Home · Register · Join Upload & Sell

Moderated by: Fred Miranda
Username  

  New fredmiranda.com Mobile Site
  New Feature: SMS Notification alert
  New Feature: Buy & Sell Watchlist
  

FM Forums | Canon Forum | Join Upload & Sell

1       2       3       4      
5
       6              9       10       end
  

Archive 2013 · Canon EF 24-70mm f/4L IS Resolution Tests!

  
 
StillFingerz
Offline
• • • •
Upload & Sell: Off
p.5 #1 · p.5 #1 · Canon EF 24-70mm f/4L IS Resolution Tests!


gdanmitchell wrote:
It turns out that this concern is vastly over-rated, and that barrel distortion and other similar sorts of lens distortions can be corrected in post with typically no visible degradation of the image.

Dan


Hate to disagree Dan, but pretty much any distortion corrected in post is going to degrade an image. What's acceptable to one's eye may not be to another...let alone too much time spent peeping. I've had fun...sarcasm... correcting my EF 24-85 in post and while it's not 'L' glass, at 24mm pregnant is pregnant and the 24-105 looks pretty much like 24-85.

The price of this new 24-70 will eventually drop a few hundred and become quite attractive; for some. With it's newer hybrid IS it will pair well with the 70-200 f4L and if the AF/IS is anything like the new 100L macro, it will be a fine bit of light glass to use, seems like the perfect match for the 6D...perhaps.

Less time in post, more time out shooting is more pleasing IMHO
Jerry



Jan 05, 2013 at 01:26 AM
Gunzorro
Offline
• • • • • •
Upload & Sell: On
p.5 #2 · p.5 #2 · Canon EF 24-70mm f/4L IS Resolution Tests!


Sorry Dan, but being the terrible pixel peeper I am in PP, I see image degradation on the sides from correcting serious barrel or pincushion. It is very obvious to see (in addition to losing serious imaging real estate). I don't like it or what it does to the image.


Jan 05, 2013 at 01:40 AM
alundeb
Online
• • • • •
Upload & Sell: Off
p.5 #3 · p.5 #3 · Canon EF 24-70mm f/4L IS Resolution Tests!


gdanmitchell wrote:
It turns out that this concern is vastly over-rated, and that barrel distortion and other similar sorts of lens distortions can be corrected in post with typically no visible degradation of the image.

Dan


I see a degradation, and that is despite that the test is cluttered by two factors:

The crop is taken from the far corner where the quality is worst to begin with. To show the true difference, it would be better to use a patch from anywhere else where the quality is better to begin with. Distortion correction does not only operate in the corner, and showing the corner really has no point.

The file size indicates a jpg compression at about 7-8 (out of max 12) in photoshop. This is barely "high" quality. I see jpg compression artifacts using that level. For pixel peeping and test presentations, I would not go below 10 ("maximum"). For printing I only use uncompressed TIFF.



Jan 05, 2013 at 04:00 AM
gdanmitchell
Offline
• • • • • • •
Upload & Sell: Off
p.5 #4 · p.5 #4 · Canon EF 24-70mm f/4L IS Resolution Tests!


alundeb wrote:
I see a degradation, and that is despite that the test is cluttered by two factors:

The crop is taken from the far corner where the quality is worst to begin with. To show the true difference, it would be better to use a patch from anywhere else where the quality is better to begin with. Distortion correction does not only operate in the corner, and showing the corner really has no point.

The file size indicates a jpg compression at about 7-8 (out of max 12) in photoshop. This is barely "high" quality. I see jpg compression artifacts using that
...Show more

The reason that the sample - and I believe you are only speaking of the "tilted building" example - was taken from the far corner is because that is where the greatest effect of the correction would be seen. In the center, where resolution would be the highest to begin with, the amount of correction is extremely tiny and only a fraction of the somewhat radical correction applied in the corners.

You are looking at a resolution that corresponds to print size of perhaps five to six feet wide. Perhaps if you made two 6 foot wide prints, snipped out the corners, set them on the table under a very bright light, and looked back and forth between the two of them long enough you might convince yourself that you could prefer one over the other.

