From Canon to Nikon (D200)

Purchase the eBook before you make any decision on the D200. He illustrates all of the features and detractors of the D200. He handles the high ISO issues very well and gives you ways of correcting this problem. D200 high ISO noise is rather unique. Since most of it is in the luminance channel, it makes it easier to get rid of it.

I am reading it from page one even though I got my first Nikon SLR in 1965 and have been with Nikon digital since the early CoolPix days. I've never read one of this eBooks so even though the early chapters contain information that I already know, he does drop hints and tips in those areas that I don't want to miss.

--
Cliff
 
...to highlight saturation. With Canons, there is a shoulder
similar to what you got with film, though obviously more shallow,
The problem with shoulders (and toes, for that matter) is that they produce a different micro-contrast in each region of the data. In the film days, for example, Ansel Adams used to warn against putting any detail outside the "linear" portion of the film curve because of that. For example, the traditional rule of thumb is to put the bottom shadow detail at Zone III, though some put it at IV. The way micro-contrast issues were dealt with in film was through burning and dodging--it was the norm to use such techniques to give highlights or shadow areas a different contrast than the midtones. Getting that "right" was the difference between Ansel Adams and Joe Consumer.

Computers give us different ways to deal with micro-contrast. In essence, you have to select data range values and manipulate them differently. Some do this using Curves, but there are dozens of techniques that can be used. But the problem is still that there is a difference between a Scott Kelby and Joe Consumer. And isolating the area in which you want to make the contrast adjustment often takes complex Curves (a single or two point adjustment means that some of your contrast change will get applied to intermediary data, not isolated just to highlights).

Further, I'm not sure that Nikon's early flat approach or Canon's humped approach to linearity is any better than the other. If anything, I'd say the opposite of what everyone else is saying: that it is the shadow areas that are the problem, not the highlights. Why? Because the way we set exposure now is to expose right up to the limit, but not over it. Unlike film, where the non-linearity tended to give you a little "headroom" (especially with print film, where overexposure is just increasing density and reciprocity failure starts to produce the flattening of response), we capture all the highlight detail and then put up with what's happening in the shadows. I personally haven't found any substantative difference in the highlights between a Nikon and Canon, and indeed tend to prefer the linear highlights of the Nikons because it's easy to control micro-contrast on.

But the problem is then at the other end, the shadows. Indeed, this is where a Canon user should be arguing that lower noise levels are indeed something useful. What happens is that, when my highlights are not blown out, the critical aspect of dynamic range for my digital image is ALWAYS at the lower end of the response. At the base ISO I usually shoot at, this isn't an issue, but at high ISO values it can be, as the shadow details then start to muddy up as noise intrudes on the data. Moreover, as I noted in my newsletter, the number of bits used to record the data in the shadows is also a problem, as we don't have as much differentiation for each actual light value difference.

So, I'm 180 degrees opposed to the views being expressed in this thread. Highlights aren't any different between the two. Shadow areas are where you'll see the difference.

--
Thom Hogan
author, Nikon Field Guide & Nikon Flash Guide
editor, Nikon DSLR Report
author, Complete Guides: D50, D70, D100, D1 series, D2h, D2x, S2 Pro
http://www.bythom.com
 
So this is were Canon have a an adventage over Nikon if I understand you right. As far as I know Nikon produce more noise in shadow than Canon.

It is hard to make perfect exposure with changing EV with Nikon becouse Canon is more sensetive and you have let say more room to adujst. If you put +0.7EV on Nikon you lose speed and also possibility to shoot from arm. Let say that you use 1600 and don't have monopod or something...
So, I'm 180 degrees opposed to the views being expressed in this
thread. Highlights aren't any different between the two. Shadow
areas are where you'll see the difference.

--
Thom Hogan
author, Nikon Field Guide & Nikon Flash Guide
editor, Nikon DSLR Report
author, Complete Guides: D50, D70, D100, D1 series, D2h, D2x, S2 Pro
http://www.bythom.com
 
Please tell me how. I tried many Canon and Nikon pictures to clean up with Noise ninja and I found very hard to clean it out. I mean you can clean it out but you lose details much faster than on Canon image.
Since most of it is in the luminance channel, it makes it easier to
get rid of it.

