A very first fully detailed A7rII review has been posted by Mobile01. It’s not in English but there are plenty of pictures and tests you can check.
First impressions by DearSusans:
A minor miracle appears to have occurred in that two of my favourite M lenses – the 18mm Super Elmar and 24mm Elmar – appear to show no colour fringing, whereas they were simply awful on the original A7. The 18mm Super Elmar and 24mm Elmar still have some problems (vignetting and softness at the edges) but are much, much better than they were on the A7r1. I would think anything from 35mm on will perform well.
A7RII 12-bit mode. How to avoid it by Fred Miranda.
The Sony A7RMII does 10 frames with auto bracket for an extreme HDR pic (Flickr by Richard Banton).
Sony A7RII 1st Thoughts, Images & Video by Steve Huff.
A7rII cage from Varavon (Newshhooter).
This is a 181mp panorama which turns out to be a 153mb file (Wayne G on Flickr).
New A7rII image samples and owners reports on SonyAlphaForum.
A7rII preorder list:
In USA: at Amazon, BHphoto, Adorama, SonyStore US, FocusCamera, Uniquephoto.
In Europe at Sony DE, UK, FR, IT, ES, NL, BE, CH, AT, SE, FI, NO, PT. Wex UK.
In Asia at Digitalrev, CameraPro.
A7rII Gariz case
The current A7II Gariz cases suite perfectly for the A7rII too:
Alcantara case in [shoplink 38168 ebay]Red (Click here)[/shoplink] and [shoplink 38169 ebay]Black (Click here)[/shoplink].
Leather case in [shoplink 38164 ebay]Orange (Click here)[/shoplink], [shoplink 38165 ebay]Red (Click here)[/shoplink], [shoplink 38166 ebay]Black (Click here)[/shoplink] and [shoplink 38167 ebay]Brown (Click here)[/shoplink].
As this is the same A7II case you can buy them all also at Amazon US (Click here).
43 Responses
12bit vs 14bit = nontroversy.
I am a little supprised how fast 3200-6400iso files lose detalils… is this ok. for 6400?
Interesting topic.
Though what’s even more interesting is where none of the samples taken to demonstrate the differences can actually show any differences… :/
That said, I seem to recall a very similar topic with Nikon camera’s awhile back, with very similar results. – 12/14bit providing unidentifiable differences in all but the most extreme cases.
Today’s sensors are better, so 12/14 difference is going to be more noticeable now.
“Though what’s even more interesting is where none of the samples taken can actually show any differences.”
Precisely, just wasted 15 minutes reading through the threads. It is a whole sport in and of itself, apparently; it really has nothing to do with photography.
“A7rII saves child from drowning after photographer notices struggling boy in high res image. Critics point to 11+7bit lossy compression could not have saved kid if it would have found itself too deep in the shadows.”
I’d imagine this is the whole compression debate, people just don’t realize why sometimes their files are like this, other times that. It took us a little bit to figure out what was what, and which modes to avoid in challenging environments, and while no camera is perfect, and the work arounds easy, I miss bulb as my go too when landscaping.
I get the impression that the noise and DR ‘limitations’ of 12bit only exist in examples where shadows have been pushed really hard. Not in properly exposed (or slightly exposed to the left) photos. 12-bit in itself does not introduce noise or limit DR. It just limits the range that those characteristics can be manipulated in post.
So if you lose a stop of dynamic range at 12 bit. That means you only have an advantage of 2 stops rather than 3 stops over a canon….doubt I will lose any sleep…
More like it drops from an advantage of 2 1/3 stops to 1 1/3 stops.
But that drop continues through higher ISOs. So at ISO 800 on up, canon will have a 1 stop advantage
? Jim Kasson (http://blog.kasson.com/) has done measurements and comparisons with the D810 to show that above ISO 200 you won’t see a difference of whether you use 12 or 14 bits as DR decreases with ISO.
This is an old topic for Nikon. On my monitor there are differences.
https://photographylife.com/14-bit-vs-12-bit-raw
http://www.diyphotography.net/12bit-vs-14bit-raw-and-compressed-vs-uncompressed-does-it-matter/
http://www.nx101.com/12vs14.html
So enabling long exposure NR is making it record in 12bits. So how is anyone going to shoot something that is within the 2 – 30sec timeframe & get nice images? Either you live with 12bits or live with hot pixels.
Darn!
Are you serious? Long exposure NR is often inconvenient.
