Advertisement
SNAPCLOUDBack up every photoUnlimited RAW storage, 30 days freeTry freeSample ad
Advertisement
Advertisement

Canon to start a new sensor tech “war” in September.

I have a good news and it’s coming via Canon 🙂 The last couple of years Sony did lead the sensor world and Canon is struggling to keep up with Sony. But according to Stefan from CanonWatch Canon will now strike back with a new kind of sensor tech that will find his way in new cameras announced at Photokina. It’s good news for us because it means that Sony will have to answer soon too and we may see a sort of 3 layer Sony sensor sooner than expected.

At least at Low-ISO the 3 layer sensor from Sigma already shows an impressive performance (see here at MirrorlessRumors). But there are many issues on noise handling and so on. I am sure Sony has the knowledge to fix them and also to match if not beat any new Canon sensor tech.

All I can tell you know is that I heard about a 54 Full Frame Megapixel sensor from Sony (with 2460 focusing points (no joke!) and the focusing area covers 78% of the entire sensor). It was actually planed for a 2015 release but maybe Canon will make them change those plans…

 

Share this story
Community

Join the discussion

0 Responses

    1. Really, why?! What is so exciting for you in always posting this when you can be first? If you don’t have an intelligent comment, or at least a comment relevant to the post in some way, don’t waste the rest of our time.

  1. as ‘announced at Photokina’ doesnt mean you will be able to buy it in a new body this year im not too excited…same goes for futuristic stuff made by sony.

  2. Awesome news! Now we can expect the competition new tech sensor to debut in Nikons! 😉

    (It will eventually trickle down to Sony cams 2 years later..)

  3. There have been multiple rumors on Canon forums about a revelutionary new sensor technology to appear on two of their new cameras soon and more later.
    Currently these rumors are vague.
    Sony has missed thier opportunity to be first with a new technology sensor and will definitely loose market share because of another very poor management decision.
    If Sony does not release thier new technolgy now…..the Barry Sensor….then those in need of higher res will jump ship to Canon.
    The new Tamron 150-600mm lens is currently available for Canon but not yet out for Sony.
    Its not too late for us to cancel our preorders for Sony mount and get it in Canon mount.
    If Sony doesnt do something bigger and better now they will regret it.
    Sony sensors have been ahead of Canon…but other than that Canon as a system is way ahead of Sony.
    Sony cannot afford to lose thier only advantage over Canon.

      1. Heres a radical idea..how about actually producing a better camera so Sony users can take better pictures now…instead of waiting till they have to to compete…of course that choice is no longer an option..now they have to introduce thier new technolgy to compete.

        1. Strange I thought Sony was improving their cameras constantly. Look at the A77m2, or look at the A7 series, of the A6000. All are cameras capable of taking stunning pictures. The most important part of the process that Sony shoot look at is the most important part: The person behind the camera. When using a camera well you get great pictures with all modern DSLRs/SLTs/and mirrorless cameras. New technology is very important and we must not forget about that, but that technology will not change a bad photographer into a good photographer.

          1. That’s cliché. Photographers are always looking for better equipment to stretch the limits of their profession or hobby.

          2. They do improve as little as they can get away with…..not as much as they can.

        2. I agree. Sony likes to take a great piece of technology and build something around it. The A99 and the A6000 are the closest things Sony has to complete cameras. IMO they’ll never build a complete camera because it goes against their mentality of leaving something to entice the next upgrade.

        3. …or we could just agree that the minimum standard in equipment (where the user rather than the camera is always the limiting factor) has already been reached some time ago, and photographers are people who display and sell their photographs… and someone who displays his lenses and camera badge instead is probably just a little confused by the concept.

          Full of radical ideas today 🙂

      1. Sony may not have hundreds of thousands of wildlife and sports photographers…but Cannon sure does…and unless Sony can win over some of them they wont.
        Furthermore why do you think a better sensor is only for wildifers?

        1. Does anyone on Sony´s side need a new sensor? What they need are more lenses and more photographer friendly bodies (ergonomics, UI etc). The 36mp sensor in a fast body with enough lenses might be all people want.
          On Canon´s side, they really need new sensor tech or many might jump ship in the long run.
          It´s all about lenses for Sony and sensor tech for Canon because there are the weak points atm. But sadly Sony doesn´t seem to get that…

          1. All Canon need is Sony’s new A7S sensor and EVF in their “5D3S” then make it record 4K internally and every pro would buy it. They could even stick the A7R sensor in the “5D3R” and sell two cameras simultaneously.

        2. What good a sensor for taking pictures? We need that in a camera, lenses, flashes and accessories to go with that. One can shout all they want about Sony “taking on” Canon. It is good for Canon. Free advertisement. 🙂

          You have to admit though that Sony are reigning supreme with their cybershots. 🙂

          1. yes do have the nex5/a6000 and still no 85mm native e-mount lens like the 35mmF1.8 …

          2. There are many many lenses for sony..and you want many more..just so you can buy yor 3.
            Yet you object to a better camera to put them on.
            Just what do you expect people to do with all these lenses

          3. I think I have an ok lens collection with Tokina 11-16 at the wide end and Sony 70-300G at the long end and several ok lenses in between, and with a house that need fixing I really can’t afford any new lenses. However, it might not be a good sign for Sony that the two lenses I wish I could buy are Sigma 50mm Art (to replace Minolta 50mm 1.7 (except for as a reverse-mounted macro)) and Tamron 150-600. I do feel that there are a decent collection of both cheap and expensive lenses for the A-mount, but often the best value isn’t delivered from Sony, though I think that situation is the same for Canon/Nikon. Especially Sigma has made some impressive lenses at decent prices lately.

