PDA

View Full Version : Why so many 100% crop comparisons?


tkbslc
20th of February 2009 (Fri), 12:20
It seems like the standard benchmark comparison for any test, is side by side 100% crops. Why do we test this way when almost nobody uses 100% pixel level images for anything? The vast majority of the time we just view images at full screen, which may be up to 1920x1080, or print at 11x14 or smaller. It seems like it would make more sense to post real world sizes, rather than 100% crops for comparison.

tonylong
20th of February 2009 (Fri), 12:45
Well, many people here will agree with you, and in the final analysis it is the final outcome that matters.

On the other hand there are elements in an image that you can't fully analyze if you view it compressed -- relative sharpness, rendering of detail and degree of noise are, if you are concerned with these things, best viewed non-compressed, i.e., 100% pixel in image to pixel in screen. This is how a lot of software for, say, sharpening and noise reduction has us view the image. It's really a no brainer.

So, if someone has a concern about these things in an image, then posts a highly compressed Web view of the full image, it's pretty impossible to really say what's going on unless it's obvious. Post a "100% crop" and you have a much better ability to help the person.

Now, many people only shoot for the Web and small prints, say a max of 8"x10", and at those size some of these qualities get equalized in the compression of the image and, understandably, could question how "picky" one should be. In fact, I put images on the Web that are not the best, but are useable at that size and that show something I like. No harm, no foul.

But I'm also very conscious of a few things. First, with much of my shooting, I often have to crop pretty closely. Looking at a close crop as a complete image all of a sudden requires the best of sharpness, noise quaility, and detail, wouldn't you agree?

Also, when I review my images, I have two ideal final outcomes I consider: I'd like images that would be able to produce gallery-size prints for one thing and would be eligible for publishing for another. Many magazine and stock publishers do examine submissions at 100% for sharpness and noise. So, many of us here can't ignore that level of quality demands.

Even if I never see a gallery or work sold to a demanding publisher, I still aspire to excellence in my photography. I use tools like a good tripod to ensure that when taking a landscape/scenic shot it will, to all my abilities come out tack sharp. I shoot sports/action shoots taking care with my technique to maximize sharpness. I work on good exposure techniques to minimize the noise problems. That way, when I'm back in the digital darkroom I can have hopefully good material to work with which will look good at any viewing size.

mrkgoo
20th of February 2009 (Fri), 14:15
Think of it this way - the vast majority of people will look at a diamond ring from an unmagnified point of view. So why buy one that is perfect? Just give your significant other a cheap one that looks the same from the distance most people see it.

justaf IREMAN
20th of February 2009 (Fri), 14:44
Think of it this way - the vast majority of people will look at a diamond ring from an unmagnified point of view. So why buy one that is perfect? Just give your significant other a cheap one that looks the same from the distance most people see it.

haha tell that to my wife.

Anders Östberg
20th of February 2009 (Fri), 14:48
Because if we did it your way we would have to admit that there is no need for better gear and we would no longer have any arguments for that next upgrade - *gasp* - :p

NinetyEight
20th of February 2009 (Fri), 14:48
It seems like the standard benchmark comparison for any test, is side by side 100% crops. Why do we test this way when almost nobody uses 100% pixel level images for anything? The vast majority of the time we just view images at full screen, which may be up to 1920x1080, or print at 11x14 or smaller. It seems like it would make more sense to post real world sizes, rather than 100% crops for comparison.

Completely agree, there is far too much pixel-peeping and micro-analyzing going on and not enough shootin' :D

Some people are so anal about it I doubt they actually take pictures of anything but rulers, batteries and test charts and then view them at 100% :rolleyes:

<RANT OVER> :confused:

mrkgoo
20th of February 2009 (Fri), 14:54
Completely agree, there is far too much pixel-peeping and micro-analyzing going on and not enough shootin' :D

Some people are so anal about it I doubt they actually take pictures of anything but rulers, batteries and test charts and then view them at 100% :rolleyes:

<RANT OVER> :confused:

But what's a real world size? That's dependent on the final viewing output. Maybe most people are taking pictures for billboards, and viewing them from 3 feet.

John_B
20th of February 2009 (Fri), 14:57
It seems like the standard benchmark comparison for any test, is side by side 100% crops. Why do we test this way when almost nobody uses 100% pixel level images for anything? The vast majority of the time we just view images at full screen, which may be up to 1920x1080, or print at 11x14 or smaller. tkbslc,
One reason might be, because many crop there photos, the more you crop the more you see what is at 100%. It also really depends on what the photo is printed on and what printer was used. Where a 100% crop more likely shows what the camera/lens delivered.

Think of it this way - the vast majority of people will look at a diamond ring from an unmagnified point of view. So why buy one that is perfect? Just give your significant other a cheap one that looks the same from the distance most people see it.mrkgoo,
Isn't that what Cubic Zirconia are for? :lol:

NinetyEight
20th of February 2009 (Fri), 15:00
But what's a real world size? That's dependent on the final viewing output. Maybe most people are taking pictures for billboards, and viewing them from 3 feet.

Ah, that must be it... :D

tkbslc
20th of February 2009 (Fri), 16:01
tkbslc,
One reason might be, because many crop there photos, the more you crop the more you see what is at 100%. It also really depends on what the photo is printed on and what printer was used. Where a 100% crop more likely shows what the camera/lens delivered.

mrkgoo,
Isn't that what Cubic Zirconia are for? :lol:

Cubic zirconia are actually perfect diamonds. Since they are lab created, they are flawless. So, see, honey, I didn't go cheap I got you the BEST diamond. I can't help it is was only $49.99.. :)


And the cropping argument is fine IF you crop a lot. Most of the time we are lookingat %100 crops of half a beer bottle label and that is supposed to tell me that one camera/lens is superiour. I guess when I roll out my calendar called "bottle labels at 10x life size", I will know which camera to choose.

Anyway, just like the 50/40D noise thing. If you look at 100% crops, one looks a little noisier. If you actually print or view the whole photo, who can tell?

Anyway, I'm hungry so I'm going to worry about more important things. Like sandwiches.

S.Horton
20th of February 2009 (Fri), 16:07
I think the only way to "compare" two image results on the web is 100%.

tkbslc
20th of February 2009 (Fri), 16:27
I think the only way to "compare" two image results on the web is 100%.

