Canon EOS 7D Mark II Specifications Confirmed

[PureClassA]

Playing 6 page catch up now, but folks on here saying the 7D2 "has image quality already lagging behind he competition" has got to be joking me..... None of us has seen the actual camera yet let alone a SINGLE frame from it. Good Lord!! Get a grip. Why don't we all wait to see what it really is and what it can do before we bring out the cross and nails!! And btw, I have a T2i... And the 7D. Same camera? Uhh No......

 

[LetTheRightLensIn]

It doesn't seem to be the codec that is doing the damage though.

The 5D3 uncompressed HDMI out looks almost exactly the same as in cam compressed footage.

It's got to either be that DIGIC is terrible at de-Bayer and fine processing to retain detail and too tuned for a wax-works, smooth, no noise look or that Canon has decided to use parameters that apply too much wax works although DIGIC could do better if fed different settings, or that marketing has them apply some blur function to make sure the quality doesn't threaten their high-end stuff.

I agree, Canon's codecs could use some major overhaul. I shot a video without ML on my mk III and it pissed me off in many ways. However, it's understandable they would try to differentiate between their DSLR's and their cinema camera lineup. Enough pressure from Nikon, however, and Canon will respond with better native codecs at that time through a firmware update. Sony doesn't have count entirely as a fully competing system because they either don't have the same lens assortment, esp. supertelephotos, or you can use their FS100/700 series cameras with adapters, so you'd end up using the same Canon or Nikon lenses.

That is not true at low ISO when shooting scenes with a lot of DR. In those cases Canon is like 2-3 stops behind and that amounts to a heck of a lot more than a dime.

Eagerly awaiting your real world sample photos which illustrate this ;D

So am I. I hope the prove is not DxO scores though. I challenge anyone to find me an example where the 70D's sensor is outperformed by any APS-C sensor from any manufacturer by 1 full stop in terms of anything, lowlight performance, dynamic range at any ISO, and resolution. Just one stop, not 2 or 3 as you people claim.


____________________________________

Regarding video: Do we know anything about the 7D's video to the point of determining it's a bad video camera and crippled and not suitable as a hybrid?

No. We don't know squad about the 7D's video. Nothing. Just that it does 60p. that's it.

The factors that will determine whether this is an awesome video camera or nor are all unknown:
What exactly is the sensor?
How does the camera downscale the full resolution to 1080p? Line skipping or pixel binning?
How does the processor compress the feed to H.264, what's the codec, bit rate, 90mbits ALL I or higher, or less?
How fast does it read the sensor lines for the rolling shutter effect?
Does it have a clean HDMI output? Or not?
Does it have a headphone jack, good preamps? Are the audio meters controlled whilst recording or just prior to?
Is reading the sensor at 60p going to introduce worse IQ than normal speed, or will it be just as good for slowmotion?
Are the new processor affecting the noise performance by two stops as Canon claims? Are they affecting the downsampling process therefore the detail?
How well does the autofocus work in video mode?

Come on this is just getting ridiculous, you're concluding somehow that it's bad and crippled for video based on having 60p? That's all you've got about the 7D mk II video mode!

and concerning 4K, just because it's not 4K doesn't mean it's a bad video camera. This is ridiculous. 4K is the next step forward but there still is a place for a great 1080p camera in the market for the next few years.

What is the most successful camera in the video world? The Canon C300. It dominates the broadcast/news/rournalism industry, and the documentary world. Is it a 4K camera? No. 1080p. It does perfect 1080p. And the highest-end professionals are fine with that, but not consumers on the forums, for some reason.
Another example, the Arri Alexa? The most successful cinrma camera in the industry. All the films you see in cinema are shot on the Alexa, is it 4K? No. 2K. The Arri Amira, which is a 60 thousand dollars professional camera that's announced this year, is it 4K? No, 1080p. Yet it's creating the biggest interest among video professionals, but not us. 1080p is too little for us.

