5D Mk III vs D800/E, is the 5D3 better at anything?

[Bruce Photography]

It is interesting how much the ADL (Active D-Lighting) does to the D800 images , as seen in DP reviews dynamic range tests (images below). Source: http://www.dpreview.com/reviews/canon-eos-5d-mark-iii/19

It can be seen in the last image, albeit less dramatically.

Thanks for the DPR graphs. By the way, I did go to the DPR site and what your last graph did not show is where the D800 was set to ADL Extra High. I've been shooting my D800 with this setting and I've been amazed on the DR that I can pull out of my RAW images. I also set my 5d3 to with HTP on and off. But the HTP forces the ISO up to 200 instead of my normal ISO of 100. Both cameras have the same highlight EV when the 5D3 is set to HTP On and the D800 is set to ADL Extra High. However on the shadow side I'm getting alot of shadow detail that I haven't seen before. Part of that is due to upgrading to Lightroom 4.1 and CS6 Photoshop with their new and effective shadow controls. When I do comparision shooting between cameras I now set both at ISO 200 so the shooting is more comparable. I'm now using a long RRS rail to hold the two cameras and make the exposures at the same time. Before I was getting inconsistent results due to changes in lighting from one shot to the other. This is an exciting time for Digital Photography.

 

[Bruce Photography]

It recently occured to me after Nikon announced that with a grip the D800 shoots 6 FPS that I can't find a single advantage to the 5D3 over the D800. Here's what I know so far:

5D3 vs D800/E by catagory:

Dynamic Range: Inferior
Color Fidelity: Inferior
Noise: Equal
FPS: Equal (with a grip)
Video: Inferior
Autofocus: Equal (both are pro level)/Inferior (no f/8.0)/Superior (better with fast lenses due to special f/2.8 sensors)
White Balance: Inferior
Auto Exposure: Inferior
In Camera Lens Correction: Inferior
Pop up Flash: Inferior
Price: Inferior
Resolution: Inferior
AA filter: Inferior


I realize that the 5D3 has a better menu system, so that's one thing it can possibly do better.

Also please don't smite me. I'm not trying to bash the 5D3, as I'm actually a Canon fan and will be getting the 5D3, but can anyone think of one major feature that the 5D3 does better than the D800? If so feel free to post, I'm just trying to get informed here.

Thanks.

Well I can tell you at least one thing that is MUCH MUCH better with the 5D3 than the D800. If you were a landscape person using tilt shift lenses, Live View focusing of those manual lenses is a very preferred technique. Using the Canon's excellent back LCD screen at 10X it is easy to do manual focus. Using the Nikon D800 is like looking through a bowl of jello and trying to focus at one mag factor less than Max. Max mag on the Nikon for focusing is horrible and un-usable. I've tried a cheap Marshall field monitor on the Nikon but that was even worse. I'll go over to the Nikon forum and ask for some help on this problem.

By the way, I love Canon lenses. My 17 and 24 tilt shift lenses are both one generation ahead of anything Nikon has. With the exception of the 14-24, my Nikon lenses are good but not up to the latestest generation of Canon glass. I really wish I could fit Canon glass on the D800 but I'm told many times that it is a physical impossibility.

 

[well_dunno]

Surely his point is that resolution/DR is "all" alot of people seem to care about. Personally I find myself in that position mostly taking landscapes and would probabley buy a D800 if I upgraded to FF(most of my investment is in EF-S lenses anyway) but it does seem to me that alot of amature reviews are focused on these areas to the almost total exclusion of everything else.

The net photographic community generally has a bias towards these areas IMHO, partly i'd say because its something that easier to quantify for users who'll never own or never use these cameras in situations were features like FPS and AF will be important factors.

Yes, the 'net is fixated on DR and resolution at the moment, but prior to the 5D3 announcement, much if not most of the talk revolved around AF. There was the occasional resolution debate, but many people wanted to talk about was how many AF points they needed. "Cross-type points" was the catch phrase back then, and pulling an arbitrary number of cross-type AF points out of your ass was en vogue ;D

Indeed! Mainly AF improvements were on demand for the successor of the mark 2. Though I think Canon came under fire because of the price tag they put on the mark 3. Many, including myself, feel there is not 3,5 years +$1300 difference between the mark 2 and the mark 3. Unfortunately, D800, selling $500 cheaper than the mark 3, became the reference point for this argument and thus it's strong sides got more and more mentioned... IMHO anyways...

Cheers!

 

[awinphoto]

It is interesting how much the ADL (Active D-Lighting) does to the D800 images , as seen in DP reviews dynamic range tests (images below). Source: http://www.dpreview.com/reviews/canon-eos-5d-mark-iii/19

It can be seen in the last image, albeit less dramatically.

