Some experience with the A7rII

[John Hollenberg]

yes, this is the one... there you are far away from UniWB as RED is 2+ times more than both greens and BLUE 1.5+ times more... so you need to drive both RED and BLUE multipliers to something close to 1024...

the usual one though camera's menu to set a custom WB by shooting a target... in this case an image on your LCD screen, which is a simple magenta color filled area.

OK, finally figured out what was going on... still fairly new to A7r2.  I got the numbers to 1024 1024 1024 1020--good enough 

Thanks for your help.  Only problem is now my photos look kinda green. 

[Jim Kasson]

For a really quick and real dirty approach to a7RII UniWB, you can set custom WB to 3400K, A=7, G=7. Gets you about 1500/1024/1500.

Better than nothing.

Jim

[AlterEgo]

OK, finally figured out what was going on... still fairly new to A7r2.  I got the numbers to 1024 1024 1024 1020--good enough 

yes, very good

Only problem is now my photos look kinda green. 

by design - this is only if you use raw files, OOC JPG and/or embedded thumbnails will be green with UniWB (or greyscale if you use picture profile that desaturates it).

[John Hollenberg]

by design - this is only if you use raw files, OOC JPG and/or embedded thumbnails will be green with UniWB (or greyscale if you use picture profile that desaturates it).

Yeah, I was just kidding.  I only shoot raw.  Thanks a lot for the help.

[Hans van Driest]

do not know what type of sharpening you use, or how accurate you focus, but in my eyes your 100% examples look horrible. jaggies, black spots and total lack of texture at other places. If have a R2 myself and never see these problems. My usual starting point with LR is 50/0.5/50/0, although it depends on the lens and f-stop. for f11 50/0.7/50/0 seems to work better and some lenses at f5.6 need less sharpening. am not suggesting you should use my numbers, all very personal. it is just that these numbers give me much nicer (in my eyes) results.

[ErikKaffehr]

Hi Hans,

Your point is well taken, most grateful. I will reprocess the images as soon I can. I think I used something like 45/0.7/100/15, but I have actually considered sharpening for a long time, so you comment is most helpful.

Best regards
Erik

do not know what type of sharpening you use, or how accurate you focus, but in my eyes your 100% examples look horrible. jaggies, black spots and total lack of texture at other places. If have a R2 myself and never see these problems. My usual starting point with LR is 50/0.5/50/0, although it depends on the lens and f-stop. for f11 50/0.7/50/0 seems to work better and some lenses at f5.6 need less sharpening. am not suggesting you should use my numbers, all very personal. it is just that these numbers give me much nicer (in my eyes) results.

[Bart_van_der_Wolf]

Hi Hans,

Your point is well taken, most grateful. I will reprocess the images as soon I can. I think I used something like 45/0.7/100/15, but I have actually considered sharpening for a long time, so you comment is most helpful.

Hi Erik,

The optimum radius to use for Capture sharpening, mainly depends on the aperture used and the lens quality. The LR/ACR Detail slider often starts to produce artifacts once it gets above 50. The amount depends on the subject and how prominent the USM sharpening halos are.

LR/ACR sharpening functionality could use a significant upgrade, otherwise disable it and use a PS plugin like FocusMagic (only from an editor that is PS plugin aware) or Topaz Detail (which can also be called directly from LR, via their free FusionExpress plugin).

Cheers,
Bart

[Torbjörn Tapani]

Erik, what has happened to the image comparing the A7rII to the Blad?

Under the heading: Can the Sony replace the Hasselblad? > Sony A7rII Central crop

Slanted lines are all jagged whereas the Blad looks distinctly smoother.

And the next shot with leaves. That leaf to sky transition is not looking good. Hope the uncompressed raw can fix that.

[ErikKaffehr]

Hi Torbjörn,

A previous poster pointed out that I apply an inappropriate sharpening, and he may very well be right. I am looking into that.

The jaggies are almost for sure an aliasing artefacts, lens outresolving the sensor, but they are enhanced by sharpening. Sharpening should be the same for both the Blad and the Sony. The effects are still there with less sharpening.

