Nikon D200 or Canon 5D?

I recently had my hands on the Sea & Sea 5D housing, and the ergonomics seem very good (better than my Aquatica at first glance). It's being put through it's paces this week in the Bahamas, but it seems quite nice - the dials for shutter and aperture are reachable without removing your hand from the handle (although I have large hands) and overall balance seems very good.

Haven't tried the 5D, but I am enjoying my D200.

Wow, what's that goby? He is cool.

Alex

Hi All,

Wow!! Great discussion!

Thanks stu for your detailed post answering my main questions.

Thanks also to Alex, James, Karl and Jordi and everybody else for your contributions.

I think one thing that comes across loud and clear is that

that both cameras are very good. The evidence is there in the pics.

(Though the 5D has been around a little longer, so more pics I think.)

Key Factors

I think choosing between the D200 and 5D is a bit tricky because (as Stu and Alex noted) it involves more than just the camera body. It also a decision about choosing between:

a. Sensor size – i.e. Small Frame vs Full Frame, and

b. Nikon v Canon.

c. Housing availability.

a. Sensor Size – Small vs Full Frame

The benefits of FF are: (i) wider angle lenses available, (ii) more telephotos available.

The benefits of SF are smaller telephotos for equal power, good macro performance.

The D200 takes all F type lenses while the 5D takes EF lenses (not EF-S).

The key point coming across in the posts (Stu, James) is that the 5D

is probably better in low or natural light, since it has better noise reduction at high ISO settings. No one is coming out to vigorously defend the D200 performance in low/natural light, but presumably, (if not quite as good as the 5D) it is pretty darn good? Yes/No?

At http://www.photo.net/equipment/canon/20D_v...D_vs_5D_vs_D200

Oliver Gallen posted Mar 15, 2006 responded to Bob Atkin’s review comparing different cameras. Gallen said:

“A side-note, though: "Smaller, but high pixel count, sensors tend to be

noisier" and "the cameras discussed here represent higher resolution, lower

noise, options in the $1000-$3000 range". That's an approximation, at best.

Strictly speaking, a smaller sensor generates slightly LESS Noise...

It's because comparisons tend to be assumed under a same f-number that

the larger sensor gets a higher Signal, thus a better Signal/Noise Ratio. This

also means smaller/cheaper lenses for the smaller sensor (ie: a 200mm f/2.8

on APS costs less than a 300mm f/2.8 on 24x36). If you do that kind of

comparison, you can't factor out the lenses and consider the body alone.

That 'sensor centric' behaviour is widely spread and, unfortunately, also

convey some strong bias toward larger sensors.â€

Wondered what people make of this?

I notice that most of the best D200 photos on wetpixel so far are macro with flash. There are not many D200 “sunball†shots. See topic "5D Sunball, Best Ever" http://wetpixel.com/forums/index.php?showtopic=13035. I guess D200 users are unlikely to contribute to a topic “5D Sunball…†so that might explain it. These type of shots are good test of dynamic range so it would be great, if any D200 aficionados posted some “Sunball†shots. Perhaps, Alex will tell us more about the D200 when he gets back from the Red Sea.

Would I be correct to assume that most people shoot raw files and then adjust colour balance, saturation, contrast and sharpness in post production. If so that leaves mainly lense selection, filters and then speed, aperture and ISO to play around with in camera. Even noise reduction can be done PP. So does this mean that better in camera noise reduction is no big deal? i.e. if the same algorithms are available in camera or post, what is the difference?

b. Nikon v Canon

Bit hard to decide here. I get the impression that Canon is a bit like Microsoft and Nikon is a bit like Apple Mac. It might help to know whether professional photographers prefer one brand v the other? Although, if professionals get more freebies from one brand that might explain a preference. Also professional mode of use may vary from my intended use (i.e. wildlife, UW, etc). So I cannot at present see any easy way to resolve this question, short of taking note of the opinions of those who have used both.

c. Housings.

There will be plenty for the D200, not as many for the 5D (Aquatica, Sea & Sea, Seacam, Ikelite). One person was dissatisfied with the 5D housing selection available at present. Presumably, there may be one or two more 5D housings coming out. It will be difficult to check the ergonomics, so I will to some extent have to trust reviews.