I noticed that also... I mentioned that in another thread and someone said it's because they use jpegs and not raw... uh huh... even looking at that, if you tally the difference in highlight for canon and shadow for nikon, DR is only maybe a stop off if that..

 

[Ivar]

I'm sorry but Canon would not have done such a thing, it just was not possible to get by without doing so.

What would have left of the 5D3 without its AF in the marketplace?

Other common declarations:

"There's no way Canon will ever put a 1-series AF system in a 5D3."

"We'll be lucky if Canon puts the 7D AF system in the 5D3."

"Canon will never put the 1Dx's AF system in anything other than a 1Dx."


Welp, Canon proved those claims wrong, addressed the most glaring weaknesses of the 5D2, but now the hot topic is DR and Canon still sucks.

 

[awinphoto]

It is interesting how much the ADL (Active D-Lighting) does to the D800 images , as seen in DP reviews dynamic range tests (images below). Source: http://www.dpreview.com/reviews/canon-eos-5d-mark-iii/19

It can be seen in the last image, albeit less dramatically.

Examining the DPReview results, the 5D MkIII has almost half a stop more DR than the D800 with HTP and ADL switched off, but the overall DR with HTP and ADL switched on is the same, the difference is in the priorities. Canon as usual has prioritised recovery of highlight detail, while Nikon has given priority to shadow detail. In contrast, the D7000 has noticeably greater DR judging by DPReview's data.
Perhaps now we can lay the DR arguments to rest, but I somehow doubt it. I still maintain that it is about making use of the equipment available to overcome the compromises than worrying about detailed specification lists. If whichever camera does the job you need it to do, then you don't need to worry what another camera might do. If it doesn't, then get the one that does, if you can afford it and the overall system also gives you what you need.

For what it's worth regarding priorities... Since the beginning of digital, it was widely understood that highlights got lost quicker than shadows. In film you shot for the shadows, printed for the highlights, with digital, you had to be very careful with the highlights and recover the shadows for print... This is partial to why HDR got so much traction, it gave you the best of both worlds and let you be creative as well... I think this is why canon added the ALO (auto light optimizer) as well as the HTP, like the old film ideals of d-min, d-max, but like film, you had to deal with grain, now we deal with noise... I'd like to see it where you can use both ALO and HTP without adding noise and essentially being able to achieve massive DR at low ISO... perhaps i'm dreaming or asking too much... it is what it is

 

[Marsu42]

Examining the DPReview results, the 5D MkIII has almost half a stop more DR than the D800 with HTP and ADL switched off, but the overall DR with HTP and ADL switched on is the same, the difference is in the priorities.

As far as I've researched it and asked here on the forum, htp is nothing more than a tone curve applied in camera or in the raw converter. As you may or may have not suspected, htp or any other gimmick does not exchange the sensor in your camera when you're not looking. In the best case it's just a convenience, in the worst case marketing has put it in there to make you think what you obviously think

 

[Kernuak]

Examining the DPReview results, the 5D MkIII has almost half a stop more DR than the D800 with HTP and ADL switched off, but the overall DR with HTP and ADL switched on is the same, the difference is in the priorities.

As far as I've researched it and asked here on the forum, htp is nothing more than a tone curve applied in camera or in the raw converter. As you may or may have not suspected, htp or any other gimmick does not exchange the sensor in your camera when you're not looking. In the best case it's just a convenience, in the worst case marketing has put it in there to make you think what you obviously think

And with the added downside that you are shooting at ISO 200 instead of 100, which is why I always switch it off, then I can control what I want to recover and what I can compromise image by image. I can't imagine Nikon is any different with ADL, apart from the different philosophy.

 

[Marsu42]

And with the added downside that you are shooting at ISO 200 instead of 100

No, you aren't ... the reason htp starts at iso 200 is that when it's on, the camera deliberately underexposes 1 ev and then applies a tone curve. So you can get the same effect with underexposing 1ev yourself or shooting full m at 1/2 iso and apply a curve in post, but you don't get to see such a nice preview on the camera display.

Actually, htp isn't that bad and I'm using it when there are extreme highlights like the furry edges of an animal with the sun behind the animal body - if I underexpose manually, the preview would be too dark.

 

[ablearcher]

Both cameras are fine, but one thing i do not get is why such a price gap? $500 extra for Canon in the US and $700 in Canada? WTF? Somehow Nikon manages to offer Canadians the same price as in the US, but Canon thinks its OK to charge Canadians extra $200 on top of its US price $500 premium. Disgusting.

 

[LetTheRightLensIn]

It is interesting how much the ADL (Active D-Lighting) does to the D800 images , as seen in DP reviews dynamic range tests (images below). Source: http://www.dpreview.com/reviews/canon-eos-5d-mark-iii/19

It can be seen in the last image, albeit less dramatically.