          4. I agree..especially about the new Tamron 150-600 that I too am waiting for.
            You may want to check out my comments elsewhere in this post about how a stacked sensor can extract double the resolution from your existing lenses than a bayer sensor can.

          5. Yes, that’s the good thing about the RX-1/R and the RX100III–they are (essentially) complete systems, with a fixed lens, so one is not dependent on Sony to enable future system capabilities, except for firmware upgrades.

            The same thing applies to other manufacturers, of course, with regard to P&S’s.

            The RX100III and the rumored RX2 are the Sony offerings that have me really jacked. I have been waiting forever for a new A-mount flagship…so long, that it’s now out of my everyday thinking.

            Yes, handling is an issue, but for travel/hiking, the diminutive RX100III is a winner.

            And if they can make a curved-sensor pocketable RX2, and hopefully a companion curved-sensor high-res pocketable RX3 with a telescoping single FL lens, I might forget about an A-mount flagship altogether.

    1. Multiple rumours?!? You mean everyone repeating the same rumour…..With usual Canon style. It will be the same old technology renamed and repackaged…

      1. Check Canon rumors..2 different cameras rumored to have this new sensor..A C size.and. a ff..2 seperate rumors.

    2. The rumors claim it’s not based on Foveon. The best explanation I’ve seen is it is based on the tech found in an extremely low-light Canon CMOS sensor they announced March 2013 for video. Of course, this is just an educated guess. A sample of the new sensor’s capabilities is pretty amazing.

      https://www.youtube.com/watch?v=-MW58xsRO5Y

    3. Why do people think Sony, Nikon and Canon are secretly sitting on a bunch of new technology waiting until the competition forces them to release it????

      It doesn’t work that way. If Sony had a better sensor they would be selling it. And management cannot just wave a magic wand and take a year out of a development schedule.

      1. not really, its like with intel and amd. intel could deliver already 5Ghz CPUs, but why should they …

        same goes with Sony. im pretty sure they have the next generation ready, but as andrea posted at some point, they are waiting to answer to canon/nikon.

        anyway, no matter what canon releases, im not interested, for today standards (LCD/print) current sensor tech is more then good enough, what i really need is solid video performanace, and even here sony is lacking behind panasonic (GH4). where is the sony APS-C 4K answer to that? and please dont mention the F5/F55 😉

        g.

    4. Why would you start to cancel your Sony pre-orders when:

      a) These are just rumors for the Canon camera – even Canon Watch says take all of this with a large pinch of salt.

      b) We have loads of rumors for Sony cameras as well – we do not know for sure if the A88 or A99II are coming out at all. We have no rumors for what Sony “might” be doing for Photokina.

      Cancelling orders just because of some vague rumors is stupid. If you are actually worried then wait until Photokina or at least until August for all of the alleged announcements to come out.

      Panicking doesn’t help. Everything is just rumors at the moment and most of them are all vague.

    5. So you say that the rumors from Canon are vague, but Sony has missed the opportunaty to be first. But first in what? First in a vague rumor? And you call that a poor management dicision?
      Well how many people will realy need higher ress then 36 MP. How many lenses will be capable of resolving a higher res. sensor? And yes when Sony rushes the new sensor technic to market and we will find all kind of little problems then what? Everybody will claim that Sony management is making a poor market decision on bringing out the sensor too soon! Fact is we know nothing about the new Canon sensor and we know nothing about how far Sony is with their new sensor technology.
      And then having the newest technology will not automaticly draw users to Conon (or any other brand). This will only happen when the new technology is realy ground breaking and very capable.
      You ask Sony to do something bigger and better now, but bigger and better then what? A vague rumor from Canon? So Sony must make a more vague rumor? I don’t get it.
      Let we all wait and see, Sony has showed some new technology on sensor tech, like the curved sensor. Canon has shown nothing at this moment but rumors.

      But I hope that Canon will come with something realy amazing, so Sony have to come with something even better. And remember the one in front is most likely the last to change.

      1. I’ll take all the resolution you throw at me for cropping.

        I’m not convinced the Canon announcement will all be about being king of the hill in the MP race. Canon has done well squeezing out IQ for lower MP sensors and selling lower MP bodies than competitors. Sure, they probably will boost their MP to catch up but I doubt they will set much of a higher bar. I suspect what Canon has developed is a significantly better low-light sensor. Sure, this is mostly speculation but it’s also based on a video sensor they announced in March 2013 which is pretty amazing.

        1. Cropping eh? Even at f/6.3, diffraction starts to creep in on a 36MP sensor. Do you have long lenses that can resolve over 400lines/mm at apertures wider than that?

          1. Who cares? OK so cropping might result in a slight loss of quality when pixel-peeping.

            An equitable trade, for easily getting the composition you want, which is vastly more valuable IMO than is pixel perfection.

            Unless you are a technocrat, that is…not that there’s anything wrong with that…

          2. Hey I shoot with my phone too. But there comes a point beyond which cropping isn’t going to give more reach. Just magnification… dont look at me, take it up with physics 😉

          3. A stacked sensor will effectively double the resolution…4x the mp…as a bayer sensor because the photosites are much larger.

          4. The photosites on stacked sensors are 4x as large as the same resolution bayer therefore whatever the lens limitations..including diffraction…will be encountered at double the resolution…4x the mp ….. as on bayer sensors.

          5. I am trying to explain the differrences between stacked and bayer qu as lity in a simple enough maner that some of the sar readers unfamiliar with it can understand.
            As you will see in the comments some people are having some difficulty understanding what I am saying.
            Perhaps you can explain it differrently so they can.

          6. That’s the future. Single lens cameras with cropping for other FLs. Or the same with ILCs, but requiring lugging less lenses.

