Home · Register · Search · View Winners · Software · Hosting · Software · Join Upload & Sell

Moderated by: Fred Miranda
Username   Password

  New fredmiranda.com Mobile Site
  

FM Forums | Canon Forum | Join Upload & Sell

1      
2
       3       4       end
  

Archive 2012 · Full Frame v. Crop Factor Images
  
 
kevindar
Offline
• • • •
Upload & Sell: Off
p.2 #1 · Full Frame v. Crop Factor Images


Jeff Nolten wrote:
Kevindar, thanks for the explanation. Could you expand on the tradeoff of "5d2+1.4x, is about same pixel on duck at 60d" if you've experimented with this? I have both 5D2 and 7D to use with my 100-400. I find the images with the 5D2 are superior to those from the 7D when not reach limited. So the question, hopefully relevant to the OP, is whether using a 1.4x and an extra stop of ISO on a 1DX or 5D3 will yield a better or equivalent image quality to using a straight 7D? I'm considering replacing my 7D with a
...Show more
5d2 images are superior to 7d when not reach limited, b/c a lower pixel density is less demanding of lens sharpness and the high iso is better.
7d for me produced consistently better images with 100-400L, than 5d3 with 1.4x tc, and 100-400L. of course the autofocus is the very obvious difference. however, the tc causes decrease in contrast, and sharpness, and the bare lens does better there. 5d3 has 1.5 or so of noise advantage over the 7D. 1 stop of that is lost by the tc.
Now, of course, this may be a little different with my 70-200 2.8 ISII, where not much image quality is lost by using the TC (or one of the canon's super tele primes) however, here is the thing. you can always use the tc on the 7D also to add additional reach on those super teles. There is a test somewhre on the net comparing 1d3 plus tc and I think 300 2.8, vs 50D, and the detail is pretty close between the two.



Dec 30, 2012 at 06:18 PM
Jeff Nolten
Offline
• • •
Upload & Sell: On
p.2 #2 · Full Frame v. Crop Factor Images


Thanks for sharing your experiences Kevindar. I was going to skip the 5D3 until the f8 AF was announced and use of the 1.4x became a possibility. Lots to think about. Cheers.


Dec 30, 2012 at 06:38 PM
kevindar
Offline
• • • •
Upload & Sell: Off
p.2 #3 · Full Frame v. Crop Factor Images


You are welcome Jeff. thats honestly another reason I have seriously looked at the d800. its not that I need the extra pixels in my landscapes, its that as a wildlife camera, it gives you a 15mp 1.5x crop equivalent, so is dual purpose.
that said, if not focal length limited, 5d3 is an amazing wildlife camera. awesome AF, and clean files.



Dec 30, 2012 at 06:55 PM
kevindar
Offline
• • • •
Upload & Sell: Off
p.2 #4 · Full Frame v. Crop Factor Images


Jeff Nolten wrote:
Not sure what you are saying here, if both cameras have the same photo-site density don't they essentially have the same sensor? 7D v 60D?

Jeff, to expand on Dan's point, which I also mentioned in my original post, its not the crop factor that matters, its the pixel density. the reach of the crop factor is all about pixel density.
so a 1d3, with its 1.3x crop factor, actually has less reach than the 5d2. which has no crop factor. 5d2 and 20D have the same reach, even though 20d is a 1.6x crop. 5d3 has more reach than the 1dx. d800 has about the same reach as the canon 50D, etc.



Dec 30, 2012 at 06:59 PM
curious80
Offline
• • •
Upload & Sell: Off
p.2 #5 · Full Frame v. Crop Factor Images


Comparing a 1DX crop to 7D, the 7D will clearly give you more pixels as jcolwell mentioned earlier. Whether those extra pixels result in a real advantage depends on two factors:

1. The size of print / output that you need
2. Are you shooting in good light or not.

1Dx will have fewer but higher quality pixels. In ample lighting the pixel quality would be good from both and you will should get an advantage from 7D in terms of say ability to print larger, or crop even tighter. In lower light levels as you need to push the ISO, 7D will start losing its resolution advantage over the 1DX crop, because its smaller pixels will deteriorate faster.



Dec 30, 2012 at 07:11 PM
ggreene
Offline
• • •
Upload & Sell: Off
p.2 #6 · Full Frame v. Crop Factor Images


galenapass wrote:
I think point C is really important:
"c: as you move up high iso, the gap becomes less, but the larger pixel density for the same final output size, will always do at least as well, many times better in detail"
I do a lot of shooting in low light conditions and this is an important point to consider when that is the case.


