Are These The EOS 7D Mark II Specifications?

[jeffa4444]

The points Jrista has made about AA filters and oversampling are bang on the money. For a sensor at super35 which in size is between an APS-C & APS-H sensor to be able to fill a cinema screen with sharp images is no mean feat. The current cameras mainly used are the Arri Alexa and the Red Epic / Dragon as well as a fair few still on film. The point is digital cinema cameras have AA filters and they are part of the overall optical design. New cameras in development also retain them. Yes they rob a tiny amount of resolution but so many factors affect that anyway. The cinema lenses are held to tolerances far in excess of stills lenses even great ones like the Zeiss Otus never more so with zooms.

Oversampling is the only way forwards the only downside is file size everything else is a gain and oversampling means Raw files will not nessisarily be the only option with wider gamut Rec2020 RGB will make a comeback for video which is an open format and not subject to manufacturer raw conversion and easily conforms to MXF or DPX.

The subject of pixel size, resolution, colorimagery, dynamic range & nyquist can get very complicated and for designers the criteria is always about the compromise. Systems designed to match like the Olympus / Panasonic 4/3rds (it never had to use legacy lenses) will always be broken by creatives and that is far better for diversity. The holly grail is the widest DR and the highest resolution both can be limited but not expanded in a set system.

 

[jrista]

The 7D was the softest Canon DSLR I've owned (and I've owned a bunch). So any move away from that buttery, waxy look is progress.

You weren't using the right ACR settings...or were using DPP. (Seriously.)

That said, I agree with you on RAW sharpness and taking more post processing. I wouldn't miss an AA filter.

I use Lightroom. Trust me when I say I had several 7D's, and all of them performed this way regardless of settings. My 70D is much, much better.

I'm not sure what the exact cause was. I believe it was a result of several factors, perhaps a "sot focus" combined with a buttery AA filter. My keeper rate, with several 7D's was considerably less than any other Canon DSLR I have owned, and across the same set of lenses. I shoot mostly nature, and the 7D just hated organic colors.

Let me ask, what do you think of this IQ:

Is that waxy and horrible and hateful? Because that's a 7D. So is this:

And this:

And all of these:

In practice...I've never really experienced the supposed "waxy, extremely softening effect of the 7D's AA filter." Only at particularly high ISO settings, where the 7D does not perform well for numerous reasons, have I ever experienced a loss in color fidelity and sharpness that I really did not like.

 

[preppyak]

65 AF points “All Cross-type”. Dual cross on the center point.

I like the 19 point system over the 45 point system. Easier to actually identify a tracking point AND it makes you work. I like cameras that reward you putting some work in.

Can't tell if this is serious or not. Both because the 19pts are contained within the 65pts (so you can opt not to use the extra 46), and because by that logic, you should really just want one focus point. Or a film camera...

Just seems odd to want all the video specs that don't really belong on a traditional DSLR (but would be great to have), yet scoff at the one reasonable spec

 

[ScottyP]

As for photographers, if you think the removal of an AA filter is better than oversampling, then yes, I absolutely DO BLAME YOU for forcing a ludicrous trend on camera manufacturers.

So rather than removing obstacles to improve detail and simplifying, you want to convolute the process?

Your an idiot if you think a lower resolution sensor without an AA filter is ever, even remotely, going to be better than downsampling an oversampled image that doesn't NEED an AA filter (because by oversampling, you ARE anti-aliasing!)

As for sharpening soft images...are you refuting the claim that you can restore detail by sharpening? Seriously?! I've proven this case so many times before, do I really, truly, need to prove it again?

In my response, I claimed that soft images can be sharpened. But the problem with soft images is they are much less malleable than sharp, clean images out of camera.

The better the sensor, the less post processing you have to do.

I'm not sure there is any actual evidence for that. And again, I'd point you to all the artifacts that occur with sensors that lack an AA filter entirely. You could spend DAYS trying to correct moire or extensive aliasing in an image, and still never get rid of it. As for a sensor with an AA filter...run it through a light sharpening filter and your done. Maybe that's 5 seconds of additional processing...ooh, that's just so much time. In the grand scheme of things, I'd say that you still have to spend time sharpening an image without an AA filter...you just use less sharpening. So there really isn't any major difference in processing time period.

Now, regarding oversampling. You seem to be misunderstanding that. A sensor that oversamples lenses, at their best resolution, say f/2.8 as a round-about high quality aperture for lenses the likes of the Otus. You still wouldn't have an AA filter. However, you wouldn't NEED an AA filter, because your anti-aliasing by oversampling. You do understands what that means, right? A sensor that is capable of oversampling is going to be of MUCH higher resolution than any sensor that isn't oversampling and lacks and AA filter.

