Are These The EOS 7D Mark II Specifications?

[LetTheRightLensIn]

The "huge difference" you are referring to in 5D3 vs. D8x0 online tests is not DR per se (the 5D3 clips to black about the same time as the D810) but latitude: the ability to push shadows without image destroying noise.

which is also DR

Except for the part where it's not.

and mostly importantly regardless of how you feel like defining terms...

I didn't define the terms. These are standard terms which were in use in the photographic industry long before I was born.

And it's a "huge difference" which can only be seen by turning all NR completely off for the Canon sensor :

yeah whatever sure

Try processing the files yourself sometime. When you intelligently use the NR sliders the difference is nothing like the drama tests. There is a difference, Exmor is better, but the difference becomes a more subtle one.

it doesn't seem subtle to me

and how exactly is it that a bye the eye DR rating gives a higher score than an engineering rating which goes deeper in the noise floor than some eyes would like?

 

[LetTheRightLensIn]

I read the posts and I didn't see anything about clean HDMI out. My thinking is that it should be a standard feature on the newer Canons. At least it would give the user the ability to record at a higher quality if they want to do it.

clean HDMI out actually proved to do very little for the 5D3, most of the damage was done at a pre-compression stage it seems, only ML RAW which avoided whatever it is they do to mangle the image quality produced radically better image quality

 

[LetTheRightLensIn]

"Photographic DR" would likely be less than DR measured this way, for a simple reason - we don't usually tolerate image detail that's near or at the noise floor.

DxO overstates DR for sensors with noise below their arbitrary threshold, and understates it for sensors with noise above their arbitrary threshold (especially in light of RAW conversion NR).

Perhaps more importantly, they're not measuring detail or steps in either case. If you photograph a step wedge with two cameras and one has slightly less noise in some of the otherwise pure black steps, DxO will count those as more DR! This is why when they "normalize" an image to 8 MP they report more DR.

It's silliness created by hardware nerds and not photographers. Someone needs to send DxO a few Ansel Adam's books.

Funny you mention Ansel Adams since I'd bet a lot of money that he'd be.... not on your side here.

 

[x-vision]

Does make you wonder though, if these end up being the final specs, why didn't they bang out a 7D replacement right after the 70D was announced? Why did we have to wait (still waiting) for a 2fps bump from the previous model and what seems like the exact same sensor from the 70D?

+1000

But will it still be awesome in 5 years time?? ???

Nope. In fact, the notion that the 70D is the best that we'll see from Canon
in the next 5 years in terms of 1.6x image quality is quite ... sobering.

I thought that they would do better.

 

[dash2k8]

I'm disappointed that the highest ISO setting is 51200. Yes, I'm a low-light geek. My work puts me in a lot of dim situations. Yes, I could use artificial lighting, but sometimes there's not enough manpower and/or time for that (the assistant is already doing audio). Right now I have a A7s and it's great. I was hoping that the 7D2 would, if not match, then at least have a huge increase in low-light capability. Judging from this rumored ISO spec, I'm guessing it's still more geared for sports and not about conquering dark caves (which I don't blame Canon for). If 51200 is the ceiling, then obviously Canon wasn't trying to push this area. The new sensor will help but I doubt the highest setting is going to be much useful (hope to be proven wrong).

 

[Chuck Alaimo]

Define prosumer? Quantify extra cost..

Cost is negligible to the point of being non-existent to the consumer. They put wifi in $200 point and shoots. The reason I mentioned pro and prosumer levels is because Canon already puts wifi in the bottom end, cheapest cameras they make and for some reason don't in the top level, most expensive cameras despite it being an incredibly useful feature for lots of working, professional photographers. I can't even begin to imagine why anybody would be opposed to features, even if you think you wouldn't use them, especially when they have no measurable cost to the end user.

