Who said Canon cameras suck?!?

[AmbientLight]

If the expected sensor improvements are similar to autofocus improvements between 5D Mark II and 5D Mark III all the whining may at least lead to something worthwhile.

[Zlatko]

After 4 years what we see is basically a resolved AF. Thats great, but given the resources of Canon for R&D, it's just laughable how they advanced sensor-wise. Basically they didnt.

You seem to be uninformed about the list of improvements going from the 5D2 to the 5D3.  Canon's R&D was busy addressing numerous requests from professional photographers, and they delivered brilliantly.  The AF is not just "resolved" as you say; it is upgraded to 1D-series level.  And you're incorrect about Canon not advancing the sensor.  I shoot both 5D2 and 5D3 and I see the advance in the sensor every week.  The 5D3 has significantly better high-ISO performance, which makes a big difference for my work.

[nightbreath]

If the expected sensor improvements are similar to autofocus improvements between 5D Mark II and 5D Mark III all the whining may at least lead to something worthwhile.

For whom? For people who care more about technical part than about the photography itself? Are those who whining really able to use at least capabilities of previous generation bodies? For me it looks like:

[Chuck Alaimo]

on one of the local wedding forums I visit, a nikon user was asking for some upgrade advice and this is what was said, which I find quite interesting and relavant to the whole 5d3 vs d800 debate:

" I've spent around 2-weeks exhaustively researching my next camera. I do not think that the D800 is good for weddings... it is extremely slow shooting and you are paying for a lot of resolution that serves minimal practical purpose in retail photography (i.e., a 12mp cropped file makes a gorgeous 24x36" canvas).

If you can handle a single card slot, I think the D700 is presently the best value camera in Nikon's lineup for weddings. I owned one and have used many others and never really had an issue with it.

I think the D3s is the best wedding camera on the market today. Fast shooting, exceptional AF, amazing high ISO capabilities and ample resolution... even in 14-bit uncompressed RAW, I have never hit the buffer (and I am one fast shooter)!

When I shoot events, I use a D3s as my primary camera and a D3 as my backup and feel very comfortable with that set up."

And,

"As an owner of the D800, it never comes out at weddings, I stick to the D3s. As Brady said, it is just far too slow of a shooter. Plus, the files sizes are too much of a hassle to drag around and edit. A 16 bit one layer tiff is 289.2 mb per file. I would be looking at a D700, D3 or D3s."

And,

"The D700 is amazing! All the features of the D3 that meant anything and was about $2k less. I love that camera. I still use mine today and it has been three years. The D800 is overkill for weddings. It should come out for portraits, maybe some details and that is it. It is total overkill for the rest of a wedding..."

There ya go, for all the rave reviews, there are many who are opting out of the d800.  Grass is always greener!

[Bosman]

I rarely have a need for iso 100 so i like to use highlight tone priority, it saves me sometimes as i like to shoot to the right of the Histogram.
Per Canon
Highlight tone priority extends the dynamic range of highlights by about one stop and improves gradation within highlight areas.

[jrista]

Some of you Nikon D800 fanboys really make me sick.
You think the D800 is the be-all and end-all of all DSLR cameras, when in fact it has some glaring faults like poor autofocus (especially on the left side) and a rear screen that makes everything look green.

I do a lot of commercial shots where colour is very important, and if I had to go by the rear screen, a D800 would drive me bonkers!
I can and do go by the screen on my 5D3 and it is very accurate, unlike the Nikon screen.

Put properly exposed shots of the D800 beside shots from a 5D3 on a good quality computer screen and you would have trouble picking the differences, except the Canon's colours are more true to life.

Even if the D800 had a thousand megapickles, the 5D3 is still a better all-around camera and is certainly my tool of choice for the jobs I do.
In fact, with all the types of photography I do for a living, or for my own fun, it has always done a sterling job, even in very difficult and demanding situations.

You guys who carry on about how much better a D800 is than a 5D3 remind me of the immature little boys who say that their car is better than yours, because it can do 0-60 1/10th of a second quicker, yet it rides like a buckboard and handles like a limp rag.
There is far more to a good camera than a heap of megapickles, just as there is a lot more to a good car than a quick 0-60 time.

we are talking sensor only here nothing more

other than low ISO DR the 5D3 really is quite awesome, 6fps FF, 1 series AF, compact body size, now with ML the video is quite usable and soon it should offer better compression and be a really nice video solution, nice UI (more MP would be nice but you can't have it all yet perhaps so it's really just the low ISO DR that was the one unfortunate thing, other great)

I'm sorry, but no. I'm talking about the Canon SYSTEM. I used Sensor in my original title, but its not just sensor capabilities I'm talking about. I've corrected the title.

[Chuck Alaimo]

Technology is a race.  You don't lead every lap.

of course .. but when you are second place you normaly don´t have the most expensive products... 

Ok, lets think about this.  Imagine if, and we don't know if this is true, if, both canon and nikon put the same $$$ into R&D on both the d800 and the 5d3.  Dollar for dollar from sensor dev on through parts and manufaturing to shipping and advertising.  Dollar for dollar match.  Now, nikon, who is trying to catch up to canon, decides to sell their body at a lower profit margin in order to climb their way back.  Nikon needs to do that to catch.  Canon has no incentive to under-price their products --- BECAUSE THEY ARE THE MARKET LEADER!

