p.3 #1Sony A7Rii compared with Sony A7R and Canon 5DSr
timparkin wrote:
The base exposure was calculated using raw digger to ettr (ie not affected by tint). The exposure was then dialed down 6 stops. So the test is on 6 stops below typical ETTR practise.
If we assume that the author's methodology was correct, this would suggest that the A7RII is effectively operating at a base ISO of ~40 which would be a pretty big win for the A7RII.
p.3 #2Sony A7Rii compared with Sony A7R and Canon 5DSr
Fred Miranda wrote:
This test is a complete joke. However I do expect the A7RII to do substantially better when recovering deep shadows.
Why do you think so?
I see no flaw in the testing. Sure they used different shutter speeds but the reason for that is that the a7rII does not only have much cleaner shadows but also seems to retain the highlights better. Here is what they did:
"We exposed such that the brightest area of non-sky was exposed to the right (using rawdigger to check) and then stopped down 6 stops and took our ‘under’ exposure."
I think this is a valid method to test a camera's dynamic range.
I think everybody should decide for themselves how important DR is to them. I own a couple of photo books with absolutely beautiful pictures in it, all of them were taken with Velvia 50 which has somewhere between 6 and 10 stops of DR. My own pictures don't come close to the quality of the images in that book but none the less, some times I really enjoy the flexibility of my Sony's files .
p.3 #3Sony A7Rii compared with Sony A7R and Canon 5DSr
RustyBug wrote:
It sounds like they were equitably dialed down 6 stops ... but they were NOT equitably given the same amount of light (which has a direct influence on noise, iirc) @ capture to achieve that (same) ETTR.
Essentially, you are saying that a 50 ISO film has better noise characteristics than a 125 ISO film. In order to achieve a "correct" exposure in each, you are applying more light to the 50 ISO film than you are to the 125 ISO film and thus get less noise. Gee, where is the surprise in that.
Despite both cameras (A7R II & 5DsR) being rated at 100 ISO ... IF one requires more light to achieve the same level of exposure (the ETTR referenced in the test), then the actual ISO is lower.
If anything, this suggests (to me) that the Sony ISO may be even more of a "stretch" than their previous stretching.
As to the tint being "not effected" ... hmmm, -24 tint means a reduction in the green channel. Given that the green channel carries half of the weight in a bayer array, you really think that it would have no effect in producing additional underexposure, nor account for the color noise amplification we see after all the ratcheting is complete.
Maybe I'm off here, but whenever I use the Color Balance and take my green channel -24, things tend to get darker (and magenta) as the green channel moves left....Show more →
Any tint only gets recorded as metadata but doesn't affect the raw file. We used the raw data to work out which frame was just about clipping whites.
p.3 #4Sony A7Rii compared with Sony A7R and Canon 5DSr
They go to one of the most beautiful places on earth (Lake District) and shoot nasty images of an ugly cliff?? I wouldn't take anything from this test. By their own admission, it's not a real world (or realistic) test and we shoot the real world.
p.3 #5Sony A7Rii compared with Sony A7R and Canon 5DSr
Bones74 wrote:
They go to one of the most beautiful places on earth (Lake District) and shoot nasty images of an ugly cliff?? I wouldn't take anything from this test. By their own admission, it's not a real world (or realistic) test and we shoot the real world.
If you want to see nice pictures check out the author's flickr stream
Or have a look at http://www.onlandscape.co.uk, as a subscriber I have learned a lot from Tim and the other authors. The magazine is the least gear oriented and most manufacturer independent (well maybe a little focused on Fujis's Velvia 50 with it's 6-8 stops of DR;-)) I know of and full of nice or thought provoking images.
The test clearly demonstrates that the a7rII has better DR than the a7r and thats important information for me. Which is quite relevant in the field.
p.3 #6Sony A7Rii compared with Sony A7R and Canon 5DSr
Bones74 wrote:
They go to one of the most beautiful places on earth (Lake District) and shoot nasty images of an ugly cliff?? I wouldn't take anything from this test. By their own admission, it's not a real world (or realistic) test and we shoot the real world.
Jeez, people who try to test camera equipment really can't win. What have you provided for us to discuss, Bones?
Guy stays in and shoots controlled-lighting test-charts in a hermetically sealed cave... and gets panned for not just using the thing in the real world.
Gal goes out and shoots purely real-world images... and gets panned for not just staying in the cave and shooting a better, more controlled study.
Intersex person goes out and shoots semi-controlled semi-real world images... and gets panned for not adhering to any dogmatic view of how to test out a new camera.
Personally, I'm happy for each new test that someone is willing to go out of their way to conduct, document, write-up, and discuss with us. It's one more datapoint that I can then use my own needs and judgement to parse in developing whatever opinion I can before I have the said gear in question in my own hands.
p.3 #7Sony A7Rii compared with Sony A7R and Canon 5DSr
justruss wrote:
Jeez, people who try to test camera equipment really can't win. What have you provided for us to discuss, Bones?
