The GetDPI Photography Forum

Great to see you here. Join our insightful photographic forum today and start tapping into a huge wealth of photographic knowledge. Completing our simple registration process will allow you to gain access to exclusive content, add your own topics and posts, share your work and connect with other members through your own private inbox! And don’t forget to say hi!

Sony A7r sensor not the same

Guy Mancuso

Administrator, Instructor
Went to Samys here in La to pick up the Sony 43 flash works a charm. Anyway I actually talked to someone that knew something. From the reps mouth the A7r is NOT the same sensor as the D800e it's 3 years newer and specifically designed for this cam and the new processors and image chain. Honestly just looking at the files I never thought it was myself. To me it's actually better.

Thought I would share that
 

k-hawinkler

Well-known member
Went to Samys here in La to pick up the Sony 43 flash works a charm. Anyway I actually talked to someone that knew something. From the reps mouth the A7r is NOT the same sensor as the D800e it's 3 years newer and specifically designed for this cam and the new processors and image chain. Honestly just looking at the files I never thought it was myself. To me it's actually better.

Thought I would share that
Comparing images between my D800E and A7R files I get the impression the D800E files have more bits per pixel.
 

GrahamB

New member
DXO measures the color depth of the a7r at 25.6, identical to the d800e, and .9 bits behind the 16 bit Phase One IQ 180 back.

"The Sony is in joint fourth place for color depth alongside the Nikon D800E. These are very high scores indeed and are only fractionally behind the Phase One P65 Plus models with their higher 16-bit A/D conversion. However, bear in mind that the P65 Plus models aren’t the latest from the firm."

Note: as the Sony and Nikon have identical color depths, they share 3rd place, only bettered by medium format backs.

Sony A7R sensor performance - DxOMark

Graham
 

k-hawinkler

Well-known member
DXO measures the color depth of the a7r at 25.6, identical to the d800e, and .9 bits behind the 16 bit Phase One IQ 180 back.

"The Sony is in joint fourth place for color depth alongside the Nikon D800E. These are very high scores indeed and are only fractionally behind the Phase One P65 Plus models with their higher 16-bit A/D conversion. However, bear in mind that the P65 Plus models aren’t the latest from the firm."

Note: as the Sony and Nikon have identical color depths, they share 3rd place, only bettered by medium format backs.

Sony A7R sensor performance - DxOMark

Graham

Thanks Graham. The reason I had the impression was that when processing a file with CS6 it showed intermittently a pixelated view. I had never seen this kind of behavior with an D800E file.

Thanks for the correction.
 

fotografz

Well-known member
Went to Samys here in La to pick up the Sony 43 flash works a charm. Anyway I actually talked to someone that knew something. From the reps mouth the A7r is NOT the same sensor as the D800e it's 3 years newer and specifically designed for this cam and the new processors and image chain. Honestly just looking at the files I never thought it was myself. To me it's actually better.

Thought I would share that
It seems the A7R sensor assembly with angled micro-lenses and gapless pixels has to be different given the distance to sensor compared to a 35mm DSLR ... no?

No matter ... that you can get D800 sort of performance in a camera this small seems quite a feat.

Have to go back and find Jack's thread where he recommended the D800 over the D800e because you could match the e version in post. Don't remember what adjustments he recommended, but it'd be interesting to see whether they would apply here.

- Marc
 

Shashin

Well-known member
Thanks Graham. The reason I had the impression was that when processing a file with CS6 it showed intermittently a pixelated view. I had never seen this kind of behavior with an D800E file.

Thanks for the correction.
If you are processing, then the preview could very well be 8-bit until you apply the change and the software recalculates the values. This is how Photoshop works.
 

Jack

Sr. Administrator
Staff member
To be clear, I never recommended the D800 *over* the D800E as I believe they each have their place. I only indicated that with a little extra work in post, you could get the D800 to the same practical level of resolution performance as the E, or get the E to the same level of smooth tonality as the non-E. (At that time, I probably indicated they were close enough I personally did not see a strong need for owning both, but did anyway for convenience.) What the non-E it requires to meet the E in high-frequency detail rendering is roughly 30% more amount on your high-freq capture sharpening routine (0.4 to 0.6 pixel radius at low threshold ~ 1); AND increasing clarity about 6 to 8 points to subdue the slight AA filter veiling. (Reverse that with the E to be smoother toward the non-E rendering.)