Maybe it's because what I shoot is vastly different than what you shoot - but - I would think some kind of bluetooth connection would be better all around than built in wi-fi. I'm only chiming in here because you blanket the field of working professionals and state that it would be incredibly useful and why would anybody be opposed to those features? Well, I have a 6d and a 5d3. the wifi was a fun thing to play around with when i first got the 6d. I haven't even turned the wifi on in maybe 10 months or so, and that was only because I was on my honeymoon and thought ohh...maybe I'll post a few...couldn't though because the signal at the resort in jamaica kind of sucked!

so i just am at a loss to see the uses of wi-fi for pro level functions - and if your working for a media outlet that demands on the spot reporting, my guess is they'd pay enough for the add on transmitter...

 

[dtaylor]

I'm not talking about DxO's reporting of DR, I'm talking about computing real well capacity and read noise from DxO's measurements. DxO's interpretation of their own data is pretty much total crap, and I never visit their site. But the raw measurements are useful if properly interpreted.

I don't disagree with this statement. The problem is that people go to their site and accept their DR and overall score values at face value.

 

[dtaylor]

I used to prefer Photography DR until I couldn't find a consistent means of computing it.

Step wedge test.

 

[dtaylor]

The 70D is at 13 stops. There isn't room for "significantly better" DR with a 14-bit ADC.

Except that the 7D has 11 stops of DR, not 13 stops.

Except that I was talking about the 70D. It's a good guess that The 7D markII sensor will be at least as good as the most recent APS-C sensor from Canon.

The D7100 and the Exmor sensors have 13 stops of DR.

13.3 to be precise.

Canon is still stuck at 11,

EOS M 12.4 Total RAW
http://www.imaging-resource.com/PRODS/canon-eos-m/canon-eos-mA5.HTM

70D 13 Total RAW
http://www.imaging-resource.com/PRODS/canon-70d/canon-70dA5.HTM

Myths die hard, don't they.

 

[Chuck Alaimo]

yikes...16 pages....

rumored specs. I find this funny because 16 pages later ---- it's still just a rumor!

with that said, was not in the market for a crop camera before this rumor, and still am not after. I like FF. In terms of the specs I'd say this camera sounds like a beast of a sports and wildlife camera. Improved AF, newer better sensor (newish? Maybe, who knows...) - all things sports and wildlife folks will have lots of fun with.

am I angry that this doesn't make me want one? Nope, not one bit. The 7d was never the go to cam for landscape work. not that it won't be able to do fine landscape shots, but the 7 series was about giving pro action grade features to a crop camera for less cost. These specs sound spot on

Of course, I bet this is just the first of many possible spec sheets we'll see floating around the interwebs. Maybe this is a tease and we're actually looking at a 24MP sensor with a totally new design. Or, maybe they considered going that route and ended up with a camera that was great but with a price tag of $2800 and came to the conclusion that the crop sensor market just wouldn't accept that kind of price for a crop camera. So they backed it down to a more conservative level which could retail in the $1700-1900....

bottom line is....nobody knows nothing about nothing right about now!!!

 

[dtaylor]

it doesn't seem subtle to me

and how exactly is it that a bye the eye DR rating gives a higher score than an engineering rating which goes deeper in the noise floor than some eyes would like?

For the nth time...DxO is not measuring photographic DR. They are measuring sensel SNR, running those values through a 'black box' formula, and predicting DR.

Understand that direct observation always trumps theory and prediction. Always. Never the other way around. No matter how much math, effort, belief, or faith you put in a theory, nor how many so called 'experts' trust in the theory, direct observation always wins.

DxO measures sensel SNR and concludes that a 70D (for example) only has 11.6 stops DR. Someone photographs a step wedge and sees 13 stops between black and white. The question is not "how exactly is it that a bye the eye DR rating gives a higher score than an engineering rating"? No, the question is "where did the engineering testing, model, or rating fail?"

 

[dtaylor]

Funny you mention Ansel Adams since I'd bet a lot of money that he'd be.... not on your side here.

Come back when you've read some of his work.

 

[expatinasia]

I am sorry, I do not have the energy to read all 16 pages!!

