A7RII first impressions and LOTS of RAW files

[tashley]

Without ado and after about 100 frames, therefore highly provisional...

I like it but it is a bit noisier than I was expecting. There's even a hint, tiny, of banding at mid (3200) ISO but that might be an exposure issue, because the 'blinkies' panic much much too early, encouraging one to underexpose. In general, using the multi meter seems never to actually overexpose but often to warn you that it has.

Files don't respond well to Clarity in LR - they quickly break up in not-a-good-way.

For some weird reason I find that 60/0.7/70/20 for sharpening works well but that even at lower ISO, a touch of NR seems to clean the files up without removing significant detail. I need to do a lot more exploration on this.

WB seems to me more than usually tricky and I will use a WB card when it matters.

It shows the weakness in lenses like crazy. Even my treasured 16-35 looks asymmetrical and a bit stressed in ways I never noticed before. But the 55mm F1.8 works superbly.

It feels great and the shutter is SO much nicer.

More to come, if anyone is interested, as I explore - though I do think that these things take weeks rather than days to clarify in one's mind....

http://tashley1.zenfolio.com/img/g909131451-o374332360.dat?dl=2&tk=DKb6OdiHI8z0sFHBqWoQDDU9_zZW6agLabOJiWzGw3U=

http://tashley1.zenfolio.com/img/g651464162-o374332360.dat?dl=2&tk=lESsW62K48Phrs4QiAmhVgHN8xGFyuqTsNW6pTm37hI=

http://tashley1.zenfolio.com/img/g739220745-o374332360.dat?dl=2&tk=uzamRQGtq8Wm3iVwpRjeAkYkmuSArZusnGqz2I4V5Jw=

http://tashley1.zenfolio.com/img/g723662388-o374332360.dat?dl=2&tk=Q-MCcXdTOzpT_GNdcItAswcLdVVBBuGJ3Oy0ZgUvjto=

http://tashley1.zenfolio.com/img/g558601346-o374332360.dat?dl=2&tk=GN4_gNwHlKDRc_2XEFDfGcFqSPkzHwe-h2wjerLBprA=

http://tashley1.zenfolio.com/img/g1001742675-o374332360.dat?dl=2&tk=5V1HxRhLBWALAorLECa6raRm4ZD5GlX5TehC-NNHkRA=

http://tashley1.zenfolio.com/img/g1011674541-o374332360.dat?dl=2&tk=eD9wet1CG_pWvYVEYm779x1Kb0WG-RA0DPEG2eiW0bc=

http://tashley1.zenfolio.com/img/g752736460-o374332360.dat?dl=2&tk=Rki1wrrBZoIKe2Vb8lpBdp7djQFo3VZHcaczEme7aDk=

http://tashley1.zenfolio.com/img/g745109402-o374332360.dat?dl=2&tk=q-gGDieZvM5eVjb6fz1vCGBv2-qZQrbV9ox2xPi1ADU=

http://tashley1.zenfolio.com/img/g884995149-o374332360.dat?dl=2&tk=rrJwm1vm7X-UeP6D1vo3C1DAYlBDKKwMmaEdFOnO1pI=

 

[Leigh]

Even my treasured 16-35 looks asymmetrical...

How does a lens look "asymmetrical", and what does that mean ?

- Leigh

 

[Annna T]

How does a lens look "asymmetrical", and what does that mean ?

- Leigh

I think he means decentered, aka some corners looks sharper than others.

 

[UHDR]

i would love to see these high MP sensor stress Otus and ZF/ZE 135mm. but probably not there yet. :grin:

 

[tashley]

@ Joe, nice to hear from you!

I have been resisting the temptation to take out the D810 and try to shoot them side by side with the same lens because in a way, though I too want the answer, you'd have to take so many shots under so many circumstances to 'prove' anything and even then, you'd likely have proven that thing only to yourself. My first impression being that the files were a little noisier and a little sensitive to the clarity slider do seem to have been because of the panicking blinkies. I totally get the 'expose to the right' school of thought but frankly, until these cameras have RAW histograms and highlight warnings, I will always tend to underexpose on the basis that you can generally save a shadow or lose it to darkness but you can never save a non-specular blown patch.

Now I've got that under consideration as I shoot, the files are very nice. I would say that at a pixel level and from ISO 800 onwards, there may be a tiny touch more noise than the D810 but that DR seems better. I would also say that that with the exception of the video features, I still feel that the D810 has a better feature set and better ergos but that it's size and weight outweigh those advantages so I provisionally prefer the Sony. But honestly, this is angels on pinheads. The IQ on both is fabulous.

As for WB I'd need to shoot it a lot more to have a view. But it feels a touch trickier maybe?

@Leigh, as Anna says, I am referring to a lens which has some de-centered element or group which manifests as files with one side of the sensor being in less sharp focus than the other on a planar target. It's a lot more complex than that in real life though...