Why?

darosk
20th of February 2009 (Fri), 16:28
The answer is porn, pure and simple.

Dooms_day
20th of February 2009 (Fri), 17:02
Some people are so anal about it I doubt they actually take pictures of anything but rulers, batteries and test charts and then view them at 100% :rolleyes:

haha perfect quote, its so true

S.Horton
20th of February 2009 (Fri), 17:16
I think the only way to "compare" two image results on the web is 100%.

Why?

Varying sensor sizes, for one.

For another, if it is < 100%, then you're looking at something compressed, which means apples-to-oranges comparisons.

I'm not saying that 100% is perfect, just that it is the only way on the web.

In the end, it is the print that matters. ;)

vwpilot
20th of February 2009 (Fri), 17:33
100% is just that, 100%, its what the image ACTUALLY is. In the same way that people looked at slides with loupes, we should look at our digital files at 100% because anything less than that is not showing the actual image, but a compressed version of it.

I can take a photo that is just plain out of focus. Its no good, but looks good at 800x600 for the web because the compression has removed the softness out of it. However, its not a sharp image and was not taken properly.

The only way to tell that is if you view it at 100%. Its not pixel peeping, its viewing it as it was meant to be viewed. For me, any shot that is not perfect at 100% goes tot he recycle bin as that is the only way I can be confident it can be used in whatever way it needs to be used, not just to show off in web forum where no one can really tell if it was ever in focus or not.

JeffreyG
20th of February 2009 (Fri), 17:39
A 100% crop is the internet sharing equivalent of you and I looking at a very large print together. It is the only way to meaningfully share performance details of various equipment in a way that would matter if we are making large prints.

Now, the funny thing is that some people who decry 100% crops and pixel peeping will also own a camera like a 50D or 5D2.

This is funny because if you make only small prints and web images then you do not need to look at 100% crops, but you also do not need to own high pixel count cameras or lenses capable of very high resolution. The act of making a small image throws away all that detail anyway.

Nortelbert
20th of February 2009 (Fri), 17:48
Cubic zirconia are actually perfect diamonds. Since they are lab created, they are flawless. So, see, honey, I didn't go cheap I got you the BEST diamond. I can't help it is was only $49.99.. :)


No... it's a crystalline form of Zirconum Diocxide. Diamonds are Carbon. :-)

I don't generally look at pix at 100% but as previously mentioned, sometimes it helps to check for softness, et cetera.

Radtech1
20th of February 2009 (Fri), 18:00
It seems like the standard benchmark comparison for any test, is side by side 100% crops....It seems like it would make more sense to post real world sizes, rather than 100% crops for comparison.

100% crops give information about the image in its native resolution. Without anything else interpolating. Perhaps the word "Pure" works here. For many of us, myself included, it allows valid hardware assessments, "Do I buy this lens or that one?" Or more importantly on a day in-day out basis, I use it as an editing tool, "Do I follow up on this image or that one?"

I've been doing it for 35 years, and it has worked so far.

Rad

zincozinco
20th of February 2009 (Fri), 18:11
In order to see if a print holds for print in a magazine or an enlargement its vital to see it at 100% There are very few cases where I would let an image go if it was not A ok at 100% Like radtech said as well if the image is not good at 100% very seldom it will even be retouched and probably never printed.

tkbslc
20th of February 2009 (Fri), 18:47
You still aren't convincing me. If the image looks good at the magnification level you are going to view and/or print it at, why does it need to look good at full magnification?

If anything it means I am making comparisons that may cause me extra work or expense to get an image that does not come out any better. Again, if you print or crop to near full resolution, then you should be comparing those kinds of crops. For most of us, it just makes more work for no gain.

JeffreyG
20th of February 2009 (Fri), 18:55
You still aren't convincing me. If the image looks good at the magnification level you are going to view and/or print it at, why does it need to look good at full magnification?

It doesn't.

What you seem to be missing is that people looking at 100% crops are doing so because they are planning to print the images large.

If you are never going to print large then you do not need to view 100% crops. You also don't need a camera with higher resolution than the 1D classic and you only need L lenses if you specifically need their maximum apertures or focus speeds for what you are shooting.

A dRebel with 18-55 lens will look just peachy as a 400x600 pixel web image.

If anything it means I am making comparisons that may cause me extra work or expense to get an image that does not come out any better. Again, if you print or crop to near full resolution, then you should be comparing those kinds of crops. For most of us, it just makes more work for no gain.

Not only that, think of all the money you can save. Web-only shooters can get an awesome 1D body for $500 with no loss in IQ. They can use any Canon lens they want without worrying about sharpness.

You cheated yourself getting that XS. You could have picked up a 4MP used 1D for the same money. It would have all those 1D features and the 4MP images would look just as good (or better I suppose) as the 12MP images from your XS. You traded features for megapixels, and now you are tossing the pixels away.

S.Horton
20th of February 2009 (Fri), 18:59
You still aren't convincing me. If the image looks good at the magnification level you are going to view and/or print it at, why does it need to look good at full magnification?

If anything it means I am making comparisons that may cause me extra work or expense to get an image that does not come out any better. Again, if you print or crop to near full resolution, then you should be comparing those kinds of crops. For most of us, it just makes more work for no gain.

I'm not trying to convince you of anything.

You asked why, you have answers.

What you do with them, that's up to you. ;):cool:

Radtech1
20th of February 2009 (Fri), 20:11
For most of us, it just makes more work for no gain.

Then don't do it. Those of us who find it a useful means of quickly assessing the quality of an image will continue to do so, with or without you understanding why.


Rad

kitacanon
20th of February 2009 (Fri), 21:18
A 100% crop is just a standard, a known quantity useful for comparisons. As been said, without knowing the amount of enlargement you'd not know if you're comparing a postcard to a billboard...
...on the net, at full frame, printing nothing larger than 8x12 (or cropping a 4x6 from it, a 25% crop), the 3.3megapxl D30 did great work....it's the same for lenses...small full frame pix don't take advantage of "L" quality lenses...still, we upgrade, and compare...and get better lenses to take advantage of the higher resolution of the new bods...

BestVisuals
20th of February 2009 (Fri), 21:36
The answer has already been stated but the OP refuses to understand.