Guy stays in and shoots controlled-lighting test-charts in a hermetically sealed cave... and gets panned for not just using the thing in the real world.
Gal goes out and shoots purely real-world images... and gets panned for not just staying in the cave and shooting a better, more controlled study.
Intersex person goes out and shoots semi-controlled semi-real world images... and gets panned for not adhering to any dogmatic view of how to test out a new camera.
Personally, I'm happy for each new test that someone is willing to go out of their way to conduct, document, write-up, and discuss with us. It's one more datapoint that I can then use my own needs and judgement to parse in developing whatever opinion I can before I have the said gear in question in my own hands.
I'm not particularly interested in discussing anything with you Russell and I'm not obliged to please you either. I don't have to like or appreciate every "test" undertaken and I'm quite within my rights to say as much on this forum without your attempted censoring. To me this particular test is unrealistic and pointless. Considering the area of the UK the testers were in, their choice of subject is also disappointing, with so many interesting and beautiful locations to choose from. Before you honk, I know the beauty of the test subject is not high on the list of test criteria. Dan, however is providing us with great test images and edited samples to peruse.
p.3 #8Sony A7Rii compared with Sony A7R and Canon 5DSr
Phillip Reeve wrote:
If you want to see nice pictures check out the author's flickr stream
Or have a look at http://www.onlandscape.co.uk, as a subscriber I have learned a lot from Tim and the other authors. The magazine is the least gear oriented and most manufacturer independent (well maybe a little focused on Fujis's Velvia 50 with it's 6-8 stops of DR;-)) I know of and full of nice or thought provoking images.
The test clearly demonstrates that the a7rII has better DR than the a7r and thats important information for me. Which is quite relevant in the field.
I will go check out the beautiful images, thank you I'm sure you already knew the Sony sensors have more DR before this "superfluous" test
p.3 #9Sony A7Rii compared with Sony A7R and Canon 5DSr
Bones74 wrote: Phillip Reeve wroteIf you want to see nice pictures check out the author's flickr stream
Or have a look at http://www.onlandscape.co.uk, as a subscriber I have learned a lot from Tim and the other authors. The magazine is the least gear oriented and most manufacturer independent (well maybe a little focused on Fujis's Velvia 50 with it's 6-8 stops of DR;-)) I know of and full of nice or thought provoking images.
The test clearly demonstrates that the a7rII has better DR than the a7r and thats important information for me. Which is quite relevant in the field.I will go check out the beautiful images, thank you I'm sure you already knew the Sony sensors have more DR before this "superfluous" test
I think you will enoy them.
What I didn't know was that the a7rII has significantly more DR than the a7r, thats why the test is useful to me.
p.3 #10Sony A7Rii compared with Sony A7R and Canon 5DSr
Bones74 wrote: justruss wroteJeez, people who try to test camera equipment really can't win. What have you provided for us to discuss, Bones?
Guy stays in and shoots controlled-lighting test-charts in a hermetically sealed cave... and gets panned for not just using the thing in the real world.
Gal goes out and shoots purely real-world images... and gets panned for not just staying in the cave and shooting a better, more controlled study.
Intersex person goes out and shoots semi-controlled semi-real world images... and gets panned for not adhering to any dogmatic view of how to test out a new camera.
Personally, I'm happy for each new test that someone is willing to go out of their way to conduct, document, write-up, and discuss with us. It's one more datapoint that I can then use my own needs and judgement to parse in developing whatever opinion I can before I have the said gear in question in my own hands.I'm not particularly interested in discussing anything with you Russell and I'm not obliged to please you either. I don't have to like or appreciate every "test" undertaken and I'm quite within my rights to say as much on this forum without your attempted censoring. To me this particular test is unrealistic and pointless. Considering the area of the UK the testers were in, their choice of subject is also disappointing, with so many interesting and beautiful locations to choose from. Before you honk, I know the beauty of the test subject is not high on the list of test criteria. Dan, however is providing us with great test images and edited samples to peruse.
Sorry I hit nerve! (Side note: birth certificate says "Russ," hence, just...; I do understand that Russell sounds a bit more patronizing, so I'm willing accept it as a gesture of friendship!)
In any case, I think we got off on the wrong foot, as it wasn't intended as an attempt to censor you... just a "let's keep it in perspective that someone went out and tried to provide something useful for us and maybe we'd be better not discouraging those folks from trying again just because the tests they went out of their way to perform don't match our hydra-headed (and disparate) views of what a great test is."
On that note, I'm happy to discuss, debate, etc with you in the future, Stuart. I'm all for viewpoints other than my own.
