Fred Miranda wrote:
This RAW file was a little underexposed and it was not tack sharp but even with a little noise reduction, it would look terrific in a large print.
And it will do so because of the beautiful lighting, composition, and subject. In this case, noise and sharpness matter less than any of those three elements.
jcolwell wrote:
Thanks for posting the FJ photos, Conrad. Here's a few of and/or near my Rugged Rentals jeepster. First photo is sunrise over "The Ancestors" at Torrey UT. Second is on the Burr Trail. That was crazy hot! 109 deg F. I would perish if I wandered 10m from the road, maybe 3m. Last two are from Boulder Tops.
Sorry, these photos are definitely OT, as I sold my 7D last year. OTOH, the 1DsIII I used was very quiet.
TeamSpeed wrote:
Nice Romy, my results come out to be very close to yours as well on the one raw. Here is a really quick run though, nothing very elaborate, no masks, etc. I think I tried to hard to bring the detail out, but we are really only looking at the noise for this example, i believe. The image is indeed a bit underexposed, and this will produce undesirable effects on the 7D. If there is one shortcoming with the 7D, it is that it does not like being underexposed by 2/3 or more, if you do, you have quite a bit of work on your hands.
Here are 4 of my raws from recent shoots at ISOs > 800 (a couple at 1250 and a couple at 3200), since it seems nobody ever posts unadulterated raw images and just processed websized images (even though through the years, I have posted many 100% cropped untouched images, then the post processed images, especially in the tutorial thread over on POTN).
See what you think compared to what others have posted, and let's see your handiwork on polishing these turds.
n0b0 wrote:
My turn. Just wanted to see what can be done with mostly blur tool and a little sharpening... Well, a lot of sharpening actually, maybe too much, but the original RAW is a bit soft IMO.
I agree 100% crop never lie but then again it's not important unless you intend to print large poster size, ESPECIALLY unedited 100% crop. What for? Who doesn't edit their photos? Only the pretentious elitist snobs.
Romy's and others here have proven then when you've got yourself a great shot, what camera you used to take it is irrelevant.
I personally prefer to check how my photos look at 50% crop. If it looks good, I'm a happy camper....Show more →
Oh yeah the shot is soft, just loved the BG and subject
Fred Miranda wrote:
I have a love/hate relationship with 100% crops...
This RAW file was a little underexposed and it was not tack sharp but even with a little noise reduction, it would look terrific in a large print.
Below is a Lightroom conversion showing a 50% crop and the full frame reduced. Then focal length limited, the Canon 7D is a great choice.
Hey Fred! Happy to see you gave my RAW file a shot. Looks good man! I'll share a more challenging shot next week. Dark darks, white whites, everything in between.
mikeengles wrote:
100% crops never lie, but up to this thread, almost no one would actually post them, hoping that showing downsampled and oversharpened images would be sufficient to show the qualities of any camera.
Mike Engles
Here you go, Mike - a full res (5184x3456) quality 10 jpeg of my Honeybuzzard image. This was with a 7D + 500 f4 IS + 2x II.
Liquidstone wrote:
Amazing that our RAW conversions are very similar, eventhough I used ACR and you (I presume) used DPP.
Yes I use DPP. The trick is to not underexpose at all if possible, and don't let the raw converter sharpen the image, IMO.
As to the other question between the 1D4 and 7D, there is just over one stop improvement of the 1D4 from the 7D. This is pretty nice especially if you shoot at 3200/6400 on the 7D alot, the 1D4 gives you that extra stop. However, there are several features on the 7D are very nice, and I wish Canon had added them to the 1D4.
mco_970 wrote:
That is my 7D strategy as well. If the 100% crop looks like crap, don't look at it. See how the 50% looks. If it looks good at 50%, it will print well and downsize well for web viewing.
I use 100% view to assess critical focus and do my sharpening, and use 50% view to assess what a print will look like. It's pretty good measure of how the print turns out, but I think if the print is large enough, eventually you'll want the shot to look good at 100%.
I'm not going to try and work Conrad's bird shot, but to my eye the noise looks worse since the shot is not critically focused; it's missed by a hair and this seems to have a deleterious affect on IQ, especially when the subject itself is small in the frame. I've seen the same thing in my 7D, one shot the detail is excellent and noise is low and next shot under same settings seems to have fallen apart.
TeamSpeed wrote:
As to the other question between the 1D4 and 7D, there is just over one stop improvement of the 1D4 from the 7D. This is pretty nice especially if you shoot at 3200/6400 on the 7D alot, the 1D4 gives you that extra stop. However, there are several features on the 7D are very nice, and I wish Canon had added them to the 1D4.