ANYTHING other than 100% crop is an approximation of the true image. 200% looks fuzzy because you're interpolating the missing data to show you a magnified image. 50% (for example) has interpolation going on to choose which pixels to show, which to omit, because you aren't seeing all of them, only half.

100% crop has nothing to do with final print size, I have no idea what people are thinking with that one. It's only used to make a side-by-side, apples-to-apples comparison of an image chip, a lens, flare, etc.

Oh and by the way, pixel peepers are GREAT. By doing so the camera manufacturers will have to improve their glass. For TOO LONG most of the camera makers (yes, Canon and Nikon) have made very poor lenses but we had no clean way to guage their lens performance. With digital, we do. I fully expect to see improvements in lenses, if just their ratings, with the advent of digital cameras.

rdenney
20th of February 2009 (Fri), 21:47
It seems like the standard benchmark comparison for any test, is side by side 100% crops. Why do we test this way when almost nobody uses 100% pixel level images for anything? The vast majority of the time we just view images at full screen, which may be up to 1920x1080, or print at 11x14 or smaller. It seems like it would make more sense to post real world sizes, rather than 100% crops for comparison.

Because such comparisons are intended to evaluate one aspect of a lens (resolution). Showing a whole image on a computer screen doesn't do that.

I put up a web page a while back showing the bokeh of a range of lenses. You won't see any 100% crops there, because for that lens characteristics, the whole image is needed.

The problem isn't the comparison. The problem is the disproportionate conclusions people draw from those comparisons.

Rick "who sometimes crops and often makes bigger prints than you list" Denney

DC Fan
21st of February 2009 (Sat), 10:19
Why do we test this way when almost nobody uses 100% pixel level images for anything?

Because it's easier than judging a picture on composition, subject or impact.

CAL Imagery
21st of February 2009 (Sat), 11:13
Not only that, think of all the money you can save. Web-only shooters can get an awesome 1D body for $500 with no loss in IQ. They can use any Canon lens they want without worrying about sharpness.

You cheated yourself getting that XS. You could have picked up a 4MP used 1D for the same money. It would have all those 1D features and the 4MP images would look just as good (or better I suppose) as the 12MP images from your XS. You traded features for megapixels, and now you are tossing the pixels away.
I was just going to suggest the 1D. I'd probably have it by now, but, I do print off big prints, such as a 20x30, which happens to be 7.5x larger sq in wise than an 8x10...

hollis_f
21st of February 2009 (Sat), 12:29
Because it's easier than judging a picture on composition, subject or impact.
bw!

tonylong
21st of February 2009 (Sat), 18:19
Because it's easier than judging a picture on composition, subject or impact.

?

That's funny, suggesting that people who need excellent image quality don't also want good composition, subject matter and impact. We can also throw out good exposure, since it's not as important your big three, right?

I can understand how some, because of the photography and display/presentation requirements that they practice, don't find the want or need to pay attention to the finer details of image quality that make use of 100%/magnified views.

What I don't understand is why these photogs adopt an attitude of superiority over those that do. Whats up with this type of response? Have all the reasonable, rational and justifiable reasons for using this tool given in this thread gone totally like "swoosh"?

Radtech1
21st of February 2009 (Sat), 19:11
Because it's easier than judging a picture on composition, subject or impact.

I find your implication offensive. The idea that those of us who do find this tool useful (zooming to 100% as a means of checking the resolution of the image) - are doing so at the expense of the "composition, subject or impact".

Climb off your high horse and learn something: You can have an image of perfect composition, intriguing subject and dramatic impact all day long, but if you miss the focus you can throw all that away. You have nothing. Zip. Nada. THAT is the value of the 100% crop.

I zoom to 100% because sometimes I print that big. (24" on the short side with an iPF6100.) I suppose if all you do is print your little 4 x 6s on your little Pixma, then a 100% crop is more than you need - but don't begrudge those of us who actually make use that particular tool.

I imagine you look down your nose at checking the display after taking a shot, too.

Rad

sjones
21st of February 2009 (Sat), 19:26
It doesn't.

What you seem to be missing is that people looking at 100% crops are doing so because they are planning to print the images large.

If you are never going to print large then you do not need to view 100% crops. You also don't need a camera with higher resolution than the 1D classic and you only need L lenses if you specifically need their maximum apertures or focus speeds for what you are shooting.

A dRebel with 18-55 lens will look just peachy as a 400x600 pixel web image.



Not only that, think of all the money you can save. Web-only shooters can get an awesome 1D body for $500 with no loss in IQ. They can use any Canon lens they want without worrying about sharpness.

You cheated yourself getting that XS. You could have picked up a 4MP used 1D for the same money. It would have all those 1D features and the 4MP images would look just as good (or better I suppose) as the 12MP images from your XS. You traded features for megapixels, and now you are tossing the pixels away.

I have no proof of this, so everyone can dump on me to thy heart's delight, but I would assume that the majority of hobbyists who scrutinize photos at 100 percent not only never make large prints, but also never make prints of any size.

Don't get me wrong, if people want to check out their stuff at 100 percent or even 1,000 percent for whatever reasons, including removing dust spots, it's not going to ruin my day, but yes, what the photo looks like under normal viewing conditions, whether as a highly compressed jpeg on a web site or a 4'X4' fine art print, is what generally matters.

JeffreyG
21st of February 2009 (Sat), 19:52
I have no proof of this, so everyone can dump on me to thy heart's delight, but I would assume that the majority of hobbyists who scrutinize photos at 100 percent not only never make large prints, but also never make prints of any size.

Beats me. I assume that they make prints because I do. More than anything I print 11 x 14 photo books which demand pretty good resolution (especially if I need to crop 20% or so). These books are excellent for documenting all sorts of things and they make awesome gifts.

What I do try to point out to people is that if they are not printing, or are not making at least 8x12 prints then 100% crops, sharp lenses and lots of pixels are pretty much meaningless to them.

But if peering at 100% crops of images they will never print makes them happy, who am I to say otherwise?

Don't get me wrong, if people want to check out their stuff at 100 percent or even 1,000 percent for whatever reasons, including removing dust spots, it's not going to ruin my day, but yes, what the photo looks like under normal viewing conditions, whether as a highly compressed jpeg on a web site or a 4'X4' fine art print, is what generally matters.

As long as they are not printing.

