Today I started testing the new Sony A7RII against the A7R and have not seeing much difference in DR (shadow recovery) between them at base ISO.
Update: Also check A7RII vs A7R HIGH ISO test here.
Here is a sample from an image that was exposed to the highlights (underexposed about 1 1/2 stops) and then pushed 5EV in Lightroom. The A7RII image was downsample to 36MP using bicubic.
All exposure settings and histogram were identical
Lens compensation set to OFF
Steadyshot set to OFF for the A7RII
White balance set to Daylight
Exposure: 1/320 f/8 ISO 100
Lightroom Default settings
Lens used: FE 35 f/2.8
See the crops below and let me know what you think.
The A7RII seems to have a bit more contrast. Fred, I think the advantage is in the high ISO region, especially past ISO 800, where the A7RII switches to a different signal converter to gain more DR/SNR. Base ISO 100, I suspect the A7RII would be same as the A7R and D810.
hiepphotog wrote:
The A7RII seems to have a bit more contrast. Fred, I think the advantage is in the high ISO region, especially past ISO 800, where the A7RII switches to a different signal converter to gain more DR/SNR. Base ISO 100, I suspect the A7RII would be same as the A7R and D810.
hiepphotog wrote:
The A7RII seems to have a bit more contrast. Fred, I think the advantage is in the high ISO region, especially past ISO 800, where the A7RII switches to a different signal converter to gain more DR/SNR. Base ISO 100, I suspect the A7RII would be same as the A7R and D810.
I will run higher ISO tests next. Perhaps we will see a DR increase there.
I was worried there'd be a drop...so a wash is fine with me. Quite a huge architecture change for this sensor, it's somewhat amazing it's so close. I think you'll see it pull ahead of the a7R at higher ISOs though looking at some of Jim Kasson's analysis.
nntnam wrote:
Nice test, Fred. Thanks for sharing it!
They look pretty close on my display (iMac 2012 & macbook 12"). I thinkk the A7RII shows a little bit more details in last picture.
This was a shadow recovery noise test though. It's likely the A7RII would show even more detail if I would upsample the A7R 36MP image to 42MP and compare both.
I did not check to see which is which, but they do look close. I feel like the examples in the left column may show ever so slightly more noise, but it is hard to tell.
Read the summary of 4th (A7R), 5th (A7Rii) and 6th (A7Riii/Axxx?) generation model if you want to know the different architecture between FSI EXMOR, BSI EXMOR R and BSI EXMOR RS versions. Pay attention on deeper pixel well to raise the DR of the 4th generation and whatever missing on the 5th which to be filled in by the 6th generation in near future: http://www.framos.com/en/news/news/single-view/article/what-is-sonys-technology-anyway.html
We don't expect much differences on low iso range but lot of DR improvement on higher end based on BSI structure.
mttran wrote:
Read the summary of 4th (A7R), 5th (A7Rii) and 6th (A7Riii/Axxx?) generation model if you want to know the different architecture between FSI EXMOR, BSI EXMOR R and BSI EXMOR RS versions. Pay attention on deeper pixel well to raise the DR of the 4th generation and whatever missing on the 5th which to be filled in by the 6th generation in near future: http://www.framos.com/en/news/news/single-view/article/what-is-sonys-technology-anyway.html
We don't expect much differences on low iso range but lot of DR improvement on higher end based on BSI structure.
I think you're reading this incorrectly, or at least predicting the future incorrectly.
My understanding is that the RS sensor will allow for more resolution scaling without reducing DR... or maaaaybe better DR scaling without reducing resolution. So it will more or less follow the pattern from the A7r -> A7r2 (but a bigger resolution jump possible). If Sony continues down the A7s/A7r split route, we might get an A7r3 that has similar DR to the current model, but with an 60-80mp sensor, and an A7s with a 24-36mp sensor with maybe a stop better DR than the current model.
What I don't expect-- if we see this in the next 24 months-- is to get an 80-100mp model, in 135-format size, usingTHIS tech to expand DR by much. That's pie in the sky-- or a handful of generations down the line, or some other technology-- stuff.
All of this at base ISO, of course. I expect that Sony may push it's high ISO DR more towards the potential shown in its base ISO... just like they did with the A7s and A7r2...
Fred Miranda wrote:
This was a shadow recovery noise test though. It's likely the A7RII would show even more detail if I would upsample the A7R 36MP image to 42MP and compare both.
Keep in mind that the crops posted here are JPEG files with minimum compression (Photoshop JPEG quality 12).
In Lightroom it seems that the A7R has a very slight advantage in DR at ISO 100 but it should be imperceptible in real world images.
Thanks for doing this review it has made me do some more testing and am amazed at the quality of the images produced at 800 ISO. Very little noise and definitely better density range to my eyes. Can't wait to see your tests in this area. Heading out now to do some shooting at 800 ISO to see how it handles the files more diverse lighting situations. Would be fantastic to be able to shoot at 800 ISO so many situations. I also tried 640 with very similar results.