What I have seen is that Lightroom, the tool I am using is about the worst raw converter in creating aliasing artefacts, so I guess I will look into the issue using other raw converters, too.

But really, the Sony Images seem a little "brittle" to me. Interestingly, Tim Parkin (OnLandscape) has looked at the "orange peel" artefacts that Lloyd Chambers reported and he attributed it to aliasing/lack of OLP filter to a great part. I have actually seen that image but I don't feel I can discuss my findings due to constraints when getting that image.

I will reprocess both shots without sharpening in LR and that applying sharpening in FocusMagic and republish, but it will take some days until I get time to do that.

You may check this posting: http://forum.luminous-landscape.com/index.php?topic=94812.0

Sandy McGuffog, the author of AccuRaw made a recommendation on that posting, here:
http://chromasoft.blogspot.se/2014/11/mini-comparative-review-of-accuraw.html so I guess he found it relevant.

What I have noticed is that LR/ACR adds a halo around high contrast structures on non OLP filtered images. That halo is always there but will be enhanced by sharpening.

Thanks for the comment, really much appreciated!

Best regards
Erik

Erik, what has happened to the image comparing the A7rII to the Blad?

Under the heading: Can the Sony replace the Hasselblad? > Sony A7rII Central crop

Slanted lines are all jagged whereas the Blad looks distinctly smoother.

And the next shot with leaves. That leaf to sky transition is not looking good. Hope the uncompressed raw can fix that.

[eronald]

What is the bottom line here? Does the Sony replace the Hassy for Erik?

I guess Erik is not a bleeding edge adopter, as all Nikon users have basically had the same rez for years.

Edmund

[ErikKaffehr]

Hi,

One issue with the "Blad" is lack of ultra-wides as the lenses are really made for 6x6 (56x56 mm), so all backs are really cropped frames. Another issue with the P45+ back is lack of magnified live view. To that comes portability.

My guess right now is that the A7rII replaces the "Blad" as soon as aerial travel is involved and also for "serious shooting". But I still think the blad may have an edge in image quality. That difference may not be visible in prints of any size.

One advantage of the A7rII is essentially infinite choice of lenses. Right now I am looking into ordering HCam Master TS that would allow T&S with all my Hasselblad lenses and all Canon lenses.

I have observed some raw conversion issues that need a deeper study.

Best regards
Erik

What is the bottom line here? Does the Sony replace the Hassy for Erik?

I guess Erik is not a bleeding edge adopter, as all Nikon users have basically had the same rez for years.

Edmund

[AlterEgo]

My usual starting point with LR is 50/0.5/50/0

I always use for A7R2 = 5/0.5/100/0 or zero amount for anything where nominal ISO + push in converter > ISO6400 and finish in PS (focusmagic or even smartsharpening)... PS: in C1 I simply always switch sharpening off for output to .tiff

[ErikKaffehr]

Thanks!

You really mean amount = 5?

My gues is right now that the issue is twofold. I lift shadows a lot and I did lighten up the church using Auto Mask, I think that may cause the black spots. Will heck when I get home.

Best regards
Erik

I always use for A7R2 = 5/0.5/100/0 or zero amount for anything where nominal ISO + push in converter > ISO6400 and finish in PS (focusmagic or even smartsharpening)... PS: in C1 I simply always switch sharpening off for output to .tiff

[AlterEgo]

You really mean amount = 5?

yes, 5, very small amount... not claiming that it is somehow the best number, just what I use

[AlterEgo]

also NR:

I always use 5 (five)/50/50 for chroma with A7R2 where nominal ISO + push in converter <= ~ISO6400, may be up to 10/50/50 after that...

[ErikKaffehr]

Hi Torbjörn,

I did reprocess the images using RawTherapee, just with standard sharpening. In this case neither image is showing the staircase artefacts. So it is a demosaic issue in Lightroom.