Yep, before I sold my 1D4 I ran some quick tests at higher ISO's against my 1DX and starting at around 2000 the 1D4 begins to lose it's advantage. By 3200 the 1DX is much better.



Dec 30, 2012 at 07:19 PM
Jeff Nolten
Offline
• • •
Upload & Sell: On
p.2 #7 · Full Frame v. Crop Factor Images


kevindar wrote:
Jeff, to expand on Dan's point, which I also mentioned in my original post, its not the crop factor that matters, its the pixel density. the reach of the crop factor is all about pixel density.
so a 1d3, with its 1.3x crop factor, actually has less reach than the 5d2. which has no crop factor. 5d2 and 20D have the same reach, even though 20d is a 1.6x crop. 5d3 has more reach than the 1dx. d800 has about the same reach as the canon 50D, etc.


I didn't understand Dan's sentence, but I guess he means a 5D2 will yield the same image as a 20D if cropped the same given their equal pixel density. I've been reading and thinking about these kinds of posts so I think I understand this concept. My problem is that, similar to Galenapass' opinion above, I think the 7D has pushed pixel density a bit too far so a 7D image needs to be resize/sharpen down to 12 MP to be equivalent to the 5D2 at pixel level for sharpness. Qualify this with the fact that I'm always using the 7D with the 100-400 so I'm often pushing shutter speed and ISO to get the shot. I use a 5D# when I can. But with wildlife, pixel peeping and detail is important.

Put another way, there are two ways to gain resolution or reach, increasing pixel density or optical magnification. I think increasing pixel density degrades the image. Increasing optical magnification does not if you can put the money into a big white, but increasing magnification via a teleconverter does degrade. You answered my question of which degrades more in my situation. I'll have to do more comparing 5D2 + tc vs 7D to see if the 5D3 is still a worthy upgrade: 5D2 + 7D vs 5D2 + 5D3 + tc. Thanks for the help. Hope this isn't too OT to those reading this thread.



Dec 30, 2012 at 08:09 PM
kevindar
Offline
• • • •
Upload & Sell: Off
p.2 #8 · Full Frame v. Crop Factor Images


Jeff, increasing pixel density does not degrade image for the same output size, esp with todays technology, with gapless microlenses and minimizing dead space/non light collecting segment of the sensor.
So a 24mp apsc will always produce as good if not better 16x24 prints as a 12mp crop sensor. of course, when viewed at 100%, the 24mp sensor would show far more noise at higher ISO, but if final output size is the same, they should be very similar.
You are correct that there is no substitute for longer lens. is size, weight and price is not a limiting factor.
The only other thing I would say, is many of the differences, though there, are inconsequentail in most prints. you can crop the 1dx to size of 1.6x, and make perfectly good and detailed 16x24 images from it. If your goal is to print 20x30, then no.



Dec 30, 2012 at 08:54 PM
gdanmitchell
Offline
• • • • •
Upload & Sell: On
p.2 #9 · Full Frame v. Crop Factor Images


This is a bit complicate, and unfortunately some want to try to reduce it to "A is better than B!" or "B is better than A!" simplifications... so I chose my words very carefully in that first post so that it would only say what is true and not debatable. The context means more regarding this question that trying to come up with some simple rule about the relative quality of crop and full frame in this situation.

Here is what we know:

- If you were to put lenses covering the same angle of view on a cropped sensor and full frame camera and those lenses produced identical lens resolution, the full frame lens/camera system would be capable of producing a "sharper" print. Lens resolution is often described using terms such as "line pairs per millimeter" - which is a bit more complex concept than it sounds like, but I'll leave it this way for now. However, the larger the sensor or film, the "more line pairs" that can be captured in the image. Using made up numbers, let's say that we use lenses with 20 line pairs per millimeter (20 lp/mm) resolution two cameras. Let's say that one has a sensor that is 20mm wide and the other has a sensor that is 40mm wide. The smaller sensor can theoretically resolve 400 (20 x 20) line pairs while the larger can resolve 800 (20 x 40) across the frame width, and that is what determines the better resolution potential of the larger system.

- A simple way to distill this is to realize that a particular lens resolution results in higher image resolution as the frame size increases.

- If you compare a APS-C (or "cropped sensor") size image from a cropped sensor camera and a crop of the same size taken from the middle of the frame of a full frame camera, there are essentially two possibilities that we could be interested in. (There is a third possibility, but I'll ignore it since it is very unlikely.)