Meanwhile, back in the real world, there is no difference in resolution or anything else in these images that would be remotely visible in, say, a 24" x 36" print. Except that the corrected version would obviously look much better than the tilted and skewed uncorrected version. :-)

Of course, the other "flaw" with this "test" is that people who know that it involves an assessment of the amount of degradation caused by correcting barrel distortion or perspective distortion in post and who have a pre-existing bias towards a belief that this method of correction is a bad thing will tend to determine that the example that seems to have been corrected has flaws since that is consistent with their belief system. They may even look for other possible flaws in the test procedure (using the corners rather than the center, the level of jpg compression, etc) as a way to distract from the demonstration that they cannot really see the difference that they believe they should be able to see.

The issue of correction in post is one of those that challenges the orthodoxy of many who are utterly certain that only certain very expensive lenses that capture "perfectly" in camera can produce excellent image quality. For those who invest themselves in such beliefs, it can be tremendously difficult to look objectively at alternatives in the context of real world photography and accept the possibility that the orthodoxy is flawed. I also started out believing (and even telling others) that "correcting in post can only degrade the image quality." Now, after actually investigating this issue and making a lot of prints of images that have been corrected in ways such as those shown in the samples, I am happy to report that the orthodoxy is wrong.

I notice that you did not offer a guess as to which half of the other example was the corrected one and which wasn't. You have a 50% chance of guessing correctly. ;-)

Dan

(By the way, when I look at the 50mm prime "barrel distortion" example very closely in the side-by-side presentation, I can just barely see the difference in resolution in the 100% crops. If you showed me either one by itself, or showed them to me one after the other instead of side-by-side, I'm pretty certain I couldn't pick out the resolution difference. I know I can't see this sort of difference in a 18" x 24" print. When I look at the other set of samples of the "tilted building," they both simply look pretty much equivalent to me - aside from the difference in alignment of the image elements I cannot see any significant difference in resolution, even in at 100% magnification equivalent to looking at the far corners of a 60" print.)



Jan 05, 2013 at 11:28 AM
StillFingerz
Offline
• • • •
Upload & Sell: Off
p.5 #5 · p.5 #5 · Canon EF 24-70mm f/4L IS Resolution Tests!


Ah yes, the need to label, categorize one's fellow humans as some elitist gang of thugs with their perfect bits of glass yet unable to see anything objectively for they must be blinded by bias, not of independent thought or possible experience. Oh no no, it can't be, they're just the mob...what a bunch of fresh road apples, smells a bit of ego...or bias perhaps.

Really Dan, forumography opinion, how dare you be like...the rest of us...all knowing



Jan 05, 2013 at 11:47 AM
alundeb
Online
• • • • •
Upload & Sell: Off
p.5 #6 · p.5 #6 · Canon EF 24-70mm f/4L IS Resolution Tests!


gdanmitchell wrote:
The reason that the sample - and I believe you are only speaking of the "tilted building" example - was taken from the far corner is because that is where the greatest effect of the correction would be seen. In the center, where resolution would be the highest to begin with, the amount of correction is extremely tiny and only a fraction of the somewhat radical correction applied in the corners.


There are two sources of image degradation from geometrical correction. One is proportional to the amount of correction, and one just follows the resampling that has to take place anywhere some amount of correction is applied. The visible impact of the resampling per se is larger than the visual impact of stretching as long as the stretching is below a few percent.


You are looking at a resolution that corresponds to print size of perhaps five to six feet wide. Perhaps if you made two 6 foot wide prints, snipped out the corners, set them on the table under a very bright light, and looked back and forth between the two of them long enough you might convince yourself that you could prefer one over the other.

Meanwhile, back in the real world, there is no difference in resolution or anything else in these images that would be remotely visible in, say, a 24" x 36" print. Except that the corrected version would
...Show more

I am not arguing that the difference alone is big or even significant at some print sizes. It is just that your (valid) point about using the next best can be close enough to the best, loses strength as we combine the effects of many factors. You use the same arguments about sensor resolution, lens resolution, diffraction and software correction algorithms.

When I print an image from a 36 MP camera, taken with the best lenses, at f/8, and avoid software corrections that rely on resampling, the result is clearly better than using a 21 MP camera with a good zoom lens at f/16 and corrected.



