Peet van den Berg: "iTR struggles to accurately track moving subjects, especially fast ones" I have found exactly the opposite - its fast and spot on in half a heartbeat! Are the reviewers used to taking BIF pictures? Love this beast already!
I'm curious if dpreview is going to downgrade ANY camera in this field of semi-pro cameras that can't AI servo moving objects up to the level of a 1 series or a Dx series camera body from Nikon now.
rnclark: The example on above (page 13, real world raw dr) is an example of differences in processing. While the Nikon does have a slight advantage, the actual difference between the two cameras is much smaller, less than a stop in dynamic range. The main reason that the nikon data look so much better out of camera is that nikon clips the low end and runs a filter on the raw data before recording the data. With the canon, there is an offset and no apparent filtering in the raw file. Nikon's raw filter is very very good. Apply similar filtering on the canon raw data (luminance filter done in the raw converter), then adjust the offsets in post processing and one can get very close to the same dynamic and look of the nikon image.
I have downloaded the above raw files and processed the canon data to properly deal with the offsets and the results between the two cameras are very close. I would post the results, but being new here I do not see a way to do that.
the problems aren't offsets - Nikon / sony are still fixed position offsets with no variability (from what I recall). canon has floating offset on the row and column since around the 40D timeframe.
dpr4bb: Some of the comments here are so crass, they border on hooliganism. I trust that the DPR team will not be unduly influenced by them.Many thanks, again, for taking the time and the effort to bring us the DR comparison in this review. The technical depth that this type of analysis brings to the reviews here is indeed very much appreciated. Please keep up the good work.
the dr comparison that basically shows .. that even with a D7000, learn to be a photographer versus a post processor?
Pritzl: I have looked really hard and I really cannot see the advantage for any of these cameras (maybe the NX-1, but a tiny one at that) at 12800 ISO. They all look to be about the same. If there is a difference, it's one that requires a loupe and someone far more nit-picky than me.
It doesn't matter much anyway, as they're all a smudgy mess at that ISO anyway.
look in low light - especially in the darker areas ..
Just another Canon shooter: Thank you for adding a page about DR. It was time to make this "more official". Canon cannot pretend anymore than they do not know what everybody is talking about.
BTW, why comparing the 7DII to the D7000? Why not to the D7100? The latter is newer, and ... suffers from banding (still better DR than any Canon).
high ISO invariance doesn't give any credence's to the fact that sensors can also adjust their bias voltage depending about ISO amplification which can change how the sensor behaves at higher ISO's.
Most of the theory surround ISO invariance assumes that everything is constant as the sensor amplifies the signal for higher ISO's.
Jackinthebox: just what does "optional" mean for wifi? I would very much like this camera but do want GPS and wifi to be able to control the camera from an Android device.
camranger works, tethered usb works to your android (with less lag than wifi), you can go all out and get the 7DII communications upgrade which gives you gigabit lan, ftp, http host, wifi,etc as well. lots of options. if you just want to flip small jps over onto your phone, use a eyefi card in your sd slot and cf card to store the raws.
or you could sit in your mom's basement like Barn and attempt to say that wifi accessory for 800 is simply a wifi dongle and act outraged.
Ozyxy: where's support for my 60D and 5D Classic and SX50?
they are still supported on current versions of DPP 3.x - I fail to see what the problem is.
not only that, but the DPP 4x still has it's issues, the excuse could very well be; supportability and managing the software rollout.
ozturert: DPP 4.X is a parallel development with 3.X. I think development of 3.X still continues and 3.X supports all Canons.I think Canon will cease 3.X when 4.X supports all Canons.
4.x is 64 bit only
DPP 3.x is getting regular updates as well for older cameras.
JoEick: Rishi, you need to stop being defensive and actually read what is being criticized. Everyone knows Sony exmor sensors have better low ISO DR. We knew this years before this article tried making it into something groundbreaking.
People (including myself) are taking offense to those who try to overstate the effectiveness in being able to get a shot or not, based on the low ISO DR. It's just pushing people to let their gear dictate everything for them, without any thought or skill required by the photographer.
Extra DR is nice and never hurts (except in photographer skills), but it's not really the feature that is making photos possible that were previously not possible with some basic photography skills in merging exposures.
I am aware that this is DPR, a gear praising website, where suggesting features are not needed or are overstated, is like breaking all the 10 commandments in one shot. Photographer skills are going backwards, while tech marches forwards. :(
odd. it's not even my pet camera - what's staggering is the lack of skill and technique and the reliance on poor technique from many photographers.
Trubbtele: I earn quite a lot of money on my D7100, no one complaining if I use dito Nikon or D810. Even with my old Samsung X1 pictures I earn money on :-) Todays cameras is on a high level, Canon or a other brand, who cares in the end of the day. The person behind the camera is much more important...
