Is the new Sony 5 axis image stabilization really a deal breaker? Just watch Matt Granger’s video above to hear his opinion on it!
And Diggloyd is asking Sony why the new Sony A7R II has no Pixel Shift:
Conspicuously missing from the Sony A7R II feature set is Olympus-style pixel shift, which in my testing can deliver roughly a 32-megapixel image from a 16-megapixel sensor. On the 42-megapixel Sony A7R II sensor, a similar technology would deliver ~63 megapixels, or maybe more like ~80 megapixels using a Zeiss Otus.
It’s a pity that Sony is not offering pixel shift, since it has Olympus-style sensor stabilization. Perhaps it will appear in a firmware update, but that seems dubious, since it is a terrific feature to brag about at a product release (and would be a first for a full frame camera).
A7rII and RX camera preorder list:
In USA:
A7rII at Amazon, BHphoto, Adorama, SonyStore US, FocusCamera, Uniquephoto.
RX10m2 at Amazon, BHphoto, Adorama, FocusCamera, Uniquephoto.
RX100m4 at Amazon, BHphoto, Adorama, FocusCamera, Uniquephoto.
In Europe at:
A7rII at Sony DE, UK, FR, IT, ES, NL, BE, CH, AT, SE, FI, NO, PT. Wex UK.
RX100m4 at DE, UK, FR, IT, ES, NL, BE, CH, AT, SE, FI, NO, PT. Wex UK.
RX10m2 at DE, UK, FR, IT, ES, NL, BE, CH, AT, SE, FI, NO, PT. Wex UK.
In Asia at:
A7rII at Digitalrev, CameraPro.
RX10m2 at Digitalrev, CameraPro.
RX100m4 at Digitalrev, CameraPro.
91 Responses
The stabilizing from in body compared to in lens is different with longer focal lengths :I
Also, the way the video highlights in body stabilization in general as a new feature for persistent users of cameras without such: canikon, makes it impossible for important qualifications to be addressed. This is a clear case of good news, bad news, and the bad news which will follow may even make the somewhat useful wide angle stabilization presented seem unnecessary and the long focal length lens stabilization ineffective. In fact, Sony A7 cams don’t need it for great shots (not least with regards the new bsi sensor sensitivity!) and it can lead to over reliance on longer shutter speeds than would stop motion, and we will see this next as a reported ‘problem’ lol. Also not covered is the matter of how the new bodies effect balance of handholding with lenses and of course that will be the BIGGEST ‘problem’ invented where it can, and obsessed about where it can even be found, while the in body and lens stabilization actually of course offer great all round stabilization for our camera systems not offered by other 35mm camera systems at all.
Digilloyd is particularilly pernicious in that we have hardly got the new camera out and he demands something done by olympus and only effective with COMPLETELY static shooting like its an omission or faux pas on the part of Sony offending his greater marketing and camera system knowledge :I and then suggests using the Otus for it XD
The only thing conspicuous about that, is that having such a high resolution and quality image producing cam from Sony on a tripod for product shots and image copying and such would have the Sony cams considered as a very serious system indeed for a small partition of users, does he consider it serious? NO, it doesn’t even have 14bit like he says, like medium-formica-nikon hahahahahahahaha
One clear benefit of stabilizer in general aside from for video, is focusing through the viewfinder for instance, as it helps keep the subject still with longer lenses 😀
The fact is already, with all the new features of A7 cams, and whether all are implemented in a new body or not, the cameras are almost unbeatable for the likes of canikon and rather you could consider Pentax’s new 35mm sensor cam, with weather sealing, optical viewfinder and so on alongside in body stabilizer and perhaps those similarly great Sony sensors lacking in say canon? And I wonder how well that in body stabilizer works with the new lenses, perhaps with image circles made for it? 😀
Do you really want us to read that? o_O
either crap or too much, difficult this guy
I must say this is his better post, you should have seen what he wrote on other comment section~
Some of you real identity or not think I am not a photographer, least of all the best here or best anywhere, but that will also ‘surprise’ you won’t it. Photography seems to be a lot about the gear and I love lenses and photography gear so it has been a natural progression to understand that as well as I do photography, after all, I have more gear than subjects at the moment.