So...where, exactly, is your lower resolution AA-less sensor actually getting higher IQ than a high resolution oversampled sensor? The higher resolution sensor, even it it may look "soft" at 100% pixel peeping, is STILL resolving FAR more detail than the lower resolution sensor that lacks an AA filter. You want a sharper image? Well, if your 2x oversampled, downsample by a factor of two (reduce it to 1/4 area). If your 3x oversampled, downsample by a factor of three (reduce it to 1/9th area.) The oversampled image will be sharper, out of camera, without any sharpening or noise reduction, than the lower resolution image that did not have an AA filter.

When it comes down to sensors at today's resolutions, I'll take the one with an AA filter over one without an AA filter any day. It might take me an extra five seconds to dial in a slightly stronger amount of sharpening than one without an AA filter, but at least I won't have to spend an extra day trying to get rid of aliasing and moire.

What if they just did a non-Bayer pixel arrangement, like Fuji? Isn't that supposed to let you ditch the AA filter?

 

[jrista]

What if they just did a non-Bayer pixel arrangement, like Fuji? Isn't that supposed to let you ditch the AA filter?

LOL. We had a very extensive debate about that not too long ago. The 6x6 pixel interpolation of the Fuji results in soft detail, and it obliterates a lot of fine detail. Technically speaking, if you oversampled enough, you could indeed interpolate more pixels together...but you would really have to be significantly oversampling. The problem with Fuji's sensor is it does not oversample the lenses at all in most cases, or enough in the best cases (those cases being when you are either at a very narrow aperture, or just have a crappy lens.)

Oversampling is a pretty specific use case. No one has really done that these days...not at any wider aperture...you would have to be at f/22 at least on most lenses these days before you really just barely begin to oversample with any existing APS-C sensors...and you would need to be around f/32 or so before FF sensors begin to oversample.

 

[ajfotofilmagem]

These specs sound like an update of 7D very conservative, with the exception of autofocus. ??? The big question will be: The picture quality is the same for 70D, or have visible improvements? :-\

 

[pablo]

65 AF points “All Cross-type”. Dual cross on the center point.

I like the 19 point system over the 45 point system. Easier to actually identify a tracking point AND it makes you work. I like cameras that reward you putting some work in.

Can't tell if this is serious or not. Both because the 19pts are contained within the 65pts (so you can opt not to use the extra 46), and because by that logic, you should really just want one focus point. Or a film camera...

Just seems odd to want all the video specs that don't really belong on a traditional DSLR (but would be great to have), yet scoff at the one reasonable spec

1. Yes entirely serious.

2. No. 19 larger points filling the centre third of the frame is not the same thing as 45 or 65 pints filling the same area with some points ignored. In you model you would have larger gaps between the af points where things like say, adjacent point tracking weren't happening. One of the great things with the EOS 7D was the ability to select one AF point, but for the camera also to have cognisnece of the adjacent 4. Just in case the subject was faster, or the camera smarter than the user.

3. "By that logic.." you lost me. That is the vocabulary of the flamer. "I don't understand what you said, so I'll apply my misunderstood logic to try and make you look ridiculous." It's a false dichotomy.

By my logic, I only want one AF point at any one time. Sometimes that is the centre point. For say a steam train pummelling right down the lens at me. Sometimes it is on an upper left third cross section or upper right third cross section, for say portraiture. Sometimes it is on a lower third, when I'm composing a nice skyscape at sunset against a still firth of clyde.

I won't attempt to understand your logic. I use manual AF point selection a lot of the time. I find 19 enough.
Shock horror, I sometime use MF. But I never ever ever use auto AF point select.

So in answer to your poser. Yeah, sometimes I only want one AF point. (Quite how you make the uber patronising leap to film use I don't comprehend) But I would like a choice of AF points to use.

Sometimes when I'm tracking a moving subject, I find 45 tiny af points just too many. I got on great with the EOS 7D system is all. I'm sorry that provokes so much ire.

Maybe you should use a compact in green square mode if you want to surrender all control to your camera?

Interested in how you made the leap that I should use film. Please explain.

 

[dtaylor]

Sure but a 20MP vs a 24MP is 10% difference on X and Y axis approximately. Are Canon that much behind the competition?

No they are not. Not too long ago I spent an unreasonable amount of time comparing files from the Sony A7 (24 MP AA filter) to the A7R (36 MP no AA filter). I always resized the 24 MP file to 36 MP with very light sharpening (scaling tends to soften) so that size (and therefore magnification) differences didn't affect my judgement.

You're looking at a nearly 25% axis gain in this comparison vs. 10% for the APS-C sensors.

My conclusion? Occasionally an area of very, very fine detail would be rendered better on the 36 MP sensor. This would be visible while pixel peeping but never in print.

