controlled via the menu. High ISO noise reduction only becomes an automatic feature above ISO 800.

[brianb]

Same applies to the newer Canons

Do you have any evidence/reference for this? I have not seen any suggestions that Canon clip their stars in the way Nikon do (did?).

The critical test is what happens when the focus is slightly off. With a true "raw" the limiting magnitude will drop off very rapidly with a small change from perfect focus. With the newer Canons, the limiting magnitude drop is small ... another way of putting this is that the response at the bottom end of the curve varies with blob size, which is an indication of processing in the camera.

For astro work, the processing you need to do is far better done outside the camera & after stacking, so the rawer the image is, the better.

[jgs001]

The Nikon D80 does indeed have NR, which quite rightly as John says can be switched off and controlled via the menu. High ISO noise reduction only becomes an automatic feature above ISO 800.

I guess the Canon slightly beats the Nikon there then, as the High ISO nr is optional at all ISO settings up to 1600.

[Beyond_Vision]

The critical test is what happens when the focus is slightly off. With a true "raw" the limiting magnitude will drop off very rapidly with a small change from perfect focus. With the newer Canons, the limiting magnitude drop is small ... another way of putting this is that the response at the bottom end of the curve varies with blob size, which is an indication of processing in the camera.

For astro work, the processing you need to do is far better done outside the camera & after stacking, so the rawer the image is, the better.

Are these tests you have carried out or can you provide a reference to them. As far as I'm aware the noise reduction application is a menu option. It is on by default but can be turned off for astro work

[dark knight]

If the camera sensors/ccd best speed is as Brianb says at ISO 200, noise reduction will not take place to the Nikon Raw files provided the noise reduction is turned off in the menu. Basically if you use a Nikon just keep at ISO 800 or lower and no noise/detail will be lost due to the camera's noise reduction system.

[skye at night]

I am at heart a Nikon Man.....

Partly due to the D70 being the first dslr I purchased and then ended up with a nice set of lenses...

What you need to remember is that 'we' are looking at these camera brands in terms of our very 'niche' application....

Neither Canon or Nikon really ever thought too much about the needs of astrophotography users.. they are thinking about daytime and low light (hence much of the high iso trend) use and putting their product into as many pro, semi pro and casual consumer hands as possible....

If you look at the 'top end' cameras the two companies are continually 'one-upping' each other with the release of new D3's D3x's , 1Ds Mk3's, 1D4's etc at eyewatering prices... In fact they are at it throughout the range with new models from the bottom end up and bringing in new features on cameras often to some disgust from those who bought the previous or another model (much talk of 7d v 5d at present) . Few Pro Photographers seem to run Canon and Nikon kit but you do hear of them completely changing kit from one to another just because of a new body or lens that is perfect for them... Crazy..

The sale of one body could result in the sale of a much greater value in lenses should the buyer become even a slightly keen enthusiast...

I think it has been by luck rather than design that Canon cameras have lent themselves better to our needs....

The noise of my D70 on even short night-time shots drove me mad as did turning off the power to 'avoid the NR process...

It prompted me to buy a Canon 20Da - which IS Canon's one short lived 'nod' towards astrophotography and mainly due to feedback from early days japanese astrophotographers. It is a camera I will keep and use until it either stops working or falls apart.

I still love the build and quality of Nikon gear and my decision one Birthday to give my wife a Nikon D40 (fantastic little camera) was in part a desire to keep Nikon in the family and make use of a couple of the good lenses I held on to...

I now have some really decent Canon lenses so will be sticking there for now..

The real Holy Grail for me (daytime anyway) would be a Nikon body that could take Canon L lenses and still autofocus... Not going to happen due to sensor spacing ...

What if one day they just agreed on a common lens fit and went from there..

Hmmm..

Steve

[brianb]

Are these tests you have carried out

Yes. I've done quite a lot of experimentation trying to get precision photometry out of my Canon DSLRs.

[Beyond_Vision]

Yes. I've done quite a lot of experimentation trying to get precision photometry out of my Canon DSLRs.

What Canon cameras models are you refering to ?

[salmja79]

Thanks all for the info, it's been very helpful

Sal

[brianb]

What Canon cameras models are you refering to ?

5D, 40D. The 5D has problems with dust on the sensor, the essential problem being that it moves around with shutter & mirror action & is hard to flat-field out. The cleaner in the 40D is a big benefit. A lot less vignetting with the smaller sensor, too.

Under good conditions I can now get measurements to ~0.003 magnitudes though 0.05 to 0.01 magnitudes is more typical.

[Beyond_Vision]

I found this article on the 50D. Quite an interesting read and it states that there is no noise reduction applied to the raw image. It also looks like the Nikon D90 has less noise at higher ISO settings.

Canon EOS 50D Digital Camera Hi_iso_nr - Full Review - The Imaging Resource!

Regards

Kevin

[dph1nm]

None of these photographic reviews are really any use to astronomy.

The problem with the Nikons is/was median filtering being applied to the RAW files. This is quite independent of any of the noise-reduction settings the user can tinker with. See here

NIKON D3 / D300 ET ASTRONOMIE

As far as I know this has only been spotted by astronomers, so it rarely, if ever, gets discussed in regular photography channels.

It is commonly said on this forum, and others, that Canons do not suffer from problems like this, and produce true RAW data - and hence are best for astro work. If this is no longer the case in the newer models I think we should be told!

NigelM

[skye at night]

It is commonly said on this forum, and others, that Canons do not suffer from problems like this, and produce true RAW data - and hence are best for astro work. If this is no longer the case in the newer models I think we should be told!

NigelM

I suppose it again comes down to US being a very niche market and this not being something that they are concerned with..

ie... they are about cutting out noise esp in low light high iso conditions with a view to not using flash...

Steve

[George]

If your using purely for astro use the 1000D is cheaper than the 450D.

[salmja79]

I'm after a camera that can do both Astro-imaging plus the usual weddings/Bar Mitzvahs/Holidays etc etc. I think i'm going to take the plunge and get the 450D as it's in the right price range for me.