You r really going to town with the comments 🙂
…
Everyday @sonysonysony:disqus sucks someone into a long-winded debate. Why even argue??? I think it’s funny.
I’m laughing my head off. This guy is merciless. At first I thought he was serious, now I just glance his comments across and reading the highlights make me laugh – a new type of troll?))))
Yes, I even intend to use constant lighting as main light source for portrait….everywhere….wooooooh :U
Is it really a person? – it looks like one of those Article Writing Programs that is not properly tuned.
That’s a really log-winded way of saying the article is mostly nonsense and that criticisms of the Sony system are just through people inventing problems.
This is one of the very few posts by ‘I.O.’ that I have understood and, mostly, agree with!
No, its not like that at all. All gear made for any purpose has a certain level of utility and there are not many things that work exactly for purpose. With scissors you may have good and bad and you may even find some too sharp for what you want to cut with them. I want to shoot great images and so my idea for a camera and lens systems assumes that first. You assume a lot of nonsense instead of that. Sony makes a fantastic camera system and knows it better than we do, well, maybe I know something of it too, as my ‘predictions’ proved 🙂
Sorry, I thought for one brief moment that you might have had a lucid moment.
Unfortunately, it appears I was wrong.
yes another fail on all accounts for you lot.
Quite the contrary: I never argue with myself.
Even if I did find myself in the very strange position of disagreeing with my own beliefs, I’m sure that I would make excellent arguments on both sides!
Or always find yourself accepting someone else’s argument instead of having your own. We can consider hypothetical cameras and base upon the features we understand already and can estimate cost of, or just work our ideal camera solution out in the manner of supposing it in the upcoming camera before it is revealed: by 5 months.
So, there is no time in your life when, having heard an excellent and well thought out argument, you have accepted it as such?
If you haven’t, then you’re either a monumental fool, stubbornly contrary, or a pointless troll. Which is it?
I argued correctly for all present tech in discussion and for the future tech as you will find out. Medium format cam, even before Sony 35mm sensor system is complete. APSC going… I wonder what might replace it at top? No one will want it and you won’t get it. Great, so only those interested can speak of what it should comprise of till that point.
Why do you repeatedly edit your posts?
They make little sense initially but, having altered them, their relevance is even more obscure.
I asked a question relating to your criticism of me. However, you have changed the post and not answered me as a consequence. Why do you bother at all?
Did you accept my arguments then, no you didn’t, you were proven wrong and a charlatan. The argument for best lenses and cameras is based upon complex matters, and I can do it, you can’t. I made a lot of sounds with modular synthesis and seemed to have learnt from that process. I have said this and mentioned that the complexity may overload the cpu or certain components so I had to master using best fit arrangements or forgoing reduction to keep something yet sounding interesting. I rearranged the interfaces a lot, which is itself complex as the modules themselves had to represent the interface also, and perhaps that’s how I learnt so much. I did it because I enjoyed it. Its not just that though is it, I have been working on these matters for many years. It returned to the start with digital images and boom. Everything I know or learnt from extraneous sources about music tech, I did so by reading music tech magazines, in fact the more reader friendly ones like future music.
I’m no mathematician or even technician as such, i’m the artist who proves mastery over other disciplines, for itself. Representing photography, now 😀
I found problems even in those situations, designing, and those form implications for considering best practices in design, AND tool making.
Please, seek help. Your medication is wearing off.
Two whack jobs having a pointless, protracted argument on a forum.
This is what the internet was made for! lol
One real person, a master photographer, and a lot of not real, fooling not so good real photographers hahahahahaha exactly!
If Sony has the better system in A7 cameras, then vouch for that with only some qualification and never see arguments made for competition contrary to that better system. This assumes you haven’t learnt wrong permanently already: you have, collectively. This Isn’t for you
They are costing all real photographers, not just me personally.
Matt, (if you are reading this) why didnt you use inbody with lens stabilization to achieve far greater stabilization?
Exactly.
Try reach him on youtube, he reads some coments.
The A7II doesn’t let you selectively choose IBIS vs OSS as your stabilization method – if you turn off SteadyShot on the body then it disables both IBIS and OSS, so Matt’s IBIS vs OSS comparison is invalid – he’s actually comparing no stabilization vs IBIS vs no stabilization again.