After having shot with both the 7D and the 1D Mark IV each for about 3 years, I would agree that there is about a one stop advantage for the 1D Mark IV with regard to noise, but I would also add the caveat that, in focal length-limited situations, if you are going to have to crop the 1D Mark IV image down to the same field of view as the 7D image, that one stop advantage basically disappears and the noise comparison between the two cameras pretty much becomes a wash, but the 7D image will show greater detail than the cropped 1D Mark IV image does. In the very controlled tests that Skibum has done and which have been referenced in a number of threads here, he compared the 7D to the 5D Mark II and 5D Mark III, and, in the same way, when those 5D Mark II and 5D Mark III images were cropped to the same field of view as the 7D images, whatever high ISO noise advantages that the 5D Mark II or 5D Mark III may have started out with were essentially gone, but the 7D image showed more detail than the cropped 5D Mark II or 5D Mark III images.
The bottom line is that, if you are going to have to rely on severe cropping to get the framing that you want, you are going to have to deal with noise to a greater extent as you crop to a greater extent. And, if you have to crop more with an APS-H or full frame body with lower pixel density than you would be cropping with the 7D, you are not likely to be happy with the amount of noise from those more heavily cropped images from the APS-H or full frame camera either.
Les.
Liquidstone wrote:
Wow.... 11 pages now. Have we agreed on something already?
I don't expect agreement, but I certainly have learned a number of things.
First, even though the sample size is small, this thread certainly gives indication that "bad copies" of the 7D are not the reason why some are dissatisfied with the pictures that they are taking with the 7D. It would appear that there are wide ranges of techniques being used to take images, that some are more careful about proper exposure than others, and that some have developed better, even if very simple, processing techniques for working with the 7D files. And these latter factors seem to be much more important in producing excellent results than any copy-to-copy variation in the 7D.
While I don't expect that I will see a need to alter my normal workflow for most images, I can see that, when/if I push the 7D a bit harder, I should be able to use the processing tips provided in this thread as a good starting point for working with those images in ACR. My normal choice of raw converters has, however, for years, been Capture One. Thus, I would be interested in learning if there are any other Capture One users who might have some starting point settings for working with higher ISO images in Capture One.
Finally, even though I know that it has been everyone's goal in working with the sample raw files during the last several pages to eliminate noise, I think that I've also confirmed that, to me, completely noise-free images are not the ones that have the highest visual appeal. I'm not sure how to describe why that it is the case, but, somehow, when the images are completely noise-free, they seem less "real" to me.
msalvetti wrote:
I think a lot of the comparisons in this thread were probably taken in One-Shot. I do shoot some wildlife, but not with the frequency of Conrad and Les. Most of my images are in AI Servo: ice hockey, soccer, auto racing, and aviation. I primarily use a 70-200II and a 100-400.
Those of you that shoot both cameras, am I correct that as ISO's climb and the light gets dimmer, the 1D4 should significantly out-perform the 7D? I've always wanted a 1-Series, but $3K is a lot of money for me, and I need to start talking myself out of this if this isn't a good idea.
Mark, as long as you aren't severely focal length limited, the 1D4 outperforms the 7D by about a stop as far as noise is concerned. With f/4 lenses + 1.4x TC, the 1D4 is also snappier in AF than the 7D. With bare f/4 lenses, f/2.8 + 1.4x combos, or bare f/5.6 glass, the AF performance is very close.
uz2work wrote:
I don't expect agreement, but I certainly have learned a number of things.
First, even though the sample size is small, this thread certainly gives indication that "bad copies" of the 7D are not the reason why some are dissatisfied with the pictures that they are taking with the 7D. It would appear that there are wide ranges of techniques being used to take images, that some are more careful about proper exposure than others, and that some have developed better, even if very simple, processing techniques for working with the 7D files. And these latter factors seem to be much more important in producing excellent results than any copy-to-copy variation in the 7D.
While I don't expect that I will see a need to alter my normal workflow for most images, I can see that, when/if I push the 7D a bit harder, I should be able to use the processing tips provided in this thread as a good starting point for working with those images in ACR. My normal choice of raw converters has, however, for years, been Capture One. Thus, I would be interested in learning if there are any other Capture One users who might have some starting point settings for working with higher ISO images in Capture One.
Finally, even though I know that it has been everyone's goal in working with the sample raw files during the last several pages to eliminate noise, I think that I've also confirmed that, to me, completely noise-free images are not the ones that have the highest visual appeal. I'm not sure how to describe why that it is the case, but, somehow, when the images are completely noise-free, they seem less "real" to me.
Excellently written as usual, Les. For somebody who churns out amazing nature images on a regular basis, your open-mindedness to new ways of doing things should be an eye-opener to others.
Excellently written as usual, Les. For somebody who churns out amazing nature images on a regular basis, your open-mindedness to new ways of doing things should be an eye-opener to others.
Thanks Romy.
I think that Conrad's image is very enlightening. Here, we started out with an image that was under exposed, that was slightly out of focus, and that, even with the 7D, required a severe crop. Yet, even with all of that going against it, people were able, with relatively simple processing and post processing, take the file and make it into a very pleasing image.