The specs sound nice and the 61 AF points all cross type gets me very excited if only because it makes me wonder what the future 1D X ii will have.

Does Servo AF for video shooting mean it will track subjects like the 60D or 6D does, where you can easily switch focussing points or maintain face focus even if subject moves?

Good it has a headphone jack.

Will it be APS-C?

Is the 10fps with the mirror locked up or its maximum standard speed?

Nice to be having new rumours but should they not be CR something, like CR1 or CR 2. If you confirm something later then just adjust the CR rating.

 

[jrista]

The "huge difference" you are referring to in 5D3 vs. D8x0 online tests is not DR per se (the 5D3 clips to black about the same time as the D810) but latitude: the ability to push shadows without image destroying noise.

which is also DR

Except for the part where it's not.

and mostly importantly regardless of how you feel like defining terms...

I didn't define the terms. These are standard terms which were in use in the photographic industry long before I was born.

And it's a "huge difference" which can only be seen by turning all NR completely off for the Canon sensor :

yeah whatever sure

Try processing the files yourself sometime. When you intelligently use the NR sliders the difference is nothing like the drama tests. There is a difference, Exmor is better, but the difference becomes a more subtle one.

The differences are subtle to non-existent at ISO 400 and up. However, I have processed Exmor images (some of them from people on these very forums who were willing to share theirs), and there is no question that you do have considerably more editing latitude with an Exmor. A D800 allows more shadow lifting. Now, I can deband a 5D III image, and the differences drop...however, the D800 images still contain more detail and usually still have less noise in the shadows. Here is an example of a 5D III vs. a D800 from Fred Miranda's examples back when he first reviewed the 5D III and D800 (he was the first guy to clearly demonstrate with actual images the difference, and this was the first time I believed the D800's editing latitude advantage):

The 5D III looks pretty bad, but strait out of the camera, that's the real difference. That is the editing latitude difference I'm always talking about. Now, things don't have to stay that bad. The D800 doesn't need anything done to it, it's fine as it is, but with a minute or two worth of debanding and denoising with a tool like Topaz (what I used in this case), you can close the gap considerably:

The D800 obviously maintains the detail lead, and it still has less noise, but the 5D III image doesn't look like crap anymore. I could probably reduce shadow noise even further, however if I did I would start eating away at even more detail, and at this level, it isn't "chew your hand off at the wrist" annoyingly bad. It's actually quite decent in the grand scheme of things...a considerable amount of DR has been recovered (maybe a stop or so, definitely not the full 2.2 stop difference between the two cameras.)

 

[jrista]

I used to prefer Photography DR until I couldn't find a consistent means of computing it.

Step wedge test.

That only gets you raw data to work with. You still have to compute the actual dynamic range, engineering or photographic or whatever, FROM that image. Just saying "Oh, I shot a step wedge" doesn't tell you enough...you have to measure the noise levels in the darkest wedges (from the original RAW data itself), measure the white point (from the original RAW data itself), THEN COMPUTE dynamic range between the RMS noise level and the white point. Just shooting a step wedge and looking at the resulting image leaves you with a purely subjective interpretation of the result. Only when actually measuring the noise and actually calculating a real number for dynamic range do you end up with an objective result.

 

[jrista]

The 70D is at 13 stops. There isn't room for "significantly better" DR with a 14-bit ADC.

Except that the 7D has 11 stops of DR, not 13 stops.

Except that I was talking about the 70D. It's a good guess that The 7D markII sensor will be at least as good as the most recent APS-C sensor from Canon.

The D7100 and the Exmor sensors have 13 stops of DR.

13.3 to be precise.

Canon is still stuck at 11,

EOS M 12.4 Total RAW
http://www.imaging-resource.com/PRODS/canon-eos-m/canon-eos-mA5.HTM

70D 13 Total RAW
http://www.imaging-resource.com/PRODS/canon-70d/canon-70dA5.HTM

Myths die hard, don't they.