          7. Such a future won’t appear out of no where one day after you get out from bed 🙂 pixel densities have been going up and lenses have been getting better. Cropping has improved since what it used to be a few years ago but optics still beat digital zoom.

            And again, diffraction is still an issue. You can’t have a 1000MP sensor, close your eyes and really wish to make it resolve more than physics would allow 🙂

          8. Well, there’s a lot to be said about diffraction, it’s not just some impenetrable barrier, but that would be a long and boring discussion 🙂 No need for a 1000MP sensor. Even today I can crop a file from a Ricoh GR 28mm down to 35mm and get an extremely fine 10MP file, perfectly enough for the vast majority of people. Same goes for the RX1 with 35mm and 50mm. Or crop an a7r file to APS-C size and get 16MP of goodness. That already can cover two very different FoVs. For my needs, where I use mostly primes, that’s pretty much half the lenses to carry. Today.

          9. I avoid it when I can. But sometimes I just don’t want to carry all the gear. Mind you, I am not presenting this as a solution for everyone (It often isn’t one for me either :), it is just one thing we’ll be seeing more of in the future, I think.

    6. WTH is a Barry Sensor? At 2014 Sony already has great cameras, if you can’t take a great shot with that cameras, something is not right with you my friend.

  4. I have doubts that this rumor is correct. There seems to be a lot of false rumors around at the moment like the Fuji x-t1 replacement one.

  5. (SR5) – A99II will have the a7r’s 36mp FF sensor. Sadly, it will use the same 79 points SLT focusing module found in the current A77II.

    A99II will be the last A-mount camera. After that, Sony will focus on E-mount and on sensor phase detection technology only.

    1. (SR5) There will be no mirrorless A-mount camera at all, no 5-axis IBIS, no z-shift nonsense… Sony will slowly replace the fast A-mount lens (F2.8 zeiss zoom telephoto lens) with E-mount OSS version.

      1. Really Troll. Cause my playbook has them updating their entire Zeiss lens line-up starting with the 135mm starting this fall following the A99II, A88 launch.

      1. Don’t pay attention to the troll.

        He’s been extremely active since elementary school let out for summer break. A coincidence? I don’t think so. ;^)

      2. It’s a Troll post. Sony’s A-Mount represents over 30% of their camera sales profit, and it’s sales and market share are growing. See the A77II launch presentation.

    2. (SR6 ish) Sony will also bring out some new “Flying Pig” tech which will involve a floating mirror system to help stabilisation. It will also not require any lenses at all, since they will all be built on the sensor itself. Goodbye to all a-mount & e-mount lenses.

      Haven’t heard of the SR5 above but mine is definitely true.

    3. Your a moron. Sony is launching its A88, A99II, and a 400mm G-Series at Photokina for the A-Mount, cause they still earn over 30% of their camera profits from the A-Mount.

      If anything from Sony is going to receive a massive sensor upgrade this fall it’ll be an A-Mount.

  6. Didn’t Sony just display their curved sensor tech? Unless Canon promises a radical change in design and availability soon, I’d say Sony’s still ahead.

    1. Canon marketing is leaking to freeze the market.

      They haven’t upgraded their sensors in eons and have to fire a shot across the bow of Sony, Nikon and Panasonic since Canon is slipping far behind. Their EOS 5D Mark III has been the standard everyone else is measured against in both stills and video. It’s even been used by Hollywood for productions. Now it’s getting smoked by the Sony a7S and Lumix GH4 4k on video and Sony’s a7R And Nikon’s D8xx on stills.

      Yet, it seems Canon still won’t have a response until fall after Photokina.

        1. I think that’s the way the game generally is played. The flip-side is it freezes it for their legacy bodies as well as keeping people from running to the competition.

          1. Well said, but when the new Canon sensor is not ground breaking and the other brands show better cameras then Canon they wil be running every where but Canon.

          2. Yeah if Canon don’t deliver at Photokina…….I think well will start seeing their market share drop.

      1. lol I wonder how many people will shoot and continue to shoot with a7s or gh4 in comparison to mkiii. I guess “somking” it just don’t cut it///

      2. Think about it for just a second Canon with its admittedly mediocre sensor especially in APS is the biggest player by a long way in interchangeable digital cameras. With their entrenched user base ,huge selection of native mount AF lenses and a genuine advance in sensor tech they would be a very attractive proposition

    2. Its far for display the curved sensor until its in a new camera. If Sony wants to massproduce the curve sensor. Start with A99 mk II and some new lenses that can use this technology. No poìnt in having a sensor that nobody can use

          1. Why a DSLR? the curved sensor technic will need new designed lenses that can use the curvation. It makes lots more sense to use it on a compact camera with fixed lens first to show the world what the curved sensor is capable. And to be honnest, did we see any Canon camera with the new sensor technic? Did we see specs of the new sensor at all? How do we know this is not a lot like with the Foveon sensor, a very good sensor but with major drawbacks for most of us… Look at the great video sensor they made that can do very low light video. It is great but at a very low resolution, making it not even capable of doing 4K let alone stills…

          2. That’s a horrible idea, at least with one of the current mounts they have. It would render whole lens collections useless and alienate lots of users with legacy equipment.

      1. It’s only suitable for fixed lens cameras. Change the exit point from the lens and you need a whole new curve.

  7. I’m guessing that it will be the 100 megapixel canon full frame sensor that was rumored years ago. Canon rumors is saying that it will be very big and the only thing I could think of that will be bigger than Sony’s 50mp medium format sensor is the 100mp ff sensor that Canon has been cooking up for years.

    The thing is..I still find no use for a 100mp sensor. Isn’t it too big to process? The 36mp from A7R is already choking most computers when it comes to processing multiple raw files.