This guy is looking for lots of clicks now he got some early samples of the Batis lenses. However he uses the Sony system the first time (Sony lend him bodies) and he is clearly lacking knowledge about the system. In his last videos that’s really obvious, he didn’t even know the A7r and A7s only have contrast detect AF.
He might be a good photographer I don’t know, but he’s talking too much about gear in my opinion.
He is very good but he was known as “that nikon guy” or something close to it only months ago. He changed it to his regular name.
This is not what a professional should do.
When I saw this video some time ago I asked why he did not compare any stabilized lens like Nikon 70-200 VR on Nikon Body vs A7II.
D750 vs A7ii? The former focus down to -3 eV and latter around -1 eV. They even have different lens (ultrasonic vs linear motor). How are you going to fairly compare the two?
If you didn’t get it – Matt tested image stabilization, not autofocus.
How could we miss it?! If there’s one thing the a7rii is missing, it’s resolution…
“It’s a pity that Sony is not offering pixel shift, since it has Olympus-style sensor stabilization.”
Sony does NOT have Okympus style sensor stabilisation, so Diggloyd’s “conspicuousness” is completely invalid. The systems are completely different.
Absolutely. It may even be that he is the ONLY person to have noticed this ‘conspicuous’ omission!
For the occasions that such a facility would be useful I’m much happier they keep costs down as much as possible. This is one hellish expensive camera anyway, so including this type of ‘extra’ would make matters much worse. As you say, the Olympus system is different and not FF.
They better have the same IBIS system on next A-mount body~!!!
(A mount death flame war begins)
We all know this won’t happen 🙂
Dam straight!
Who or what is A-mount?
I am glad that Sony focuses on other firmware features at the moment, that seem to have much higher priority for most users…
I agree. This sensor shift stabilistation is a gimmik that nobody is asking for in this camera
At least I hope to see it in form of an App, if it’s possible. They need to stay open to every possibility.
How could IBIS be a deal breaker?
Maybe he meant deal maker?
IBIS? Down with this sort of thing.
Careful now.
Its a deal breaker me!
Why Pixel Shift on an A7R II?
A camera who’s sensor already resolves 42 MP natively doesn’t need a function like pixel shift. It simply doesn’t make sense as there exists no lens that would deliver more than 30-40 MP resolution on its projected image. And here we are talking of the very best fixed lenses available. Zooms tend to reproduce not more than 20-25 MP worth.
So what should pixel shift be good for here?
“there exists no lens that would deliver more than 30-40 MP”
This is completely and demonstrably wrong… otherwise there would be no APS-C cameras with more than 18 megapixels.
Lenses for smaller sensors are designed for higher resolution than full frame sensors. That’s why they are comparatively large in relation to the image format they are able to cover. If you would design a full frame lens with the resolution of a MFT lens you will end up with Otus-like 1 kg monsters for a 50mm fixed lens….
Usually you are limited to peak performance at a certain f-stop. Wide open aberrations limit resolution, stopped down diffraction. We know MTF is product of lens*sensor MTF, so you gain by increasing sensor resolution. The Otus mounted on the 5ds (lensrentals.org) showed only 20% resolution increase over 5diii, a factor 1.5 were optimum. Weaker lenses show inferior performance especially in the corner region. But for half pixel shifts your lens needs to resolve about twice as much to take full advantage. So I expect especially better color accuracy and with the best lenses at a certain f-stop a 20-30% increase in resolution. Nevertheless, if we would get the resolution for free and considering the fact that moire/aliasing is significantly reduced, why not take it?
Pixel shift is desirable for still life photography like product’s. It’s not so much about more pixels as it is about truer color data captured.
It’s a consumer feature at the end of the day, with inherent compromises. If you need more than 48mp, you’re going to be looking at medium format.
Not if I could get similar results from an a7r ii for much less. pixel shift originates from hassleblad medium format. It’s a professional feature. Also again don’t focus on mpix, it’s about correcting flaws native to bayer sensors.