Weather Proofing

I have read some conflicting reviews. One that say that the D200 is weather proofed, while the 5D (& other Canons) are not. Another review said that the Canons are also weather proofed. Since I intend using this camera outdoors (on boats, treks, jungle, travelling, UW etc) I wonder what the if the weather proofing is a really significant difference?

Built in Flash

Also the 5D does not have a built in flash. So I wondered, do 5D owners not use it topside (i.e. as a walk around camera), or do you just have to carry a portable flash? (Ok, I see that some people with a 5D use a different camera topside while others carry a portable flash.)

Both cameras are about the same weight so there is nothing to compare there.

James said that the 5D is probably better if you are going to use it as your main camera topside. That is the case for me.

D200 - Wireless Commander Mode

At http://www.photo.net/equipment/nikon/D200/ the “reviewer†talks about -

i-TTL Speedlight Technology. Anyone use this? Like it, not like it?

Other Discussions

Found this discussion comparing dSLR offerings, it’s probably worth a look for anyone following this thread:

http://www.photo.net/equipment/canon/20D_v...D_vs_5D_vs_D200

Speculation on Future Releases

This is a bit of crystal ball gazing (i.e. wild speculation), so skip this if you know better.

Given that the higher end digitals at Nikon and Canon are all full frame, would it be reasonable to assume that the medium entry camera models will converge towards the higher end, i.e. Full Frame? Hmmm, perhaps not, can’t see Canon or Nikon abandoning the Small Frame sensors. So they must intend to sustain the different streams.

One thing about the Canon choices at present is that there is a big jump between the 30D and 5D both in terms of price and the type of camera it is (FF v SF). The Mega-pixel difference 8 v 12.8 is not such a big deal and the functionality is fairly similar. Given the Nikon D200 sits in the middle, I am guessing that Nikon are taking market share from Canon for these types of cameras. (See for example, the number of people going for the D200 at wetpixel and other sites.)

If I was Cannon I would be thinking of plugging the leak caused by the D200. How would Canon go about this? Would they bring out say a 6D or 5DMkII and reduce the price of the 5D to compete more directly on price with the D200? But, the life cycle of the higher end cameras is longer, i.e. more than 12 months more like 18 months to 2 years. So anything new on this front would be at least

12+ months off, probably more??

Another way to compete would be to bring out a 10 or 12Megapix 40D, but since the 30D came out only 4 or 5 months ago a 40D is unlikely to come out before 7 or 8 months.

Why all this? Just want to avoid buying say a 5D one day before the 5DMkII comes out and the price of the 5D drops by 20%. In any case it this will eventually

happen to any electronic purchase, so no big deal.

Cheers

Matt

Hi Matt

To try to answer a few of your queries:

Nikon have traditionally been used underwater for historical reasons - the validity of these is now unimportant (other than being irritating - the Nikonos flash socket is such) but the legacy still lingers - hence why there will be more users of D200s than 5Ds on this site.

The 5D will work with the 60mm EFS macro lens if a 12mm Extension tube (MkII) is used - this gives a very usable and fast focus range but it is limited.

Very few professional photographers have switched from Canon to Nikon but in my experience many have switched to Canon from Nikon. This reflects the regard in which a FF sensor is held (rightly or wrongly sometimes) and the belief that larger is potentially better - especially in terms of noise. There are differences in image production though and I personally prefer the Canon's image rendition. However the D200 and 5D are still both superb camera and I'm not sure that a current choice would be sensible based on this sort of info! Freebies are not really relevant (and from what I gather few are offered) in determining people's likes/dislikes.

Wireless flash isn't relevant underwater!

And in my opinion it is not relevant to speculate on near future potentialities - the current D200 and 5D are both capable of producing stunningly good images NOW! Any upgrades are likely to feature nuances of improvement rather than dramatic steps forward.

Given that the higher end digitals at Nikon and Canon are all full frame, would it be reasonable to assume that the medium entry camera models will converge towards the higher end, i.e. Full Frame?

The higher-end Nikon digitals do not use full-frame sensors. Only Canon's 1Ds series and (I believe) Kodak's 14n use full-frame sensors.

Very few professional photographers have switched from Canon to Nikon but in my experience many have switched to Canon from Nikon.

Nikon to Canon, right?