And at least we agree on one thing, Dan (and Tim, for that matter) does provide some nice images to look at!
p.3 #11Sony A7Rii compared with Sony A7R and Canon 5DSr
Fred Miranda wrote: retrofocus wroteFred Miranda wrotezlatko wroteIt's obvious the Sony sensor is better for these extreme scenarios -- I don't doubt that. But why would someone choose to base their camera comparison on this one extreme test (and then rig the comparison by underexposing the Canon an additional 1.3 stops)? Is 5-stop underexposure recovery the new standard for camera comparisons? The author may as rationally have chosen some other extreme comparisons:
- comparison of viewfinder lag
- comparison of battery life in the Arctic
- comparison of image stabilization at 800mm
- comparison of speed of manual selection of autofocus points
- comparison of reliability of recording to dual cards
- comparison of durability when getting trampled by wildebeest
- etc.You are being too rational. Once DXOMark posts the overall score for the A7RII, that is the magical number the media will focus on...and it's meaningless.Not sure I would say it is meaningless - problem is that the algorithm how this number is calculated for each camera is not known. But I suspect that all cameras of the same category like full frame are determined by the same calculation method. Of course we can argue one feature should have a higher or lower weight - for example DR. Seeing the recent rating of the 5Ds(R) series, it appears that DR is more heavily weighted in this score - not a bad thing for newer FF camera models IMO.I say meaningless because it's subjective to what the tester believes is important.
For example, you mentioned DR. A sport shooter or wildlife photographer would not care about base ISO dynamic range. If that carries too much weight to calculate the overall score number, it would be misleading for that type of photography.
In case DxOMark weightens the DR to a certain degree in this score, I agree with it independent on the usage of the camera and for what kind of photographers it is mostly made for. Because it acknowledges technology innovation. Such score cannot take into account the type of photography it will be mostly used for - it should be measured purely on technology impact (which includes resolution, DR, AF capability, image stabilization, low/high ISO performance, noise reduction etc.). The more innovative a camera is in those and other specs, the higher the score. I find this approach absolutely valid and those scores justified.
p.3 #12Sony A7Rii compared with Sony A7R and Canon 5DSr
justruss wrote:
Sorry I hit nerve! (Side note: birth certificate says "Russ," hence, just...; I do understand that Russell sounds a bit more patronizing, so I'm willing accept it as a gesture of friendship!)
In any case, I think we got off on the wrong foot, as it wasn't intended as an attempt to censor you... just a "let's keep it in perspective that someone went out and tried to provide something useful for us and maybe we'd be better not discouraging those folks from trying again just because the tests they went out of their way to perform don't match our hydra-headed (and disparate) views of what a great test is."
On that note, I'm happy to discuss, debate, etc with you in the future, Stuart. I'm all for viewpoints other than my own.
And at least we agree on one thing, Dan (and Tim, for that matter) does provide some nice images to look at!
It's all good Russ, sorry I had a toddler screeching in my ear so I was probably a little more "nervy" than usual! I agree we should appreciate that folks do get out and test gear "on our behalf" and they do so to the best of their ability and hopefully with no agenda. This particular test just didn't sit well with me, but this guy sure does have some great Lake District shots
p.3 #13Sony A7Rii compared with Sony A7R and Canon 5DSr
No surprise that the Sony A7RII is THE Best choice for Landscape/Architecture photography at base iso. No question about it. A bit surprised that the A7RII has MORE DR than the A7R from the looks of things since the A7R already had significantly more DR at base iso than any Canon ever made.
The sensor in the A7RII is just amazing (again from what has been posted so far including RAW files that I have played with) and cutting edge. Best sensor out there bar none.
Does that make the 5DS/R useless? Of course not and if I did a lot of work using iso800 and above it would be my choice. Also for very long exposures the Canon is superb. Best Canon made so far.
p.3 #14Sony A7Rii compared with Sony A7R and Canon 5DSr
Phillip Reeve wrote:
What I didn't know was that the a7rII has significantly more DR than the a7r, thats why the test is useful to me.
This.
I must say, though, my initial reaction to the test is that it feels like 2012 all over again. Here we are in 2015 doing a DR stress test comparison between a Sony and Canon sensor. It's like racing a cheetah vs an african wild dog over 100m and stupidly asking, "Hmmmm, I wonder who will win?" Canon is simply not going to be competitive in a low-ISO DR stress test against Sony sensors. In that way I cannot help but regard the test as rather idiotic. It is a test that says something overwhelmingly obvious that didn't need saying. It's like a lens review that "discovers" that the Canon EF 11-24mm f/4L USM has a fantastic build quality (wake me up when an L lens is reviewed to have a poor build quality, otherwise: snooze).