    1. I don’t think most of us need more than 16mp, 20+mp will be helpful if we do massive prints.

      1. You are incorrect. I dont need a massive print to use high megapixels. I can always use more and it helps with cropping and zooming in among other things. I also believe it creates that pop in my image. So stop thinking for me.

  8. Clearly Sony does not need to release their barry sensor since there are no major challenge from Canon at Photokina this year. Its just a ‘significant improved sensor’, does not mean any hyped! In sort, i speculate something like these from next Canon:

    – huge MP (40mp-54mp)
    – improved ISO performance (but still behind to A7S) 2400-2600
    – increment in AF points (but still behind to A6000)
    – improved AF speed in Still and Video
    – 4K internal

    1. Canon apparently already has tested a new CMOS video sensor that’s better than the a7S (there’s a comparison in the comments on this topic on Canon Watch). It’s not 4k, though.

      If Canon could come out with a pro DSLR with internal 4k that alone would be pretty huge and they’d regain leadership there. Sony’s choking badly on internal 4k and I’m hoping they figure out their overheating issues with the a99II.

      Canon – for whatever reason – is the market leader with a massive user base. Good-enough AF and resolution combined with better low-light and 4k internal would catapult their sales through the roof.

      1. I really doubt we will see 4K in a Canon DSLR anytime soon. Canon generally dumb down the DSLR video to help there motion department to keep selling the C-series cameras. 4K DSLR would pretty much kill the C500

      2. That will be interesting if new CMOS can beat A7S even without 4K. I kinda like this idea whereby I guess mostly pro STILL will buy pro dslr, pro videographer will buy BM, GH4, A7S for 4K. 1080P is enough for hobbyist and will available in all devices 🙂 but i hope 120fps will be incorporate as one of the basic feature too. lovely slomo!

        Canon and Nikon still remain market leader in dslr, even though we have seen mirrorless attack from A7 (sport), A7r (lanscape, studio) & A7s (4k video with 0.4mil ISO) Sony lately. That still not seeing an impact to Canikon so far. Guess their users still waiting what Canikon will offer on Photokina. jump ship is a tough decision, you need to know what you need exactly! ie: Dave lately jump from Canon to Panny GH4 🙂 Curious why he did not wait for this year Photokina event…

      3. What overheating issues with the a99II? There has been no confirmation or even any more rumors about an a99II. Ditto for the “wished for” a88.

        Don’t get me wrong I am seriously looking at the a99 and if the a99II is coming out great. But so far there has only been one SR3 rumor for the a99II & many trolls / rumors about the death of a-mount (I’ve lost count the number of these rumors on SAR).

        I believe that Sony will / should have something for Photokina but at the moment an a99II replacement is just a rumor, like the Canon sensor and not a very strong rumor at that.

          1. Good point – heating issue fixed with the release of the a99II. However a99II (or a88) has not been confirmed at all. Again all just rumors….

      4. You mean like the 1DC already on the market for near enough to 2 years? Yeah morally catapulted their sales through the roof… not.

    2. I think your right, but Canon is probably going to Jack the price of the 5D MK IV to reflect those changes/advances. I’m guessing 4K, at a minimum.

      Sony’s probably going to be caught flat footed, but the A99 MKII will probably sell for half the price, I’ll probably have the faster AF system, impressive burst rate (12fps), and some matching Zeiss glass (hopefully 35mm).

  9. Admin, It looks like the Leica mount wides do very well on the A7s. Steve Huff has been reporting…

  10. New sensor??.
    …. Mirrorless from Canon still crap.
    Most people choose Canon because it’s Canon,not by image quality.

    If Canon can’t attack to mirrorless market, Sony still the best choice.

    1. Canon’s sensors and glass have always had a really nice, warm look. Better than Sony’s imo, though I have jumped ship to Sony based on their exciting pace of advancement. If Sony releases a Foveon sensor and a Monochrome sensor, well I’ll be with them till the end.

    1. It also sounds like the approximate pixel density of an aps-c 24mpx sensor, upscaled to full frame with a bayer sensor

          1. This is correct
            If the same type of sensor were scaled up from 24mp C to ff it would actually be 56mp not 54mp
            However the numbers are close enough to be considered as the same.

          2. I’d have to agree on that. We are dealing in rumors not fact so for all intents and purposes 56 or 54 may as well be the same #

      1. Partly correct, partly incorrect.
        Resolution wise..considering demosaicing…when interpolated to the comparable level 54mp stacked is equivelent to 72mp bayer.Equivelent to more pixel density..1.3x
        The size of the photosites however are about 2.5x larger for the the 54mp stacked sensor.
        In addition the larger photosites generating more effective pixel density effectively allow much more resolution from the lens.

        1. All I was pointing out Barry was that this info could be interpreted as a large bayer or a barry sensor. I think the Bayer is more likely as it needs no new tech. It’s basically already there in apsc they just need to upscale.

          1. You may be correct however since there are widespread rumors of a sony 54mp stacked sensor this seems more likely.

    2. At least it sounds better than sigma’s 20+5+5?
      Sigma’s already got a usable iso1600, that’s pretty much enough for APS-C. If this 18×3 is a FF with a two stop advantage over the x3q, it’ll definitely be worth an upgrade from a current Bayer FF IMO.

  11. Canon and Nikon rumor sites are talking about Sony a7s and we are talking about Canon new sensor

    1. well, it seems we are at least then talking serious matter here. This old shrines that get re-polished eternally start getting boring. We are now beyond 50 mpix, as it seems, in discussion, 12 mpix, that was 8 years ago.