I totally agree that pixelshift is for correcting bayer grid flaws but I do not get why people ask for it before the product has been released. This is a feature that everyone can mimic by superresolution technique on every goddamn camera even a cell phone with a little more inconvenience. Hte hardware is in place in a7II and a7rII. This will be a firmware patch in the near future with a very high probability. In the meantime you can just take 8-12 pics handheld and stack them to a multilayer image later on in post.
I sometimes wonder why people want super high resolution for everyday use.
Before or after release who cares? I just want it. Is that so terrible and wrong of me to want something beneficial to my work? If it’s just a matter of firmware update, I’d rather pressure Sony sooner rather than later.
Pixel shift is far more convenient to me than hand holding in studio, especially with a bellows and t/s in play. It also gives a better final result because it takes the guesswork out of the algorithms.
It is totally fine for you to want this feature. It is very convenient in some situations, and I myself would absolutely use it. I guess you do not work with flashes if you are so eager for this feature.
But I can understand that it is not a top priority for Sony from an economic perspective. Sensor Shift is gimmicky in the way that you have many restrictions applied when using it. Imho it is more viable to release a solid and polished product than having loads of features as well as loads of bugs.
No.
Not really true. Tests with the 5Ds demonstrate that even mediocre lenses show at least some resolution gain on higher resolution sensors, though of course better lenses show more gain. But even if they didn’t, oversampling and downsizing would still be better for resolution and for color accuracy.
Go do some reading mate. Lenscore.org has a good article about lenses vs sensor resolution. No one said anything about DxO but before some one does I wan to point out that DxO resolution tests are flawd so it’s pointles to make raferences to their test scores.
Rumour has it that the Pentax FF DSLR will have pixel shift and they are expected to be using the new 42MP sensor. Although the feature is for me not a deal breaker, it would be very useful to have for still life work. There is a jump in image sharpness with pixel shift as the Olympus users have reported, its a shame Sony are unable to give pixel shift on the A7rII.
It’s not a deal breaker but shame not to have it? I am confused!
Every reference to ‘still life’ and photography is because through all the bs I missed what was directly related to it. This and the near deadline to prove not only i’m the best but a form that can work at an exponential rate, instead of help, they hinder, and in hindering, cause great wrong only. Why do they have to remove these ones, oh because i’mback in touch now. It can be the ideal reference but also the cursor just changed again too.
It all changed with the internet. Suddenly we are overwhelmed by the thoughts of the illiterati as they spew across the web like stinking turds from a broken sewer, through blogs, Facebook, Twitter and the comment sections of news websites.
No sooner has an idea dawned with feeble glow of a five-Watt bulb than it’s there on the internet for the world to see, forever. Speak You’re Branes: there’s no one to stop you.
Why hide when its so ‘obvious’? much like the last century internet pirate meme.
http://dirtyhorror.com/wp-content/uploads/2013/02/aba4506a90ff29296e_dwm6bx0f8.jpg
This theatre you mean, or the film lighting?
IlluminatedObserver has got to be a piss take. The guy writes like a bad Google translate of Japanese or Bulgarian into English! These forums need some moderation to cut this kind of crap out – it just gets too hard to wade through it all and remains funny for about a nanosecond!
you just wrote that, read all the context and don’t demand posts be removed along with context for other posts just because you have only just posted after reading that one and not read them all.
Not Bulgarian. I listened to only Bulgarian folklore music, reached around level 1000 (and easily beat nilo’s who refuse to listen to music more than once, a later time even without my favourite songs playing) and still hadn’t tired of it, maybe I was a little expressive in my personal musical arrangement but what actually ended it was I hadn’t recorded the songs I loved so much from the start and ended up heartbroken and still am.
You really should take your meds on a daily basis.
You started 2 years ago, 1 year after I got camera, and at the time or before I managed to get a pentax K01. Started in doubt and go in doubt.
Now I have A7R and those doubted photographic understandings from theory to gear such as the 90mm 2.8 I just tested. You will always doubt all apart from your politics. Which also coincides with your circular interventions for which now its clear to everyone I am smarter and serious about the inversion.
That Nikon guy is going dark side…
Nikon IS the dark side with Sony being the Trade Federation.
It will catch up with D850/900 and D5. It’s more likely these 2 companies will ride along each other.
Their K-3 II has it, but only does so for full colour sampling, not extra resolution.