Of course, the other "flaw" with this "test" is that people who know that it involves an assessment of the amount of degradation caused by correcting barrel distortion or perspective distortion in post and who have a pre-existing bias towards a belief that this method of correction is a bad thing will tend to determine that the example that seems to have been corrected has flaws since that is consistent with their belief system. They may even look for other possible flaws in the test procedure (using the corners rather than the center, the level of jpg compression, etc) as a
...Show more

Why would I have to guess? There are geometrical giveaways for both kinds of correction.



Jan 05, 2013 at 12:05 PM
gdanmitchell
Offline
• • • • • • •
Upload & Sell: Off
p.5 #7 · p.5 #7 · Canon EF 24-70mm f/4L IS Resolution Tests!


alundeb wrote:
Why would I have to guess? There are geometrical giveaways for both kinds of correction.


I first published that "barrel distortion" test over five years ago and it has been viewed many, many times - despite which no one has ever identified which half of the image is "worse" on account of the correction. (See my added footnote to my earlier post in which I acknowledge that I can barely see it if I look very closely at 100% crops.)

As for prints, the notion of "better" is a funny thing that presupposes that there is some thing that might be identified as "perfection" in a print. There isn't. Sharpness or resolution is one of many factors that affect the perception of the quality of a print and the sense of sharpness is created by a number of means other than image or lens resolution... and by comparison to these other issues the essentially invisible differences that no doubt might be measured at 100% are virtually always utterly insignificant on their own, and more than overwhelmed to the positive effect (in at least some cases) of correcting for various sorts of image distortion (tilting buildings! curving lines!) that are more likely to be seen.

It turns out that I have both the 24-70 II and the 24-105 lenses, so I'm well aware of the strengths and weaknesses of both. If pure resolution and lack of barrel distortion and vignetting were the whole story, I would put the 24-105 away and never use it again since the 24-70 measures "better" on all of these counts - though not necessarily by as much as we might imagine. But I don't, for a whole range of reasons. (One is that I share the 24-105 with someone else... ;-) Today, in fact, I'm heading out for some serious photography, some of which will be tripod based and perhaps done at night. I'll take the 24-105.

Dan



Jan 05, 2013 at 12:19 PM
Gunzorro
Offline
• • • • • •
Upload & Sell: On
p.5 #8 · p.5 #8 · Canon EF 24-70mm f/4L IS Resolution Tests!


Dan -- I don't really want to make an enemy here, but the following is my evil response to a number of your points here and on other thread.

I don't really know where to start to respond to your points. You've done a clever debating maneuver, which I've seen you use numerous times, which is to throw out so many points full of so many half-truths and arbitrary measures that those who see the fallacies are left floundering trying to demonstrating why what you say is incorrect. The technique is every bit as annoying as the "straw man" complaint you so often raise when being criticized.

My opinion is that you often raise artificial and arbitrary measurements to support your position, with repeated mantras regarding printing and print sizing and viewing distance. It's fine that you are print-centric in your approach, but to use that as a yardstick of acceptable resolution is foolish. Printed images, despite being desirable for display, offer some of the worst resolution and reproduction of digital images -- well, dot-gain in "4-color" litho printing is worse -- my point is, printed material should not be the yardstick for anything but exhibition prints. Apparently "print rule" is part of your real world observation, but it is not part of mine or many other photographers.

I have no problem with you printing, or being interested in prints, or making a decent living off your prints. I'm a rarity I suppose and on the "other side" -- I never print, not since olden film days when clients final uses came from prints or transparencies. I never sell prints any more and don't get involved in the subject. More power to anyone who does, but it's not my thing. I have no special printing interest to protect that I'm making money off of. For me, it's all about computer screen viewing at X%, internet/email, and DVD burning for client use.

To throw "belief system" in the face of people looking for technical refinements and precise observations of resolution is mildly insulting (at the least). You practice a form of reverse snobbery, continually praising older proletarian gear as being more than adequate. That's fine as an opinion, but your tone embraces a type of closed-minded pedantry. If you feel the 24-105L is a wonderful lens, that's totally fine, and the 5D2 -- excellent choices. Just don't continually spout-off how [newer, pricier, techno] gear is not needed -- for YOU it is not needed or desired. Got it! Right! Move on, nothing more to see here.

