gdanmitchell

<p>I'm actually sympathetic to the "just use your camera and don't worry" sentiment - it isn't a great idea to go looking for problems.</p>

<p>However...</p>

<p>... knowing your equipment is a Good Thing. All lenses vignette - what are the vignetting characteristics of your lens? All lenses have sharpest apertures - where is yours sharpest? And so on. So, so running some initial controlled experiments with your new gear can be a fast way to start to understand its personality. I do this with all new equipment, and the advantage is that I can more quickly use it to full advantage.</p>

<p>The basic idea of any test is to try to remove variables that can skew the results. For a lens test a simple approach is to:</p>

<p>1. Put the camera on a tripod; use a remote release; uses mirror lockup.</p>

<p>2. Point at a subject that has some detail and contrast and which provides a plane of focus parallel to the sensor of your camera. (In other words it is flat.)</p>

<p>3. Put the camera in aV mode and let it automatically meter the scene and autofocus. Then turn AF off.</p>

<p>4. Make an exposure at some focal length at your largest aperture. Switch to the next smaller aperture and make another. Continue through each aperture.</p>

<p>5. Repeat at several representative focal lengths.</p>

<p>Dan</p>

<p>Just a thought... did you have your white balance problems while shooting in artificial light? If so, was it fluorescent lighting? The color of that lighting varies at a rapid rate, and a succession of shot under such light can have different color balance.</p>

<p>Dan</p>

<p>There's a popular myth that you must have f/2.8 (or f/2 or f/1.4 or f/1.2) to shoot portraits. While there are a few situations in which you might need this, most portrait photography is done at smaller apertures and for this work f/4 is fine.</p>

<p>If you must have the larger apertures for portrait work but you want the 70-200mm range for your landscape work, consider getting the excellent non-IS f/4 70-200mm L zoom and an appropriate focal length prime for you portraits. You could actually get this zoom and a couple of very useful portrait primes for the price of the f/2.8 70-200mm zoom, and you would arguably be set up better for portrait work and just as well for landscape.</p>

<p>Dan</p>

<p>Unless I overlooked it, you didn't tell us what camera you have. If you have a cropped sensor DSLR (a digital Rebel or a 20D, 30D, 40D, or 50D) I think that the best large aperture zoom for the sort of shots you describe could well be the Canon EFS 17-55mm f/2.8 IS lens. f/2.8 is a large a maximum aperture as you will find on a zoom lens. The image stabilization (IS) feature can let you hand hold the camera a slower shutter speeds - though you'll need to try to get the kids to hold still if you really push this.</p>

<p>A flash unit might be even more useful for you and less expensive.</p>

<p>Dan</p>

The latest word (from Jeff Schewe and Michael Reichmann, or at least that's where I saw it

articulated) is that up/down-sampling is not the way to go.

Rather than doing that, simply adjust the dpi in the PS image size dialog so that the print

dimensions end up being what you want. Yesterday I printed a few images from my 5D -

one at 6 x 9 inches and another at 8 x 12 inches. If memory serves, for the 6 x 9 I simply

used something along the lines of 420 dpi or thereabouts and 300-something dpi to get

my 12 x 18 size.

This corresponds to the notion (described by Schewe) that higher dpi is important for

smaller prints since they will be examined more closely, and that lower dpi is fine for

larger prints.

Try it. It really works nicely.

Dan

This does NOT sound like the diopter adjustment is the cause. In that case the view through the viewfinder would never be

sharp, but the photograph might well be. In other words, the symptoms would be the opposite of what we see here.

Sometimes when a problem like this is reported it is due to an autofocus problem - often one of two. First, the camera/lens

could be out of adjustment causing the autofocus system to set the lens incorrectly and giving front or back-focus. A second

type of autofocus problem can be user error - the photographer is not using AF correctly and perhaps the system is not using

the focus point that it on the subject or the photog is using the focus and recompose method at a wide open aperture (narrow

DOF).

Not sure that this is what is happening here.

For one thing, don't trust the distance scale on the lens barrel. There are reasons that it might not register accurately, even if

the lens is capable of good manual focus. Try to focus manually using the (somewhat small and dim) viewfinder. Perhaps take

a few photographs with the focus displaced slightly in front of and behind the "best" manual focus setting determined this

way. This should rule in or out a problem where the lens simply is incapable of focusing correctly.

Dan

Shay:

I had a fairly specific socially-useful idea for improving the situation that I though could make a positive difference without compromising the anonymity that seems

to be so important at this site. I tried to mail Josh, but just got an error message which I reported here. That seemed to be the end of it.

I'm not going to sweat it. I can post my photos elsewhere. As I wrote above, I'll will still contribute to the discussion threads.