All things being equal, do I want 4K vs. 1080p? I do. It's better.
But all things are rarely equal, and many times perfectly done 1080p is way better than poorly done 4K. A GH4 at 4K is not a better video image than a 1080p C300. Not by any means. There are MANY different aspects to image quality than just resolution. There is dynamic range, sensor size, noise charachter and pattern, high ISO performance, colour depth, colour science, etc.

There are certain things the 7D need to nail to be a great video camera, but 4K is not one of them. There still IS a place for a great 1080p camera other wise they wouldn't be selling 1080p cameras at 10K$ to 60K$!
Just give me a perfect 1080p image on the 7D mk II and I am VERY happy. The sensor size in the 7D is more suitable for video/cinema applications than the 5DIII, APS-C equals super 35mm which is the absolute professional standard in the video world. I want the 7D to be a perfect s35 video camera upgrade for all Canon video shooters. It really makes sense. But we don't know a thing. Nothing.

Yeah, but so far Canon DSLR give poor 1080p fo ranything less than 1 series (they all had bad aliasing other than the 5D3 and they are all waxy, poor detail, poor DR).

The only one that gives great 1080p is the 5D3 but only if using ML RAW (it is great in that case, which also means that there is no way the 7D2 can deliver better video unless they went 4k with it and then you'd hope).

Also if they gave 7D2 4k it means they'd be a bit freer to not cripple whatever 4k they put in the 5D4.

If it delivers C300 1080p detail and has 10bits then maybe the 7D2 does decently well and that is easier to deal with than 5D3 RAW. But the way they are already talking about the advanced NR in the Digic6 makes me think it will be more wax works (and 8bits only of course), we'll see.

 

[joejohnbear]

Yeah, that troll with the t2i deleted his comment in response to me before I could post my response and tear him apart. I'm considering reposting his comment and my response to it, but I won't do so unless he starts it up again.

Playing 6 page catch up now, but folks on here saying the 7D2 "has image quality already lagging behind he competition" has got to be joking me..... None of us has seen the actual camera yet let along a SINGLE frame from it. Good Lord!! Get a grip. Why don't we all wait to see what it really is and what it can do before we bring out the cross and nails!! And btw, I have a T2i... And the 7D. Same camera? Uhh No......

 

[PureClassA]

Yeah, that troll with the t2i deleted his comment in response to me before I could post my response and tear him apart. I'm considering reposting his comment and my response to it, but I won't do so unless he starts it up again.

Playing 6 page catch up now, but folks on here saying the 7D2 "has image quality already lagging behind he competition" has got to be joking me..... None of us has seen the actual camera yet let along a SINGLE frame from it. Good Lord!! Get a grip. Why don't we all wait to see what it really is and what it can do before we bring out the cross and nails!! And btw, I have a T2i... And the 7D. Same camera? Uhh No......

Ahh. Well for the record I loved my T2i!! A dear friend now has it (mine) and loves it. I bought my 7D about a year after is bought the 2i. They are both great but very different rigs.

 

[dtaylor]

DR is driven in part by pixel size,...

If that were true, the G15 wouldn't have more base ISO DR than the 1DX:

http://sensorgen.info/CanonPowershot_G15.html
http://sensorgen.info/CanonEOS-1D_X.html

Oh look...regurgitated DxO used dog food.

If I look it up, will the G15 have a higher score then a Hasselblad? :

I get more DR from an EOS M on a step wedge then they claim for the 1DX. Hmmm...hey Neuro, if you're still reading this, want to trade my M for your 1DX? You want maximum DR don't you? ;D

 

[joejohnbear]

Yeah, I didn't deconstruct what was doing the damage, but I'm sure it can be solved with a firmware update IF someone came out with a product that could compete at the same price point. Even the D800 requires an external capture device (i.e. that Ninja thing) that costs an additional $1-3k, I can't remember the price. This becomes an especially niche video market. Most pro studios are spending $12k and up for raw video anyways, and Canon's not going to sacrifice this market so that an indie filmmaking crew whose producer couldn't gather enough funding could buy their top of the line features. Right now, I think there are a few third party options at the same price point, but not with the same lens system and professional service. Working pro's and studios aren't going to gamble on a blackmagic, regardless of how well it works, because the return times required when the device needs to be serviced. Obviously I'd love if Canon gave me clean video with ML, but for now I don't do enough video work as my main source of income to complain, and I always ML to back me up if needed. I understand that for the price point ($3500), I'm getting less features and quality than a full cinema camera. If Canon releases anything better and sends it my way, I'm more than happy to accept it, but for now, I'm not clamoring for pitchforks and fire and I carry on with what I have.

