Gary’s Parries 20/08/06

August 20, 2006 | Mark Goldstein | Gary's Parries | 17 Comments | |

Gary's ParriesThis week’s Gary’s Parries topics are:

1. Nikon D80 Got The Shakes
2. Digital Versus Film Cameras In Baghdad
3. Once A Photographer, Always A Photographer
4. Five Megapixel Images Look The Best

Introducing this week’s Gary’s Parries column. Everything you always wanted to know about digital cameras, but were afraid to ask. No question too difficult, or too easy. As a Senior Principal Software Engineer, and a former Assistant Professor of Computer Information Systems, as well as a recording studio owner/operator, inventor, and now, a digital camera enthusiast, GARY has more digital camera knowledge in his entire brain than most people have in their little finger. In the unlikely event that GARY would not know the answer to your question, he will answer it anyway, true to the spirit of the word “Parries”, a fencing term which, in this context, implies “cleverly evasive answers”. So let your imagination run wild. Email all your nagging digital camera questions to: , and then, En Garde!

You may also attach to your email an ORIGINAL PHOTO of your choosing. A preview of the photo will be displayed with your question, and a full-sized version will be just a click away. No personal information will be published with your question unless you specifically include it in the text or attached photo of your email, which may be further edited for grammar, content, or other reasons.



The new Nikon D80 digital SLR doesn’t have anti-shake. Neither, for that matter, does Panasonic’s first digital SLR, the Lumix DMC-L1. Why do you think these manufacturers have chosen to ignore such a great technology?

Regards (en gardes?),
Ben Rothfeld

*** ANSWER 1

Good question, Ben. I have no idea of the answer, but that has never stopped me before, so ...

First, however, we need to establish some terminology. When you say ‘anti-shake’, I assume you are talking about the Konica Minolta type, which uses a built-in gyroscope to detect the presence of camera movement during a shot and then compensates for it by applying tiny adjustments to the position of the camera’s CCD, as opposed to the Casio type, which uses digital signal processing (DSP) to compensate for the camera movement. The Konica Minolta Anti-Shake is a form of image stabilization, whereas the Casio Anti-Shake DSP is not.

Getting back to your question, the ONLY form of image stabilization that Nikon has ever used is their Vibration Reduction, which is similar to the Konica Minolta Anti-Shake in that it utilizes gyroscopic control, but instead of applying the tiny adjustments to the position of the camera’s CCD, it applies them to one of the camera’s lens elements. Given this history for Nikon, it is certainly no surprise that their new D80 does not include image stabilization as part of the camera body, but rather, requires the addition of a Vibration Reduction lens to implement that feature.

Panasonic’s image stabilization, Mega OIS,  has also always been ONLY in the form of gyroscopic control of a lens element, and their first DSLR, the Lumix DMC-L1, is no exception. However, unlike Nikon, Panasonic has chosen to provide their image stabilization feature with the L1’s kit lens, which I’m sure has nothing to do with anything other than marketing considerations.

But if I understand your question correctly, Ben, your point is why not use the gyroscope to control the camera’s CCD rather than a lens element, which would avoid the expense of building image stabilization into a multitude of lenses? The only companies I know, besides Konica Minolta, that have done this are Ricoh, with their R3 and R4 point-and-shoot models, and Sony, with their new DSLR-A100 model. My guess is that, for Nikon and Panasonic among others, incorporating image stabilization into a lens is a much more tried-and-true technology than incorporating it into a CCD.


I’m sure this is a stupid question for you, but here goes:

Do American journalists covering events in Baghdad use digital or film cameras?

Marie Fowler

[Note: U.S. Army photo by Specialist Teddy Wade, January 10, 2006. Engineers from Troops Battalion, 4th Brigade Combat Team, 101st Airborne Division, search for improvised explosive devices that may be hidden on the side of a main road in east Baghdad.]

*** ANSWER 2

Marie, for this column, there is no such thing as a stupid question, only stupid answers, especially when I have no idea what the answer is, but that has never stopped me before :), so ...

The truth is, Marie, the American journalists in Baghdad use every kind of camera imaginable, including plastic throwaway film cameras and mini-DV digital camcorders. Due to the severe dust conditions in Baghdad, I would think that: (1) a point-and-shoot would have a better survival rate than an SLR due to the lens changes that could expose the SLR to the dust, (2) a digital camera would have a better survival rate than a film camera due to the film changes that could expose the camera to the dust, and (3) unless money is no object, an inexpensive and easily replaced camera would be preferred over an expensive one.