      1. I had this same debate with people at 3mp is enough, when we were talking about 4mp cameras.

        1. in some way, 12 mpix is enough if you downsize to album print and do not look close at full size. I downed full sized 7S shots, and seen from far they look ok, but if you look closer, there is a lot of detail you do not see, once you see the same shot made with a 36mpix, you see what you miss. The fact is that for a newspaper shot or a magazine photo, this is not that important and definitely, with silent shutter mode, and low light ability, 7S is the camera for the normal filed and political event, even shows, par excellence. I can also make good for the home family album, but we are inhabited in the mean time to something more subtle, to a resolution level that makes it just hard to accept the step back in real IQ that A7S is indeed. The IQ subtility became reality with the 10 mpix chip from the D200, this sensor made amazing good detail and a remake oft hat one in FF at 20 mpix could have made the A7S a better choice for stills, or maybe a new 7 xx with such a sensor, we had then 1 stop gain instead of 2, but for stills a much better IQ, and, they could do it without video, something that targets still shooters only, high IQ, silent shutter mode, and better low light ability, simply in the direction of Nikon’s DF.

      1. Lets put things in thier proper perspective.
        Yes I use Sony equipment…but if someone uses a differrent camera that does not make them my enemy.
        Sony, Canon, Nikon and the others compete with each other…but as a Sony user that does not put ME in competition with users of other cameras.

  12. Canon really let their sensor tech get old and tired, even the fanboys are having trouble pretending. Given that they are a very solid company – we could only imagine that there is something very good in the works. I’m happy for them, the DSLR folks deserve something new.

    1. Not with sensor tech. You either spend a truckload of money on R&D and let that build up, or you don’t move forward. CMOSIS was a leading fab in terms of technology (although not sensor size) long before Leica came in.

      I’m not holding my breath for what Canon has to offer. Most likely better than what they now have, but still 1-2 generations behind the D800E and A7S sensors.

    2. I haven’t heard any news of Canon building/acquiring a new sensor fab, so it would be difficult to pull ahead in resolution or dynamic range, or producing a layered sensor with a more convincing low light than sigma’s.

      My guess is the “technology war” will be based more on something achievable with the current process like their dual pixel AF.

  13. Layered 54 mp sensor sounds nice so long as the noise performance is as good or better than their 36mp sensor.

    However, perhaps Sony should take a page from Canon’s book and worry less about sensors and devote more resources to the software / firmware , support services and flash / accessory ecosystem (wireless trigger and 3rd party native accessories are conspicuously absent).

    1. A layered 54mp sensor will have photosites that are twice as big as a 36mp bayer sensor!
      54 mp is 3 layers of 18mp therefore 2x as big photosites as one layer of 36mp

        1. The 54mp stacked sensor will produce a choice of 18 or 72mp color files.That is 1/3 more mp than a 54 mp bayer sensor.
          As for iso…the only stacked sensor you have to compare it to is foveon..this was the very first stacked sensor and no one is talking about using Fovion…we are all talking totally new technology so we dont yet know the iso.

          1. The native res would be 18mp…non iterpolated.
            Bayer is interpolated 4x.
            When 18mp is interpolated 4x to be comparable the yield os 72mp.
            A 54mp stacked sensor will NOT produce a 54mp color file size.
            54mp could be done but would be an odd size scalling therefore yield substantially poorer quality than either 18 or 72mp.

          2. A 24mp bayer sensor only produces a 6mp color image..
            This is then interpolated from 4 different locations to produce an INTERPOLATED 24mp color file.
            None of the pixels are actually native.
            If I am confusing you look up demosaicing to understand what is going on.

          3. Yes, you get more colour information, though not as much as you might think…. BUT you can’t magically gain spacial resolution by having greater colour sampling.

          4. In simplest terms terms compared to a bayer sensor a stacked 54 mp sensor would have photosites the same size as a 18mp bayer and produce a 72mp file comparablel e to a bayer.
            In addition it would produce a non interpolated 18mp color file that could then be scaled up 200% to 72mp of the same quality as the natve 7wmp bayer.
            The reason for this huge difference in effective resolution is that bayer requires demosaicing and stacked does not.

          5. Where are you getting this from? It’s totally incorrect.

            Yes the photosites are the same size as an 18MP bayer, but that doesn’t mean you get to magically generate 4x as many pixels without loss of quality.

            Thought experiment for you: imagine you are photographing a grey on grey scene. Each one of the 18 million locations on the sensor captures has the same value for red, green and blue (the definition of grey). This image is in fact identical to the one you’d get with an 18MP bayer sensor, the fact that the bayer only records one colour per location is irrelevant if all the colours are the same value!

            In your bizarre theory you get to upsample the stacked image 2x in each direction, despite the fact that it is identical to the Bayer image!

          6. You are getting confused by my description.
            My claim is 1/3 better resolution with a stacked sensor than bayer of the same pixel count.
            A 54mp stacked will match the resolution of a 72mp bayer.
            Thats 1/3 more resolution.
            The Stacked sensor is 18mp x 3 layers
            The Bayer would be 27mp green, 13.5mp red and 13.5 blue.
            The extrapolation is 4x…from 4 positions.
            The stacked produces perfect native 18mp pixels
            Extrapolating 4x…200%…yields 72 mp at the same degree of extrapolation as the 54mp bayer image.
            You would ACTUALLY extrapolate to that level in order to produce the cleanest file.
            For a 54 mp stacked sensor you would actually produce 18mp and 72mp files….you would NOT produce 54mp files.
            It is confusing because we are comparing a 4pixel array in bayer to a 3 layer stack that is then either left as is at 18mp or extrapolated 4x to 72mp.
            You would NOT extrapolate 3x to 54 mp unless you wanted inferior quality.
            Your grey on grey example is meaningless as the subject has no detail to resolve.