Examining the DPReview results, the 5D MkIII has almost half a stop more DR than the D800 with HTP and ADL switched off, but the overall DR with HTP and ADL switched on is the same, the difference is in the priorities. Canon as usual has prioritised recovery of highlight detail, while Nikon has given priority to shadow detail. In contrast, the D7000 has noticeably greater DR judging by DPReview's data.
Perhaps now we can lay the DR arguments to rest, but I somehow doubt it. I still maintain that it is about making use of the equipment available to overcome the compromises than worrying about detailed specification lists. If whichever camera does the job you need it to do, then you don't need to worry what another camera might do. If it doesn't, then get the one that does, if you can afford it and the overall system also gives you what you need.

DPR is just testing jpgs and doing weird things.
This is another example of the difference without resorting to numbers:
http://forums.dpreview.com/forums/read.asp?forum=1032&message=41585485

 

[LetTheRightLensIn]

Both cameras are fine, but one thing i do not get is why such a price gap? $500 extra for Canon in the US and $700 in Canada? WTF? Somehow Nikon manages to offer Canadians the same price as in the US, but Canon thinks its OK to charge Canadians extra $200 on top of its US price $500 premium. Disgusting.

If you care about 6fps, the price difference does go away though.

 

[LetTheRightLensIn]

Surely his point is that resolution/DR is "all" alot of people seem to care about. Personally I find myself in that position mostly taking landscapes and would probabley buy a D800 if I upgraded to FF(most of my investment is in EF-S lenses anyway) but it does seem to me that alot of amature reviews are focused on these areas to the almost total exclusion of everything else.

The net photographic community generally has a bias towards these areas IMHO, partly i'd say because its something that easier to quantify for users who'll never own or never use these cameras in situations were features like FPS and AF will be important factors.

Yes, the 'net is fixated on DR and resolution at the moment, but prior to the 5D3 announcement, much if not most of the talk revolved around AF. There was the occasional resolution debate, but many people wanted to talk about was how many AF points they needed. "Cross-type points" was the catch phrase back then, and pulling an arbitrary number of cross-type AF points out of your ass was en vogue ;D

Other common declarations:

"There's no way Canon will ever put a 1-series AF system in a 5D3."

"We'll be lucky if Canon puts the 7D AF system in the 5D3."

"Canon will never put the 1Dx's AF system in anything other than a 1Dx."


Welp, Canon proved those claims wrong, addressed the most glaring weaknesses of the 5D2, but now the hot topic is DR and Canon still sucks.

It was actually the pure fanboys who were doing all of the talk about how Canon would never do that and how dare you image to ever get nice AF in anything other than a 1 series.

So far 5D3 AF working well for me for one shot, def better than 5D2 and 7D, although I've only hit a subset of situations so far. I'm not quite sure about for macro yet compared to 7D and didn't test it for sports yet.

Anyway DR got talked about first since it was easy to measure and safe to talk about having never even touched the camera yet and it was kind of a shock that there was zero improvement, maybe exmor level would not have been realistic, but the fact that at ISO100 it delivers nothing better than 3 years ago shocked people into lots of talk. AF takes time to test out and you need the cam in hand for a while. But everyone had been mentioning how awesome it was that on paper they finally did something fantastic below the 1 series in terms of AF.

 

[LetTheRightLensIn]

It recently occured to me after Nikon announced that with a grip the D800 shoots 6 FPS that I can't find a single advantage to the 5D3 over the D800. Here's what I know so far:

5D3 vs D800/E by catagory:

Dynamic Range: Inferior
Color Fidelity: Inferior
Noise: Equal
FPS: Equal (with a grip)
Video: Inferior
Autofocus: Equal (both are pro level)/Inferior (no f/8.0)/Superior (better with fast lenses due to special f/2.8 sensors)
White Balance: Inferior
Auto Exposure: Inferior
In Camera Lens Correction: Inferior
Pop up Flash: Inferior
Price: Inferior
Resolution: Inferior
AA filter: Inferior


I realize that the 5D3 has a better menu system, so that's one thing it can possibly do better.

Also please don't smite me. I'm not trying to bash the 5D3, as I'm actually a Canon fan and will be getting the 5D3, but can anyone think of one major feature that the 5D3 does better than the D800? If so feel free to post, I'm just trying to get informed here.

Thanks.

Well I can tell you at least one thing that is MUCH MUCH better with the 5D3 than the D800. If you were a landscape person using tilt shift lenses, Live View focusing of those manual lenses is a very preferred technique. Using the Canon's excellent back LCD screen at 10X it is easy to do manual focus. Using the Nikon D800 is like looking through a bowl of jello and trying to focus at one mag factor less than Max. Max mag on the Nikon for focusing is horrible and un-usable. I've tried a cheap Marshall field monitor on the Nikon but that was even worse. I'll go over to the Nikon forum and ask for some help on this problem.