1. If the two sensors start out with the same number of megapixels (e.g. - the same number of photo sites), when you crop the full frame image down to an APS-C sized section, the sensor resolution of the image is more limited in the case of the cropped full frame image, and its potential resolution is diminished. (In the real world, it might not make a difference unless you print quite large.)

2. If the full frame sensor starts out with higher photosite density such that an APS-C size crop from its center ends up with the same number of photo sites found on the cropped sensor, the image quality should be equal. (This is a pretty unlikely scenario, though with 36MP full frame sensors now on the market, we can imagine it.)

When you are trying to parse out this choice between a full frame sensor and cropping versus a cropped sensor camera and using its full image, the "right" answer depends on a lot of stuff.

If you are only interested in filling the frame with the subject when you use your longest lens, and you are not all that interested in other kinds of photography in which you don't do this - say landscape or portrait photography at more typical focal lengths, then you fall into a special case where you might as well get a cropped sensor body. You'll get more "reach" (e.g. - ability to fill the frame of your camera) with a given lens, and you might be able to use less expensive long lenses.

If you are very critical about image quality - perhaps you frequently make very large, high quality prints - and you do not just shoot at the extreme ends of your long focal length lenses, you have a trade-off to consider: are you willing to give up image quality potential in your large prints for the ability to fill the smaller frame more without cropping? Only you can answer that.

Dan



Dec 30, 2012 at 09:12 PM
Jeff Nolten
Offline
• • •
Upload & Sell: On
p.2 #10 · Full Frame v. Crop Factor Images


I don't disagree with you Kevindar. But I do like to view my images, especially birds, flowers, other wildlife, at 100% on my 27" iMac to see detail. My biology background When I upgraded from a 5Dc to the 5D2 I saw no deterioration in the 100% view even though I was effectively zooming in further. I did see deterioration in the 7D 100% view compared to my 40D 100% view. This deterioration was fixed by a 7D image reduction to 12 MP (15 MP if I really nailed it). So yes still an improvement on the 40D but not quite what a 10 MP to 18 MP increase should yield. Also, the 7D takes a lot more work in Lightroom balancing the noise reduction and sharpening than my other bodies. I love my 5Ds, they've always just been so easy to work with; perhaps I should say forgiving which benefits an amateur like me. But I digress. Good exchange. Like the OP I'm just agonizing over my kit trying to get the most out of it.

BTW, I like to use an iPad to share images with friends and fellow travelers. I've noticed that if you hand an iPad to someone they will likely double tap the image to go 100%. So now part of my post processing is to make sure my images look good at 100% even if I have to reduce.



Dec 30, 2012 at 09:49 PM
 

Search in Used Dept. 



gdanmitchell
Offline
• • • • •
Upload & Sell: On
p.2 #11 · Full Frame v. Crop Factor Images


Viewing at 100% is not really of much value aside from clinical inspection or closely monitoring the effect of things like sharpening operations. You can't actually see the whole photograph at 100% unless you have a monitor that is something like 5 feet across with computer monitor resolution.

Sharpness and resolution at 100% is only meaningful to the extent that it relates to sharpness at the sizes you will actually display.

And, of course, once the so-called "retina displays" become more common, the whole notion of what a 100% view is will have to change, won't it...?

Dan



Dec 30, 2012 at 10:27 PM
StarNut
Offline
• • •
Upload & Sell: On
p.2 #12 · Full Frame v. Crop Factor Images


FWIW, my purely subjective impression, based on a lot of shooting but nothing more, is that the IQ of my 5D3 is noticeably better than the IQ of my 7D, so I use the 5D3 when I'm not focal length limited.

However, the IQ advantage is not so pronounced that I would use the 5D3 even when it's focal length limited.

The more I'm cropping (i.e., the less of the sensor the target is filling), the more this is true.

If the light is so poor that I have to use ISO 3200 or higher, I don't bother with either body.

I assume that the IQ of the 1DX is comparable to the 5D3, so I hoped this would be a useful bit of information.

Again, this is VERY subjective; doing your own test is a good use of your time.