Cliff
 
Fortunately I don't take photographs indoors without artifical light (including sports). Thom wanted to see what would happen if he tried to shoot a basketball game with available light. He used a D200 for the event and set it at H1.0. On Page 253 of D200 he talks about what could happen and how to mitigate the noise. When she showed the 175% crop, it was pretty noisy but I didn't notice a lot of color noise just a lot of grain like noise (except it was white not black). After applying Neat Image, the noise was virtually gone. I was impressed. If I ever get caught where I have to go above ISO 400, I'll try out Thom's technique.

I highly recommend Thom's D200 book. You might get Thom to explain further his technique.

--
Cliff
 
As being a former 1D owner, I would like you to comment on an article I read in the Nikonians. According to them, a full frame digital sensor would experience a gradual decline of captured photons because of the way digital sensors are designed and the way conventional lenses deliver the photons to the photo site. What this would result in is an image that gradually fades towards the extreme edges of the sensor. The effect is a slight vignette. Have you ever noticed this on a 1D using conventional lenses?

The solution to this issue is a lens that could collimate the entire photo site.

Thanks

--
Cliff
 
I quite agree with you. I have seen this myself with my digital camera experience but what is the solution? Overexpose to bring in the shadows at the expense of the highlights? If you are shooting still life and landscape (which I don't) you could take two images, one to cover the highlights and the other to cover the shadows. Merge them with Photoshop and now you have nine or more zones. With the five stop limitation of the modern DSLR, what can we do? Graduated filters? Lower contrast by using artifical lighting or a reflector?

Thanks

--
Cliff
 
Hi Nikon people

I come from the Canon side, and have been using a Canon 1Ds and
various L zooms. While I loved the 1Ds' image quality, resolution
and build quality I never really cared much for Full Frame or
whatever buzz is beeing pulled on that account. Besides, a 1Ds does
balance heavy lenses quite nicely, but carrying the thing around is
cumbersome and I have for a long time been throwing envies looks to
the D200.
Now things got a whole lot easier since my 1Ds and most lenses was
lifted from my car :( Annoying, but fortunately for me my insurance
will cover.
So on the bright side , I am now "a free man" , and I can choose to
stay with Canon or jump ship, if that is what will work best for me.

The thing is, I want that D200. It is so close to beeing my dream
camera in terms of size, weight, build quality, "feel",
features... even the popup flash I'll much rather have than not. I
have been toying a lot with one at the local shop, I even like the
viewfinder although it does lack compared to the Canon FF's.
BUT ... high ISO performance does worry me. 1Ds was considered one
of Canons weekest models when it comes to high ISO performance, and
it topped out at 1250. My particular 1Ds was maybe better than the
average 1Ds (it was also one of the last bodies produced), and I
used it whenever I had to at 800 and even 1250 without hesitation.
For small prints I could even pull it a stop or two, to get the
equivalent of around ISO3200, although chroma noise would be quite
overwhelming and usually I would apply Noise Ninja and just live
with the reduction in details.
I guess this is the one area that Canon 5D would really shine - the
question remains , does the D200 advantages make up for this one
advantage of the 5D.
Does anyone have some samples of what the D200 is capable of at
ISO1600/3200 ? And what are your experinces of making prints of
high ISO pictures ? How much noise am I going to see at say , 10 x
8 ?
For what its worth , I dont care much for pixel peeping at 100% but
I'd like at least usable ISO1600 prints at normal-large print size.

Now another thing I have considered is my lens options. On the 1Ds
I used the Canon 24-70/2.8L as my walk around. I tried that lens on
a 5D once, and didnt care much for it. Without battery grip it is
too heavy. So should I decide on the Canon 5D my new walk around
would be the 24-105/4L. I guess on Nikon D200 the one lens in this
range to go for would be the 17-55/2.8. Thats already one full stop
advantage to the Nikon in this setup + I dont have to worry so much
about DOF restrictions of the FF 5D.´
Does anyone knows how the 17-55/2.8 compares to the Canon 24-105/4L
in regards to sharpness, color, contrast ?
On a side note, the 17-55 will cost me the double of the 24-105 and
it will eat up the price difference on the bodies... though I'll
rather spend my money on glass than on bodies that will be
obsoloted in 2-3 years.

