The phase-detect AF sensor is separate from the imaging sensor, it uses a semi-transparent mirror to bounce light up to the Focus/metering sensor array. This is not stabilized with in body stabilization as only the imaging sensor shifts around to compensate for the camera moving, this does affect subject tracking.
Ok.. but last I checked, all forms of image stabilization are basically useless for a moving object. So OSS is no better than IBIS.
13.4 - 11.7 is an increase of 12%
24.4 - 22 is and increase of 9.8%
Again neither those numbers are huge increases.
Those are noticeable differences. You are making a weird argument -- That since 24.4 is only 10% better than 22, the means 22 is better?
By your logic, all cameras are the same. The full frame Canon 5Diii has a dynamic range of 24. The old APS-C Canon 50D has a dynamic range of 21.8. Again, only a 10% difference. It's hard to find any 2 cameras with more than a 10% difference in dynamic range.
Again there is not a single cross-type point in the right third, or left third of the frame on the Sony, all 15 cross-types points are in the middle section of the frame. Both the 7D and 70D have 4 cross type points in the right third, and left third of the frame, and 11 in the middle third.
You're quite hung up on that. I'll wait to see objective testing of the total effect. There are areas on the A77ii with focus points, where the 7D and 70D don't have any focus points at all.
The AF range limit is nice, but isn't needed as there is a physical switch on a lot of EF lenses. ie. EF70-200mm f/2.8 is choice of 1.5 m or 3 m minimum distance. And I couldn't even figure out what "Flexible Spot AF" even did.
Not all Canon lenses have focus limiters --- You picked a very expensive professional lens as your example. Furthermore, the choice of 1.5m or 3m may be nice, but it's a very limited feature compared to the Sony focus range limiter. I can make it exact --- Focus between 10 feet and 15 feet only. Focus 20 feet to infinity. Focus only 7 to 14 feet, etc. So it's much more customizable than the physical switch.
You do realize the 70D has the exact same 7D focus system, with the addition phase detect pixels on the imaging sensor, making AF in Live View, and video mode quicker.
Yes, and even the original A77 out-performed that AF system based on professional testing. And while live view AF is faster on the 70D than prior Canon models, it is still MUCH slower than the Sony dSLT cameras.
Imaging-resource measured the AF/shutter time in the 70D live-view mode to be just over 0.5 seconds. Not bad for a traditional dSLR. But still much slower than the A77 -- 0.124 seconds. Four times faster. (400%)
A guy took the mirror out of his A55, and noticed a loss in detail and light.
http://thesybersite.com/sony/a55/index.htm#slt_mirror_image_quality
Oh.... a guy blogged about it. Okay... I'll rely on professional reviews from trusted sources. And such sources have not found any resolution penalty for the dSLT mirror. It effect is only light loss, about 1/3rd of a stop usually.
As far as DXOMark, If you actually look at the graphs, DR is almost identical between ISO 100 and 3200 (the ISO you actually shot in most of the time), and you see the same with colour, and Noise almost identical. The only reason the Sony has a better scores is because it has a higher max ISO (which is almost unusable because it's too noisy).
Huh? No.... That's not how DXOMark does their testing. You should read the explanation of their testing. Not sure where you are finding the graphs either. But they test DR and color at base ISO.
In fact, here is what the reviewer from DXOMark commented:
"Although technically the ageing but still hugely capable and popular Canon EOS 7D is the closest to the new Sony, the sensor in the Canon EOS 70D is the latest from the firm. Like others from the firm the Raw data output doesnt fair that well with rivals in our tests, particularly at base ISO. The Sony has +1.3 improvement in color and a sizeable +2 stops wider dynamic range, however in low-light capabilities the two are evenly matched."
Now, does DXOMark consider those differences to be significant?
This is what DXOMark says about DR:
"A value of 12 EV is excellent, with differences below 0.5 EV usually not noticeable."
So according to DXOMark, the Canon 70D does not reach "excellent" level of DR. While the A77ii breaks the excellent level of 1.5 stops.
For color depth, DXOMark claims:
"A color sensitivity of 22bits is excellent, and differences below 1 bit are barely noticeable."
So according to DXOMark -- Canon 70D barely reaches an excellent level, while the Sony is noticeably better.
Now, I'm not claiming that DXOMark is the sole judge of such criteria. They are but one source. There are other professional testers, using different methodology, which have favored Canon in other ways. (For example, at high ISO, Canon better preserves the dynamic range)
the A77 ii is not the second coming, it's a stop gap, until Sony can convice people to stop buying Alpha mount cameras.
Wow. Someone is a hater. I never did claim it was the second coming. I suggested it, because you made a comment about considering the A6000, and the A77ii is still a better camera for most uses, apart from size.
What we know from objective testing so far, is that the image quality equals or exceeds all other APS-C dSLRs on the market. (With high ISO performance trailing behind by about 1/3rd of a stop). We also know it has a host of "mirrorless" type features lacking in traditional dSLRs (like focus peaking, faster live view, etc). And we know that the AF system -- which was already faster than most rivals --- has been significantly improved.
May not be the second coming. But as stated in the DXOMark review, may be the best semi-pro dSLR on the market right now.
But as I stated, I suspect that Canon and Nikon have new cameras in the works. Especially as Canon hasn't made any real progress in sensor performance in several years, I kind of think they are working on something big, soon.