Mark



Dec 30, 2012 at 10:34 PM
skibum5
Offline
• • • • • •
Upload & Sell: Off
p.2 #13 · Full Frame v. Crop Factor Images


rcm123 wrote:
I believe that this has been discussed previously, but I couldn't locate the thread(s). I am going to Alaska this summer to photograph the brown bears at Lake Clark and I will be shooting with a 1Dx. I will have a 7D as a second body. If I were to capture an image of the same bear from the same distance with each body & 500 mm lens, then crop the full frame image so that the bear is the same size as in the 7D image - or crop both images so that the bear is the same size
...Show more

7D tosses a LOT more pixels onto it, sure the 1DX pixels are higher quality, but mostly that won't come close to making up for it (although the bears are sort of dark and if you were shooting in winter against snow and such they might fall into the shadow realm of the captures and if you needed quite high iso as well then the 1DX result might be better in the end). If the subject falls in a reasonably mid-tone or above portion of the image I found that even at ISO3200 and 6400 the 7D tended to give better final quality than even the 5D3 (and always versus the 5D2).

I sold my 7D but I do miss the reach. When I had 7D+5D2 or 5D3 I almost always used the 7D for distance limited stuff (and almost always the 5 series for everything else other than occasionally sports or macros).

Edited on Dec 30, 2012 at 10:55 PM · View previous versions



Dec 30, 2012 at 10:53 PM
Gunzorro
Offline
• • • • •
Upload & Sell: On
p.2 #14 · Full Frame v. Crop Factor Images


gdanmitchell wrote:
Viewing at 100% is not really of much value aside from clinical inspection or closely monitoring the effect of things like sharpening operations. You can't actually see the whole photograph at 100% unless you have a monitor that is something like 5 feet across with computer monitor resolution.

Sharpness and resolution at 100% is only meaningful to the extent that it relates to sharpness at the sizes you will actually display.


Dan -- This is of course strictly your opinion and has no basis in objective facts. You are welcome to that opinion, and I'm sure many people share it, but I'm not one of them. If I could look at 1000% with no raster edges and minute details, I'd like that option and more. It as much value to me, although I'll admit it's an arbitrary opinion I hold.



Dec 30, 2012 at 10:54 PM
skibum5
Offline
• • • • • •
Upload & Sell: Off
p.2 #15 · Full Frame v. Crop Factor Images


kevindar wrote:
I recearched this extensively, did my own testing with a 7d, 1d4, 5d2, 5d3, using extenders, various ISO's and searched for other tests.
1. Under LAB CONDITIONS if you are focal length limited (have to crop to the size of apsc or smaller, for the snesors of same generation, you are always better off with higher pixel density (obviously not crop factor, since 5d2 has same pixel density as the 20D even though 20d has 1.6x crop factor. its all about pixel density).
a: you wont be shooting in a lab
b: you are talking about densor designs of different
...Show more


I will second that.

(other than for #1 if the high density camera used worse technology than, under bad enough lighting it would start to lose to the lower density camera with better tech)



Dec 30, 2012 at 11:03 PM
Jeff Nolten
Offline
• • •
Upload & Sell: On
p.2 #16 · Full Frame v. Crop Factor Images


gdanmitchell wrote:
Viewing at 100% is not really of much value aside from clinical inspection or closely monitoring the effect of things like sharpening operations. You can't actually see the whole photograph at 100% unless you have a monitor that is something like 5 feet across with computer monitor resolution.

Sharpness and resolution at 100% is only meaningful to the extent that it relates to sharpness at the sizes you will actually display.

And, of course, once the so-called "retina displays" become more common, the whole notion of what a 100% view is will have to change, won't it...?

Dan


I think I'm in the minority because most folks seem to speak of higher resolution in terms of how large one can print. I don't print except rarely to give an image as a gift. My main use is to look and share images on my iPad, computer screen, and TV. I do like to zoom in to see the detail in a carving at an archeological site or the detail in a tree frog's eye, etc. So 100% view is valuable to me. YMMV.

The concept of a Retina display is interesting because it suggests an upper limit on how high a display resolution one will ever have to accommodate. My iPad is 264 ppi which is supposed to be better than the human eye can differentiate. (Note that the 15" laptop retina display is only 220 ppi as compared to my iMac's 109 ppi.) So a 30" 16:9 aspect retina display (220 ppi) would be 5752 pixels wide essentially the pixel width of a 5D3. In other words an uncropped 5d3 image would be viewed 100% on a 30" retina display. I look forward to that day and hope I can afford it when it comes. In the mean time I double tap my iPad and scroll around.

Edit: Back in the 1980s I worked at the Visibility Laboratory at Scripps Institution of Oceanography doing satellite image processing. I remember what a great day it was when we could work with 1024 x 1024 images instead of 512 x 512. Imagine that!



