Are These The EOS 7D Mark II Specifications?

Lightmaster said:
LetTheRightLensIn said:
It certainly sounds good and yet if it doesn't do more than just give a 70D sensor without AA (and maybe it does do more) and no 4k it seems like it's something they could've produced as is some time ago already.

my first thoughts exactly.

where are the video features that are hyped for years?

does the 20.2 MP 70D sensor replaces the 18MP sensor now for the next 5 years?

+1000

If the 7DII sensor is a tweaked/re-purposed 70D sensor, I'd be very disappointed.

There's no point in a 7DII, IMO, if image quality is only marginally improved vs the 7D - and that's exactly the case with the 70D sensor.
Is this really the best that Canon will have in the next 3-5 years for APS-C image quality ??

Well, I have a 70D already, so I guess I'm all set then.
Bummer.
 
Upvote 0
Steve said:
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.

Do you want the lens from a $200 p&s on your DSLR? Do you want the flash from a $100 powershot? The screen?

Yes. Lets look to the consumer market for inspiration.... compact pixel pitch on a 135 DSLr sensor... must be in the 100's of MP. yeah. great? right?

Or do you want stuff that works properly?

A powershot that talks to your printer in the suburbs, or a high bandwith wifi connect in amongst 30 other high bandwith sports shooters.

Yeah, lets have consumer grade built in pish.

Or let the folk who need it set it against tax and buy something with dual channels. security. bandwith, seperate battery units and reliability.

Go.

For.

It.


The bottom line is, canon will launch what they launch. It's very late here. Night xxxx
 
Upvote 0
pablo said:
Yes. Lets look to the consumer market for inspiration.... compact pixel pitch on a 135 DSLr sensor... must be in the 100's of MP. yeah. great? right?

Absolutely, it would be great. I'd kill for that. Image quality would destroy current DSLR sensors (at high ISO too), and it would act as a built-in optically-perfect teleconverter as well.
 
Upvote 0
LetTheRightLensIn said:
dtaylor said:
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.
 
Upvote 0
Marauder said:
Secondly, "79 AF points, granted not all cross type..." That in itself speaks volumes. The D7100 has 51 AF points vs the 70D's 19. But only 15 of the D7100's are cross type. There is a HUGE difference between a big number of AF points, vs their type. Also, take a look at the Servo AF comparison between the D7100 and the 70D here: https://www.youtube.com/watch?v=bOM4r1gxsbs
Note the 70D's buffer is substantially better as well, which is a factor when shooting action. Unfortunately, the tester only shot both cameras in JPEG and RAW+JPEG, and not in RAW only. The 70D not only outperforms the buffer of the D7100 in JPEG by a large margin, it also does so in RAW only. Neither is stellar RAW+JPEG (original 7D after fw update beats them both handily there) but the 70D is still better in buffer depth.

This part of your post doesn't mean anything. I was talking about the A77 II, not the D7100. The A77 II has a pretty damn good AF system by all accounts, a deep buffer, and 12fps - all of that feeding into a pretty fantastic sensor for a very reasonable price. That was my point - somebody already made the 7DII that most people around here were asking for, it just wasn't Canon. At least not yet. These specs CanonRumors put up may not be correct and, even if they are accurate, may indicate a fantastic camera if the price is right, but so far they aren't much to get excited about. That was the point of my post.

As for gimping framerates, that part wasn't directed specifically at you. Just because I quoted your post to discuss one point doesn't mean that the entire post is specifically addressing points that you have personally made. I wrote that last line because it has been a general vibe around here that when someone would say something like "I just hope the marketing guys don't gimp the 7DII!" many people would jump in with "I've worked in engineering for 3489054378924 years and that NEVER happens!" Mostly, it seems like its jrista but there are others as well. I just didn't feel like quote mining for a throw away line.
 
Upvote 0
Steve said:
Sony already dropped a crop camera with 79pt AF system (granted, not all cross-type, but still excellent by all accounts), 12fps burst, and a better sensor earlier this year for less than $1700.

Every time I've looked at user reports Sony's DSLRs could not keep up in AF tracking with Canon or Nikon bodies. Not the 7D, and certainly not the 1DX or 5D3.

It's not just point count. Many Nikon bodies have the same point count as the D3 or the D4. But talk to a Nikon shooter and the actual performance varies considerably. Didn't the D800 have the same "AF module" as the D4? I know a guy who will tell you straight out that it could not track like one. Not even close.

There is no reason that Canon should be falling behind Sony of all companies.

Have you compared lenses? ;D

At this point a "revolution" in sensor tech will require multiple layers (for DR or for color) or at least 16-bit ADCs and sensel characteristics to produce meaningful bits beyond 14. This is true for Sony as well. I got excited when the rumors were for a multilayer sensor. Now that it appears this is not the case, I expect incremental improvements.
 
Upvote 0
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.
 
Upvote 0
Lee Jay said:
"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.
 
Upvote 0
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.
 