          7. I didn’t mean uniform grey, but instead imagine a grey wolf sitting on a grey rock. Each pixel in the stacked sensor image would have equal R,G,B values, and thus you get an 18MP greyscale image. The Bayer sensor would, on the other hand produce a 54MP greyscale image (demosaicing will work very well, the fact that each red pixel is missing a green and blue value is irrelevant, because these values will be the same as the red value because the object is grey!)

            Which do you think has more detail, an 18MP greyscale image or a 54MP greyscale image?

          8. Both are 54mp sensors.
            The Bayer is only 13.5mp,13.5 mp and 27mp..none of which are alighned and every single pixle must be imagined becsuse they dont really exist.
            This process degrades the quality of the bayer image.
            The 54mp stacked is 18mp red 18mp green an 18mp blue perfectly alighned producing a perfect native 18mp color image.
            When that image is interpolated 200% to 72mp everything is clean and alighned.
            The 72mp image produced by a 54mp stacked sensor will be at least as good pixel for pixel as the 13.5mp bayer image interpolated to 54mp that the 54mp bayer would produce.
            You dont seem to understand that the interpolation greatly reduces the quality and resolution possible with a bayer sensor.
            The stacked sensor produces a real image with real pixels.
            Bayer will not produce real native pixels AT ANY RESOLUTION.

          9. I understand plenty about interpolation having studied the algorithms in detail.

            Demosaicing a Bayer sensor is nothing like upsampling an image. When upsampling you have literally nothing to go on so you have to make an assumption of smoothness and fit a curve to the surrounding pixels, which results in a substantial loss of detail compared to a higher native resolution. This is what is happening when you upsample your 18MP stacked image to 72MP.

            Demosaicing involves generating the 2 missing colour values at each pixel, so you’ve got one piece of information to start with and the colour values are often highly correlated which makes estimating the other values much easier. In short it is wholly incorrect to say “the interpolation greatly reduces the quality and resolution possible with a bayer sensor”.

          10. You have a very limited grasp of the technical side of demosaicing.

            “Bayer is interpolated 4x”

            Each pixel in the final colour image will be interpolated from far more than four pixels, probably in the region 16-20 depending on the algorithm. That doesn’t mean you can increase the resolution of your stacked sensor image 20x!

            The interpolation process is complex but fundamentally it exploits statistical correlations between the colour channels, e.g. a pure yellow object will show the same variation in both the red and green pixels thus the green values provide a suitable substitute for the missing red pixels.

            A neutral grey object has (in theory) 100% correlation between colours. Thus a stacked sensor gets three copies of the same brightness information for a single location, whereas a bayer sensor (with the same total number of sensels) gets three different brightness values for three different locations and thus resolves more detail.

            In reality colours vary between 0%-100% correlated. In testing you need around 50% more sensels in a stacked sensor to get the same level of detail as a Bayer.

          11. The stacked sensor can of course only capture the resolution that is there.
            Demosaicing…necessary for bayer sensor…loses about 75% of the maximum resolution that a stacked sensor should be able to capture

          12. Demosaicing loses some resolution, that is true. So an 18MP bayer will not be as sharp as a 18x3MP stacked sensor.

            However a 36MP bayer sensor will be pretty close.

            Please, please stop saying an 18x3MP stacked sensor is equivalent to a 72MP bayer sensor. It has no basis in theory or practice.

          13. Exactly. Canon and Sony have way more R&D to throw at three chip sensors. Sigma has done a great job with the Merrills (and we’ll see how the Quattros look). It’s obviously the future.

          14. ummmmm… no, I’m sorry you don’t get to magically upsample an 18MP 3-colour file to 72MP colour image.

            In tests the 15MP 3-colour (call it 45MP if you will) Sigma DP2 Merrill produces about the same detail as the 36MP D800e, so an 18MP 3-colour (call it 54MP) so is equivalent to at most a 43MP bayer.

          15. My numbers are based on what the sensor should be able to do based on the pixel type and math.
            The actual performance of specific sensors will of course vary.
            I would expect both Sony and Canon stacked sensors to outperform the much older Foveon sensor.
            Certainly buying a Sigma camera is not a consideration.

          16. It’s based on the wrong math since you are not taking correlations between colour channels into account (and even then you are multiplying by 4 for some reason instead of 3). A 54MP Bayer sensor will in most cases produce a greater information content (due to correlation, knowing the red value in one location gives hints about what the blue or green values might be) than a 3x18MP stacked sensor (which you know the red green and blue values, but for many fewer locations).

            So with a Bayer you have a very good idea of the R, G & B values in 54 million locations across the sensor plane, and with a stacked sensor you know for sure the R,G,B values but in only 18 million locations.

            In the best case the Bayer image contains 3x the information, in the worst case the Bayer image contains the same amount of information.

    2. That would be a big mistake; Sony needs to improve in some areas, but they should always be sure to manufacture the very best sensors in the business. Image quality is number one, everything else although important is secondary. Currently Sony is making the best image sensor and they should keep doing so above everything else.

      1. Making top image sensors is important for their sensor business. Making a great camera system should be the goal of the Sony alpha division. Currently they have good camera hardware available but the potential held back by the system as a whole.

        I do not do my job with a CMOS sensor. I do it with cameras, lenses, flashes, wi-fi transmitters and computers. My main work is done with Canon-branded gear (except the computer ) because they can do things (even an old 5d2) that my a99 cannot (either efficiently or at all), which is limited by accessory availability / functionality or the software / firmware.

  14. Let’s wait to see what they come up with before we go calling it a war. Most likely, based on their previously week attempts to play catch up with Sony “The King of Sensor Tech”, we’ll be lucky to have a skirmish.