Many years ago I used to have time to take nice pictures with an SLR camera and a couple of lenses.  After many years of photographic inactivity (children, life, ...) I have decided to try and take some more, but this time with a digital outfit.  Do you have suggestions of what I should look for? - I would like a camera body with a couple of lenses.

Alison Copeland

*** ANSWER 3

Alison, I’m sure it wasn’t ‘that’ many years ago; however, until just recently, my recommendation would have been quite different from what I am about to recommend, namely, do NOT go with a digital SLR (DSLR) unless you absolutely intend to do much more than just casual shooting. The quality of today’s digicams (i.e., a digital camera that is not a DSLR) is good enough to take pictures just as nice as those from a DSLR. If, however, you would still prefer to go with a DSLR, let me know your price range for the total package, and I will be more than happy to provide DSLR outfit suggestions for you in next week’s column.

If you are looking for a camera with both wide angle and telephoto capabilities, then I would definitely recommend a digital ‘bridge’ camera, which is an ‘SLR-like’ digital camera, but with a single, multi-purpose, non-removable zoom lens that covers the entire range from wide to telephoto. One that looks real nice is the Panasonic Lumix DMC-FZ50, but it has not yet been reviewed, and I would not buy it until it is. The FZ50 has 10 megapixels, a 12x (35 - 420mm) image-stabilized optical zoom lens, a 55mm lens thread (for an optional filter, wide angle or telephoto converter, or close-up lens), a 2.0” flip-out and twist LCD, a TTL flash hot-shoe, 4:3, 3:2, and 16:9 aspect ratios, and if it’s anything like its FZ30 predecessor, excellent handling capabilities and fast response times, all for the price of a comparable quality DSLR kit ($650 US).

In case the FZ50 does not get such good reviews, there are similar digital bridge cameras, such as the Canon PowerShot S3 IS (12x optical zoom lens, $450 US), the Fujifilm FinePix S6000fd/S6500fd (28mm wide 10.7x optical zoom lens, but no image stabilization and also not yet reviewed, available 10/2006, $600 US), and the Sony DSC-H5 (12x optical zoom lens, $450 US).

If you can get by without such a wide-range zoom, and would prefer something smaller than a bridge camera that could be easily slipped into a pocket, purse, or handbag, then I would recommend a compact or ultra-compact point-and-shoot style digicam, but one with the addition of manual controls. Some good ones are the Fujifilm FinePix F30 (3x optical zoom lens, ISOs up to 3200, but no image stabilization or manual focus), the Canon PowerShot A700 (6x optical zoom lens, but no image stabilization), the Canon PowerShot S80 (28mm wide 3.5x optical zoom lens, but no image stabilization, also very difficult to find), and the Panasonic Lumix DMC-LX2 (28mm wide 4x optical zoom, 16:9 native aspect ratio, but not yet reviewed).

And, if you are willing to forego the manual controls, then I would ‘highly’ recommend the Canon PowerShot SD700 IS, a.k.a. Canon Digital IXUS 800 IS (see our Canon Digital IXUS 800 IS Review).

[Note: Much thanks to Nick in Japan for submitting the accompanying photo, which he shot with an ultra-compact point-and-shoot style digicam, the Kodak V570.]



I have several digital cameras with various numbers of megapixels, and I have noticed that, when viewing a picture at its real size, the five megapixel camera’s images look the sharpest. Sure, the more megapixels, the better the resolution, but at the same time the image becomes larger (in real size), and everything gets bigger, so the demands for what is looking sharp rises.

Could my observations have been faulted by other factors like sharpening, etc., or is there in fact an ideal size in this respect?

Anders Tiberg

*** ANSWER 4

Thank you, Anders. Finally, a question to which I actually know the answer. :)

It all comes down to comparing apples with apples. In order to judge the effect of a camera’s increased image resolution, it goes without saying that you need to keep all image conditions the same, as well as all camera settings (except image resolution) the same, but you also need to print and/or display the images AT THE SAME SIZE.

So, Anders, your observations were correct, but the experiment was at fault for allowing the higher resolution images to be printed/displayed at larger sizes. Like I always say, you don’t want to be comparing Apples with Windows. :)

[Column photo “The Photographer” by Brenda LaFleur of Brenda LaFleur Photography.]