Upvote 0
Based on the specs and having the camera arsenal I already have, I won't be buying the 7DII. I have been thinking about the 7DI though. I do a fair amount of time lapse and the 7DI is rated at about a 200K shutter count. That is pretty stout for the current price of the 7DI.
 
Upvote 0
x-vision said:
+1000

If the 7DII sensor is a tweaked/re-purposed 70D sensor, I'd be very disappointed.

There's no point in a 7DII, IMO, if image quality is only marginally improved vs the 7D - and that's exactly the case with the 70D sensor.
Is this really the best that Canon will have in the next 3-5 years for APS-C image quality ??

Well, I have a 70D already, so I guess I'm all set then.
Bummer.
My thoughts as well so add my +1 to that pile lol
I do like the IQ improvements from the 60D/7D to the EOS-M/70D age (yep used them all), but they were only really minor sharpness tweaks probably due to them doing a bit better on the AA filter front. I really do like my 70D but I'm not blind that it lacks on the DR front vs the competition (luckily I can make do). Like many others I was waiting for this to their comeback to at least match them in the sensor game, but if that "20.2MP" indicates what we're thinking ... then heck I'd rather keep my 70D with the flippy screen (great for low angle) and WiFi.

Then again I'm obviously a stupid n00b shooter because I prefer a feature not put onto "Pro" bodies ::)

Who knows, maybe they have improved things like the read noise significantly again to give a similar result to higher DR (a bit like how neater ISO grain not necessarily less still makes it a little more usable)
 
Upvote 0
dtaylor said:
LetTheRightLensIn said:
Put another way the 1/3 stops difference between 70D and aps-c exmor sure doesn't match what ones sees with their own eyes when out taking photos or the huge difference between D810 and 5D3....

So we're going to draw conclusions from an entirely different format???

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.

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

The 5D III does not clip to black at all...it uses a 2048 bias offset.
 
Upvote 0
Steve said:
Marauder said:
Secondly, "79 AF points, granted not all cross type..." That in itself speaks volumes. The D7100 has 51 AF points vs the 70D's 19. But only 15 of the D7100's are cross type. There is a HUGE difference between a big number of AF points, vs their type. Also, take a look at the Servo AF comparison between the D7100 and the 70D here: https://www.youtube.com/watch?v=bOM4r1gxsbs
Note the 70D's buffer is substantially better as well, which is a factor when shooting action. Unfortunately, the tester only shot both cameras in JPEG and RAW+JPEG, and not in RAW only. The 70D not only outperforms the buffer of the D7100 in JPEG by a large margin, it also does so in RAW only. Neither is stellar RAW+JPEG (original 7D after fw update beats them both handily there) but the 70D is still better in buffer depth.

This part of your post doesn't mean anything. I was talking about the A77 II, not the D7100. The A77 II has a pretty damn good AF system by all accounts, a deep buffer, and 12fps - all of that feeding into a pretty fantastic sensor for a very reasonable price. That was my point - somebody already made the 7DII that most people around here were asking for, it just wasn't Canon. At least not yet. These specs CanonRumors put up may not be correct and, even if they are accurate, may indicate a fantastic camera if the price is right, but so far they aren't much to get excited about. That was the point of my post.

As for gimping framerates, that part wasn't directed specifically at you. Just because I quoted your post to discuss one point doesn't mean that the entire post is specifically addressing points that you have personally made. I wrote that last line because it has been a general vibe around here that when someone would say something like "I just hope the marketing guys don't gimp the 7DII!" many people would jump in with "I've worked in engineering for 3489054378924 years and that NEVER happens!" Mostly, it seems like its jrista but there are others as well. I just didn't feel like quote mining for a throw away line.

Well it's not really a throw away line if you're aiming it specifically at something I'd posted as speculation (a speculation that the dichotomy in frame rates might mean that the camera has latent capabilities beyond what it might ship with. And the then imply that the same people who are saying Canon marketing has no bearing on specs are "now sagely acknowledging that they are probably purposefully gimping frame rate purely for marketing reasons." Seems fairly direct.

At any rate, my pondering is pure speculation of course, but it's possible. I also do concur that it would be far better for Canon to overreach and give 12fps over 10--but 10 is nothing to sneeze at!

Regarding my D7100 vs 70D, I'm well aware that you weren't referring to the D7100. My point is that the D7100 vs 70D is analogous to the 7D2 vs the A77II comparison you were making. I wasn't as clear as I ought to have been. I used the D7100 vs 70D since both are now known quantities. We can do direct comparisons, not just on paper figures, but on actual comparable results. The reason it springs to mind is I've seen many such comparisons made in less extensive reviews where the 51 point AF system is given the win to the D7100 in terms of AF, simply based on the fact that it has 51 vs 19 points. Simply looking at the total number of AF points, and not just the type of AF point can be very deceptive. The 70D may only have 19, but they are all of the far more accurate cross type, compared to 15 cross type on the D7100. Moreover, in a more detailed review, such as the one I referenced, not only does the reviewer point that fact out, they also do a real world test of Servo capability, which the 19 point 70D system (which is a much older AF than the D7100's) wins. They also compare buffer depth, which the 70D wins. (D7100 wins it's share of contests too--not all one sided and a very good review).