By the way, I love Canon lenses. My 17 and 24 tilt shift lenses are both one generation ahead of anything Nikon has. With the exception of the 14-24, my Nikon lenses are good but not up to the latestest generation of Canon glass. I really wish I could fit Canon glass on the D800 but I'm told many times that it is a physical impossibility.

I hear that max mag is also slow frame rate, so focusing for movies can be tricky. (on d800)

 

[LetTheRightLensIn]

And with the added downside that you are shooting at ISO 200 instead of 100

No, you aren't ... the reason htp starts at iso 200 is that when it's on, the camera deliberately underexposes 1 ev and then applies a tone curve. So you can get the same effect with underexposing 1ev yourself or shooting full m at 1/2 iso and apply a curve in post, but you don't get to see such a nice preview on the camera display.

Actually, htp isn't that bad and I'm using it when there are extreme highlights like the furry edges of an animal with the sun behind the animal body - if I underexpose manually, the preview would be too dark.

yeah it doesn't actually do anything for you in RAW that you couldn't do anyway but when you do need to favor highlights 1 stop then it certainly makes reading the histogram and image previews easier than looking at underexposed ones (And for jpgs/movies, those are cooked, so it's a real mode for jpgs since you can't add HTP back in afterwards in that case)

 

[Kernuak]

And with the added downside that you are shooting at ISO 200 instead of 100

No, you aren't ... the reason htp starts at iso 200 is that when it's on, the camera deliberately underexposes 1 ev and then applies a tone curve. So you can get the same effect with underexposing 1ev yourself or shooting full m at 1/2 iso and apply a curve in post, but you don't get to see such a nice preview on the camera display.

Actually, htp isn't that bad and I'm using it when there are extreme highlights like the furry edges of an animal with the sun behind the animal body - if I underexpose manually, the preview would be too dark.

yeah it doesn't actually do anything for you in RAW that you couldn't do anyway but when you do need to favor highlights 1 stop then it certainly makes reading the histogram and image previews easier than looking at underexposed ones (And for jpgs/movies, those are cooked, so it's a real mode for jpgs since you can't add HTP back in afterwards in that case)

But in my experience, by underexposing by one stop, you are increasing shadow noise. Also, the EXIF states ISO 200 (or at least it did on my old 40D), regardless of whether or not it's underexposing. Personally, I'd rather shoot in RAW at the optimum ISO for the circumstances, so that I can control how much exposure I want (not the camera) and so that I can push the highlights as far as possible without losing detail to maximise DR. For landscapes, with the use of the correct amount of grad filters, I can mostly get as much DR as I need. For wildlife, shadows are rarely important enough for me to worry too much about them. There is also the argument that I just need better lighting.

 

[Marsu42]

But in my experience, by underexposing by one stop, you are increasing shadow noise.

Of course you are, and it's the same with htp.

Also, the EXIF states ISO 200 (or at least it did on my old 40D), regardless of whether or not it's underexposing.

Like a couple of people explained: The iso number is wrong, actually the picture is shot @ 1/2 iso you're seeing.

Personally, I'd rather shoot in RAW at the optimum ISO for the circumstances, so that I can control how much exposure I want (not the camera) and so that I can push the highlights as far as possible without losing detail to maximise DR.

htp applies to raw since a tone curve is applied in the converter. But htp does *nothing* to the dr potential of the sensor, you cannot increase it, you can just fully use it (= histogram filled and raw recovery) or distribute the data differently (expose to the right for shadow detail, to the left for highlight recovery aka htp).

 

[awinphoto]

To me the proof is in the pudding... Some landscape/scenic shots I took this weekend... Yes I know the 5d3 supposedly blows for landscape work, but here it is...

 

[Kernuak]

But in my experience, by underexposing by one stop, you are increasing shadow noise.

Of course you are, and it's the same with htp.

Also, the EXIF states ISO 200 (or at least it did on my old 40D), regardless of whether or not it's underexposing.

Like a couple of people explained: The iso number is wrong, actually the picture is shot @ 1/2 iso you're seeing.

Personally, I'd rather shoot in RAW at the optimum ISO for the circumstances, so that I can control how much exposure I want (not the camera) and so that I can push the highlights as far as possible without losing detail to maximise DR.

htp applies to raw since a tone curve is applied in the converter. But htp does *nothing* to the dr potential of the sensor, you cannot increase it, you can just fully use it (= histogram filled and raw recovery) or distribute the data differently (expose to the right for shadow detail, to the left for highlight recovery aka htp).

I didn't say it increases DR, I just said I'd like to maximise DR, big difference. I also said I'd[/] prefer control not the camera, I didn't refute anything you or anyone else said.

 

[iso79]

Yep, the 5D Mark III has shitty dynamic range