Reality is that nikon didn't put in all the $$$ canon did in R&D due to the deal with sony for sensors.  All that savings rolls into a lower retail price, which if canon matched then they'd be taking a loss on the first few production rounds.

And you know what?  I am happy for all of this!  The better nikon can be the better canon will be, and thus the cycle repeats itself.  So yeah, I am routing for nikon because the more they advance the more canon will too --and this benefits all of us!

[jrista]

Lets stop talking about the D800 as it its simply and solely a trophy to be compared, and start talking about it from a real world context. No one cares how it compares if you upscale a 5D III image to 36.3mp size. Neither does anyone who uses a 5D III or any other Canon camera really care how it compares if you downscale a D800 image to 22.3mp size. They care whats possible in the real world, with real-world software...tone curves and all.

real world it's simply not fair to compare cameras on a non-normalized basis to one another,it doesn't make any sense

And yet...no one actually lives in the limited reality wherein technical comparisons between hardware actually create photography. Sorry...people live in THE REAL WORLD, and in the REAL WORLD, people don't "objectively" utilize their cameras with unmodified, linear import to "see" all the dynamic range their camera has available. In the REAL WORLD, people apply base tone curves to their photos, to compress the considerable dynamic range...from either a Nikon or a Canon camera...which tends to be far greater than the dynamic range of either our computer screens or anything in print, into the much smaller dynamic range of those devices so our photos actually look good.

Tone curves are a real-world thing, they actually exist, and they are utilized by the very, very vast majority of photographers. The differences in tone curves and the distribution of levels in those curves between manufacturers is a meaningful topic.

[jrista]

Why people mistake DR for exposure latitude?

I always thought they were the same. Care to explain the difference?

Dynamic range refers to the total maximum physical range of tonal levels a camera sensor is capable of recording. Exposure latitude refers to the ability of an actual exposure taken with a camera to be tuned or adjusted. A true RAW photo strait off the sensor is very dull, flat, lifeless, lacking a significant amount of contrast. When we import our RAW photos, most RAW editors apply a tone curve. Usually one of the manufacturer defaults (such as Camera Standard or Camera Neutral, etc.) These tone curves adjust how levels are allocated in the final image you see on your screen.

In a linear image, levels are distributed equally (hence the dull, lifeless, low-contrast appearance). With a tone curve applied, more levels are allocated to the shadows and the highlights, effectively "compressing" the wide dynamic range into a narrower contrast range. That brightens and adds life and color to an otherwise dull original exposure. The side effect of that is you have a lot of levels "bunched up" in the shadows and in the highlights around the roughly linear growth of the midtones. It's thanks to these tone curves that we have the ability to "recover" highlights and "lift" shadows.

Technically speaking, LetTheRightLensIn is correct...there is no such thing as highlight recovery or for that matter shadow lifting. Not with a true RAW image that has not yet had tone curves applied. But we generally don't work with our RAW photos in their true form. When it comes to the shape of tone curves, Nikon tends to allocate a lot more levels to the "foot of shadows" than Canon (and, for that matter, most other manufacturers, including MFD manufacturers.) They have more freedom to for sure, thanks to their lower read noise. That doesn't account for the ability to push shadows around by as much as 6 stops though...Exmor sensors only offer about 2 stops of additional DR in the shadows. Examining Nikon's tone curves indicates they allocate more levels to the shadows than their low read noise offers alone with their curves.

Similarly, Canon allocates more levels to the "shoulder of highlights" in their tone curves. They don't allocate as many more levels to highlights as Nikon seems to do to the shadows, however in Canon's newer cameras their highlight shoulder tends to be a little longer and fall off more into the highlight range than Nikon cameras. This is part of the reason you can overexpose by four stops with a modern Canon camera and still be able to recover (although its doubtful you could overexpose by 6 stops and still recover...Nikon still has around a 2-stop DR edge in the end.)

Exposure latitude is benefited by these tone curves, and the ability to recover highlights and shadows from "beyond the foot and shoulder." Exposure latitude is enabled by DR, and the more DR you have, the more you can tune those curves to allow greater and greater latitude.

[jrista]

You're right, it does have 14.4 stops of DR, despite a 14-bit ADC.  It's also powered by an internal perpetual motion machine, floats in the air when released, and basically defies many other laws of physics and thermodynamics.
 

at the pixel level, the D800 DR at ISO 100 is 13.23EV.
the 14.4 is because of the normalizing in the DXO print mark.

Yes, I know. 

If a data analysis method includes a normalization step which forces data to fall outside of the range that's physically possible for the measurement, that data analysis method is flawed, and by extension, any conclusions based on that method are also flawed.  If a hospital reported to parents that their newborn infant had a population-normalized length of -4", you'd say WTF, a negative height is impossible, right? 