  15. The most rented/used cameras on commercial productions are Canon’s, saying they are “struggling” makes you seem like an incompetent sony fanboy. Adjust your self accordingly 🙂

    1. Someone is too incompetent to understand the concept of context, judging by the clear fallacy above.

    2. He didn’t just say Canon is struggling. He said Canon is struggling to keep up with Sony on sensor tech advances. This is %100 factual truth, just check DPR reviews from the past year and see their comments, or notice how other companies, like Nikon, are now using Sony sensors. Canon hasn’t innovated much in sensor tech for at least 2 years now, that doesn’t mean they’re doing badly in the market, it just means their technology is slightly outdated and most people buy their cameras because of a) brand recognition or b) their truly great lens selection.

    3. No dumbass you have the “incompetent fanboy” market cornered. Now adjust yourself and find another site to Troll.

    4. A sensitive bunch you are. I would say 50% of my work is done with sony, and 30% is done with canon. My opinion above represents my experience on various productions and rental houses. The f5 is cheaper to rent then the c300, why do you think that is? One is IN demand, the other is NOT, at least not that this moment. My point is they are FAR from struggling in the sensor market.

      1. What cameras you use does not excuse your lack of reading and reasoning skills, further seen in your second comment. Learn how to make an argument.

        1. If their sensor tech was old, why would anyone use them? Why would they cost a premium at every rental house in the world? Why are they the most sought after cameras in the documentary world? Why did the f55 JUST begin taking a foot hold in the commercial industry?

          There is no argument, there is just a very simple fact. Canon, even with their high prices and “old sensor tech” seems to be reporting RECORD earnings, why is that? Is it because their sensors are “struggling” or is it because there are that many fanboys buying them? They haven’t made a huge leap in technology in quite some time(I agree) but do you think it’s because they “can’t” or because they “don’t have too”.

          1. I see an explanation is needed. You are presenting a straw man argument. You are talking about demand, about how many people are using their cameras and using “struggling” in that context. That is not the context of the post. It is pretty obvious that Admin is speaking about keeping up in terms of the qualities of the sensor, such as resolution, dynamic range etc. Which you apparently agree with – “they haven’t made a huge leap in technology in quite some time”. Whether they can’t or don’t have to is entirely irrelevant here, although if the latter is the answer, it means they are just ripping off their users.

            The reason why many use Canon gear, even with the older tech, is that (1) even though this here discussion is about sensor technology specifically, in real usage the sensor isn’t everything, or (2) inertia, people are either used to their gear or changing it involves important costs and inconvenience (particularly true for the commercial industry) or (3) they don’t know better.

    5. Canon are struggling with the quality of their products (in certain areas) but not with the sales of their products. Most people are blissfully unaware of the difference say in dynamic range, therefor Canon will continue to do just fine with the status quo. But for people who can tell the difference there are other choices.

  16. What is the purpose of a 3x sensor? Color rendition? But that’s already great with Sony.
    Sony already has the best sensor on the market (12Mp high ISO king), just need to put it into a proper SLT camera, and the ‘sensor war’ is won before is starts.

        1. 12Mp is not ‘low resolution’, it actually a lot. Most monitors are 1920×1080, which is only about 3Mp.

          I think for most 24Mp, 36Mp is more of a marketing gimmick than an actual need.

          1. Nice of you to decide for others what they need. I like to choose what I need for myself.

          2. I wrote ‘for most’. Sure, some shoot Sigma DLSR, some buy a Hasselblad and some enjoy big format film. But that’s not mainstream.

      1. Turns out that the 12mp is no better than the A7r’s sensor… total noise per sensor is the same. Scale down the image to 12mp and you end up with the same. This was shown on DPreview this week.

        1. Yes, we can scale down huge 36MP files and get ‘close’ to A7S with A7R. Except it takes extra work, there is no advantage then to the huge file sizes, it’s still no ‘as good’ (but closer) and A7R will struggle to focus… But hey, if that is what floats my our boat. I mean it’s only 1 puny stop better right? I mean why not shoot APS-C as it’s only 1 stop away from FF. And why not m43’s it’s only a stop away from APS-C… then why not 1″ it’s only half a stop away from m43’s… and 1/1.7″ is only a stop away from 1″… and so on.

    1. Have you seen the output from Sigma’s Foveon Merrill sensor? It everything else out of the water at low iso.

      1. ‘At low ISO’ I can shoot with my smartphone, thank you very much.
        Didn’t really buy DLSR to shoot at low ISOs 🙂

        1. lol what did i just read?

          So you buy an DSLR to shoot at ISO 200k and get Smartphone Image Quality but be able to shoot at complete darkness wheres no space for any creativity or light play?

          dumbass

          Photography is about creating art with Light.

          Light! Do you know that? Its the stuff which comes from light-sources like the sun.

          There is absolutely no sense in taking photos when there is no light.

          1. Why ISO 200k? I meant ISO1600-3200, which you really need when shooting events. ‘Low iso’ is usually ISO100-400. What did you mean by ‘low ISO’?

      2. Only due to the massive amount of large radius sharpening applied by Sigma’s conversion software, use a third party conversion and the files look much worse. There’s a thread on FredMiranda which demonstrates all of this. When the same processing is applied to D800e files the detail is almost exactly the same (the D800 has some colour moire).

        Hence 46MP foveon is roughly equivalent to 36MP bayer. And that’s at ISO100, at higher ISOs the bayer runs rings round the foveon.

        1. I process with the sharpening all the way down, so no (and sigma’s sharpening isn’t that aggressive anyway). And it’s not just about detail; it’s about color detail and color and richness. But a 15mp Foveon file is far more impressive than a 36mp file coming out of my A7R and D800. Detail os higher (though nearly the same) but micro detail is way way higher. And again, color rendition and the avoidance of interpolation. All of the pixels in a CMOS sensor are theoretical. All of the pixels in a Foveon Merrill sensor are actual. The files have a much thicker look. They look like film because the sensor is based on the design of film emulsion. If you had one, you’d be blown away too.