I think that about 30% of my 'keepers' wind up getting printed in a book, and since I generally don't know which of them will be printed full bleed I tend to sort them assuming that any and all need to meet the requirements of this print size.

I've kept mildly OOF shots in the past though, but I've added tags indicating that they are 'web only'. It's funny but you can put mildly OOF shots up on the web and nobody can tell.

tonylong
21st of February 2009 (Sat), 20:20
I have no proof of this, so everyone can dump on me to thy heart's delight, but I would assume that the majority of hobbyists who scrutinize photos at 100 percent not only never make large prints, but also never make prints of any size.

Don't get me wrong, if people want to check out their stuff at 100 percent or even 1,000 percent for whatever reasons, including removing dust spots, it's not going to ruin my day, but yes, what the photo looks like under normal viewing conditions, whether as a highly compressed jpeg on a web site or a 4'X4' fine art print, is what generally matters.

I don't get the point here. Are you encouraging people to not aspire to develop their skills and utilize their great gear to be able to achieve technical excellence as well as creative excellence? That doesn't make sense to me.

Why, for instance, use a tripod and good technique to get a good, sharp landscape/scenic photo or a macro or still life photo if you don't care if you maximize sharpness and resolution of detail? If that's the thinking, then using a P&S for such shots would be just as good and much more convenient -- in fact, I've gotten some fine scenics with a compact camera, but even then good craftsmanship has to come into play.

Acting like hobbyists as a whole don't want good enough quality to make the most of their images seems quite, well, not even sensible to me, a hobbyist.

sjones
21st of February 2009 (Sat), 20:55
I don't get the point here. Are you encouraging people to not aspire to develop their skills and utilize their great gear to be able to achieve technical excellence as well as creative excellence? That doesn't make sense to me.

Why, for instance, use a tripod and good technique to get a good, sharp landscape/scenic photo or a macro or still life photo if you don't care if you maximize sharpness and resolution of detail? If that's the thinking, then using a P&S for such shots would be just as good and much more convenient -- in fact, I've gotten some fine scenics with a compact camera, but even then good craftsmanship has to come into play.

Acting like hobbyists as a whole don't want good enough quality to make the most of their images seems quite, well, not even sensible to me, a hobbyist.

Well, that's fair, because I don't get your point, or at least how you reached the conclusions that you did.

No, this is what I am saying, as I was simply agreeing, for the most part, with JeffreyG's overall point. The issue is a matter of frivolity, and of course, let me throw out the obligatory disclaimers: live and let live; I'm OK, you're OK; do as you please, doesn't bother me, and so on…

Now, with all that aside, my point, to clarify (as I can see where some misunderstandings might have arisen), is simply that sweating over details and spending five hours meticulously, obsessively, fastidiously performing a photographic autopsy at 100 percent crop might, just might, not be necessary if the photo is only going to be used as a compressed jpeg on a website. On the other hand, if you are going to print a 4'X4' for the Guggenheim, then yes, diving into the minutiae for hours on end might, just might, be worthwhile.

My admittedly unproven comment about hobbyists not making prints was to point out that while many of these folks will dive into a 100 percent crop interrogation simply because, frankly, it is the thing to do, it just maybe, might be, a perhaps possibly somewhat unnecessary exercise depending on the demands of the final output.

On the contrary, I am not arguing against quality but for it; and if one is not even going to print out their photos (or at least selected favorites), never mind large prints, which immutably offer the best quality, then slaving over 100 percent crop images can potentially be, in some cases, an irrelevant effort…but that's OK, "it's not going to ruin my day."

tonylong
21st of February 2009 (Sat), 21:30
Well, that's fair, because I don't get your point, or at least how you reached the conclusions that you did.

No, this is what I am saying, as I was simply agreeing, for the most part, with JeffreyG's overall point. The issue is a matter of frivolity, and of course, let me throw out the obligatory disclaimers: live and let live; I'm OK, you're OK; do as you please, doesn't bother me, and so on…

Now, with all that aside, my point, to clarify (as I can see where some misunderstandings might have arisen), is simply that sweating over details and spending five hours meticulously, obsessively, fastidiously performing a photographic autopsy at 100 percent crop might, just might, not be necessary if the photo is only going to be used as a compressed jpeg on a website. On the other hand, if you are going to print a 4'X4' for the Guggenheim, then yes, diving into the minutiae for hours on end might, just might, be worthwhile.

My admittedly unproven comment about hobbyists not making prints was to point out that while many of these folks will dive into a 100 percent crop interrogation simply because, frankly, it is the thing to do, it just maybe, might be, a perhaps possibly somewhat unnecessary exercise depending on the demands of the final output.

On the contrary, I am not arguing against quality but for it; and if one is not even going to print out their photos (or at least selected favorites), never mind large prints, which immutably offer the best quality, then slaving over 100 percent crop images can potentially be, in some cases, an irrelevant effort…but that's OK, "it's not going to ruin my day."

OK, I find Jeff a pretty sensible fellow, so if you are agreeing with him, you are trying to be a sensible fellow too:)!

Anyway, I can agree that some people get too anal about things, especially when pushing their gear to the limits of its capability. For example, I occasionally set out to push the limits of my gear either out of necessity or out of the exercise of testing my gears limits. So, for instance, while I had my 300 f/2.8 IS lens, I would occasionally go out of my way to push it by putting a 2x TC on it for wildlife shooting.

Now, the 300 is one of Canons sharpest lenses, and shooting with it alone or with a 1.4 TC gives you shots of utmost quality, and, guess what, each shot with that gear has to stand up at 100% view because I know anything else woud be my error or mistake.

But, with a 2x TC the tack-sharpness that I'm used to is softened. Detail is less than I would wish. Well, I understand those limitations. I will typically not crop too closely (after all, the whole purpose of using this combination is to more closely frame a shot). I'll post shots taken with this combo even with the given limitation, and I won't be picky about that limited sharpness.

Than being said, any movement on my part that shows at 100% is a reject, even if compressed at Web size looks pretty decent.