Tracker Pixel for Entry

Your Comments

17 Comments | Newest Oldest First | Post a Comment

#1 nick in japan

Mr. Rothfeld, Gary built you a clock, then told you exactly what time it was with his last sentence! Whatta guy!
No wonder 10.7 million viewers eagerly await his "Parries" every week!

9:39 am - Sunday, August 20, 2006

#2 Gary's Parries

Nick, are you doing something in PhotoShop to your images when
you send them to me that is causing the purple noise in the sky? I
thought it was because of the noisy LX1, but I see the exact same
thing on the V570 image in question #3.

10:54 am - Sunday, August 20, 2006

#3 nick in japan

I don't keep a record of tweaking, I always play in PS, the exact abnormality is kinda hard to determine after the image has been sent around the world, opened and closed countless times.
Purple casting is beyond me, lottsa blue haze due to the high pollution here and because I dont use a UV filter. The Sony 828 will give a noticable cast from time to time. I may be over saturating them for your likes, or maybe it's the high contrast image combined with my tweaking that's causing you concern.
Kinda hard to evaluate, I'll look a bit closer and see if I can find out where it may be coming from.
I have one idea that may be the cause, I'll check into it... I sometimes will use the "Styles" tools, found in the top tray in "Windows" , specifically "Color Burn", which is usually used for text variations, but works well on alot of bland images for that extra "snap". Works well for Black and White too. With any kind of color in the clouds in the sky, either visible or not initially, I suspect that that "Color Burn" may increase the colors in the sky to a level that is a bit unnatural.
I usually dont blow up images to check closely if they are just for e-mail. If I'm going to print something I'll put some extra time in, I tweak about 20-40 images a day, by the time I gotta start supper I'm almost blind!
Thanks for your help!

12:15 pm - Sunday, August 20, 2006

#4 Gary's Parries

That 'Color Burn' sounds like a good candidate. Before you mentioned
it, I was thinking too much contrast and/or saturation.

When you look at my column photo (in the column), does it look like it
has a slight 'Portrait' orientation? It's supposed to be square, and when
I copy it to a different application it is square but in the column it looks
slightly compressed horizontally. First I thought it was my monitor, but
then I noticed that the picture in question #1 looks perfectly square.

3:20 pm - Sunday, August 20, 2006

#5 Jason

The VAST majority of journalists are using Digital SLRs, including pro-level sealed models like the Canon 1d and Nikon D1/2 series cameras.

There are still some of your traditionalists using Leica M's for those not on tight deadline, and Alex Majoli won World Press Photo awards using Olympus C-5060's but the day-in, day-out reporters are using digital SLRs, hooked to laptops with portable BGAN satellite transmitters, if they don't have network connectivity.

8:20 pm - Sunday, August 20, 2006

#6 nick in japan

You are correct, Brother Ben is square ( But a hell of a nice Guy!!) and the last 3 pix are more horizontal. I first thought your column image was an optical illusion due to the verticals within, but it, surely is, more vertical.

10:13 pm - Sunday, August 20, 2006

#7 terry chay


You forgot that Pentax with the K100D will have anti-shake in the body.

Here is my summary of the technology:

If I had to guess why is less political than Gary’s. I believe it is simply one of patents and profit. Sony (K-M) and Pentax sit on the patents for CCD-shift anti-shake and Canon and Nikon (and I guess now Leica) feel they can charge more for IS/VR/MegaOIS lenses. Pentax and Sony need to compete so by innovating here (allowing you to turn your cheap wide prime into a IS lens), is a good way to penetrate the market.

4:56 am - Monday, August 21, 2006

#8 Gary's Parries

Jason, thank you for all the 'inside' info. I am sure that was the answer
Marie was looking for.

Terry, I agree that patents, profit, and market penetration all contribute
to the situation, but I think it is even more than that. Many professional
photographers prefer the lens shift method of image stabilization to the
CCD shift. I am not exactly sure why but I suspect it gets back to "don't
fix it if it ain't broke."

And what if Nikon did come out with an anti-shake DSLR? Would you be
able to still use your VR lenses with it? :)

Thanks, Nick for the observation. I will have to ask Mark what's going on.

10:13 am - Monday, August 21, 2006

#9 Gary's Parries

Oh, and BTW, that was not a picture of Ben.

10:17 am - Monday, August 21, 2006

#10 nick in japan

OK! Then, are we to assume that is a picture of you?

11:56 am - Monday, August 21, 2006

#11 Daveed V.

The benefit of moving antishake to a DSLR sense is obvious: You only pay for one instance of the mechanism.