Thanks Romy for the 100% image. A small 100% crop would have sufficed. It is a fine photograph and looks good because you managed to nearly fill the frame. What was the ISO? Excessive cropping and excessive noise are the main factors in bad image quality. The removal of noise is in essence interpolation of nosiy pixels with genuine image pixels. The result is a loss of resolution and detail. This can never be recovered, no matter what software is used. There comes a point at high ISO, when it is impossible to tell what is noise and what is an image pixel. With a 7D this happens at quite low ISOs, but all cameras have the same problems.
In my opinion one can only truly assess an image at 100% and it is essential to do all NR, sharpening and exposure adjustments at 100%, because the software we use is optimised for 100% viewing.
I have found in all the threads about cameras and their qualities in this and other forums, that there seems to be a shocking lack of objectivity. For any prospective buyer of a camera, it is pretty useless to show a 18mp image downsized to 1mp. Almost anything will look good and the result is a completely misleading to such a person.
The only way to fairly represent an image from a camera that someone is considering buying is to show a context, ie how much of the total image is taken up by the main subject. Next to show a 100% crop of the main subject with the default settings of the software used. Next a 100% crop of a finished image and lastly a web sized version. Obviously a listing of lens and TC ,exposure details and focal lenght is also essential.
That is in essence the purpose of such a forum. We all wish to show the capabilities of our chosen equipment and our own prowess in using them and impress others including prospective buyers. We should not be misleading or to put it more strongly dishonest.
There is no magic to all of this, it is a scientific process and we should always try to be objective, try not to fool ourselves and in the process others.
mikeengles wrote:
because the software we use is optimised for 100% viewing.
I have no idea what this means.
mikeengles wrote:
In my opinion one can only truly assess an image at 100%.
There is no magic to all of this, it is a scientific process and we should always try to be objective, try not to fool ourselves and in the process others.
Mike Engles
While my interest is to assess image quality at the image level, I have no problem with those who want to look at images at 100%, but comparing image quality of different cameras needs to be done in a valid and meaningful way. And what you fail to show any willingness to do is to do that assessment in a valid way.
If you want to compare the ability of the 7D, for example, to that of the 5D Mark III in focal length-limited situations, comparing 100% crops from both cameras is totally meaningless. The only meaningful comparison is to take the 5D Mark III image, crop it to the field of view of the 7D, and then upscale it to the 18 mp of the 7D. At that point, and only at that point, does comparing the two images at 100% become valid and meaningful. And well-controlled tests done by many over and over again show that, when this kind of comparison is done, the 7D image will show more detail than the cropped and upscaled 5D Mark III image, and noise will become a virtual wash.
And that is the very "scientific process" that you call for but that you continue to reject repeatedly.
mikeengles wrote:
I have found in all the threads about cameras and their qualities in this and other forums, that there seems to be a shocking lack of objectivity. For any prospective buyer of a camera, it is pretty useless to show a 18mp image downsized to 1mp. Almost anything will look good and the result is a completely misleading to such a person.
Considering the fact that most of those posting in this thread have and use, not only the 7D, but also a 1D Mark IV and/or a 5D Mark II or 5D Mark III, or so some other camera, I find it hard to believe that there would be any cause/motivation for lack of objectivity. In my case, I appreciate the strengths of both my 7D and my 1D Mark IV, but I also recognize that each one can be a better tool for getting the job done depending on the situation at hand. And I would be making the case just as vigorously against someone who tried to make equally closed-minded arguments that the 1D Mark IV does not have advantages over the 7D in many situations.
And, frankly, with regard to lack of objectivity, I think that we may well have a case of the pot calling the kettle black. I would have to look for quite a while before I saw a better example of someone with a "my mind is closed, and I will not be confused with the facts" mindset.
mikeengles wrote:
That is in essence the purpose of such a forum. We all wish to show the capabilities of our chosen equipment and our own prowess in using them and impress others including prospective buyers. We should not be misleading or to put it more strongly dishonest.
Finally, I take offense with the notion that others are somehow being misleading or dishonest.
Pixel Perfect wrote:
I use 100% view to assess critical focus and do my sharpening, and use 50% view to assess what a print will look like. It's pretty good measure of how the print turns out, but I think if the print is large enough, eventually you'll want the shot to look good at 100%.
Right! What I didn't add to my opinion about this topic is that who wants ten thousand bird prints in their house at 16x20? I think the biggest one I have hanging is 11x14, and it's from a 1D2. The horror of 8 MP.
I have scads of 8x10's of wildlife displayed, and the noise is irrelevant at that scaling. I don't even denoise many images that I plan to print anymore.
So while others may need it to look perfect at 100% for the humongous print work, I have not found the time & equipment-needed (1DX? ) tradeoff to make sense for my own shooting.
It's a real bummer for the gearhead to think such thoughts, though!
.