I also shoot at high ISOs a lot. I typically avoid shooting at higher than ISO 1600 (which is very useable with my two bodies) and viewing at 100% I apply a bit of noise reduction and sharpening and find the shots quite useable, even given larger prints. But occasionally I'll push it because of conditions or whatever and go to ISO 3200. Well, I don't care what body you shoot with, ISO 3200 is in the range of pretty noticeable noise when you are talking about any kind of close crop/close view. Well, I know that, and, after doing what I can with noise reduction/sharpening (which I do at 100%) and ensuring that focus and proper lack of motion, etc, is good at 100%, well, I'm fine because at a reasonable print size the noise will be very manageable.

What I don't do is "spend hours slaving at 100%" -- I'd agree with you that counter-productive. And, guess what? I don't hear people in this discussion and on the forum describing that as their workflow. So, I'd suggest that this is a kind of straw-dog argument -- where are these people?

Whenever we suggest to someone that they post a 100% crop as an example of a question/problem it's to try to help them analyze their question -- sometimes, for instance, a perceived lack of focus is simply a lack of contrast or need of basic input sharpening that we can't tell from a Web-compressed full image. Who's to know if you can't see the actual uncompressed detail of the shot?

nicksan
21st of February 2009 (Sat), 21:41
I got a 30" monitor and I hardly ever print out my images.
Where does that put me? :-)

JeffreyG
21st of February 2009 (Sat), 21:43
I got a 30" monitor and I hardly ever print out my images.
Where does that put me? :-)

You are in the camp that does not need to view 100% crops or use top end equipment.

Even a 30" monitor has far less than 10 million pixels.

sjones
22nd of February 2009 (Sun), 01:35
OK, I find Jeff a pretty sensible fellow, so if you are agreeing with him, you are trying to be a sensible fellow too:)!

Anyway, I can agree that some people get too anal about things, especially when pushing their gear to the limits of its capability. For example, I occasionally set out to push the limits of my gear either out of necessity or out of the exercise of testing my gears limits. So, for instance, while I had my 300 f/2.8 IS lens, I would occasionally go out of my way to push it by putting a 2x TC on it for wildlife shooting.

Now, the 300 is one of Canons sharpest lenses, and shooting with it alone or with a 1.4 TC gives you shots of utmost quality, and, guess what, each shot with that gear has to stand up at 100% view because I know anything else woud be my error or mistake.

But, with a 2x TC the tack-sharpness that I'm used to is softened. Detail is less than I would wish. Well, I understand those limitations. I will typically not crop too closely (after all, the whole purpose of using this combination is to more closely frame a shot). I'll post shots taken with this combo even with the given limitation, and I won't be picky about that limited sharpness.

Than being said, any movement on my part that shows at 100% is a reject, even if compressed at Web size looks pretty decent.

I also shoot at high ISOs a lot. I typically avoid shooting at higher than ISO 1600 (which is very useable with my two bodies) and viewing at 100% I apply a bit of noise reduction and sharpening and find the shots quite useable, even given larger prints. But occasionally I'll push it because of conditions or whatever and go to ISO 3200. Well, I don't care what body you shoot with, ISO 3200 is in the range of pretty noticeable noise when you are talking about any kind of close crop/close view. Well, I know that, and, after doing what I can with noise reduction/sharpening (which I do at 100%) and ensuring that focus and proper lack of motion, etc, is good at 100%, well, I'm fine because at a reasonable print size the noise will be very manageable.

What I don't do is "spend hours slaving at 100%" -- I'd agree with you that counter-productive. And, guess what? I don't hear people in this discussion and on the forum describing that as their workflow. So, I'd suggest that this is a kind of straw-dog argument -- where are these people?

Whenever we suggest to someone that they post a 100% crop as an example of a question/problem it's to try to help them analyze their question -- sometimes, for instance, a perceived lack of focus is simply a lack of contrast or need of basic input sharpening that we can't tell from a Web-compressed full image. Who's to know if you can't see the actual uncompressed detail of the shot?

Who said I was referring to anyone specifically on this particular thread, or even in the forum: read it in the broader context of my point, and it makes perfect sense, if for no other reason, because I am sensible…don't need anyone else on this site to validate this (said with a degree of facetiousness).

Moreover, because I do not operate with a black and white mindset (despite my photos), I am not even arguing against people who indulge in 100 percent crop molestation. I work with digitized film negatives, using 200 percent crop to clean them up, particularly in regards to scratches, but that is because I might print them out larger than my current 8X10 jobbies. As I noted, if you are working on a large print, I don't think it is overkill to spend as much time as needed to clean up or proof a photo.

"Five hours," of course, was supposed to be an obvious exaggeration, but fretting over minute details at 100 percent view even if for just five minutes might be a tad excessive, or, as some might say, not very sensible, if the photo is only going up on a web site…it all depends….and just as a general side note for the collective, I've already spit out all the appropriate disclaimers regarding "why should you care."

Also, my comments were not directed at people who use 100 percent view to analyze a new lens; to ensure that haloing (which can be detected on a web-based photo) doesn't occur during sharpening; or something of that nature; that's fine.

But as much flak as DC Fan received, he has a point. I don't think his remark was directed at everyone who dabbles in 100 percent crop voodoo, but there is a tendency among a few, one I think fueled by the digital medium, to become more concerned about what a photo (never mind one that won't even be printed out!) looks like at 100 percent than almost any other aspect of the photograph. This observation is not directed at anyone on this site, let alone this thread, but it would be a bit naive to think otherwise.

And finally, I am curious as to what 100 percent crop viewing has to do with maximizing the full capabilities of one's gear, unless you need such viewing to confirm the limits. But then again, if the photo as a whole will only be viewed on the web or even a small print, then maybe some of the flaws detected at 100 percent are irrelevant anyway, as they won't be seen when the photo is displayed in its natural setting.

tonylong
22nd of February 2009 (Sun), 02:07
Heh! I see you are trying to approach things from a perspective in a reasonable way:)! And, I hope that these discussions add to some kind of understanding and perspective.

Let me just say, I have nothing "bad" to say about people who prefer to...

Well, let me back up and assign labels. To those who spend at least some time regarding many if not all of their images at a 100% level, we can assign the often-prejorative label "pixel peepers". Those who do not, "non-pixel peepers".

So, I have nothing bad to say about non-pixel peepers. Period. Each of us gets to know what we are after in the craft and maybe/hopefully the art of photography.

Now what I see in the forum is that in fact none of the pixel peepers here rant against non-pixel peepers here. I've never seen a rant in which a pixel peeper says that non-pixel peepers are essentially wasting their time or pursuing a "false" idea of image quality.