It doesn't seem to be the codec that is doing the damage though.

The 5D3 uncompressed HDMI out looks almost exactly the same as in cam compressed footage.

It's got to either be that DIGIC is terrible at de-Bayer and fine processing to retain detail and too tuned for a wax-works, smooth, no noise look or that Canon has decided to use parameters that apply too much wax works although DIGIC could do better if fed different settings, or that marketing has them apply some blur function to make sure the quality doesn't threaten their high-end stuff.

I agree, Canon's codecs could use some major overhaul. I shot a video without ML on my mk III and it pissed me off in many ways. However, it's understandable they would try to differentiate between their DSLR's and their cinema camera lineup. Enough pressure from Nikon, however, and Canon will respond with better native codecs at that time through a firmware update. Sony doesn't have count entirely as a fully competing system because they either don't have the same lens assortment, esp. supertelephotos, or you can use their FS100/700 series cameras with adapters, so you'd end up using the same Canon or Nikon lenses.

That is not true at low ISO when shooting scenes with a lot of DR. In those cases Canon is like 2-3 stops behind and that amounts to a heck of a lot more than a dime.

Eagerly awaiting your real world sample photos which illustrate this ;D

So am I. I hope the prove is not DxO scores though. I challenge anyone to find me an example where the 70D's sensor is outperformed by any APS-C sensor from any manufacturer by 1 full stop in terms of anything, lowlight performance, dynamic range at any ISO, and resolution. Just one stop, not 2 or 3 as you people claim.


____________________________________

Regarding video: Do we know anything about the 7D's video to the point of determining it's a bad video camera and crippled and not suitable as a hybrid?

No. We don't know squad about the 7D's video. Nothing. Just that it does 60p. that's it.

The factors that will determine whether this is an awesome video camera or nor are all unknown:
What exactly is the sensor?
How does the camera downscale the full resolution to 1080p? Line skipping or pixel binning?
How does the processor compress the feed to H.264, what's the codec, bit rate, 90mbits ALL I or higher, or less?
How fast does it read the sensor lines for the rolling shutter effect?
Does it have a clean HDMI output? Or not?
Does it have a headphone jack, good preamps? Are the audio meters controlled whilst recording or just prior to?
Is reading the sensor at 60p going to introduce worse IQ than normal speed, or will it be just as good for slowmotion?
Are the new processor affecting the noise performance by two stops as Canon claims? Are they affecting the downsampling process therefore the detail?
How well does the autofocus work in video mode?

Come on this is just getting ridiculous, you're concluding somehow that it's bad and crippled for video based on having 60p? That's all you've got about the 7D mk II video mode!

and concerning 4K, just because it's not 4K doesn't mean it's a bad video camera. This is ridiculous. 4K is the next step forward but there still is a place for a great 1080p camera in the market for the next few years.

What is the most successful camera in the video world? The Canon C300. It dominates the broadcast/news/rournalism industry, and the documentary world. Is it a 4K camera? No. 1080p. It does perfect 1080p. And the highest-end professionals are fine with that, but not consumers on the forums, for some reason.
Another example, the Arri Alexa? The most successful cinrma camera in the industry. All the films you see in cinema are shot on the Alexa, is it 4K? No. 2K. The Arri Amira, which is a 60 thousand dollars professional camera that's announced this year, is it 4K? No, 1080p. Yet it's creating the biggest interest among video professionals, but not us. 1080p is too little for us.