~Matt Segal

The benefits of SF are smaller telephotos for equal power, good macro performance.

 

<{POST_SNAPBACK}>

This is not correct. We've had long discussions on this in the past. Macro shots typically bump against DOF limits and diffraction limits on resolution. For any given diffraction limited resolution the DOF on the FF will be better. I've posted an analysis here awhile back.

http://wetpixel.com/forums/index.php?act=S...t=20#entry80286

On telephoto, the comparison should be between a lens that is 1.5x longer and one stop higher for the FF vs cropped, and shoot the FF camera at one stop higher ISO. That will give you about the same DOF and field of view. I think the image quality of the 5D at one stop higher ISO is still ahead of the D200. It not clear which way the cost and the weight of the lens goes. Is a 200mm F/2.8 less expensive than a 300mm F/4 ....

Would I be correct to assume that most people shoot raw files and then adjust colour balance, saturation, contrast and sharpness in post production.  If so that leaves mainly lense selection, filters and then speed, aperture and ISO to play around with in camera.  Even noise reduction can be done PP.  So does this mean that better in camera noise reduction is no big deal? i.e. if the same algorithms are available in camera or post, what is the difference?

<{POST_SNAPBACK}>

Yes, most people shoot raw files and do post process noise reduction, but the difference in noise performance is not from the algorithm, although choice of algorithm can certainly make a difference. A bad algorithm can make things worse.

The real reduction in noise comes from physics. Bigger photosites generate better signal to noise. Here is how to understand it. The function of a photosensor is to convert photons into charge on the photosite. The charge is then collected, amplified, and then sent to an A/D converter. When you expose a sensor to a uniform illumination for a period of time, the number of electrons captured is a statistical process, which has fluctuations approximately equal to the square root of the number of electrons stored. If I double the intensity of light, on average, the number of electrons doubles, but the fluctuations increase only by sqrt(2), so the signal to noise ratio increases by 2/sqrt(2) = sqrt(2) ~= 1.4

The point is that as you make the photosites bigger, for a given illumination, you increase the number of electrons proportional to the area of the photosite. Therefore the signal to noise ratio increases proportional to the square root of the area, or directly proportional to the linear dimension of the photosite.

Of course, this assumes that we are limited by counting statistics, and not other sources of noise in the electronics. It also explains why at higher ISOs you get more noise. At higher ISO, you are underfilling each photosite (so you get fewer electrons on each site) and increasing the amplification before the A/D conversion to compensate. The amplification is nearly noiseless, but the raw signal to noise is reduced due to the smaller number of electrons.

You can simulate this easily with a coin. If you flip the coin 20 times (corresponding to a small photosite) you expect around 10 heads, but each trial (of 20 flips) has a statistical fluctuation around sqrt(10) ~= 3. That is to say some trials will be 10, some will be 12, some will be 9, etc, but the scatter will be around 3. This is like a S/N ratio of around 10/sqrt(10) ~= 3.2. If you flip the coin 200 times (corresponding to a big photosite), you expect around 100 heads with a fluctuation of around sqrt(100)=10. This gives a S/N ratio of 100/sqrt(100) = 10. The conclusion is that bigger photosites give bigger S/N ratios.

Note: For those who care, the above is an approximation. If you use the binomial distribution you can compute the exact values. For 20 trials S/N= 4.48 and for 200 trials S/N=14.14. Also the statistics of a coin are binomial and that of a photosensor are Poisson.

Sorry about the math.

Ok, thanks again.

Paul , thanks for your comments, the UW history argument (re Nikon) sounds reasonable. I knew someone would jump on my speculation, so fair enough. Actually, if something new came out from Canon I would still be looking at the 5D, not the “new thingâ€.

Also thanks Herb. I have read through some of the Macro posts you suggested. Looks interesting. As I see it (correct me if I am wrong) the conclusion reached was that with the right lenses the 5D is fine for Macro.

Hi Dr Steve, thanks for the S/N explanation, quite persuasive.

(The maths is no problem for me. B) )

BTW, I really like the Florida Springs Cave Shots on your website.

Anyway, you should put one of them in this weeks blue photo comp at wetpixel.

Are these caves similar to the freshwater caves in Yucatan? (Unfortunately,

I believe that the tour operators in Yucatan who have the rights to the site, won’t let people take cameras. )

Just one question on Digital ISO.