You are correct in the matter of distortion: the center of images is least disturbed or reduced in resolution. But the areas most disturbed are the outer areas, which are already at the greatest disadvantage in quality and suffer from cropping. Reducing the IQ of that large surrounding surface area is not to be done lightly. You obviously have little interest in the surround areas (nearly 50% of the viewing area of the image) and instead base your opinion on the central portion, which in fairness is usually the greatest area of interest -- but not always, and probably should not be the basis of a broad rule. Many photographers strive for edge-to-edge sharpness. Being this is a gear forum, championing mediocrity is not going to be a terrible fruitful strategy.

Your comments are not all without merit, but your emphasis is to minimize some hard won technical points and interests among members, which not surprisingly raises hackles. In short, your balance and perspective seem often out of touch with "real life".

Your views are your views. But when so many of us are trying to improve our imaging and gear, we don't want to hear why we are part of a resolution cult, convincing our brains that improvements are real and not imaginary.



Jan 05, 2013 at 01:32 PM
gdanmitchell
Offline
• • • • • • •
Upload & Sell: Off
p.5 #9 · p.5 #9 · Canon EF 24-70mm f/4L IS Resolution Tests!


Gunzorro wrote:
My opinion is that you often raise artificial and arbitrary measurements to support your position, with repeated mantras regarding printing and print sizing and viewing distance. It's fine that you are print-centric in your approach, but to use that as a yardstick of acceptable resolution is foolish. Printed images, despite being desirable for display, offer some of the worst resolution and reproduction of digital images -- well, dot-gain in "4-color" litho printing is worse -- my point is, printed material should not be the yardstick for anything but exhibition prints.

I have no problem with you printing, or being interested in
...Show more

Resolution of current high end printers greatly exceeds that of computer screens, even when "dot gain" is taken into consideration. I can see more detail in a print that has a 27" diagonal measurement than I can see in the display of the same image so that it fills my 27" monitor. (Yes, I know about the new "retina displays" and similar, but the available versions are still relatively small.)

As my friend Charlie Cramer likes to say, an image that looks great as a print will almost always look good on the screen, but that the opposite is not anywhere near as certain.

As for the idea that the goal is to be able to view photographs at 100% resolution on the computer screen, I'll just say a couple of things:

1. To me, and I'm confident to most viewers of photographs, the point of all of the effort we put into composing and creating photographs is not to have small sections of them viewed at great magnifications on the screen, but rather to be able to view the photograph as a whole. (When I email a photograph or share one on the web, I cannot imagine sending or displaying a full resolution version, nor can I imagine that many people would want me to, since they want to see the photograph in its entirety on the screen.)

2. Since I've heard otherwise from a few people in photography forums, I'm will to concede that some of you may actually find that viewing the 100% magnification crop of a small section of the whole photograph brings you your greatest joy and strongest aesthetic response. I also admit that I do not understand this at all.

... your tone embraces a type of closed-minded pedantry...

Lovely thought... Thanks for that! ;-)

... championing mediocrity is not going to be a terrible fruitful strategy.

I most certainly do not "champion mediocrity" in any way or on any level!

Your comments are not all without merit, but your emphasis is to minimize some hard won technical points and interests among members, which not surprisingly raises hackles. In short, your balance and perspective seem often out of touch with "real life".

I believe that the other perspective is out of touch with "real life," at least in the context of making photographs. I ground my opinions - which others are certainly free to disagree with - in my experience with real photographs, and I try to check my theoretical presumptions against photographic reality. This has led me to give up some of my "beliefs" on many occasions. Some that I used to hold, but which I have given up after testing them include:

1. Printing must be done at 300 dpi or higher.
2. File resolution should be equal to printer resolution or some whole division of it.
3. Compensating for various kinds of distortion in post (barrel/pincushion, "perspective" tilt, CA, etc.) will degrade the image in significant ways.
4. Zoom lenses cannot produce high enough image quality for excellent prints.
5. Eliminating noise from photographs is always a good thing.
6. Sharpness is the most important lens characteristic.
7. "Perfect" technical quality is achievable and necessary.*

(* To deflect the expected response, this does not mean that technical quality does not matter nor that mediocre is good enough. )

In fact, right now as I write this, it occurs to me that there are probably other presumptions that I should reconsider and check against reality. :-)

Your views are your views. But when so many of us are trying to improve our imaging and gear, we don't want to hear why we are part of a resolution cult, convincing our brains that improvements are real and not imaginary.