Dec 30, 2012 at 11:27 PM
Paul Mo
Online
• • • •
Upload & Sell: On
p.2 #17 · Full Frame v. Crop Factor Images


StarNut wrote:
FWIW, my purely subjective impression, based on a lot of shooting but nothing more, is that the IQ of my 5D3 is noticeably better than the IQ of my 7D, so I use the 5D3 when I'm not focal length limited.
Mark


Mark, if you shot a well-lit static image with s 5D MKIII and a 50mm, and a 7D with a 35mm, apart from slight focal length disparity, what qualitative differences would you be seeing?

Thanks.



Dec 31, 2012 at 12:12 AM
StarNut
Offline
• • •
Upload & Sell: On
p.2 #18 · Full Frame v. Crop Factor Images


Paul Mo wrote:
Mark, if you shot a well-lit static image with s 5D MKIII and a 50mm, and a 7D with a 35mm, apart from slight focal length disparity, what qualitative differences would you be seeing?

Thanks.


I'm interpreting your question as follows: If you use different focal lengths to give the same framing in the 7D and 5D3, and assuming equal optical quality in both lenses, what (if any) differences in image quality would I expect?

Again, purely subjectively from having used both cameras for a lot of photos, of similar types, I would expect the 5D3 to have less noise (even at ISO100), and otherwise just to "look" better (e.g., I find the colors rendered by the 5D3 to be excellent). I would choose the 7D over the 5D3 only when the significantly higher pixel density of the 7D is an advantage (basically, only for sports/wildlife when significantly focal-length limited). Also, to repeat, I haven't done any carefully controlled shooting to check this out; I'm just always thrilled with the IQ of photos taken with the 5D3 (and the 5D2 before that), while feeling that the 7D landscapes just don't quite have the same "WOW" factor. That's all.

That's not to say that there's anything wrong with 7D's IQ; I just think that the IQ of the 5D2 and 5D3 is better.

Mark



Dec 31, 2012 at 12:51 AM
veroman
Offline
• • • •
Upload & Sell: On
p.2 #19 · Full Frame v. Crop Factor Images


Jeff Nolten wrote:
.... When I upgraded from a 5Dc to the 5D2 I saw no deterioration in the 100% view even though I was effectively zooming in further. I did see deterioration in the 7D 100% view compared to my 40D 100% view. This deterioration was fixed by a 7D image reduction to 12 MP (15 MP if I really nailed it). So yes still an improvement on the 40D but not quite what a 10 MP to 18 MP increase should yield ....


Jeff's responses, experiences and results mirror mine, although the camera I used for comparison was the 60D, not the 7D. I've been of a mind for several years now that quality of the photo sites has deteriorated with each reduction of pixel size in favor of resolution and pixel density. This was amply demonstrated to me when I shot with the 60D for several weeks and then gave up on it. At 100% on-screen or in large prints, the overall image integrity of the 60D was inferior to that of my other Canon cameras, including the 1Ds II, 5Dc and 40D. It was also inferior to up-sized images from my lovely 5MP Panasonic LC-1, twin of the Leica Digilux 2 ... at ISO 100 anyway.

However, there's little question that the high pixel density of the new generation of crop cameras will yield/retain more detail than the same size image cropped from a full frame sensor of similar or equal resolution, i.e. the 1Dx and the 7D. But retention of detail that comes as a result of "more pixels per duck" doesn't necessarily mean better or even equal image quality. One must ask: "What am I willing to give up in order to preserve the detail?"

The recurring phrase when discussing issues like this is "It's all about pixel density." I don't agree. Pixel density is obviously one factor out of many. For my money, I prefer to invest in the lens needed to achieve the image size AND image quality I'm looking for. It's the costly (and heavy!) route, but to quote another recurring phrase: "There's no such thing as a free lunch."

Even if and when I don't have the right lens with me for the reach I need, if I'm carrying a 5D III or new 6D or my 1Dx, I would not hesitate to compose in-camera as best I can, crop later, then re-size for the size print I'd like. I think the overall integrity of the picture would still be better.

- Steve



Dec 31, 2012 at 01:04 AM
garydavidjones
Offline
• •
Upload & Sell: On
p.2 #20 · Full Frame v. Crop Factor Images


I have been very impressed with IQ of 7D when max. reach is needed compared to the 5D3.
5D3 is generally superior but not at the maximum of 100-400 mm L lens. At that point the 7D is
has the better IQ in my opinion.



Dec 31, 2012 at 01:11 AM
1      
2
       3       4       end




FM Forums | Canon Forum | Join Upload & Sell

1      
2
       3       4       end
    
 

You are not logged in. Login or Register

Username   Password    Retrive password