 

[Vivek]

I would be most interested in side by comparisons with a D810. There are rumors that the (A7R II) DR suffers with longer exposures and continuous use.

I want to be convinced that it (A7R II) is really a 3500 Euro camera and not a 2500 Euro camera with marginal improvements over the A7R.

 

[tashley]

I would be most interested in side by comparisons with a D810. There are rumors that the (A7R II) DR suffers with longer exposures and continuous use.

I want to be convinced that it (A7R II) is really a 3500 Euro camera and not a 2500 Euro camera with marginal improvements over the A7R.

And in a way, I think that's why I don't want to run the comparison. It sounds so simple but the exact methodology would be disputed along more or less 'religious' lines and even then, there would be things 'left out' such as long exposure, continuous use, processing both in camera and in post, pre-cooking of RAW, concepts of the meaning of base ISO, etc etc ad infinitum and then rapidly ad nauseum. It would be a foolhardy person who put themselves in that firing line.

Maybe I should?

But one thing that people are overlooking somewhat is that the A7RII has a base ISO of 800 and the D810 has a base ISO of 64. That means that there is no such thing as apples to apples here: the sensors are fundamentally rather different animals. That's around a three and a half stop difference - and if base ISO is the ISO at which a camera has the most DR (is it?) then one might even find that the Sony has its best DR at an ISO which doesn't optimise its noise performance. Just thinking out loud here....

 

[Guy Mancuso]

Tim I have to agree with you. Honestly I would not do it, it really has gotten so weird in the industry with reviewers. People are listening to some real inexperienced players, bloggers and taking some of the junk they are throwing out as Gospel and the few experienced players not getting a cut from it all are being ignored and put on the back burner. Honestly I have yet to see anyone do a in depth review of anything. It's more come down to one hit wonder posts. To be honest from my seat it's really frustrating to watch really bad info being passed around like gospel. I'm so furious right now at the lack of reporting Canon lenses after 2006 can really only be used and that data just came out a couple days ago. Where the hell was those reviews last week or older not reporting this critical data. That's irresponsible reporting.

From me personally as a guy that does very good reviews , not sure I want my name associated with this anymore. I love doing them for our members but after our doors I have zero interest. i will still do what I think our readers need to know but no manufacture has knocked on my door to even thank me. My rant could be very long here but I'll head for that second espresso instead.

Glad to see you hanging out again with your friends here.

 

[tashley]

Thanks for the welcome back Guy, and I totally agree. There's no test as useful as a well developed personal opinion based on seeing zillions of files from many many cameras and lenses. But that opinion takes a while to develop and has to be formed within the context of a pretty subtle understanding of a whole load of parameters and criteria.

Having said that, and for within these doors only, and with no opinion given, and because frankly someone's gotta do it. I shot some comparisons at ISO 64, 1600 and 3200 and from them I learned enough to know whether I should knock out my D810 and lenses while prices are still good.

A7RII with 55 FE
D810 with Sigma Art

I could have used the same lens on both but I didn't because I can't find my adaptor. Such is real world testing. That means that, assuming we are looking not for resolution but for noise and malleability, the only possible significant variable is the T stop. DXO says their T stops at max aperture are within 0.1 of each other. Who knows what they are at the aperture I used here - I trust my nose. But I won't give my conclusions, just links to the RAW files.

http://tashley1.zenfolio.com/img/g890098758-o374332360.dat?dl=2&tk=3Q2fAHMbkkosLxGCg_r7hqHJVG3VrwvfpGhju92wvtc=

http://tashley1.zenfolio.com/img/g867050477-o374332360.dat?dl=2&tk=cLnXO-V-SYt9wBtNBQhJXKbNx-ZMgyEUiSYRvvC41zc=

http://tashley1.zenfolio.com/img/g985314067-o374332360.dat?dl=2&tk=zsxtg6JuBKANcpzOwXP39gkQwdenRZGlpitdleme_s0=

http://tashley1.zenfolio.com/img/g866672919-o374332360.dat?dl=2&tk=RO59BxSPEwucM4S1zFva2WIpO2n2k3InvqLSVr8t6Qg=

http://tashley1.zenfolio.com/img/g675885246-o374332360.dat?dl=2&tk=gfd4ynUAAjKuWESqCew78MtWsMn_RdkHYruBQ3hNH8o=

http://tashley1.zenfolio.com/img/g859783141-o374332360.dat?dl=2&tk=o2Jz3j3LZGNey-uvXuQRWm6T9Msk3tpnRtYjFt6yw2o=

Anyone is free to draw their own, or no, conclusions from this but please don't have a go at me about the methodology: I am pretty well aware of what the 'test' means and doesn't mean. It's useful to me, I am sharing it for that reason and not because I think it should mean anything in particular to anyone else or in general....

 

[Pradeep]

Tim I have to agree with you. Honestly I would not do it, it really has gotten so weird in the industry with reviewers. ....................