Don’t people get tired of this nonsense about bit depth? Everyone will forget about it, same as it happened with the translucent mirror “robbing light”; I wonder what would be next?
Great news on the M lenses!
It can make a difference. Probably not to the casual amateur, but for some pros it sure does.
What is that difference that matters to pros, precisely? That it would be visible at close inspection when you are searching for it? Or, that it would ruin a photo? I would really like to see it. Not on a computer display zoomed at 400% and pushed 7 stops, but either downsized and compressed for web display or printed (at whatever size). Of course, no such test exists and guess why.
When you see your pictures on a 60inch LED tv, which is not uncommon, you can tell the difference sometimes.
If made that difference for “PROs” how come Nikon isn’t the no.1 on the market?
I can’t forget about it because it affects me a bit too often.
The fact that the shutter performs best in ff will hopefully be beneficial for still pictures in high resolution.
new avatar?
I constanly change my avater, at least once every few days.
They are all details of my paintings, they are not photographs.
LOL, you might as well start a blog on it. Barry’sAnimalRumors.
Hot pixels will be the least of your worries if you are shooting in quickly changing light at twilight or dawn. By the time you will have reviewed an image for exposure, the amount of time needed will have changed. I shoot without and use dfine hot pixel filter to get rid of the hot pixels
For now, yes.
It used to be far lower back in the day.
That said, I find the A7R II ISO6400 files to be very good in terms of detail retention.
So the RII is 13-bit and changes to 12-bit using the listed functions?
RII is 14-bit and changes to 12-bit.
Then it’s not 14 bit.
Do you understand logic? 14 is 14, and 12 is 12. It’s not hard.
To be honest, I think the difference to be really small for most use-cases, which leads us to use 12 bits at events all the time. But there are situations where you push files a lot because you didn’t have the time to expose correctly and there is a slight difference in tonal transition in certain areas, which are looking less harsh. Depends on how important that is to you. I think Ming Thein showed an example in the past. We all know that we are coding color values in a fixed point format. Consequently, half of all code points are in zone 10, 1/2 of the rest in zone 9, etc. In zone 1 are quite few distinguishable code points (16 vs. 4 for 14 vs. 12 bit, 32 vs. 8 in zone 2 etc.). That means higher possibility for quantisation noise and banding the deep shadows, especially if these shadows are lifted. For Ming Thein’s use-case of ultra-prints I can understand his strive for ultimate quality. A file looking less noisy with more leeway to push gives satisfaction, too (that’s what I read out of many comments I read at least), and reduces the time needed to post process it. Jim Kasson at his blog clearly measured the difference. Basically it showed that ignoring it would mean to use your Nikon D810 or A7r (if it had 14 bit) from Iso 250 and above only, instead of ISO 64. That’s not s.th. I would dismiss at once. Therefore, although I don’t think it to be huge, I wouldn’t dismiss it as nonsense.
I do see your point. However, as I said previously, I’d like to see how it affects final output and if it is a real problem am ready to eat my own words.
My bet is that even an ultraprint wouldn’t suffer in any significant amount and certainly not as far as to ruin the image. I am not saying that the issue doesn’t exist, but if it barely affects the final output…
It’s crazy how readily people dismiss better shadow recovery. Doing interiors photography, I lift shadows a substantial amount regularly. Of course you can bracket to recover, but that defeats the whole purpose of file size reduction and introduces an extra step of masking.
If you need to capture an impossible amount of DR, every bit counts. End of story!
Glad to see another interior photographer appreciating every improvement. Gosh, on some lower contrast scenes it could be the difference between bracketing and not- huge time saver.
https://www.youtube.com/watch?v=NP31-R5aeps
8bit video output on A7S, JPEG engines, sensor cover glass thickness, bayer array, etc. etc.
A7 series and now the A7rii
14 bit only in marketing
12 bit depending on mode
11+7 bit all other times
Still having problems with wide angle CRA smearing
I think that if you give s.o. a picture made in 12-bit mode and tell him it was done with 14bits, he would in 99,9% of cases not recognise this. It is only by comparing 14 vs. 12 for hours that you will see a difference in certain circumstances. Many photographers like to pixel-peep, of course.
I don’t understand the direction of your question. Does Canon use only 12-bits, too, or Pentax?
Sometimes you need to do this. Paying this much, a few people would enjoy being limited.
Today I stumbled upon this. It is of course a more extreme case (4.5 stops push), but one of the first really noticeable real world images showing the artefacts:
http://www.dpreview.com/forums/thread/3885061.