Take that analogy forward to Sony A77II to Canon 7D2, we are comparing a camera with 79 AF points, of which only 15 are cross type to one with 65 all cross type AF points. . On the face of it, I don't see those as comparable. Not to say the Sony system will be bad--but I don't think it "wins" because 79 is more than 65! Not when it's also 15 vs 65! There's more than just the total number of AF points to consider.

Moreover, we don't know how their AF systems will compare yet. We don't even know if the 7D2 specs we have are correct yet! And once we do, that STILL won't tell us how they compare until someone takes them out neck and neck. We don't even know if the A77 sensor will really be "better." All we can say is that the current fps shows an advantage to the A77, compared to this published spec, but we don't know anything about buffer comparisons (needed to USE that fast fps) or how their AF systems will actually work in the field.

In essence, I think the 7D2 will probably deliver an outstanding action camera. But I don't know that yet...and neither does anyone else! 8)
 
Upvote 0
Chuck Alaimo said:
Lee Jay said:
jrista said:
Actually, the use of Dual DIGIC 6 makes me really think that there is NO more magic cooked into the sensor, and that all the "magic" is happening after the signal is pulled OFF the sensor. It's probably roughly the same sensor that employs a slightly new DPAF design, and probably has a very weak or no AA filter, but is otherwise unchanged from the 70D. DIGIC 6, which actually came out before Sony BionzX, actually has a LOT of the same capabilities, and is the primary reason the IQ on their smaller form factor cameras is good. They really cook the signal coming off the sensor.

Since DIGIC 6 supposedly allows frame rates to 14fps and 60fps video, why would they need two of them? Dual pixel works okay on the 70D with just one DIGIC 5+.

Not positive but doesn't the 1dx have dual digic and one is to process images the other for for AF? Could that be what is going on for the 7d2?

Could be, but those are DIGIC 5+ chips.
 
Upvote 0
Lee Jay said:
dtaylor said:
Lee Jay said:
dtaylor said:
* DxO measures SNR which does NOT directly translate to photographic dynamic range.

DxO measures, but does not directly report, well capacity and SnR.

The method of determining DR from that is shown here:

http://sensorgen.info/Calculations.html

Great. That tells me what I already know. That they are not measuring or reporting photographic DR. And that there are formulas I would have to hunt down or reconstruct in order to verify their results. (Though I suppose that would be possible to do for anyone so inclined.)

Again, DxO DR measurements are nonsense.

"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.

This is measured to the noise floor.

Photographic DR is less, usually quite considerably less (by many stops). The problem with Photographic DR is it tries to be perceptual in nature....however it is rather arbitrary. The offset by a certain SNR, but there are rarely any specific rules about how that SNR offset is chosen. I used to prefer Photography DR until I couldn't find a consistent means of computing it. I also learned, after reading sensor design patents from half a dozen companies or so, that computing DR to the literal noise floor seems to be the way most companies compute DR:

Code:
20*log(FWC/RN)

Values are usually in electrons (e-). To convert dynamic range in decibels to stops, you simply divide by six. I have found that formula to be very simple and convenient, and it is easily comparable with manufacturers numbers, since they use the same formula.
 
Upvote 0
Lee Jay said:
Chuck Alaimo said:
Lee Jay said:
jrista said:
Actually, the use of Dual DIGIC 6 makes me really think that there is NO more magic cooked into the sensor, and that all the "magic" is happening after the signal is pulled OFF the sensor. It's probably roughly the same sensor that employs a slightly new DPAF design, and probably has a very weak or no AA filter, but is otherwise unchanged from the 70D. DIGIC 6, which actually came out before Sony BionzX, actually has a LOT of the same capabilities, and is the primary reason the IQ on their smaller form factor cameras is good. They really cook the signal coming off the sensor.

Since DIGIC 6 supposedly allows frame rates to 14fps and 60fps video, why would they need two of them? Dual pixel works okay on the 70D with just one DIGIC 5+.

Not positive but doesn't the 1dx have dual digic and one is to process images the other for for AF? Could that be what is going on for the 7d2?

Could be, but those are DIGIC 5+ chips.

The 1D X has three DIGIC chips. It has Dual DIGIC5+ that drive the sensor, 8 channels per DIGIC5+, for a total of 16 readout/ADC channels. The third DIGIC is a DIGIC 4, which is exclusively used by the AF and metering system.
 
Upvote 0
dtaylor said:
RGF said:
Details about the sensor are lacking. Great marketing buzz terms but does the sensor have significantly better DR (vs 70D and 5DM3).

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

Canon is still stuck at 11, as they use off-sensor A/D conversion (unlike the Exmors).
The hope was that the 7DII would be the first camera with on-sensor ADC - but this seems unlikely at this time.

Well, all we have is a rumor at this time.
Still, the fact that a 20mp sensor is rumored (as on the 70D) doesn't bode well.
Most likely the same sensor will be used. Likely with some tweaks, of course, but nothing of significance.
 
Upvote 0