One last question, mostly to the people who have handled and used
both cameras... I'd like to hear any observations you made in
regards to IQ, auto focus performance etc.

NB: I am shooting mostly still life and landscape stuff. But I
enjoy playing with long exposures, night photography and occasional
portraits and family snapshots too (small, fast moving kids...).
 
where can I send you image that you try to remove noise like you mention...

Thanks!
Fortunately I don't take photographs indoors without artifical
light (including sports). Thom wanted to see what would happen if
he tried to shoot a basketball game with available light. He used
a D200 for the event and set it at H1.0. On Page 253 of D200 he
talks about what could happen and how to mitigate the noise. When
she showed the 175% crop, it was pretty noisy but I didn't notice a
lot of color noise just a lot of grain like noise (except it was
white not black). After applying Neat Image, the noise was
virtually gone. I was impressed. If I ever get caught where I
have to go above ISO 400, I'll try out Thom's technique.

I highly recommend Thom's D200 book. You might get Thom to explain
further his technique.

--
Cliff
 
Neat Image is a superb product which you can try out before you buy. It works on other image raw files-I have a Leica D-Lux 2 which I used while photographing a nuclear bunker in Berlin last month. No way could you shoot lower ISO. Neat Image does a superb job of smoothing out and removing any noise. Never needed to use on D200 raw files as yet. Once you master settings which suit you-I am beginning to produce files which need a negligible amount of PP.

http://www.pbase.com/avsearle_101/root
 
Thom,

Would you consider writing and selling a book on the Canon 5D? I would -- and I"m sure others, would -- buy it in a heart beat.

There are no aftermarket good books on Canon cameras. I know you know Nikon's system by heart, and "think" Nikon, but I think it would be a great service to the photo community, and profitable, too.

I don't mean to impose; thank you

--
Ray

http://www.pbase.com/raylander
 
So this is were Canon have a an adventage over Nikon if I
understand you right. As far as I know Nikon produce more noise in
shadow than Canon.
I always have a problem with such statements. What does "more" mean? Are you referring to Std Dev tests of a fixed gray target? Even back in the film days, when you could measure the granularity (size) of the grain, there were higher grain films that I preferred over lower grain ones. But high ISO results on digital cameras aren't just about noise. There's also color saturation, edge detail, and other issues that come into play. Currently you don't see any real measurements of those.
It is hard to make perfect exposure with changing EV with Nikon
becouse Canon is more sensetive and you have let say more room to
adujst.
I've read that sentence a few times now and still have no idea what you really mean. A "perfect exposure" in any given situation will vary with every digital camera (or film) you use. That's why I'm a firm advocate of learning what the capabilities of your equipment and software are and maximizing them instead of shooting at defaults and running default conversions. Put another way, I know plenty of Nikon users who get "perfect exposures" and plenty of Canon users who don't. And vice versa. But it's not the camera or the sensor that's the issue.
If you put +0.7EV on Nikon you lose speed and also
possibility to shoot from arm.
Again, I've read that a few times and don't fully understand what you're trying to say. I think you're trying to say that Canon understates its ISO values (which, by the way, defeats the purpose of a standards body in the first place).

--
Thom Hogan
author, Nikon Field Guide & Nikon Flash Guide
editor, Nikon DSLR Report
author, Complete Guides: D50, D70, D100, D1 series, D2h, D2x, S2 Pro
http://www.bythom.com
 
If the linear curve has a shoulder and one of the color channels
reaches it earlier; doesn't that mean as you try to recover that
detail by compensating the exposure latitude when you open up the
file that you will get a color shift?
Any change in linearity between color channels would be a bit of a problem, regardless of whether there were toes and shoulders. But surprisingly, I've never really seen a significant one in any sensor to date. So, there is no problem with Canon's approach as long as they've matched the channel response.