There are other camera stress tests between the 5DS R and a7R II that I am more interested in. To re-use the analogy: like racing a cheetah against a wild dog at longer distance....who wins in that scenario (thinking analogously to overall build-quality endurance and longevity)? Sony claims the a7R II shutter is rated to 500,000 clicks, but does it really possess better longevity than a 5DS R? I truly truly wish to know (and if anyone can definitively tell me either way, please do so!).
But, back on the DR stress test. Yes, the only thing that interested me in the test is the a7R vs a7R II. Impressive difference there! This test, actually, is my first clue that Sony has improved in their own sensor dynamic range. For a while, it seemed that Sony themselves had plateaued - no major differences since 2012 it has seemed...until now. Very exciting!
Regarding the 5DS/ R, my thoughts are simple: can I produce amazing imagery with that camera? Yes, yes I can! DR stress test aside I am extremely pleased with my 5DS and it is now a valuable part of my gear. I am likewise highly enthused by the a7R II, and look forward to adding it to my bag as well. I am on the verge of taking an early retirement and will have limited funds going forward for a few years, so I couldn't be more pleased that both of these awesome cameras were released here in 2015.
p.3 #15Sony A7Rii compared with Sony A7R and Canon 5DSr
good to see many people thinking about this test.
as I understand it, the ARii was given more exposure (ETTR) than the Canon and Sony A7R because it holds the highlights better - indicating it has better dynamic range at least in the highlights. The Canon/Sony (earlier model) were exposed for shorter period because they clipped the highlights at a similar shutter speed...From that point, all the cameras were "pulled back" 5 stops and images processed.
I understand how some folks claimed that it was an unfair test since the shutter speeds were not the same - more exposure was given to the ARii than the other two sensors.
Someone over at DP Review had interesting comments:
"So basically, the A7RII has a little more than a stop more headroom at the top before clipping, which allowed him to expose 1/1000 vs. 1/2500 at the same ISO, capturing similar highlights.
"This does give it an advantage for landscapes especially, since you're not needing fast shutter speeds anyway. You can shoot slower to capture the highlights and give the shadows an extra boost.
"However, light gathered really is the main determinant of noise in the shadows and he's chosing to give the A7RII an advantage in the shadows thanks to its headroom. In reality, we're often in situations where we can't expose for as long as we wish, especially handheld. For these cases, it would be nice to see the tests normalized to the amount of light gathered. It would be interesting to normalize the other cameras to the A7RII's ETTR exposure and show that they blow out the highlights at full exposer and still don't have as nice looking of shadows at -6EV. Just glancing at the article, it's pretty easy to think that the shadow digging is at the same exposure since the article doesn't mention this. It should be made more clear IMO."
p.3 #16Sony A7Rii compared with Sony A7R and Canon 5DSr
Phillip Reeve wrote:
The test clearly demonstrates that the a7rII has better DR than the a7r and thats important information for me. Which is quite relevant in the field.
I'm going to get an a7rII myself, but I'm not totally convinced about this yet going by just this test.
I did take a look at the raw files of both the a7r and a7rII and while it looks promising for the new camera, I would prefer to check out a scene done at the same exposure so the same amount of light reaches the sensor. But maybe I'm just too Old School to get it!
p.3 #17Sony A7Rii compared with Sony A7R and Canon 5DSr
rdcny wrote:
as I understand it, the ARii was given more exposure (ETTR) than the Canon and Sony A7R because it holds the highlights better - indicating it has better dynamic range at least in the highlights.
But the A7R II shot's highlight is all washed out while the Canon one retains a fair level of color and details, indicating that he did over-expose the A7R II to get his desired results.
p.3 #18Sony A7Rii compared with Sony A7R and Canon 5DSr
OK checked out the raw files again. First you open them in ACR/LR and set Exposure to +5 and Shadows to +100. Then you move the Whites slider up until it starts to clip the highlights in the sky.
They all do that around +25 give or take. But it is a bit tricky to rely on since it clips the highlights in different areas so the skies have moved between shots. This also means that the light could be a different between shots even on an overcast day.
p.3 #19Sony A7Rii compared with Sony A7R and Canon 5DSr
Native ISO 40 would be huge especially for long exposures. It would allow double exposure times without ND filters. But if the A7rII has such exposure latitude, why is ISO 50 advertised as extended ISO? Wouldn't it be natural to extend the ISO to 25?
p.3 #20Sony A7Rii compared with Sony A7R and Canon 5DSr
alundeb wrote:
Native ISO 40 would be huge especially for long exposures. It would allow double exposure times without ND filters. But if the A7rII has such exposure latitude, why is ISO 50 advertised as extended ISO? Wouldn't it be natural to extend the ISO to 25?
It would, but long exposures... is where that 12-bit thing pops up, right? Not exactly ideal.
EDIT: On second read perhaps you're not talking about bulb exposures.