I do, however, hear rants of non-pixel peepers implying that pixel-peepers are, in fact, wasting their time and pursuing a false idea of image quality.

Do you see why I get a bit miffed about that? Especially when many of us need to pixel peep for one reason or another?

Anders Östberg
22nd of February 2009 (Sun), 03:18
Ignorance is bliss ... ;)

tonylong
22nd of February 2009 (Sun), 03:22
Ignorance is bliss ... ;)

OK, your point is, as in what point are you referring to as ignorance?

Anders Östberg
22nd of February 2009 (Sun), 03:41
That referred to those who don't believe in 100% crops. There are several reasons why looking at pictures at 100% is useful, and those include comparisons to learn how your equipment functions (resolution, focus acccuracy), comparing performance when evaluating which lens to buy (Sigma :) ), post processing (sharpening, dust removal, cloning and more), and it's as has been said here the only somewhat objective way of comparing gear performance on the web. It has an impact on the final image quality whether the images are small or large, viewed on screen or in print, but if people don't want to know about this, and are happy with what is likely less than the best image quality they could have, then of course that's their choice. Thus - ignorance is bliss - if you don't think you need it you're just as happy not knowing about it.
(Note: ignorance is not a negative word, it just says you don't have a particular bit of knowledge).

sjones
22nd of February 2009 (Sun), 04:01
That referred to those who don't believe in 100% crops. There are several reasons why looking at pictures at 100% is useful, and those include comparisons to learn how your equipment functions (resolution, focus acccuracy), comparing performance when evaluating which lens to buy (Sigma :) ), post processing (sharpening, dust removal, cloning and more), and it's as has been said here the only somewhat objective way of comparing gear performance on the web. It has an impact on the final image quality whether the images are small or large, viewed on screen or in print, but if people don't want to know about this, and are happy with what is likely less than the best image quality they could have, then of course that's their choice. Thus - ignorance is bliss - if you don't think you need it you're just as happy not knowing about it.
(Note: ignorance is not a negative word, it just says you don't have a particular bit of knowledge).

No, it does not always have an impact; it's not a perquisite for quality at all, especially if you are posting a, when viewed, 12 centimeter by 12 centimeter compressed jpeg on the web.

To suggest that it is always necessary irrespective of the output just promotes redundant activity.

Tom Reichner
22nd of February 2009 (Sun), 04:26
If you have two extremely similar images, and need to know which one to use, you want to make sure it's the best one, right? If two images look exactly alike - but that can't be, can it? One of them must be at least a tiny, tiny , tiny bit sharper than the other. Or perhaps there's just an extremely minute bit less noise. Our eyes just aren't good enough to be able to figure out which one is just a couple thousandths of a percentage better than the other, so we crop them both to help us decide. It's the equivalent of putting something under a microscope. Cropping magnifies something for us so our eyes can better evaluate and compare. By doing so, we ensure that the image we keep and use is truly the better image.

DC Fan
22nd of February 2009 (Sun), 18:02
Example: Migrant Mother, the picture of a 1930's sharecropper by Dorothea Lange. (http://en.wikipedia.org/wiki/Dorothea_Lange) It's an iconic image that represents a dreary time in American history.

Viewed as a complete picture, it tells a powerful story. But view it at high magnification (http://upload.wikimedia.org/wikipedia/commons/5/54/Lange-MigrantMother02.jpg) and there are flaws. It's soft. It's grainy. The depth of field is extremely shallow.

By some 21st-century standards, this image would be a rejected failure on technical merits.

Anders Östberg
22nd of February 2009 (Sun), 18:24
Sharpness is of course only one aspect of a picture, and for a lot of photography it's not even near the top of the list of what makes a picture great. For some pictures it's even wrong, you may want a soft look. That doesn't make sharpness meaningless though, for some types of photography it's important, and what's wrong about having great content, good composition, good light *and* a sharp picture to give you options on what the end result will look like?

Saying sharpness and 100% crops don't matter because all we do is look at small pictures on the web is a bogus argument; if this is all people plan to use their cameras for I would suggest cancelling that 5D Mark II order and save some mis-spent money by buying a cheaper model. A picture is shown here compressed and in a small size but the primary use is perhaps a large print on a customer's wall.

tonylong
22nd of February 2009 (Sun), 18:29
Example: Migrant Mother, the picture of a 1930's sharecropper by Dorothea Lange. (http://en.wikipedia.org/wiki/Dorothea_Lange) It's an iconic image that represents a dreary time in American history.

Viewed as a complete picture, it tells a powerful story. But view it at high magnification (http://upload.wikimedia.org/wikipedia/commons/5/54/Lange-MigrantMother02.jpg) and there are flaws. It's soft. It's grainy. The depth of field is extremely shallow.

By some 21st-century standards, this image would be a rejected failure on technical merits.

Ummkay, so is your point that it's somehow superior to have inferior "technical" (i.e. clean, sharp, well exposed) images? Or just that a photographer can sometimes produce an exception that is exceptional? I'd hate to see people rush out and take sloppy pictures in the name of "art"...

sjones
22nd of February 2009 (Sun), 18:46
...Saying sharpness and 100% crops don't matter because all we do is look at small pictures on the web is a bogus argument; if this is all people plan to use their cameras for I would suggest cancelling that 5D Mark II order and save some mis-spent money by buying a cheaper model. A picture is shown here compressed and in a small size but the primary use is perhaps a large print on a customer's wall.

No one is remotely, even closely, arguing that everyone only looks at small pictures for the web so therefore 100 percent crops do not matter, but are you suggesting that everyone who uses 100 percent view do so because they are all making large prints?

sjones
22nd of February 2009 (Sun), 18:47
Ummkay, so is your point that it's somehow superior to have inferior "technical" (i.e. clean, sharp, well exposed) images? Or just that a photographer can sometimes produce an exception that is exceptional? I'd hate to see people rush out and take sloppy pictures in the name of "art"...

I would hate to see people trash a photo that is exceptional under normal viewing conditions, just because flaws are detectable at 100 percent view.

sjones
22nd of February 2009 (Sun), 18:52
...if this is all people plan to use their cameras for I would suggest cancelling that 5D Mark II order and save some mis-spent money by buying a cheaper model...