I understand, but I encourage you and others in your situation to think more broadly about what it means to "improve your imaging" and what role gear does and does not have in achieving that goal. I can't speak for you specifically, but it is clear to me that we, as photographers, are often susceptible to the lure of gear as the answer to improving our photography. Gear is obviously not insignificant, but there is a real danger of focusing on gear for its own sake and at the expense of other things that make a bigger difference in our photography and which produce much larger rewards.

Take care,

Dan

Edited on Jan 05, 2013 at 02:16 PM



Jan 05, 2013 at 01:56 PM
timbop
Offline
• • • • •
Upload & Sell: Off
p.5 #10 · p.5 #10 · Canon EF 24-70mm f/4L IS Resolution Tests!


Roger - quick question about the "macro" mode. It reduces minimum FD to 2" from the front element, but what's the max working distance? That is, is focus fixed or do you have room to maneuver (with the obvious reduction in magnification)?


Jan 05, 2013 at 02:02 PM
StillFingerz
Offline
• • • •
Upload & Sell: Off
p.5 #11 · p.5 #11 · Canon EF 24-70mm f/4L IS Resolution Tests!


Dan, the reward has always been, will forever be the very act of seeing/creating. Gears, be it film, chemicals, bit n bytes are just tools... Sears makes great hammers, but a Vaughn in the right hands can inspire, bring confidence, and the craft/workmanship often betters as the experience is less tedious.


Jan 05, 2013 at 02:15 PM
gdanmitchell
Offline
• • • • • • •
Upload & Sell: Off
p.5 #12 · p.5 #12 · Canon EF 24-70mm f/4L IS Resolution Tests!


StillFingerz wrote:
Dan, the reward has always been, will forever be the very act of seeing/creating. Gears, be it film, chemicals, bit n bytes are just tools... Sears makes great hammers, but a Vaughn in the right hands can inspire, bring confidence, and the craft/workmanship often betters as the experience is less tedious.


I do quite a bit of photography, and I can say in all honesty that I get no more pleasure from using my most expensive lens than I do from using my least expensive lens, much less more inspiration. I don't even think about this when I'm making photographs. A lens is simply a tool that needs to perform well and consistently for the task at hand.

My photography is no better when I use a $2000 lens than when I use a $200 lens.

Dan



Jan 05, 2013 at 02:19 PM
spdntrxi
Offline
• •
Upload & Sell: Off
p.5 #13 · p.5 #13 · Canon EF 24-70mm f/4L IS Resolution Tests!


thread is in the crapper now...

I'm for one, am interested in the 24-70 F4IS and would like the discussion returned to that



Jan 05, 2013 at 02:24 PM
Gunzorro
Offline
• • • • • •
Upload & Sell: On
p.5 #14 · p.5 #14 · Canon EF 24-70mm f/4L IS Resolution Tests!


See, here you go again:
"Resolution of current high end printers greatly exceeds that of computer screens, even when "dot gain" is taken into consideration."
Computer screens at 72 DPI, yes, but not details at much higher magnifications -- that' s what I'm talking about you being fixated on your "real world" and stats, not THE real world and real stats. A 23" screen with an image detail viewed at 100% far exceeds, as you sort of concede -- but why must we wring your arm for a concession? Just be more balanced and less riding your high hobby horses. Different courses for different horses, but know when you are riding that hobby horse and not real one.

And
"As for the idea that the goal is to be able to view photographs at 100% resolution on the computer screen, I'll just say a couple of things:"
Where you follow with complete opinion, only minimally compromising due to recent forum backlash to your personal views on technical matters.