1. Not sure what you mean when you say that the photo sensors are not being “filled†at high ISO settings?

Also, I wondered whether you like your housings?

(I noticed Herb is using an aquatica and Matt Segal an Ikelite.)

Also, any strobe suggestions?

After the discussion on N/S ratios, I am now tending towards the 5D.

Now only if they could build in automatic composition …ha ha.

Cheers

Matt

PS.

I just noticed there is a new Nikon sunball forum topic.

Cheers

Matt

This seems like such a weird discussion to me. You are comparing two cameras with two very significant pricepoints. One would expect the 5D to be better quality at almost twice the price.

It seems to me the main consideration should be how much you want to spend, and if you are prepared to spend the amount of the 5D, is it worth the difference.

If you are prepared to spend the money, then it seems that the 5D should be compared with the full frame Nikons.

I love my D200 so far, and frankly with about $13,000 invested in cameras, lenses, housings, strobes etc, that's enough money for me.

If you are prepared to spend the money, then it seems that the 5D should be compared with the full frame Nikons.

 

<{POST_SNAPBACK}>

The problem is this thing does not exist.

About $1.5k difference in the camera body is not a huge percentage of the cost when you're buying a whole setup from scratch. It's not even clear the Canon system would cost more than a comparable Nikon system when all the costs are added up.

Nikon 2DX, Canon 1D are probably the more appropriate comparisons. Yes, I know they cost more, but if you're only going to be satisfied with the very best even though it will not be evident in more than 90% of your picture taking activities, then maybe shelling out yet another $1500 or so more, is the way to go, and your equipment envy problems will be over till the new models come out.

Just to throw another wrench into the equation...

It's worth considering that a cropped sensor camera (such as the D200) is less sensitive to the lens quality (since most "cheaper" lenses fall off at the corners) compared to a full-frame sensor camera (such as the 5D) which is much more sensitive to the quality of the lenses that you use. I've found that the L series lenses are pretty much the only ones that I'm using on my 5D, while a cropped sensor may allow you to get away with cheaper glass.

I agonized over this decision for a while and eventually settled on the D200, at least partly because I already owned some Nikon glass. A few points:

Don't stress - both are nice cameras.

I don't see much difference in noise between the D200 and 5D at ISO 400 and below. The 5D is a little better but not so much that it mattered to me. Since I don't really see a need to shoot above ISO 400 underwater noise wasn't much of an issue for me. Even in very bad vis here in California where it can be quite dark at depth I've been able to get a nice background exposure at ISO 200. Since the foreground on most shots is lit by the strobes and their instantaenous light I don't see a need for a really fast shutter speed underwater unless you're shooting natural light. Also if it's murky I'll often want to shoot wide open for backscatter control and that further reduces the need for high ISOs. If you are shooting natural light it's most likely going to be in good vis with good light anyway. If for some reason you need shoot ISO 800 or 1600 then the 5D is a clear winner.

At least with the Subal housings I like that I can use the multi-controller underwater. My friend's 5D housing doesn't offer access to the joystick thing on the back of his camera making review of photos at 1:1 pixels impossible. Other housings may give access to these controls on a 5D but it looks like it's easier on the D200.

The multiple memory banks are actually useful in that you can have one bank of settings for topside (eg TTL) and one for UW (eg manual strobes). Reduces the likelihood that you forget to set something before you jump in.

Sunballs are an interesting topic. Hopefully at some point my friend and I can shoot some comparison shots. DPReview seems to think that total dynamic range is similar with a slight edge to the 5D and also that the 5D has a somewhat softer roll off to clipped highlights. In theory this means better sunballs. I'll report back once I get some good side by sides.

Regarding lenses I don't see much of an advantage one way or the other. If anything the 5D is slightly handicapped by not having something like what I get with my 16mm fisheye. A similar lens on the 5D is going to be 180 degrees. I can shoot either way with the 10.5mm or the 16mm. Of course you can always crop the 5D and still have an 8mpix image. :-)

One downside to the D200/Subal combination is that I'm still waiting for a port for the 105AF-S.

Clinton

Hi Dr Steve, thanks for the S/N explanation, quite persuasive.