--
Thom Hogan
author, Nikon Field Guide & Nikon Flash Guide
editor, Nikon DSLR Report
author, Complete Guides: D50, D70, D100, D1 series, D2h, D2x, S2 Pro
http://www.bythom.com
 
I quite agree with you. I have seen this myself with my digital
camera experience but what is the solution?
The solution is no different than it was with film. I'll use Velvia and Provia F as examples, as those were my primary shooting stocks for some time and I'm quite familiar with them.

One primary difference between Velvia and Provia is the shadow response (hey, that's just like what I'm saying about Canon and Nikon! ; ). Velvia has less below-middle-gray response than Provia, and an abrupt and complete drop to absolute black at the bottom of the shadow range. Provia F is more like other film stocks and has a modest toe at the bottom of the shadows (coupled with its finer grain, this makes for good recovery of deep shadow detail, which is why I often shot it). But the bottom line is that I knew what my film did.

Which brings us to the answer: you do everything in the field you can to optimize the data you capture for the results you want. For me, that means contrast control in the field, with graduated NDs, polarizers, fill flash, diffusers, reflectors, and whatever else I can think of to keep my data optimized. I have been known to sometimes invert my logic because of the results I want. For example, I've taken a graduated ND and flipped it to take a shadow area completely to black (ala Velvia). Multiple stacked exposures are another possibility to maximize data, though they require a great deal of skill to optimize in software.

In short: if you didn't capture it, you won't have it in your output. It makes no difference whether you're a Canon or Nikon user. You STILL have to understand what your equipment is capable of and maximize your use of it. The kinds of minor gains one digital camera has over another that most people are arguing about are trivial, quite frankly. You're better off spending your time figuring out what your equipment can and can't do and maximizing the use of that.

--
Thom Hogan
author, Nikon Field Guide & Nikon Flash Guide
editor, Nikon DSLR Report
author, Complete Guides: D50, D70, D100, D1 series, D2h, D2x, S2 Pro
http://www.bythom.com
 
Velvia is Nikon then and Provia is Canon :)
One primary difference between Velvia and Provia is the shadow
response (hey, that's just like what I'm saying about Canon and
Nikon! ; ). Velvia has less below-middle-gray response than Provia,
and an abrupt and complete drop to absolute black at the bottom of
the shadow range. Provia F is more like other film stocks and has a
modest toe at the bottom of the shadows (coupled with its finer
grain, this makes for good recovery of deep shadow detail, which is
why I often shot it). But the bottom line is that I knew what my
film did.
 
Would you consider writing and selling a book on the Canon 5D? I
would -- and I"m sure others, would -- buy it in a heart beat.

There are no aftermarket good books on Canon cameras. I know you
know Nikon's system by heart, and "think" Nikon, but I think it
would be a great service to the photo community, and profitable,
too.
I've considered doing Canon books. But not at the expense of keeping my Nikon efforts up to date, and quite frankly, my travel schedule last year (26 weeks out of 52) put me behind on most of my Nikon projects.

--
Thom Hogan
author, Nikon Field Guide & Nikon Flash Guide
editor, Nikon DSLR Report
author, Complete Guides: D50, D70, D100, D1 series, D2h, D2x, S2 Pro
http://www.bythom.com
 
That's your wishful thinking. It's well-known that Canon DSLR clips
highlights so very often that people always put a -ve EC to
safeguard blown highlights. Canon sensor has very poor highlight
retention, quite the opposite of what you wish for, sorry.
then why do I alway overexpose +2/3 stops (If I can) KNOWING I can get full recovery without "blown" out highlights?

Steven

--
---
Spring 2006:
http://www.pbase.com/snoyes/images_spring_2006

 
They show how to take two shots of a landscape where you don't have a graduated ND filter and combine them to create the final image.
 
You can't. What I suggest you do is to purchase a copy of the D200 manual (byThom), read it through, and experement. I'm almost half way through.

thanks

--
Cliff
 

Keyboard shortcuts

Back
Top