And I can't even approach the rest of your pre-rehearsed rebuttal in detail. . . so I'll have to let some go for now and "ambush" you later -- you leave little alternative, since all my material is fresh and not regurgitated.

I am willing to concede that you don't realize that you are using the forum as a bully pulpit. So, please prove me right.



Jan 05, 2013 at 02:30 PM
StillFingerz
Offline
• • • •
Upload & Sell: Off
p.5 #15 · p.5 #15 · Canon EF 24-70mm f/4L IS Resolution Tests!


gdanmitchell wrote:
I do quite a bit of photography, and I can say in all honesty that I get no more pleasure from using my most expensive lens than I do from using my least expensive lens, much less more inspiration. I don't even think about this when I'm making photographs. A lens is simply a tool that needs to perform well and consistently for the task at hand.

My photography is no better when I use a $2000 lens than when I use a $200 lens.

Dan


Guess you've never missed a shot or had an inferior bit of gear fail at the time of snap/contact...some of us have, you are most fortunate not to have experienced such. So much for splinters and near outed eyes...onward and upward I roll, there be images to create.

When this new 24-70 drops in price, my wallet gets fat enough, I'll give it a go and perhaps retire a trusty old friend; an EF 24-85, at over 3 Benjamins it's done well.



Jan 05, 2013 at 02:34 PM
gdanmitchell
Offline
• • • • • • •
Upload & Sell: Off
p.5 #16 · p.5 #16 · Canon EF 24-70mm f/4L IS Resolution Tests!


Gunzorro wrote:
I am willing to concede that you don't realize that you are using the forum as a bully pulpit. So, please prove me right.


OK ;-)



Jan 05, 2013 at 02:46 PM
Ralph Conway
Offline
• • • •
Upload & Sell: Off
p.5 #17 · p.5 #17 · Canon EF 24-70mm f/4L IS Resolution Tests!


spdntrxi wrote:
thread is in the crapper now...

I'm for one, am interested in the 24-70 F4IS and would like the discussion returned to that


+ 1000
Me too. My decission 3 month ago was to purchase the Tamron. I tested it and it turned out to be a great lens. Now there is a 24-70 4.0 L. After what I can read out of Rogers first MTF test It looks better tha the Tammy in resolution, topped only by the 2.8 II (non IS). Non IS was never my choice.
I will wait for TDP.comīs full review to compare it against the Tamron and than decide after IQ and price.

Ralph



Jan 05, 2013 at 02:56 PM
snapsy
Offline
• • • • • •
Upload & Sell: On
p.5 #18 · p.5 #18 · Canon EF 24-70mm f/4L IS Resolution Tests!


gdanmitchell wrote:
It turns out that I have both the 24-70 II and the 24-105 lenses


This is a bit surprising to me. What prompted your purchase of the 24-70 II?



Jan 05, 2013 at 02:59 PM
gdanmitchell
Offline
• • • • • • •
Upload & Sell: Off
p.5 #19 · p.5 #19 · Canon EF 24-70mm f/4L IS Resolution Tests!


snapsy wrote:
This is a bit surprising to me. What prompted your purchase of the 24-70 II?


Since you asked:

1. Two of us share some of my lenses, so rather than ending up with two 24-105mm lenses (which I also considered) it made sense to have different lenses with different/complimentary strengths and weaknesses.

2. I am working on a long-term project photographing professional classical musicians for which this lens may have some real value.

The 24-70 seems like a fine lens from the use I've made of it so far. Time will tell how much difference in makes in the context of my own photography.


Dan



Jan 05, 2013 at 03:04 PM
jojomon11
Offline
• • • • •
Upload & Sell: Off
p.5 #20 · p.5 #20 · Canon EF 24-70mm f/4L IS Resolution Tests!


man this post is nothing but a big bowl of spinach soup now! one big confusing mess



just like the siggy 35 thread, dead!

Edited on Jan 05, 2013 at 03:07 PM



Jan 05, 2013 at 03:05 PM
1       2       3       4      
5
       6              9       10       end




FM Forums | Canon Forum | Join Upload & Sell

1       2       3       4      
5
       6              9       10       end
    
 

You are not logged in. Login or Register

Username       Or Reset password



This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.