I came to a similar conclusion when I mixed in 5D3 files, though the 5D3 file must be RAW. The JPEGs fell behind. Also, when converting the 5D3 file you have to be a bit more aggressive with the detail and sharpening settings. The Sony sensors would handle heavy processing a bit better because of this.

At the resolutions we're dealing with today I would say you need a 50% gain on each axis before differences become visible in print, all other differences (i.e. sensor size) being equal.

 

[MichaelHodges]

Let me ask, what do you think of this IQ:

In practice...I've never really experienced the supposed "waxy, extremely softening effect of the 7D's AA filter." Only at particularly high ISO settings, where the 7D does not perform well for numerous reasons, have I ever experienced a loss in color fidelity and sharpness that I really did not like.

Those are such tiny jpegs, JR. Would you mind posting the RAW file to determine out-of-camera IQ?

Also, it looks like flash may have been used. These would present completely different conditions.

The 7D is certainly capable of getting nice shots when you fill the frame and light is perfect. My problem was the lower keeper rate when compared to the numerous other Canon DSLR's I've used and/or owned.

 

[jrista]

65 AF points “All Cross-type”. Dual cross on the center point.

I like the 19 point system over the 45 point system. Easier to actually identify a tracking point AND it makes you work. I like cameras that reward you putting some work in.

Can't tell if this is serious or not. Both because the 19pts are contained within the 65pts (so you can opt not to use the extra 46), and because by that logic, you should really just want one focus point. Or a film camera...

Just seems odd to want all the video specs that don't really belong on a traditional DSLR (but would be great to have), yet scoff at the one reasonable spec

1. Yes entirely serious.

2. No. 19 larger points filling the centre third of the frame is not the same thing as 45 or 65 pints filling the same area with some points ignored. In you model you would have larger gaps between the af points where things like say, adjacent point tracking weren't happening. One of the great things with the EOS 7D was the ability to select one AF point, but for the camera also to have cognisnece of the adjacent 4. Just in case the subject was faster, or the camera smarter than the user.

3. "By that logic.." you lost me. That is the vocabulary of the flamer. "I don't understand what you said, so I'll apply my misunderstood logic to try and make you look ridiculous." It's a false dichotomy.

By my logic, I only want one AF point at any one time. Sometimes that is the centre point. For say a steam train pummelling right down the lens at me. Sometimes it is on an upper left third cross section or upper right third cross section, for say portraiture. Sometimes it is on a lower third, when I'm composing a nice skyscape at sunset against a still firth of clyde.

I won't attempt to understand your logic. I use manual AF point selection a lot of the time. I find 19 enough.
Shock horror, I sometime use MF. But I never ever ever use auto AF point select.

So in answer to your poser. Yeah, sometimes I only want one AF point. (Quite how you make the uber patronising leap to film use I don't comprehend) But I would like a choice of AF points to use.

Sometimes when I'm tracking a moving subject, I find 45 tiny af points just too many. I got on great with the EOS 7D system is all. I'm sorry that provokes so much ire.

Maybe you should use a compact in green square mode if you want to surrender all control to your camera?

Interested in how you made the leap that I should use film. Please explain.

It's actually been clearly demonstrated on many occasions, with AF systems from multiple manufacturers now, that a high density reticulated AF system is far better for tracking than a system with widely distributed AF points like the 7D. Canon, Nikon, Sony, and a few other manufacturers all have high density reticulated (net-like) AF systems now. Canon's 61pt system in particular, with so many cross-type points, has proven to be remarkably effective at locking onto and tracking fast and erratically moving subjects.

There is nothing wrong with having more points. The new AF systems support the same thing the 7D did, with allowing nearby points to "assist". It's called AF Expansion mode, where you select one, then allow the surrounding four or eight points to assist. You can also use zone mode, which unlike all points mode, allows you to chose on of a couple sizes of restricted, selectable zones for tracking.

Owning both a 7D and 5D III myself, there is absolutely no question which AF system is superior...the 61pt reticulated system kicks the crap out of the 19pt system. You can still use a single point if you wish, and in the case of focusing on slow moving subject such as a perched or standing bird or posing wildlife, that's exactly what you want. When it comes to subjects in motion...nothing beats having more AF points in higher density, especially if they are ALL cross-type.

 

[LetTheRightLensIn]

a foveon sensor does not need an AA filter because it does not use a bayer pattern.

Of course a foveon sensor needs an AA filter (unless it's like say maybe 60MP+ APS-C or something).

 

[dtaylor]

I use Lightroom. Trust me when I say I had several 7D's, and all of them performed this way regardless of settings. My 70D is much, much better.

So the fact that I don't see this means I have a special 7D? :

If it wasn't a settings issue then probably a focus issue.

 

[dtaylor]

a foveon sensor does not need an AA filter because it does not use a bayer pattern.

Of course a foveon sensor needs an AA filter (unless it's like say maybe 60MP+ APS-C or something).