Take care,

Dan

(FWIW: I have long experience with online communities, have build blogging sites before the term "blogging" existed, have taught online college courses since the

mid-1990s, worked closely with people at a Really Big Computer Company on related stuff, and so forth. I'm relatively familiar with the technical issues and social

dynamics of online communities.)

Hey, I'd get a <i>slower</i> computer. After all that work you deserve to sit and watch a

movie or two!

I use an EyeOne on my Macs (desktop and laptops) and it works without any problems at all.

Dan

Nothing is going to change here. Fortunately, there are other options for posting/hosting photographs. The PN discussions are still

interesting and worthwhile. Therefore...

I'm not going to continue to use PN as a repository for my photography but I will continue to participate in discussion threads that interest

me.

I'm disappointed that PN is obstinate about this, but it is clear that the folks running it are sure that they are right.

Take care,

Dan

Actually, I take it back. I _tried_ to send you an email by clicking on the link at your

members page, but I only get:

HTTP/1.0 500 Internal Server Error MIME-Version: 1.0 Date: Thu, 07 Feb 2008 21:21:28

GMT Server: AOLserver/4.0 Content-Type: text/html Content-Length: 536 Connection:

close

Server Error

The requested URL cannot be accessed due to a system error on this server.

Dan

Josh, I just sent you an email. However, the notion that proposals by members of a

"community" about how that discussions about making that community work more effectively

cannot take place within the community, but must instead be taken to private one-to-one

email is, to say the least, very odd.

Dan

I've found that "popularity" as measured by PN ratings has almost no relationship to which of

my photographs sell.

Dan

If FF sensors will never become less expensive, how to explain that the less dense FF sensors used in early models now come in cameras that cost a lot less than the originals, especially in

inflation adjusted dollars.

While it is probably true that a cropped sensor will be less expensive than a FF sensor, the prices of both have continuously declined, particularly in relationship to inflation, and the difference

in price between FF and crop cameras has declined as well.

Note that one can purchase a 5D now for about 2/3 of what it cost originally. Note also that competition has now come to the FF DSLR market.

BTW, the fact that many existing DSLR users may own lenses only suitable for crop sensor use might be regarded by the manufacturers as a positive factor in favor of introducing FF cameras at

lower prices - this would likely lead to an increase in lens sales.

Dan

Ack! I wrote: "Today quite decent primes are available at relatively low prices"

I meant to write "Today quite decent ZOOMS are available at relatively low prices"

(Just trying to keep up with 'Colin and himself'... :-)

<p>A couple things.

<p>I agree that testing at ISO 1600 and f/4 tests and outlier situation and thus is of somewhat limited value. I'd also be more interested in knowing optimum performance at more

typical settings, especially around f/8. (If you are testing <i>lens</i> performance, you'll be better able to gauge differences if you can limit the noise effects of the sensor at high

ISO.)

<p>Also to eliminate other variables besides the lens itself, do use tripod, MLU and a remote release. Hand holding the camera, even at higher shutter speeds, essentially negates any

sharpness conclusions you might reach from your test.

<p>The idea of simply shooting RAW and the processing the same <i>seems</i> like a good approach, but I'm not so sure. In real life we essentially <i>always post-process in ways

that maximize the image quality</i> from whatever lens we use. If lens X seems "better" then lens Y with identical processing, but lens Y can end up with a better image than lens X

after careful post-processing, which is "better?"

<p>Dan

<p>OK, can't help myself... :-)

<p>I <i>strongly disagree</i> with the old-school advice to start out with a single 'normal' prime. That made sense a few decades ago when the only affordable and decent lenses were

primes. I started out that way and gradually added a few more primes as I went along.

<p>But that was then and this is now. Today quite decent primes are available at relatively low prices. The notion that "too much zooming is bad for you" seems ludicrous to me. In fact,

without either a zoom or multiple lenses there are significant aspects of composition and so forth that you cannot learn as effectively. For example, with a zoom a beginner can learn how

to control the size relationships between foreground and background subjects by changing focal length. You cannot do that with a single prime.

<p>But let's say I'm wrong. Just get an inexpensive zoom and lock it at 50mm or 30mm or whatever you like and shoot it that way and you'll get the same effect.

<p>My strongest advice to an actual beginner is to not worry yet so much about getting the 'best' equipment since you don't really know what that might be for you and your

photography yet. Just get <i>some</i> decent camera and a functional and appropriate lens and start taking pictures. An XTi and the IS kit lens would be a great place to begin. Take a

lot of photographs. Work them over in PS. Shoot, shoot, shoot. Learn, learn, learn. This is far, far, far more important for your development as a photographer than making the "right"

choice between a 24-70 f/2.8 L and a 24-105 f/4 IS L at this point.

<p>Retiring from my soapbox... :-)

<p>... for the moment,

<p>Dan

<p>I don't think I have much to add to what I wrote in <a href="http://www.photo.net/bboard/q-and-a-fetch-msg?

msg_id=00OJ8T">the other thread</a>, Jami.