To my eyes the colours are quite a bit different, but that is mostly a profiling issue. Also, my processing skills with RawTherapee are limited.

The enclosed image is a corner crop. Left Distagon 60/3.5 at f/8 and on the right the 24-70/2.8 ZA also at f/8.

It is interesting that the P45+ image showed less artefacts in Lightroom than the Sony, but in RawTherapee both look similar. I would say there is not a lot of difference.

Best regards
Erik

Erik, what has happened to the image comparing the A7rII to the Blad?

Under the heading: Can the Sony replace the Hasselblad? > Sony A7rII Central crop

Slanted lines are all jagged whereas the Blad looks distinctly smoother.

And the next shot with leaves. That leaf to sky transition is not looking good. Hope the uncompressed raw can fix that.

[ErikKaffehr]

Hi Hans,

Regarding the black spots I found out that the main contributor was that I have used the "Adjustmen Brush" with "Auto mask" to brighten up the chapel shown in the images. With the Sigma image lens corrections were also used, significantly lifting the shadow area where the chapel is.

I switched of adjustment brush and used the "Radial Filter" instead to achieve a similar effect, and using that filter eliminated black dots.

Regarding the jaggies most visible in the leaves of castle image in the MFD comparison I played around with sharpening and didn't really matter. After that I converted both images using RawTherapee with default sharpening and now jaggies were seen. So I would say that the Jaggies are an Lightroom demosaic artefact. Just to say, I also sharpened the images using FocusMagic and I had still no jaggies in the RawTherapee images. I feel there are some issues in Lightroom/ACR that Adobe needs to address.

Best regards
Erik

do not know what type of sharpening you use, or how accurate you focus, but in my eyes your 100% examples look horrible. jaggies, black spots and total lack of texture at other places. If have a R2 myself and never see these problems. My usual starting point with LR is 50/0.5/50/0, although it depends on the lens and f-stop. for f11 50/0.7/50/0 seems to work better and some lenses at f5.6 need less sharpening. am not suggesting you should use my numbers, all very personal. it is just that these numbers give me much nicer (in my eyes) results.

[ErikKaffehr]

Updates:

2015-10-13:

Readers have pointed out some artefacts:

Jaggies in some images
Black dot artefacts in some images
I have looked into the issues and found that the jaggies were caused by demosaic artefacts in Lightroom combined with strong sharpening. I therefore reprocessed the comparison with MFD images using RawTherapee with default sharpening. Default sharpeing in RT may be regarded weak, but it applies similarly to both images.

The black dots artefacts were caused by mainly by me using the local adjustment brush in LR with automasking to lighten up the chapell against the background. In this case I used the radial filter instead giving a similar effect, that made the black dot artefacts go away.

The images affected by this observations have been (hopefully) updated.

http://echophoto.dnsalias.net/ekr/index.php/photoarticles/84-my-sony-a7rii-journey

Best regards
Erik

[Ray]

I think anyone who gets an A7RII should also get the Sony/Zeiss 50.

There is an amazing comparison image which compares with the Nikon 810 and the 5DS (reset the menu). Actually the D810 is pretty close to the A7R2 on resolution and DR alone, except the A7R2 image is much sharper in the foreground for some reason. The 5Ds seems to have DR issues.

Edmund

Edmund,
One wouldn't expect any noticeable difference in resolution due to the increased pixel count of the A7RII, compared with the D810. The resolution difference between 36.6mp and 41.2mp, when scaled down to 6mp compared with approx. 6.8mp, would probably result in a pixel-peeping difference. However, as pixel count increases, the 'law of diminishing returns' kicks in, with regard to the sensor's contribution toward resolution. The contribution of the lens is more significant in these circumstances.

The reason why the foreground in the A7RII image is sharper is because any part of the foreground in the A7RII image,  that is compared with the same part in the D810 image,  is further away from the bottom edge of the frame.

The test is flawed in the sense that the position of each camera is not identical. The D810 has been tilted up slightly.

Just thought I'd mention this. I'm always happy to clear up any confusion on any issue.