There are some other benefits to moving the mechanism to the lens. Most visible to the photographer is that the stabilization through lens elements also affects the viewfinder.

The other potential technical compromise is that moving the sensor changes (enlarges) the required image circle.

3:36 pm - Monday, August 21, 2006

#12 Gary's Parries

Well, Daveed, at least I did get the obvious one, but your last two are
excellent points. Thank you for your input.

And no, Nick, that is not me. It is a close friend that my wife and I just
visited on his beautiful farm in upstate Pennsylvania.

5:20 pm - Monday, August 21, 2006

#13 Gary's Parries

Received this email from Anders Tiberg:

{I think that my question wasn't answered, or that it was a "cleverly evasive answer". The question was: Is there an ideal size in megapixels (for sharpness) when looking at a picture in its real size? If so: What is the mathematics behind it?}

Anders, actually this was not one of those 'cleverly evasive answers',
but thanks for suggesting it was. Unless I am misunderstanding your
question, the issue is whether it's valid to compare two images of the
same scene, but with different resolutions, at different sizes. It is my
contention that it is not.

The problem is that the term 'real size' is a digital misnomer, and that
it is actually only a fact of life of displaying and/or printing an image at
the inherent resolution of a digital device. The only real size is the size
of the CCD.

Think of it mathematically. Suppose you shot an image with a camera
having a CCD of dimensions (for simplicity) 1" x 1" and a resolution of
4 megapixels (2000 x 2000 pixels). If you then enlarged the image by
a factor of 10, it would display or print at 10" x 10". Now, suppose you
took the same shot with a camera having a 1" x 1", 9 megapixel CCD
(3000 x 3000 pixels). If you then applied an enlargement factor of 10,
the resulting 10" x 10" image would have far greater detail than the 4
megapixel image and, all other things being equal, would appear as a
much sharper image to the eye.

The old story comes to mind of a man who walks into a doctor's office
flailing his arms wildly and says, "Doc it hurts when I do this!" to which
the doctor replies, "Then don't do that !!!"

5:35 pm - Monday, August 21, 2006

#14 Gary's Parries

Just received this email reply from Anders Tiberg.

{Okay, I understand that with different screens and different resolutions, you get different sizes of the image you are looking at "in real size", and hence the sharpness will vary indefinitely for different numbers of megapixels. But if you take an individual case like my screen, which is an ordinary 17" TV screen at 800x600 resolution (which means that there are only 800x600 active pixels as opposed to an LCD or TFT screen where all the pixels are activated regardless of resolution, and so there is much better sharpness at 800x600), at least to me the five megapixel cameras look the sharpest, which made me think you could work your way from there to the general case in a sort of nifty way, arriving at about 5.23 (total) megapixels.}

Anders, why didn't you say that you were viewing with an ordinary TV
screen? This changes everything due to the scan rate conversion that
must be performed in order to display a full-resolution photo image on
a television screen.

If you take your screen resolution of 800x600 (= 480,000 pixels), then
multiply by a 60 Hz TV vertical scan rate (= 28.8 megapixels), multiply
by a 15 kHz TV horizontal scan rate (= 432 gigapixels), I wonder if the
term 'gigapixels' has been used before), divide by 2 (= 216 gigapixels)
due to the TV's interlaced scan, and then divide by an average monitor
horizontal scan rate of 40 kHz (= 5.4 megapixels), you may notice that
this agrees rather nicely with your figure of 5.23 total megapixels.

So I guess you were right all along, 5.0 (actual) megapixels is the ideal
resolution for a digital camera. QED :)

7:19 pm - Wednesday, August 23, 2006

#15 Gordon McKinney

I agree on the profit front. Canon and Nikon would stop selling their premium lenses.

With the introduction of the 70-200L f/4 IS and 24-105L f/4 IS it would appear that Canon are not moving to IS sensors any time soon.

Will the IS work as well being in the body vs. the lens? i.e. does the position in the lens afford better control?

Certainly I find it nice looking at a stable image vs. a shaking image. Something that an anti-shake body can't offer.

4:17 pm - Friday, September 1, 2006

#16 theodore dolmatch

Just a very nice, helpful site! Thanks.

4:23 pm - Friday, September 1, 2006

#17 Gary's Parries

I think you are right, Gordon. Profit is the 'motive' and live viewing is
the 'rationalization'.

Theodore, right back atcha with the thanks.

4:47 pm - Friday, September 1, 2006