All things being equal, do I want 4K vs. 1080p? I do. It's better.
But all things are rarely equal, and many times perfectly done 1080p is way better than poorly done 4K. A GH4 at 4K is not a better video image than a 1080p C300. Not by any means. There are MANY different aspects to image quality than just resolution. There is dynamic range, sensor size, noise charachter and pattern, high ISO performance, colour depth, colour science, etc.

There are certain things the 7D need to nail to be a great video camera, but 4K is not one of them. There still IS a place for a great 1080p camera other wise they wouldn't be selling 1080p cameras at 10K$ to 60K$!
Just give me a perfect 1080p image on the 7D mk II and I am VERY happy. The sensor size in the 7D is more suitable for video/cinema applications than the 5DIII, APS-C equals super 35mm which is the absolute professional standard in the video world. I want the 7D to be a perfect s35 video camera upgrade for all Canon video shooters. It really makes sense. But we don't know a thing. Nothing.

Yeah, but so far Canon DSLR give poor 1080p fo ranything less than 1 series (they all had bad aliasing other than the 5D3 and they are all waxy, poor detail, poor DR).

The only one that gives great 1080p is the 5D3 but only if using ML RAW (it is great in that case, which also means that there is no way the 7D2 can deliver better video unless they went 4k with it and then you'd hope).

Also if they gave 7D2 4k it means they'd be a bit freer to not cripple whatever 4k they put in the 5D4.

If it delivers C300 1080p detail and has 10bits then maybe the 7D2 does decently well and that is easier to deal with than 5D3 RAW. But the way they are already talking about the advanced NR in the Digic6 makes me think it will be more wax works (and 8bits only of course), we'll see.

 

[dtaylor]

sure many shots didn't need to DR, but plenty enough along the way could've been helped for sure

Waiting for real world examples of what we are all missing ;D

 

[dtaylor]

I don't know what diet coke boxes you are going on about.

In one thread someone severely underexposed and then pushed a photo of a Coke box. That was a "real world example."

Still waiting for yours....

 

[dtaylor]

Photo up or shut up.

SOme of us even have put up the exact same shot at the same time and you still made excuses and said it didn't count. So bye.

Really? Because I don't remember that. You have links to these photos? Or is this just another baseless generic claim?

 

[PureClassA]

DR is driven in part by pixel size,...

If that were true, the G15 wouldn't have more base ISO DR than the 1DX:

http://sensorgen.info/CanonPowershot_G15.html
http://sensorgen.info/CanonEOS-1D_X.html

Oh look...regurgitated DxO used dog food.

If I look it up, will the G15 have a higher score then a Hasselblad? :

I get more DR from an EOS M on a step wedge then they claim for the 1DX. Hmmm...hey Neuro, if you're still reading this, want to trade my M for your 1DX? You want maximum DR don't you? ;D

holy shit! Did I just buy a 1DX for $250!?!?! Hahaha

 

[jrista]

Hell, let's just head this off:

http://1drv.ms/1pUWUGT

Dean's RAWs. Have at it.

 

[dtaylor]

And when Fred posted same shots, same time, same cam and showed a big difference then you just had more excuses for why the differnce was not really there.

Fred who? Miranda? When you use NR on the Canon RAW the difference is inconsequential. While pixel peeping the hard pushed sample the Nikon file had a little more detail and less noise. But the actual DR range was the same, and you would be hard pressed to notice detail/noise differences even in a large print. Especially since the scene was artificial as you would never push the shadows that hard in that example.

Your side by side redwood examples that are evidence for your conclusion of a 2-3 stop shortfall, please. Because absent those you are wasting everyone's time.

Yeah artificial. Right.

In that case, yes. Pushing shadows that hard...with no NR...might be fun in order to see differences in the shadows, but the entire scene becomes a blown out mess. That's not how you would actually prepare an image for print.

So the times people actually do bother to take the same shot the same way are artificial

No. The times people do things they WOULD NEVER DO WITH A REAL WORLD PHOTOGRAPH like turn off all NR, push a properly exposed photo 5 stops, or severely underexpose a photo then push it 5 stops with no NR, are artificial because YOU WOULD NEVER DO THAT WITH A REAL WORLD PHOTOGRAPH.