<p>Just one thing. If you can't decide and are really up in the air, there are good arguments for starting out with

simpler, less expensive equipment. Shoot, shoot, shoot. You'll figure out for yourself what specific features you need

for <i>your photography</i>, at which point you'll be better prepared to decide among the many excellent options

available.

<p>Few of us started with the "best" gear as beginners. And if we had, it probably wouldn't have improved our early

photography much at all.

<p>Dan

It occurred to me that you might wonder what I carry in that Rover AW pack and how it compares to what you intend to carry.

Canon 5D w/L bracket, 17-40mm f/4, 50mm f/1.4, 24-105mm f/4, 70-200mm f/4, 100-400mm f/4.5-5.6, a couple filters, extra batteries, remote release, Induro C313 tripod with

Acratech ballhead, and a few odds and ends.

While it is possible to cram all of this in the lower section of the pack, I'll frequently carry the camera either in the upper section (in good conditions and on shorter hikes) or in a chest pack

where it is more accessible.

The upper section of the pack handles the usual stuff you might need on a long hike and varies depending upon conditions, terrain, and weather: water, food, dark glasses, extra clothing,

LED headlamp, etc.

<p>For longer day hikes my Rover AW pack from Lowepro works quite well. It has a lower section that can accommodate several lenses and a camera body - or quite a few

lenses if you carry the body somewhere else. (more on that later) It incorporates a very effective tripod carrier that handles my very large Induro C313. It has a separate

upper section that I use mostly for non-photographic trail stuff.

<p>While it isn't large enough for overnight backpacking with camera equipment it is definitely large enough for day long trips, even in difficult conditions of weather and

terrain. It carries well also. It has a integrated rain cover that seem effective - not only for rain but for providing a cleaner area to put down the pack. The pack has two

weaknesses, though neither is a deal breaker for me. First, while the raincover is effective, it is not large enough to cover the pack and a very large attached tripod. Second,

for a _backpack_ of its size it is rather heavy. A non-photographic backpack of this size would likely weight no more than 2 pounds, but the padding and attachments

increase this pack's weight considerably.

<p>I frequently carry my camera (5D) in one of the large chest-strap-mounted Lowepro Topload bags - this provides ready access to the camera and perhaps even a

second lens.

<p>For actual overnight backpacking you'll need something else entirely. I've <a href="http://www.gdanmitchell.com/2007/08/26/thoughts-on-my-summer-2007-

backpacking-photography-kit/">outlined my approach in an article</a> at my web site.

<p>Dan

"Once again, I'll repeat my offer to have anyone with suggestions for the rating/critique system to contact me via email and I'll be happy to discuss it to you. I have been making this offer for the better part of

a year now, the grand total number of people who cared enough to do so...is two."

I never heard your offer before, but in the last year I offered a carefully crafted proposal for how you could combine PN's apparent need for confidentiality (more about that below) with some kind of tracking

of the performance of anonymous raters. It was dismissed. This is part of my frustration. After watching this for awhile as a paid member I believe that you'll probably just go ahead and continue to repeat

the same reasoning, but it isn't working, and the people who are most frustrated by this system are often the very people who contribute to the site. I put myself in that category.

About anonymity... PN goes far beyond what most other successful sites do. At those one can remain anonymous by adopting just about any screen name - and many do this. But at PN the non-member

posters are a complete mystery. If only they had online IDs, we would know if we were getting revisited by the same person, and whether that person did or did not have any sort of track record.

One of the features of online communities that gives them value is the credibility that members build up through their participation. On PN the fact that most ratings come from completely unknown sources

makes it impossible for the raters to develop any sort of credibility, and hampers the credibility of the rating system itself.

Note that there are three possible levels of identification that could be used. PN uses a "system" in which there is no labeling of the majority of responses to photographs whatsoever. An alternate system that

I don't expect you to adopt would require raters/posters to use their real names. An intermediate approach - and one that is common elsewhere on the net - is to let users self-assign an online name; we

don't know who they really are if they choose to conceal their real identities, but we do recognize the voices of particular anonymous individuals. Knowing this allows the community to develop some sense of

the credibility of the raters.

My earlier proposal was to assign each participant a sort of "raters' rating." Details could vary, but it could operate as a clickable link on a rating value. It could be hidden behind the current "details" listing as

a "more info" link. On this page each rater would appear on a separate line, but still anonymously if you prefer. However, one or more values could be associated with each anonymous rater: total number of

ratings, average rating, and more. As you know, these data would not be difficult to generate at all. In fact, I suspect that the data are already virtually available in the system

In any case, this is a clear issue with PN, and it has been a source of continuing frustration, especially on the part of some of the better photographers and more active participants... who have eventually

gotten frustrated and left. There are at least two ways to think about this. One, and this seems to be the response up until now, is "let 'em go. Who needs them." A more useful response, and one that would

demonstrate the responsiveness of PN and its interest in finding a way to improve this continuing source of irritation would be to try to adopt some changes to address it.