This is exactly JeffreyG's point, and one worth considering.

tonylong
22nd of February 2009 (Sun), 19:15
I would hate to see people trash a photo that is exceptional under normal viewing conditions, just because flaws are detectable at 100 percent view.

I won't speak for everyone, but for me I have various criteria and levels that I flag a photo for. As I said earlier, I have some that are less than stellar viewed at "full size" but fly well at a smaller size, say compressed for the Web or printed at 8"x10". I can and do use these, of course, within their limits. But that being said, those are their limits. Not every shot I take and keep can be used as a large print/publishable image, but I still have a standard -- I look for and aspire towards those images that do have that level of quality. I can't see advocating something less for "serious" photographers.

Let's say you've spent your time photographing things that you love, and, over time have amassed a good collection of what you feel are quality images, stored in your library and proudly displayed in a Web gallery, some maybe printed as 8x10s to share with others...it's all good, right?

Then suppose, over time, that your aspirations for sharing your photos with others expand to, say, selling, publishing, and printing at gallery-size prints. So, you enthusiastically go back to your library and begin for the first time to examine hundreds and thousands of shots more closely to process them into professional quality photos. Of course, because you've always viewed and displayed them small, so you "never labored over 100% views", which in turn never inspired you to improve your craft and, say, your lens set to give you the very best results -- you were satisfied with those nice, compressed images and had no need to do anything different. So, when you open your images to process them for these "serious" purposes, what do you find? Is it possible or even likely that a large number of what you thought were "great images" would not pass muster?

Just a possibility to consider!

sjones
22nd of February 2009 (Sun), 19:59
I won't speak for everyone, but for me I have various criteria and levels that I flag a photo for. As I said earlier, I have some that are less than stellar viewed at "full size" but fly well at a smaller size, say compressed for the Web or printed at 8"x10". I can and do use these, of course, within their limits. But that being said, those are their limits. Not every shot I take and keep can be used as a large print/publishable image, but I still have a standard -- I look for and aspire towards those images that do have that level of quality. I can't see advocating something less for "serious" photographers.

Let's say you've spent your time photographing things that you love, and, over time have amassed a good collection of what you feel are quality images, stored in your library and proudly displayed in a Web gallery, some maybe printed as 8x10s to share with others...it's all good, right?

Then suppose, over time, that your aspirations for sharing your photos with others expand to, say, selling, publishing, and printing at gallery-size prints. So, you enthusiastically go back to your library and begin for the first time to examine hundreds and thousands of shots more closely to process them into professional quality photos. Of course, because you've always viewed and displayed them small, so you "never labored over 100% views", which in turn never inspired you to improve your craft and, say, your lens set to give you the very best results -- you were satisfied with those nice, compressed images and had no need to do anything different. So, when you open your images to process them for these "serious" purposes, what do you find? Is it possible or even likely that a large number of what you thought were "great images" would not pass muster?

Just a possibility to consider!

As I stated earlier in this thread, I go through all my keepers at 200 percent, so I am not worried about what I will find should I print above 8X10. As I repeatedly stated, if the intention exists to reproduce high quality prints, particularly large ones, then ensuring quality control is important. You, to a degree, are actually reiterating making my argument.

Moreover, we're talking about the universal catalogue of all photographs, and no, many photos just aren't meant to be blown up to the size of the average American butt and slapped up on a wall.

Furthermore, many photographers in this world managed to hone their craft just fine without 100 percent view.

Here's something to consider, you don't need to view the photos in Robert Frank's "The Americans" at 100 percent to see an occasional technical flaw, they're pretty apparent even in small print, and yet, I never seen a more compelling collection of photographs in my life, and I am hardly alone in my admiration for this book: why is that?

I, and I don't DC Fan, are advocating sloppiness, and even if you never intend to print large, that's OK, quality, quality, quality…it's mostly a good thing; the issue, however, is making sure that technical consideration doesn't suffocate the importance of a photograph's overall essence…something about not seeing the forest for the trees comes to mind.

I stand by point, I would hate to see people trash a photo that is exceptional under normal viewing conditions, just because flaws are detectable at 100 percent view. And I will add that if, in certain cases, normal viewing condition is never anything larger than 8X10, then fine! A photo should not always have to prove its merit by how large it can be cleanly printed. Presentation is not shackled to just one option.

tonylong
22nd of February 2009 (Sun), 20:17
Well, since the process that you describe is pretty much what I in practice do, I guess we're pretty much in agreement. I'd rather have a useful discussion than just try to lock horns with someone who is just ranting against "pixel peepers" and ignoring the issues that we are dealing with.

rdenney
23rd of February 2009 (Mon), 00:44
Again: Who says comparison images at 100% are intended as evaluations of the photograph, rather than evaluations of the lens or technical quality?

As a musician, I might play perfectly in tune and in time a flood notes, and still not be musical. But I know of nobody who is a musically accomplished musician who cannot perform in tune and in time. Once, I asked a top professional performer, with whom I was studying, which was more important, musicianship or technique. His answer was "yes".

That argument for both artistic and technical excellence is not undermined by (1) the existence of wonderful photos that are nevertheless technically deficient in some way, or (2) my own arguments elsewhere that people focus on what's important about expression, saving the string of expensive equipment purchases for later.

Plus, I can admire great technique and the effect of superior equipment even if I don't admire the resulting art.

RIck "who evaluates his photos at 100%, but may still like them even if they show poorly in that evaluation" Denney

cdifoto
23rd of February 2009 (Mon), 00:52
You still aren't convincing me. If the image looks good at the magnification level you are going to view and/or print it at, why does it need to look good at full magnification?
Because sometimes I need to uprez. I need a file that's pristine to look good uprezzed. Ass uprezzed is uprezzed ass and I would never sell that to a client.

Anders Östberg
23rd of February 2009 (Mon), 03:12
No one is remotely, even closely, arguing that everyone only looks at small pictures for the web so therefore 100 percent crops do not matter, but are you suggesting that everyone who uses 100 percent view do so because they are all making large prints?


Of course I don't. But maybe you should ask the OP - he's the one saying "we" all only use small pictures, making pixel level quality meaningless. That's what started all this hoopla, and I'm only answering his question why someone would and should be interested in 100% crops.