I am meeting with Guy in a few weeks and will play with some Sony raw files and share whatever capture sharpening settings I think make sense here. Or in the meantime, if somebody wants to send me a well-captured raw file with significant areas of high detail, I can look at it over the weekend. It might actually be a good exercise for this group of owners -- we could have a processing contest of sorts to help each other understand what works and what won't for their needs.

PS: Just to clarify my stance re the D800, I actually still own one of each body and use dedicated capture adjustment sets for each camera to render to their respective strengths. My personal take is that for 90% of all work, they are essentially identical in practical application. There are perhaps 5% of the time the D800E is going to be superior and 5% of the time the D800 is going to be superior, but in either case the amount of superiority while seen subtly at the pixel level is almost insignificant in even the largest print save for subtleties in image undertone. Generally speaking, I use the D800E for landscape, architecture and art-repro work and I prefer the D800 for people anything, B&W conversions and long tele nature work. But as a practical matter I could easily shoot everything with one, I'd just have a difficult time deciding which it would be :D
 

fotografz

Well-known member
To be clear, I never recommended the D800 *over* the D800E as I believe they each have their place. I only indicated that with a little extra work in post, you could get the D800 to the same practical level of performance as the E. What it requires is roughly 30% more amount on a high-frequency capture sharpening routine (0.4 to 0.6 pixel radius at low threshold ~ 1); AND increasing clarity about 6 to 8 points to subdue the slight AA filter veiling.

I am meeting with Guy in a few weeks and will play with some Sony raw files and see what capture sharpening settings I think make sense, and then share them here. Or in the meantime, if somebody wants to send me a well-captured raw file with significant areas of high detail, I can look at it over the weekend. It might actually be a good exercise for this group of owners -- we could have a processing contest of sorts to help each other understand what works and what won't for their needs.

PS: Just to clarify my stance re the D800, I actually still own one of each body and use dedicated capture adjustment sets for each camera. My personal take is that for 90% of all work, they are essentially identical in practical application. There are perhaps 5% of the time the D800E is going to be superior and 5% of the time the D800 is going to be superior, but in either case the amount of superiority while seen subtly at the pixel level is virtually insignificant in even the largest print save for subtleties in the final image undertone. Generally speaking, I use the D800E for landscape, architecture and art-repro work and I prefer the D800 for people anything, B&W conversions and long tele work. But as a practical matter I could easily shoot everything with one, I'd just have a difficult time deciding which choice it would be :D
My bad Jack ... right, you didn't recommend one over the other ... just that for many, the D800 could work as well ... etc. as you explained above.

I think the whole exercise in what levels of sharpening etc. would be very helpful ... I recall the same efforts were needed with the S2 files which tended to be over-sharpened, or sharpened the wrong way for those specific files.

- Marc
 

Guy Mancuso

Administrator, Instructor
We will actually have the Leica S in our hands for a update review in about 10 days. My one complaint on the Sony is the lossless compression. I'm hoping they give us a option. Now they may be different sensors and I maybe able to see a slight difference I would not make this a one is better than the other hyper thread. What I think Sony did was just change the whole engine in camera processing and such. What they did to the actually sensor I don't know but let's remember also the marketing spin on release products. Take everything with a grain of salt if you know what I mean . I love this little thing but am I totally convinced its better. Than no I am not . I went after features now let's see how the files are is still something I need to get through. That answer won't come till I get a lot if shooting done which I have not done yet. 3 weeks and I will have a much better idea as I will have shot a lot by than.
 

Bob

Administrator
Staff member
We will actually have the Leica S in our hands for a update review in about 10 days. My one complaint on the Sony is the lossless compression. I'm hoping they give us a option. Now they may be different sensors and I maybe able to see a slight difference I would not make this a one is better than the other hyper thread. What I think Sony did was just change the whole engine in camera processing and such. What they did to the actually sensor I don't know but let's remember also the marketing spin on release products. Take everything with a grain of salt if you know what I mean . I love this little thing but am I totally convinced its better. Than no I am not . I went after features now let's see how the files are is still something I need to get through. That answer won't come till I get a lot if shooting done which I have not done yet. 3 weeks and I will have a much better idea as I will have shot a lot by than.
What is the trouble with their lossless compression?
-bob
 

Guy Mancuso

Administrator, Instructor
I just read that its not the same as the Nikon process. I have to look into this thigh. Im just always leery of compressing files. Guess we should look into this and make sure I have the correct in info. I just read some whining about it
 

Bob

Administrator
Staff member
I just read that its not the same as the Nikon process. I have to look into this thigh. Im just always leery of compressing files. Guess we should look into this and make sure I have the correct in info. I just read some whining about it
Well, if it is lossless it is a good thing since it may very well reduce the write-time to card.
I use it exclusively after convincing myself that it made not a bit (literally) of difference.