Wikipedia disagrees ;D

I was under the impression that the Foveon sensors shipped without AA filters, though I wouldn't swear to it.

 

[LetTheRightLensIn]

In the previous chain about the 7DX specs better DR, IQ, and less noise are the overwhelming drivers for interest here...But I'm not convinced Canon could make such a miscalculation.

[quote author=unfocused]
No miscalculation. What people on gearhead forums want and what the bulk of customers who will actually buy the product want are two entirely different things.

Like I said, I do NOT think they would miscalculate. The 7D line isn't meant for feature-only freak low end consumers. It was always geared at upper tier users. My point is that I think the end result will be (hopefully) something notably improved.

Okay, let's try this again.

Don't equate what people on this forum think is important with what the target market thinks is important. The 7D is targeted to upper tier users. Most upper tier users are not worried about the things that people on this forum obsess over. These specs would indicate that Canon did their market research and found that features like a top-notch autofocus system are more important to buyers than small improvements in dynamic range.
[/quote]

They probably found that their better lenses and the money people have invested help people around and that since improving AF costs them wayyyyy less than improving sensors they may as well just save the money.

It's a bit surprising they'd not bother with 4k though, but then again, with they way they've been acting, I guess not.

 

[jrista]

Let me ask, what do you think of this IQ:

In practice...I've never really experienced the supposed "waxy, extremely softening effect of the 7D's AA filter." Only at particularly high ISO settings, where the 7D does not perform well for numerous reasons, have I ever experienced a loss in color fidelity and sharpness that I really did not like.

Those are such tiny jpegs, JR. Would you mind posting the RAW file to determine out-of-camera IQ?

Also, it looks like flash may have been used. These would present completely different conditions.

The 7D is certainly capable of getting nice shots when you fill the frame and light is perfect. My problem was the lower keeper rate when compared to the numerous other Canon DSLR's I've used and/or owned.

I never use flash. I own a flash, but the last time I pulled it out of my closet was probably two years ago. And you cannot make the argument that finding and using good light is a bad thing, or unfair. I find and use good light REGARDLESS of which camera I'm using, regardless of how big it's sensor frame is. That's a part of being a good bird and wildlife photographer...to chase good light. This is my photography. This is how my photography looks. And all of those photos were taken with the 7D. I believe the examples speak for themselves...an AA filter is not a bad or highly detrimental thing that must be done away with. It has never negatively affected my work.

But your backtracking now. Your switching the problem your complaining about. Now it's the lower keeper rate, instead of the supposed waxy appearance. I will happily agree that the 7D had AF problems. It had inconsistent lock-on rates, and had an inherent jitter. But that is an entirely different argument, and has nothing to do with the supposed "waxy" appearance that the 7D, based on your otherwise unqualified statements, simply has...period. (It only sometimes gets waxy at really high ISO...however I think LTRLI has a better explanation for it...split green CFA.)

I don't want to share RAWs for some of my better works, these are photos I'm working on selling these days, however my entire point is that out of camera IQ isn't the end-all-be-all here. I've been explicitly making the argument that regardless of what impact an AA filter might have, it doesn't matter in the end. Apply a little sharpening...and bam, any softness attributed to the AA filter is simply gone.

 

[MichaelHodges]

I use Lightroom. Trust me when I say I had several 7D's, and all of them performed this way regardless of settings. My 70D is much, much better.

So the fact that I don't see this means I have a special 7D? :

If it wasn't a settings issue then probably a focus issue.

With over 100,000 actuations in the wilderness of the Northern Rockies, the 7D's just killed me. Just too many missed shots, and way too much post processing for the rest. How valuable is your time? The 70D is a winner in comparison.

Hopefully the 7D II is, too.

 

[jrista]

a foveon sensor does not need an AA filter because it does not use a bayer pattern.

Of course a foveon sensor needs an AA filter (unless it's like say maybe 60MP+ APS-C or something).

Wikipedia disagrees ;D

I was under the impression that the Foveon sensors shipped without AA filters, though I wouldn't swear to it.

Foveon sensors do ship without AA filters. They also do not suffer from color moire. However, they still NEED AA filters, because they do suffer from moire in monochrome.

 

[LetTheRightLensIn]

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.

??? Where are you getting these numbers????

 

[pablo]

There is nothing wrong with having more points.

For me, there is. It makes the viewfinder too fussy for my tastes.

For me.

For me.

For me.

If you find different great. You are not absolutely correct.

Broad church and all that.

If the 7D 2 has a dedicated AF processor like the 7D, then swell. If the VF overlays can mimick the 19 point array then great. We are all happy happy bunnies.

Neither you or I are correct or wrong. This is why I prefer photography forums to gear forums.

 

[thepancakeman]

This inspires me...to get a 5D3 (if the specs here are true.) :-\