There is almost always an inherent bias in any review. It is in the nature of human beings to be biased, it is hard-wired into our brains. I could go on about all the different biases we live with in our daily life but keeping it to camera gear, I agree with you completely.

It is often difficult for an average user to look at reviews and make up their minds. Most have already made the decision to buy or not to buy, which is largely based on affordability and lust for new gear and much less on need.

Even looking at RAW files on the net does not really provide an ideal environment, IMHO.

Yes, there truly are very few, completely independent reviewers who are not in some way influenced by their own pre-conceived notions or by financial gain.

 

[Amin]

And in a way, I think that's why I don't want to run the comparison. It sounds so simple but the exact methodology would be disputed along more or less 'religious' lines and even then, there would be things 'left out' such as long exposure, continuous use, processing both in camera and in post, pre-cooking of RAW, concepts of the meaning of base ISO, etc etc ad infinitum and then rapidly ad nauseum. It would be a foolhardy person who put themselves in that firing line.

First, Tim, thanks for your impressions and the RAW files! All very interesting and helpful. Second, as someone who from time to time tries to do these kinds of tests and reviews, I've learned to pretty much ignore the disputees. It's impossible to do a perfect test, and doubly impossible to do a test the way everyone wants it done. So I do tests and reviews that I find interesting, put them out there, and let the criticism roll off my back.

 

[mjr]

I personally don't see an issue with reviewers, good or bad, it's the people reading them that are the problem! Haha.

Honestly though, everyone has opinions, some people will believe one person, some another but it's only foolish people who make decisions based on what they read and not how a piece of equipment works in their own hands shooting their own favourite subjects, and that's down to them, doesn't bother me!

Looking at jpegs on a forum says nothing, never has done really, pointing the camera at a boring view of your garden, street or wherever with one camera and then another and putting the shots online, what on earth does that mean? Nothing, apart from how slightly sharper or more contrasty the boring scene is from a newer camera or lens. But then I think a fair percentage of people are sensible enough to buy a piece of kit they are interested in, shoot it, look at the results and see if for them the upgrade is worth while, if not just pass it on. Reviews are entertaining at best and ridiculous at worst, having to pay someone for their opinion is firmly in the ridiculous camp for me. Those that put their opinions online I can choose to read or not, I know from my own testing that it can take time but I see little point in showing it to anyone else, it's just pointless because it only relates to my needs and desires.

What I'm waiting to see from the A7rII are shots that couldn't have been produced from another camera, the only way I can do that is to buy one and see what I can get it to do, no offence to the millions of pictures that will soon be uploaded of a cat or a living room or a garden but a jump in dynamic range or sharpness on a boring shot is still going to produce a boring shot, lets see something breathtaking, original, unachievable with other equipment.

Just my opinion of course, enjoy!

Mat

 

[Vivek]

Tim, Thanks!

(FWIW, so much preamble was not necessary given the fact that you are not new here and my query was not a bait-it was sincere. As for the review industry, web outfits, blogs, etc, if it is worthwhile for them well and good. )

 

[Vivek]

Tim, not sure what you mean by the base ISO of the A7rII is 800. The specs say 100?

 

[vjbelle]

Tim..... thanks so much for the raw files. Sony vs Art (assuming both are good samples) is hard to find. Much appreciated.....

Victor

 

[tashley]

@ Vivek, the reason I say that is that the SLog2 video picture profile has a minimum ISO of 800 and that is generally set at the base ISO of the sensor. I have read this in several places too. The official 'non-pull' ISO range starts at ISO 100 but my understanding is that the ('Native' might be a better word than 'Base') ISO is 800...

 

[Ken_R]

Thanks for posting all these files! I will work on them today.

I have a question. Is it fact that the base iso of the A7RII for still images is 800? instead of 64 or 100?

 

[hiepphotog]

@ Vivek, the reason I say that is that the SLog2 video picture profile has a minimum ISO of 800 and that is generally set at the base ISO of the sensor. I have read this in several places too. The official 'non-pull' ISO range starts at ISO 100 but my understanding is that the ('Native' might be a better word than 'Base') ISO is 800...

Tim, Sony has been implementing a conversion switch tactic to squeeze the best DR out at low ISO and high ISO and it has to do with video too I guess. The switch is not exactly at ISO 800. Here is a couple of articles Jim Kasson just made:

Sony a7RII engineering dynamic range | The Last Word
a7RII read noise with EFCS on and off, plus silent, continuous | The Last Word

Edit: In a way, we have at least 2 base ISOs (or 3)

 

[Vivek]

Hiep, I agree with that. The video requirements complicate the factors.

 

[Leigh]

@Leigh, as Anna says, I am referring to a lens which has some de-centered element or group which manifests as files with one side of the sensor being in less sharp focus than the other on a planar target. It's a lot more complex than that in real life though...

OK. I just didn't understand the terminology.

I'm certainly familiar with the de-centering issue.
There's an active thread on that topic.

Thanks.

- Leigh