(The maths is no problem for me.  B)  )

BTW, I really like the Florida Springs Cave Shots on your website. 

Anyway, you should put one of them in this weeks blue photo comp at wetpixel.

Are these caves similar to the freshwater caves in Yucatan?  (Unfortunately,

I believe that the tour operators in Yucatan who have the rights to the site, won’t let people take cameras.  )

 

Just one question on Digital ISO.

1. Not sure what you mean when you say that the photo sensors are not being “filled†at high ISO settings?

 

Also, I wondered whether you like your housings?

(I noticed Herb is using an aquatica and Matt Segal an Ikelite.)

Also, any strobe suggestions?

<{POST_SNAPBACK}>

I haven't been diving in the Yucatan Cenotes, but from what I understand of the geology, they are very similar. In the Yucatan (and I think in Florida) there is a thick layer of limestone above a bedrock layer of granite. The surface water soaks into the ground. The water is slightly acidic but cannot penetrate the bedrock so it slowly dissolves the limestone until it is basically a swiss cheese sponge filled with caves and underground rivers. In various places the ground falls away into a sinkhole. The water is crystal clear assuming it hasn't been contaminated by organic matter (or pollution from urban run off). The caves range from tame cavern dives, like the Blue Grotto or Devils Den, to extreme technical dives in tunnel systems that go for miles. Sorry that this was a little off topic.

About the ISO. Each photosite has a maximum number of electrons that it can hold before it saturates. I don't know what the number is for the 5D pixels (if someone does know please point me to a link), but imagine that it is 100,000 electrons. This charge is collected, amplified, and then sent to a ADC. At ISO 100, the gain on the amplifier is chosen so that 100,000 corresponds to the maximum input of the ADC and 0 is the minimum. If the intensity is low, and the photosites have less than 25,000 electrons in them, you can use a x4 amplifier before the ADC to map 0-25,000 into the ADC range. This corresponds to ISO 400. Notice that if you now have one photosite that has more than 25,000 electrons, you haven't saturated the photosite, but after the amplifier you have saturated the ADC, and that pixel will be blown out.

About the Housing. I ordered it, but haven't gotten it yet. A couple of days after I placed the order I got a chance to play with the Sea and Sea housing. Both housings are feel pretty good in the hands, but I will stay with my Aquatica. I'll let you know when I baptize it.

I did NOT struggle with this decision because almost all the friends I have and most of the guys who willingly answer questions shoot Nikon.

My analysis was that Canon 5D was probably a better camera but the difference was insignificent underwater. The $1000 I saved went into glass.

Did I make the right choice? Yes Either is right IMHO.

Tom

"a cropped sensor camera (such as the D200) is less sensitive to the lens quality (since most "cheaper" lenses fall off at the corners) compared to a full-frame sensor camera"

Not really so! In my opinion this is a bit of an 'urban myth'.

If you compare like with like you will find that there are no fast wide-angle lenses currently available for cropped sensor cameras. Whilst this will hopefully change (and probably FF lenses will be redesigned too) currently comparable lenses (such as the macros) are exceedingly good on either format.

Lens quality is a much discussed (in terms of technicalities) topic which all too often lacks real world empirical evidence. Most of today's 'good' lenses are very good - I've never lost an image sale due to optical constraints imposed by a lens!

Having discussed this topic with a lens designer I can say that his view was that both FF and cropped sensor lenses have issues associated with both design and production. The tolerances required for cropped sensor design may well limit the final MPixel sensor and the issues of producing appropriate retrofocus ultra wide (weitwinkel) lenses for FF use may mean that they might prove to be very expensive.

Of course if you really MUST use "cheap" lenses on high MPixel cameras then they will probably be better centrally than at the edges.

If you are considering either a D200 or 5D, I think you really need to look at the practicality of what lenses are available rather than consider them in esoteric terms. Both are good cameras - but they do have differing lenses available for them.

Pgk-

My understanding is that the 14-28 Nikkor DX very good for underwater wide angle. is this incorrect?

Tom

Tom, are you talking about the Nikkor 12-24 DX?

I own the lens and only use it when I'm shooting sharks and/or larger fish in a WA context. This lens must be shot at f8 to keep the corners sharp...IMHO.

For pure WA, I prefer the 10.5mm or Sigma 15

Karl