So where is the real world photograph comparison? Where is the optimum exposure and processing to produce the best possible artistic print from each sensor? I would like to see that. I would like to see if Canon is "2-3 stops behind" in that case. (It wasn't in the FM case. Roughly the same DR, maybe 1 stop less shadow latitude.)

You have that test ready yet?

And sure, apply NR and this and that to the Canon file and THEN compare it to the Exmor, that is really fare (plus it still looks like junk in comparison anyway and lots of details are not there, even if the smoothness is much closer at that point).

Best possible print is fair because that's what real photographers do when producing real photographs for real clients as opposed to sitting in a room measurebating. And the difference in the FM test was not nearly as dramatic as any of you make the differences to be. Yes, the Exmor sensor had better shadow latitude. No doubt. Just not enough to really matter.

Now it might make a noticeable difference in some scenes. If I wanted to play devil's advocate I bet I could illustrate in a real world scene where it would be noticeable. But I could still get the shot with the Canon with a little work. But the DRoners...the DRoners seem incapable of actually showing where it matters at all.

 

[dtaylor]

Um, he did not underexpose.

I would call it underexposed.

To be very, very clear here...there was ZERO PUSHING OF ANY KIND in those Coke box images, they were strait out of camera

Oh really? Here's the link to his post:
http://www.canonrumors.com/forum/index.php?topic=22422.msg431159#msg431159

Here are the labels from his posted images (which no longer appear). Emphasis mine:

Mk3 pushed 3 stops in ACR
D800 pushed 3 stops in ACR
Mk3 pushed 3 stops in ACR 100% crop
D800 pushed 3 stops in ACR 100 crop

Just to be clear

 

[jrista]

Um, he did not underexpose.

I would call it underexposed.

To be very, very clear here...there was ZERO PUSHING OF ANY KIND in those Coke box images, they were strait out of camera

Oh really? Here's the link to his post:
http://www.canonrumors.com/forum/index.php?topic=22422.msg431159#msg431159

Here are the labels from his posted images (which no longer appear). Emphasis mine:

Mk3 pushed 3 stops in ACR
D800 pushed 3 stops in ACR
Mk3 pushed 3 stops in ACR 100% crop
D800 pushed 3 stops in ACR 100 crop

Just to be clear

This is what I was referring to:

To offset the 'why do you have to push exposure 3 stops, only bad photographers get exposure wrong' posts, here is the red channel of both, NOT pushed, just at 0 adjustment.

http://www.canonrumors.com/forum/index.php?topic=22422.msg431160#msg431160

I linked his RAWs on my onedrive. I encourage everyone to download and take a gander themselves. The banding in the Canon file is obvious. Without any adjustments.

 

[NancyP]

OK, all you sensor geeks, here's your purgatory assignment. Shoot with any Sigma Foveon sensor camera. These are odd beasts. I love my Sigma DP Merrills but they are operationally odd, and the software is s-l-o-w and buggy. The cameras are very good for rendering natural landscapes and foliage, with excellent color subtlety.

Why am I going on about Brand X? Sigma DP Merrill cameras are (big) pocket-sized landscape low ISO cameras with limited DR and wonderful resolution and color - specialty beasts. The as-yet-not-fully-specified 7D2 is an action camera, also a specialty beast. I am intrigued with the Sony A7r/A7s concepts, and these two cameras are different types of specialty cameras. Were I to do professional product photography, I might opt for the A7r, provided that it played well with tilt-shift lenses. Low light work, especially low light video, I would go for the A7s. Lots of choices out there.

 

[Loswr]

...I think DR-commentators aren't giving Canon credit for improving vastly in a lot of other areas.

Dude, pull your head out. It's clear that you just don't get what a camera actually is. "I'll explain and I'll use small words so that you'll be sure to understand, you warthog faced buffoon."

A camera is a box. A worthless hunk of plastic and some metal, its sole purpose is to contain and protect the glory that is the imaging sensor, keeping the grubby fingerprints of plebes like you from despoiling it's pristine surface.