Your being rather misleading. This is what the article says:

The Canon 70D's RAW file scored 1.1 f-stops more in total dynamic range (13 vs 11.9 f-stops) but the score at the highest quality level increased only 0.7 f-stops from 7.61 to 8.31, which isn't much of an improvement over the JPEG and below average these days. Like all recent Canon SLRs, the higher quality scores are somewhat below average for a modern sensor. For example, the Nikon D7100 managed 10.1 f-stops at the highest quality level, almost 2 stops better.

There is a difference between their total dynamic range test, and the test at certain quality thresholds. I am not exactly certain how they do these tests or how they are calculating dynamic range, but they are pretty clear that when it comes down to QUALITY (and that's what were talking about here), the high quality test still shows that the Nikon cameras have a two stop advantage. I don't know if it is necessary to use the high quality state...they also have a Low, Medium, and Medium high state. Even at low, the 70D scored 11.7 stops, and at medium it was 10.8. Those numbers seem more in line with what other sites measure.

As I just recently posted, the out-of-camera DR doesn't need to be where the story stops...you can always spend an extra couple of minutes applying noise reduction with a tool like Topaz DeNoise 5 to recover a lot of DR lost to Canon's noisy downstream components, and thanks to the fact that Canon uses a bias offset to set black level, rather than clipping to the black level, there is a LOT of detail and DR that can be recovered. The playing field, with effective NR, from a visual standpoint, isn't super terrible. But Exmor sensors still have an advantage. I did not need to apply any NR at all to the D800 example from Fred Miranda. It would be really nice to be able to skip that step with Canon cameras, too.

 

[neogomo]

How long does Canon try to milk the old cow?

If these specs turn out right, it seems Canon doesn’t want or isn’t able to compete with Sony in the foreseeable future.

To exaggerate even a bit more, if I want a good APS-C-sized sensor camera, I use my Ricoh GR.

+ way smaller and lighter than a DSLR
+ way cheaper than a (Canon 70D/7D/7DII)
+ better Dynamic Range too.

– fixed lens
– good build quality, but not like a tank

Ok, to be fair, Canon DSLR’s also have their advantages, good at skin color and tonality, or the whole accessory-system, flashes, lenses, etc. but all this will fade away, if Canon isn’t able to up their specs, sensor-wise.

I was a long-time 5D II user, and EOS 10D, 30D, 40D before that ... so I don’t just want to bash on Canon, I really want to have them come back, but maybe their business still works too well for them – even without greater technical improvements – but that won’t last forever.

 

[jrista]

This might be a stupid question, but can anyone explain to me what's the point of Dual Pixel AF when you don't have a touch screen ?

Continuous AF tracking when shooting video.

 

[x-vision]

70D 13 Total RAW
http://www.imaging-resource.com/PRODS/canon-70d/canon-70dA5.HTM

Myths die hard, don't they.

Well, here's what the IR says about dynamic range.

• While it's tempting to evaluate dynamic range as the maximum number of tonal steps that can be discerned at all, that measure of dynamic range has very little relevance to real-world photography. What we care about as photographers is how much detail we can pull out of the shadows before image noise becomes too objectionable.

You seem to be of the tempted type .

But what we care about as photographers is how much detail we can pull out of the shadows.
Not my words, btw.

 

[jrista]

70D 13 Total RAW
http://www.imaging-resource.com/PRODS/canon-70d/canon-70dA5.HTM

Myths die hard, don't they.

Well, here's what the IR says about dynamic range.

• While it's tempting to evaluate dynamic range as the maximum number of tonal steps that can be discerned at all, that measure of dynamic range has very little relevance to real-world photography. What we care about as photographers is how much detail we can pull out of the shadows before image noise becomes too objectionable.

You seem to be of the tempted type .

But what we care about as photographers is how much detail we can pull out of the shadows.
Not my words, btw.

Totally agree! That is probably the best definition of "Photographic DR" there is...the range between highlights and shadows that determines how much detail we can pull out of the shadows before noise becomes objectionable.

The trick is adequately defining what "objectionable" is, so that we can have an objective means of defining PDR.