There are many lossless compression algorithms, several which are used in tiffs.
I can't think of a software title that is not shipped in a compressed version.
The fear and loathing of lossless compressed files is baseless and probably based on the crappy jpg compression algorithms (no, no I don't want any compression please if it is like that). That is one reason that jpeg 2000 now supports a lossless compression option.
-bob
 
Last edited:

Guy Mancuso

Administrator, Instructor
Your right Bob jpegs have received such a bad rap on this subject it trickles down to other areas in the digital arena.
 
D

Deleted member 7792

Guest
Or in the meantime, if somebody wants to send me a well-captured raw file with significant areas of high detail, I can look at it over the weekend. It might actually be a good exercise for this group of owners -- we could have a processing contest of sorts to help each other understand what works and what won't for their needs.
Jack, I'll send a PM with a Dropbox link to a raw file from the A7R.

Joe
 

GrahamB

New member
My one complaint on the Sony is the lossless compression.
Guy, I'm not satisfied that the compression is lossy. Although I haven't read the current "pundit" opinions, the same discussion was raised with the introduction of the Sony a77 and Sony a99.

At that time, Sony's UK rep, Paul Genge, when directly asked about the properties of the new ARW files, replied they were "lossless".

I'm not interested in revisiting the minutia of raw discussion. I would be interested to see a side by side comparison of the output, preferably printed, from a Nikon d800e, and Sony a7r. Use the same lens on both, and in all things, make the shooting parameters equal. Prove how the a7r image is at a disadvantage from the raw compression.

I'm not holding my breath.

Graham
 
Last edited:

Bob

Administrator
Staff member
Almost all (all AFAIK) raw files are just wrapped tiffs. The compression scheme can be inspected by looking at the tiff metadata.
-bob
 

anGy

Member
It would be really interesting to share A7r development settings.
I'm comparing it with my S2 and face several issues (some colors are off with the S2 files in C1 (using the Leica S2 profile for C1) Some difficult WB adjustment with the A7r in C1 f.i. portrait with flash: if WB set for a good skin tone, the rest of the picture looks too cold/magenta / etc).
Start to think that LR best C1 for those 2 camera files (quite obvious in color accuracy, but also maybe color/tone transitions). C1 however best LR for details extraction. :confused:
 

Jack

Sr. Administrator
Staff member
Re sharpening. Joe was kind enough to send me a file with lots of high-freq detail taken with the 35 at ISO 400 in moderate light. I've only looked at this single file and for a short period of time, so it's dangerous to draw broad conclusions -- but I will do so lightly for the purposes of starting this discussion. With your input and more images it is subject change over time.

What I see initially is a file that appears to have a stronger AA filter than the D800. It requires fairly heavy-handed sharpening and a good bump in clarity to clear it up. Also, base saturation needed a slight bump -- this is likely the debayering engine design. Also, noise seems to be a little lower than the D800's, possibly due to same AA, or possibly due to a better debayering engine.

That said, here is the initial set of adjustments I worked up for the A7r in C1 to generate a highly detailed but not brittle image:

1) Saturation to +8
2) Clarity to Neutral +20 (This is higher than the D800 by almost twice, and one factor that leads me to believe a stronger AA. The other danger of going higher in Clarity is you net reduce DR.)
3) Sharpening 250/0.6/1.0 (This is very similar to the D800, and helps confirm an AA filter of some sort.)
4) NR at L20/C15/Single Pixel 1 (This is a modest 5 point higher L than the D800/E, but helped add slight smoothness back to the overall look -- could be image or ISO dependant.)

Again, much of this starting point could be due to the nature of the image itself, so let's keep it mainly as a starting point for discussion. Hope it helps and I look forward to hearing your findings!
 
Jack, in theory the A7r shouldn't have an AA filter. This notwhistanding, while performing comparison with the d800 shooting test targets, I noticed that the quantity of aliasing was lower in the A7r than in the d800, using the same lens. At the same time, in my experience, files from the A7r require less sharpening than those from the D800, so I suspected that Sony has found a way to counteract moire without an AA filter.(Raw converter ACR)
Sergio
 
Top