For real. any camera made in the last 5+ years is possible to create great imagery only limited by the IQ 8 inches behind the viewfinder.
"unreasonable to expect a photographer to have chosen an exposure closer to our 'ETTR (-1 EV)' shots for this scene, in which case the D7000 would've sufficed, but not the 7D Mark II. "
only if you're on glue - the lack of detail and resolution from the D7000 is staggering.
of course you'd never say that nor show what someone could do with exposure blending and just the staggering amount of tonality and resolution difference would be . right?
" it also means that the camera will be able to do a better job when confronted with an even wider dynamic range scene than this one. Which we can assure you landscape photographers encounter often. "
Only if you're a complete idiot and dont' use capable techniques to maximize both your captures and your resolution and don't expect to do it in sub-optimal desperate techniques such as the one demonstrated. Both the D7000 - Heck a 20D or a D70 would rock that scene with any photographer worth their salt.
commenting on the D810 / 645 - how is that relevant to this and current cropped cameras? obviously the ability to extract detail from a D810 would even make the D7000 look bad. So what?
I mean hell - you guys even went as far as to use the D7000 versus the latest Nikon camera in that class just to make this looks as bad as possible.
Nowhere did you mention that perhaps the D7100 would fare worse in this test?
and on top of that you created images on a really good sensor based camera (D7000) that looks like crap and try to pass it off as art.
pretty defensive there Rishi..
To be honest - all your examples on page 13, illustrated was both cameras performed bad and that photographers should be looking at technique versus trying to rescue such a shot via shadow lifting as you try to brag / explain about.
the lack of resolution on the D7000 in the shadow pull is significant - I question your actual goals with this, because not once in that write up on page 13, did dpreview mention that there are other ways far better to extract both dynamic range, tonality and resolution from such an easy and static scene.
Shadow pulling in that example scene is only necessary if you're in idiot.
if a photographer thinks either one of those was adequate - including the D7000 image perhaps they should stick with a smartphone. because that's what the output looks like.
grammieb14: I have a question for the Dpreview review staff. Why did the 70D get a gold award for a percentage rate 1pt lower than the 7D with a silver award. This is kind of confusing.
apparently they are scored with the imaginary cropped sensor pro/semi pro bodies that dpreview have tested lately.
DaddyG: Great review - the degree to which you highlight the 7D's poor dynamic range is long overdue. You do a great job on explaining why this is so important.
This level of negative publicity is long overdue, and can only be good for us locked into the Canon system. Surely Canon must concentrate on improving its sensor technology.
However, it is odd that you do not list it as a 'Con'. Because, for me & many, it is this camera's greatest disappointment.
D7100 would have exhibited banding and more color casts than the D7000.
not to mention one sigifnicant point. nether camera is good at lifting shadows to this extreme, both photos are crap... canon is simply more crap than the crap on the D7000.
far be it though that dpreview actually highlights skill and technique over post processing desperation.
samsamsamsam: RAW DR Test against a Camera from 2010. 1/30 and 1/25?
So why didn't you use the D7100 then?
RaghavBaijal: What I find a bit worrisome is NOT that it is lagging behind the current competition in IQ, but the fact that Canon users will have to contend with this for the next 5 Years!! At which point I am guessing that even M4/3 sensors will perform much better than this perhaps even in the High ISO department.
who says it will be 5 years? is there a rule / law that says that canon can't update sooner? it's normal update cycle for prosumer bodies in single digits is around 3 years.
mpgxsvcd: I find it very interesting that Canon basically has a stranglehold on the Astrophotography world when their dynamic range is at such a HUGE disadvantage to other competitors. However, the fact that we stack multiple RAW images to remove the noise puts the Canon cameras on a more even playing field for Astronomy.
The stacking process relies on the fact that the noise is random and the signal(Light coming into the lens) is consistent. This means that we can easily isolate and remove the random read noise that plagues the Canon cameras in single exposure shots. We also use dark frames to further identify the noise that is present if there is no signal.
canon got into astrophotography for more than this. basically: allowing tethering and mirrorlockup and then tethered usb / liveview across it's entire model lineup for years.
made the development of software, and drivers quite easy - as canon's entire lineup would just work. meanwhile to use a Nikon you had to "mode 3" it for a while, and sony .. do they even have usb tethering with liveview yet?
also with it's "shorter" by SLR standards registration distance, that allowed astrophotographers more room for filter wheels, autoguiding,etc.
then you have the canon RAW which is still the more "pure" across the board - Nikon still I believe for most cameras does LNR that removes data, and only a few bodies have floating above 0 black levels.
Pardon me for a dumb question. how'd you shoot this on a A7R in March 13th of 2013?