As for why not Sony? Well there is a big difference between shuttling around 9 x 16MP images and 9 x 42MP images.
I think Pentax using this sensor is slightly wishful thinking.
Pixel shift, for a Bayer sensor, is an operation on 4 IBIS-shifted RAW frames. Can this be done as a firmware update? Or are Sony Exmor sensors perhaps incapable of simultaneously accessing 4 frames of true RAW sensor data?
It looks like Sony might not have a very precise (and quick) control of sensor movement. It’s more like a magnetic lift system with simple counter vibration ability, which is enough for the job.
On olympus you could hand-held em5ii and use high res mode. Their hardware must be different.
https://www.youtube.com/watch?v=dZwi4ehTtiA
To me the video implies the contrary(on the point of being a magnetic lift system). This demonstration does not look like it is a passive system. Do you see any motion applied to the whole construction that would imply inertia being the source of sensor movement within it’s frame? Imho this looks like active stabilization. The wobblyness and smoothness is just caused by the routine chosen to illustrate the different stabilization axis of technology.
Yes, Sony can move the sensor to a given direction, but control doesn’t seem to be very precised. Olympus’ pixel shift is apparently very accurate and quick in movement. In handheld position 8 photos are taken at maximum aperture. These 2 companies have very different design in their stabilization system.
So his conclusion is that IBIS doesn’t really have all that much of an impact. I still find it useful, maybe its just in my head
IBIS has been proven to work. And that was an old 2-axis IBIS on A700. http://slrgear.com/reviews/showproduct.php/product/985/cat/83
It might not give you 4 stops of IS at telephoto lengths, but I even welcome those 1.5 stops at 200mm.
On my A65, i’ve consistently shot sharp images at 300mm, at 1/125. That’s 2 stops right there. Didnt test at lower shutter speeds.
Don’t worry, I’m living in the same fantasy world as you. I still use my a65 regularly with Sony’s 35mm 1.8 prime– and somehow, magically– I consistently shoot at 1/10 and 1/13 of a second, handheld, with no blur.
Since 5-axis stabilization obviously isn’t an improvement on the IBIS in my 4 year-old a65, Sony a7 II owners are probably better off just disabling sensor stabilization altogether.
For the record, if we’re just going to be real right now, I think Sony ought to just shut the f__k up about on-sensor stabilization until they sell a camera that can shoot handheld with a 600mm lens and achieve mp less than 5 to 6 usable stops of movement reduction. Really.
Looks like stabilization is very effective, but reinforces that it is mostly for A) stills photography, or B) locked off video shots that should have been done with a tripod, but if you hold steady and try to make due without, you can fake it good enough. There’s a real reason why videographers use rigs to add balance / mass / and proper handling or other specialized tools like glide cam and stedi-cam. Sensor stabilization cannot replace these tools and the subtle handheld look achieved with proper camera rigs and expert operation is not going to be replaced with the false look that video stabilization gives you… sort of akin to watching a pan and scan conversion done to a film on 4/3 playback – looks very fake and obvious when you see it.
I have to politely disagree with you. Having shot with Sony’s a65 and FS-100 (with the 18-200mm OSS kit lens), and then shooting with a trio Blackmagic cameras sans stabilization, there is a world of difference between Sony’s in-camera solutions and not having a solution at all.
I use a glidecam and a couple body-stabilized rigs with my current (no IS) cameras, and when I go back and pick up a Sony (even my RX100 mk I), it’s amazing what I can achieve (footage that is totally useable) with the IBIS or OSS.
I wouldn’t say that Sony’s stabilization solutions are the end-all to stabilizing camera movement– but having worked in both worlds for long enough, and having captured terabytes of usable footage with IBIS/OSS– I can say that stabilization will be more than a welcome feature in the next video camera I buy, Sony or not.
“Conspicuously missing from the Sony A7R II feature set is the camera’s ability to put it’s digital finger up my butt right as I reach climax. My other partner, Olympus, does it every time– and I can’t understand why Sony’s engineers left this obvious feature out of the A7R II…”
I was referring to APS-C DSLRs, Nikon have to my knowledge only one APS-C prime yet they have 24MP APS-C sensors in their cameras, which people regularly use with full frame lenses…
already did yesterday