Same thing with a 14.4 DR from a 14-bit ADC.  WTF, that's impossible.  Change the method, becasue the method is flawed.  If the analysis method is flawed, the resulting conclusions (i.e. DxOMark's Scores) are also flawed.  Note that I think (and I've repeatedly stated) that their Measurements are valid and useful - it's the Scores, which are based on the flawed normalization step (and have other problems, like undisclosed 'black box' weighting of sub-components) that are meaningless.

Couldn't have said it better myself. BTW, Love the perpetual motion machine...although in the case of the D800, I think its the aura of fanboys that surround it that have perpetual motion.

[cliffwang]

This thread is getting funny.  I use 5D3 and I am satisfied.  I haven't tried D800, but I know D800 is a good camera as well especially its high MP & DR.  We do see a lot of samples from many website how helpful the high MP & DR are.  On the other hand, we also see how useful 5D3 is in low light environment.  I thought that's common sense already.  Is that necessary to discuss 5D3 vs D800 now?

[marekjoz]

I've read somewhere (can't remind the source but can search for it), that in fact Nikon's NEF RAWs manipulate with data from sensors in the shadows area. It could be their compressed NEF.
But, I want to point to you, that in fact it is "almost possible" to get more than 14 DR stops from 14 bits data file. Since most people are most focused on shadows and highlights, you can simply forget about some information in the center of the band. It's later all the matter in implementation of interpretation of these data. Sure you loose information in the center, but you can proove that you gain more than 14bits from darkest to highest if you pay attention only on these areas

[jrista]

I've read somewhere (can't remind the source but can search for it), that in fact Nikon's NEF RAWs manipulate with data from sensors in the shadows area. It could be their compressed NEF.
But, I want to point to you, that in fact it is "almost possible" to get more than 14 DR stops from 14 bits data file. Since most people are most focused on shadows and highlights, you can simply forget about some information in the center of the band. It's later all the matter in implementation of interpretation of these data. Sure you loose information in the center, but you can proove that you gain more than 14bits from darkest to highest if you pay attention only on these areas

Well, DR generally refers to a hardware capability... the maximum range of levels a sensor can record, from its darkest (average of noise) to brightest. Dynamic range doesn't change in post... it's a fixed hardware trait. So you can't really just "exclude some of the midtones" (remember, a stop is a doubling... you would have to lose 1.2 stops of midtones, which would be noticeable.) What changes in post is how you utilize those levels... the tone curves you apply that attenuate them.

[jrista]

This thread is getting funny.  I use 5D3 and I am satisfied.  I haven't tried D800, but I know D800 is a good camera as well especially its high MP & DR.  We do see a lot of samples from many website how helpful the high MP & DR are.  On the other hand, we also see how useful 5D3 is in low light environment.  I thought that's common sense already.  Is that necessary to discuss 5D3 vs D800 now?

Well there is always something to say when the facts are incorrect. I frequently hear about how people can push D800 (or, for that matter, D7000) shadows around "by an amazing 6 stops!!" Well, sure, you can...but that doesn't mean the D800 or D7000 are capable of a full 6 stops more exposure latitude than a Canon camera. At a hardware level, the D800 is about 2 stops better than current Canon cameras (some less, as in the case of the 1D IV, which gets about 11.5 stops of DR, and some more as in the case of the 5D III, which gets about 11 stops of DR.)

The point of my post was to show that Canon's cameras aren't "six full stops worse" than a D800 or D7000. In fact, Canon cameras have for a number of years had rather amazing highlight recovery, the kind of highlight recovery that used to be relegated to MFD cameras. The kind of highlight recovery that even the D800 doesn't quite seem to reach (http://www.photographybay.com/2012/04/21/nikon-d800-vs-hasselblad-h4d-40/). I wanted to show that you can push Canon exposure around by at least four full stops when you overexpose. That greatly diminishes the differences between Nikon cameras that use an Exmor sensor, and Canon cameras. Nikon cameras till have a 2-stop advantage, but its not a 6-stop advantage as some video reviews and commentary might have you believing.

[marekjoz]

I've read somewhere (can't remind the source but can search for it), that in fact Nikon's NEF RAWs manipulate with data from sensors in the shadows area. It could be their compressed NEF.
But, I want to point to you, that in fact it is "almost possible" to get more than 14 DR stops from 14 bits data file. Since most people are most focused on shadows and highlights, you can simply forget about some information in the center of the band. It's later all the matter in implementation of interpretation of these data. Sure you loose information in the center, but you can proove that you gain more than 14bits from darkest to highest if you pay attention only on these areas

Well, DR generally refers to a hardware capability... the maximum range of levels a sensor can record, from its darkest (average of noise) to brightest. Dynamic range doesn't change in post... it's a fixed hardware trait. So you can't really just "exclude some of the midtones" (remember, a stop is a doubling... you would have to lose 1.2 stops of midtones, which would be noticeable.) What changes in post is how you utilize those levels... the tone curves you apply that attenuate them.

Yes, I know what you mean - it was just about showing more than 14DR using 14bits of data.
How about making a sensor in which, let's say even pixels in the first row work at ie ISO 100, and odd works at ISO 400 (so base ISO +2)?  In the second row otherwise of course and so on. Knowing that, you can easily interprete all the data properly and record more than now - sth like hdr directly from the sensor.