Finally, as a photographic "community" I think that it would be appropriate for the community to have some open input into this issue. I've found that request to "send an email" usually serve more to make

the issue go away than to lead to a solution.

I like the concept of PN. There are still quite a few good people here, some of whom I count as friends. That's why I joined and became a paid member. If I leave later this month, don't take it personally. But

do consider it as evidence that at least one PN supporter is not finding value in this "community" any more.

Dan

I've been a member of PN for awhile now, and I've tried to develop a thick skin about anonymous 3/3

ratings. I think I've more or less succeeded. However, I have to wonder about the value of a rating system

that doesn't differentiate between anonymous and identified raters.

What is the value of anonymous 3/3 ratings? A common pattern I see on some of my work is relatively

high ratings from members, almost always in the 5 to 7 range. And then I get the anonymous 3/3 bomb.

It is frustrating. How can I learn anything from the low rating? No idea if I've just offended someone in the

forums and they want to get back at me. No idea if the 3/3 ratings are coming from the same

person/people. No idea if the rater simply doesn't like the genres I work in.

I tried suggesting a system that would assign a "rater rating" last year. Without compromising anyone's

anonymity, it would be easy to provide info about the rating patterns of anyone who rates: How many

ratings have they given? What is their average rating? But no response.

And continuing astonishment that the random, anonymous 3/3 counts in some of the rankings while the

ratings of contributing members of the PN community do not.

If the only answer is "don't worry about the ratings," then why have them at all? In the end, if the value of

PN is supposed to be, at least in part, an opportunity for useful feedback about one's work - PN is failing

me in this regard.

My membership is up for renewal later this month, and I doubt I'll continue. I've got a Flickr site and my

own commercial gallery and blog, all of which generate sufficient traffic and leads.

I know that responses - if there are any - will include the usual: don't worry about ratings. I don't - but at

this point I fail to see the value either, and I fail to see the point of posting my work here without feedback

when there are other places to post that get more traffic and/or provide more value.

Dan

IS compensates for camera shake, allowing you to hand hold the camera at slower shutter

speeds without getting blur from the camera moving. Depending upon the lens, Canon

claims that you can shoot at a shutter speed that is from two to four stops slower. If you

could hand hold a shot without IS a 1/100 second, this means that with IS (depending

upon the lens) you could hand hold it at 1/25 second or perhaps even slower.

IS is not helpful at all for stopping a moving subject. For that you still need a faster

shutter speed. In fact, in situations with a moving shutter speed using IS can make things

worse since you may be tempted to shoot at even slower shutter speeds.

<p>Shot with the 24-105 under stage lighting only using the 24-105...

<p><img src="http://www.gdanmitchell.com/images/JamoFiddler20070331.jpg">

<p>The low light issue cuts both ways. You do get one extra stop for dealing with subject motion using the 24-

70 f/2.8, but you also get 2-3 extra stops for handheld shots where subject motion can be controlled (or

tolerated) with the 24-105 f/4.

<p>Dan

<p>Joshua, the 16-35 is better at edge to edge sharpness (but not perfect either) <i>at the widest apertures</i>. The 17-40 provides quite good corner sharpness and excellent center sharpness at smaller

apertures. All the tests and reports that I've seen tend to confirm that the 16-35 is "better" for those who need the wide apertures, but no better at all at the smaller apertures typically used for landscape.

<p>For a landscape photographer - at least those who generally aren't shooting wide open - the 16-35 is <i>not a better lens</i>. For me, "betterness" is a combination factors that determine how well a lens

performs for the purposes I use it. The 16-35 does not perform better for my landscape work, and its extra cost and size/weight reduce its value <i>to me for landscape</i>

<p>Please understand that I'm not knocking the 16-35mm L lens at all. It is a really fine lens and when you need what it provides it is the best Canon product in its class. But "best" is not an absolute thing here - it is

relative to the specific needs of the photographer.

<p>I've posted a <a href="http://www.gdanmitchell.com/2008/01/21/corner-sharpness-of-the-canon-17-40mm-f4-l-lens-on-full-frame/">slightly funky test photo</a> (no esthetic value claimed :-) that

shows somewhat clinically the corner performance of a 17-40 shot in "landscape mode." It shows the full image at reduced magnification, along with 100% crops from the center and lower left corner. You'll notice

that the corner area includes grass, one of the most difficult subjects to resolve, and that this part of the image is much closer than the focus point in the center of the frame. With all of that in mind, the quality is

really quite fine for 17mm.

<p>Take care,

<p>Dan