Anders Östberg
23rd of February 2009 (Mon), 03:17
I won't speak for everyone, but for me I have various criteria and levels that I flag a photo for. As I said earlier, I have some that are less than stellar viewed at "full size" but fly well at a smaller size, say compressed for the Web or printed at 8"x10". I can and do use these, of course, within their limits. But that being said, those are their limits. Not every shot I take and keep can be used as a large print/publishable image, but I still have a standard -- I look for and aspire towards those images that do have that level of quality. I can't see advocating something less for "serious" photographers.

Let's say you've spent your time photographing things that you love, and, over time have amassed a good collection of what you feel are quality images, stored in your library and proudly displayed in a Web gallery, some maybe printed as 8x10s to share with others...it's all good, right?

Then suppose, over time, that your aspirations for sharing your photos with others expand to, say, selling, publishing, and printing at gallery-size prints. So, you enthusiastically go back to your library and begin for the first time to examine hundreds and thousands of shots more closely to process them into professional quality photos. Of course, because you've always viewed and displayed them small, so you "never labored over 100% views", which in turn never inspired you to improve your craft and, say, your lens set to give you the very best results -- you were satisfied with those nice, compressed images and had no need to do anything different. So, when you open your images to process them for these "serious" purposes, what do you find? Is it possible or even likely that a large number of what you thought were "great images" would not pass muster?

Just a possibility to consider!

A lot of people have experienced exactly this when they start submitting their favourite images to stock photo agencies. Whether most uses of stock photos really need super quality can also be debated but that's the criteria for admission.

If all you do is use small pictures on the web (and that's of course completely OK if that's your intent) then by all means don't worry too much about good quality, spend your time worrying about good content instead.

If you envision doing more than web display with your existing and yet-to-be-taken pictures you may want to start practicing how to get the quality - small prints are easy, really great quality in a poster size print is harder. I know because I'm still a sloppy photographer and it has hurt my sales when I've turned down a sale because my soft picture couldn't deliver something I would be comfortable selling at the requested size. It takes time and analysis of what you do to become good, so why not start early and practice the craft right along with finding the right content?

Tom Reichner
23rd of February 2009 (Mon), 03:40
A lot of people have experienced exactly this when they start submitting their favourite images to stock photo agencies ...

Agreed. Unless you are very nit picky with your images and only accept the very, very finest quality, it will be very tough (and embarassing) to try to sell them to top rate publishers and stock agencies. You're competing against so many other photographers whose images do look great at 100% crop, or more. So why would a publisher choose something slightly inferior when the better ones are right there at their fingertips? It's not really about what looks good at a smaller size, it's about what a photo editor or art director is going to select.

DC Fan
23rd of February 2009 (Mon), 08:23
So why would a publisher choose something slightly inferior when the better ones are right there at their fingertips? It's not really about what looks good at a smaller size, it's about what a photo editor or art director is going to select.

I had something like that happen once.

http://www.fansview.com/racing/123005d0058.jpg

Tony Stewart celebrating after winning an indoor race in Fort Wayne. Ind.

Technically, this image I took is a mess. It's from a camera that many don't like, a Digital Rebel 300d, and from an often-criticized lens, a 75-300mm standard zoom. Exposure is wrong. Shot at ISO 1600, it's way, way too grainy and barely in focus. There was little or nothing I could do to fix this image's many faults, but I was dumb enough to send it to National Speed Sport News with a few other pictures.

And the paper ran this image.

It didn't look any better on paper than this screen version, but it got into print. There were plenty of photographers at this race and several of them regularly send images to Speed Sport, but the paper chose to run this substantially-flawed post-race picture rather than a technically superior image of a posed Stewart in victory lane.

Why? Did an editor at the paper take pity on my work? Was that week's edition assembled by someone who didn't understand imaging or had no sense of properly exposed pictures? Or was it that I was the only person with a camera who was in the right place to catch this unposed, spontaneous expression of victory? (Note the guys with cameras in the lower background - most of the night's photographers worked the race from the infield while my view was from an elevated area right behind the grandstands.)

Sometimes pictures get published because of technical quality. Sometimes not-so-good pictures get published because of an old-fashioned, pre-digital idea - that they tell a story.

Anders Östberg
23rd of February 2009 (Mon), 08:43
Content always wins over image quality, I've repeated that many times in (too) technical pixel-peeping discussions. You evidently saw this since you did submit the image, but if there had been a photographer next to you with better camera settings his image would likely have been in the paper instead.

Some pictures also get selected simply because of who took them, papers have agreements with photographers who expect to get published and paid for a job. I've taken pictures that are way better than what got published in our local paper but since I wasn't the guy with the assignment that doesn't count - the other picture was deemed "good enough" and that's that.

Bosscat
23rd of February 2009 (Mon), 16:31
Why? Did an editor at the paper take pity on my work? Was that week's edition assembled by someone who didn't understand imaging or had no sense of properly exposed pictures? Or was it that I was the only person with a camera who was in the right place to catch this unposed, spontaneous expression of victory? (Note the guys with cameras in the lower background - most of the night's photographers worked the race from the infield while my view was from an elevated area right behind the grandstands.)

Sometimes pictures get published because of technical quality. Sometimes not-so-good pictures get published because of an old-fashioned, pre-digital idea - that they tell a story.

Often times it because the editor is too lazy to look for a good pic from another photographer, or are so burnt out making deadline, they just use whatever is handy.

VWpilot hit the nail on the head, way back earlier in this thread.

Tom Reichner
23rd of February 2009 (Mon), 16:57
VWpilot hit the nail on the head, way back earlier in this thread:

100% is just that, 100%, its what the image ACTUALLY is. In the same way that people looked at slides with loupes, we should look at our digital files at 100% because anything less than that is not showing the actual image, but a compressed version of it.

I can take a photo that is just plain out of focus. Its no good, but looks good at 800x600 for the web because the compression has removed the softness out of it. However, its not a sharp image and was not taken properly.

The only way to tell that is if you view it at 100%. Its not pixel peeping, its viewing it as it was meant to be viewed. For me, any shot that is not perfect at 100% goes tot he recycle bin as that is the only way I can be confident it can be used in whatever way it needs to be used, not just to show off in web forum where no one can really tell if it was ever in focus or not.

Agreed - he's exactly right!