"There was a mighty duel. It ranged all over. The gestalt performance ran off alone, the sensor followed those color prints toward Exmor."

"Shall we track them both?"

"The loser is nothing. Only the DR matters."

;D



(You're new here, so you may not know that I enjoy quoting The Princess Bride and sarcasm...not necessarily in that order of preference.)

 

[PureClassA]

...I think DR-commentators aren't giving Canon credit for improving vastly in a lot of other areas.

Dude, pull your head out. It's clear that you just don't get what a camera actually is. "I'll explain and I'll use small words so that you'll be sure to understand, you warthog faced buffoon."

A camera is a box. A worthless hunk of plastic and some metal, its sole purpose is to contain and protect the glory that is the imaging sensor, keeping the grubby fingerprints of plebes like you from despoiling it's pristine surface.

"There was a mighty duel. It ranged all over. The gestalt performance ran off alone, the sensor followed those color prints toward Exmor."

"Shall we track them both?"

"The loser is nothing. Only the DR matters."

;D



(You're new here, so you may not know that I enjoy quoting The Princess Bride and sarcasm...not necessarily in that order of preference.)

"My name is Neuro. You insulted my camera. Prepare to die"

 

[dtaylor]

Hell, let's just head this off:

http://1drv.ms/1pUWUGT

Dean's RAWs. Have at it.

Time for you to revisit his post:
http://www.canonrumors.com/forum/index.php?topic=22422.msg431159#msg431159

And time for you to open the RAW files yourself and view the histograms. The values are literally compressed to the left third of the graph. You have to give a 3 stop push in order to even out the histogram. Which means he underexposed by 3 stops.

Now I find it impressive that Exmor can handle this situation with grace. But you would never do this. You would never underexpose a scene with a deep shadow area that has no detail, then push that shadow area hard. There's zero reason to do this in the real world. You want that Coke box with a smooth run off into the shadowed corner? ETTR and pull down. Silky smooth on any sensor.

See how these conversations go? See why we get so frustrated? An artificial 3 stop underexposure/push test turns into "Canon has terrible banding even in properly exposed photos with no pushed shadows!" It's like that game kids play where each kid whispers something to the next and you see how much the original sentence has changed by the end. You would think people could accurately repeat a sentence, but at the end you realize the final sentence is nothing like the original.

You saved the RAWs...but forgot the test!!! And are now telling people how it proved something it clearly did not. You even emphasized it...JUST TO BE CLEAR...so sure of yourself and your opinion.

This Exmor mythology has become ridiculous.

 

[x-vision]

But I could still get the shot with the Canon with a little work. But the DRoners...the DRoners seem incapable of actually showing where it matters at all.

And I can bet $1000 that any of your photos could have been taken with a phone camera or a P&S.
You certainly don't need a DSRL - and yet, you bought one.

And the reason you bought a DSLR is the very same reason why having more DR is better.

Yes, you can take award winning photos with inferior gear.
But that doesn't mean that we shouldn't want better gear.

We always want to have the best (we can afford) - even if we don't really need it.
Same for you with your Canon DSLR - and same for us, who want more DR.

So, your lame argument is just that - very, very lame.

Common now, show me any of your DSLR pictures that I cannot take with my iPhone.
If not, just stfu.

 

[dtaylor]

sure many shots didn't need to DR, but plenty enough along the way could've been helped for sure

Waiting for real world examples of what we are all missing ;D

Go shooting in mountains where you've got snow and sunshine hitting the snow and shadows deep in the valley. There you want to keep "detail in the snow" (so that you don't just have big white areas) plus you also want to keep shadows from and in trees, etc.

Is that real world enough for you or are you going to say "Post a picture or it doesn't exist"?

Since I've shot scenes like that and not had a problem: pictures or it didn't happen.

I'm sick of words. I'm sick of opinions. I'm sick of theorizing. I'm sick of people misremembering underexposure tests as "real world normal exposure and there was banding!"

Pics or it didn't happen. If I was a mod it would be: pics or you are banned for a week