          1. Even with the sharpening all the way down, SPP secretly applies a large radius sharpening pass which you can’t turn off. That’s largely responsible for the “micro detail”. Check out this Foveon image:

            http://www.flickr.com/photos/22371233@N05/8424237628/

            You can clearly see the sky get darker around the branches as local contrast has been boosted. Apply the same amount of processing to an A7R or D800 image and they will suddenly become more “impressive”, see the comparison by Alundeb on this page:

            http://www.fredmiranda.com/forum/topic/1150855/66

            Finally look at what the Foveon images look like without Sigmas careful processing:

            http://www.fredmiranda.com/forum/topic/1150855/61#11279795

            BTW all of the pixels in your Foveon image are theoretical too! Think the Foveon sensor has red green and blue layers? Nope, it has a clear (white) top layer, yellow middle layer and red bottom layer. In order to get an RGB image it has to do a lot of hefty calculations and interpolation!

            I’ve seen plenty of sample images, very nearly bought one, until I found out how much SPP cooks the RAWs to obtain that result.

          2. It’s a pale blue top layer. And the colors reflect this-they aren’t perfect, but they are richer (just like emulsion, there are different representations of reality in different sensors). It doesn’t change the fact that we get 1:1 color pixels from the foveon rather than 4:1 color pixels from a Bayer sensor. The microcontrast is a result of this, not improper cooking of the raw files. Try it out for yourself sometime, the Merrills are cheap now. The results I get are the closest thing to film in the digital world. There are tons of threads and reviews that discount that silly thread you’ve linked. Again, try it out. It’s at least more accurate than an entirely interpolated image! And by theoretical, no–the colors are adjusted, but the pixel readings are 1:1 actual.

            And I have never seen anything in the camera looking remotely like the picture you’ve linked. That’s a crock.

          3. “There are tons of threads and reviews that discount that silly thread you’ve linked” – presumably you could post one then? A actual analysis, not a “look at my beautiful photos” page.

            I’m not denying the results are good, they’re just good for software not hardware reasons. The microcontrast doesn’t come from the fact you have three colour samples per pixel, that’s not technically possible, and anyway if it did somehow then you’d see the same microcontrast in Irident Developer versions of Foveon RAW images. But you don’t.

            Oh and the next “quattro” versions of the DP cameras use different sampling densities on each layer (to counteract the fact that light defocusses as it passes deeper into the sensor, which was a problem with film!) so guess what, you have to interpolate the images and your 1:1 ratio goes out the window. Samples I’ve seen demonstrate no loss in microcontrast as a result…

          4. Micro detail is what I meant. And yes, interpolation does muddy that detail. That’s why a 15mp foveon image looks as detailed as a 36mp mos one. It’s not software sharpening achieving that detail. What do you think of the Leica M Monochrom? Again, 1:1.

            And I’ve been disappointed in the Quattro samples that I’ve seen. We’ll see how the comparisons look in August. But I’m not going to put in a pre-order.

            And even if Sigma doesn’t have it right, isn’t the idea behind the Foveon Merrill sensor more true to life and ideal than that of an interpolated CMOS sensor? If Sigma doesn’t have it yet, don’t you think a bigger camera maker will be able to perfect it with better R+D?

          5. The idea is nice but doesn’t really work in practice. Firstly the concept of a “true to life” image where you just record red green and blue values at each pixel and write that to the memory card. Unfortunately that’s not true, even though you do no spacial interpolation you end up having to do a ton of colour interpolation.

            There are two main problems, firstly having pixels side by side is far better for light sensitivity than having them on top of each other, even after taking account of the fact pixels must be smaller to be side by side. There’s a chance some new organic sensor tech could help with this.

            Secondly an image that is in focus for one layer of the sensor will be slightly out of focus at different depths (just as it would if the sensor were moved forward or backward). This ultimately limits the resolution as higher resolutions will show the slight misfocus more. This was exactly the same problem in the film days and this was solver by having a larger film area!

            All the problems of Bayer sensors, namely loss of resolution and artifacts from demosaicing go away the more pixels you have in the sensor.

          6. Well I’ll concede that you sound as though you know what you’re talking about. A lot of that makes general sense.

            Though from my experience, foveon tech seems like the better technology. And it seems more filmic (which I prefer to the sterility of current mos sensors. TBH, I preferred CCD to CMOS, too. Maybe organic sensors will be the ticket (though I’m not altogether certain what the tech entails).

        2. Most reviewers put stacked sensors as 1/3 better than bayer…plus the photosites ate bigger.
          You apparently dissagree with that but do not share the views of most camera reviewers.

      1. I would always prefer lower resolution. Rescaling from 54Mp to 3Mp (1920×1080) doesn’t seem very reasonable… And files would be huge.

  17. I hope this “War” brings some focus on the importance of real innovation. Particularly in the areas of sensor design beyond just MP’s. If it’s just a resolution war I’d imagine it to be a fail. Most photographers aren’t crying out for more resolution. I’d like to see Canon show some interest in the development of a curve sensor technology as well. I’d also like to see Sony use Photokina to showcase and expand their existing line of lenses- for both mounts. Any type of War should be a boon for us by providing a renewed emphasis on what really matters- innovation and not just incremental increases in performance.

    1. Let’s be very honest with ourselves for a moment. What kind of performance – practical features and performance are our cameras not giving us right now that we need to take the photos we need to take? We’ve got an embarrassment of riches, and further tech advances really offer us little that we don’t already have. The “we need to innovate” drumbeat is not really driven by photography anymore.

      1. You need to take a look at the output from Sigma’s Foveon Merrill sensor at low iso. It is like night and day compared to a CMOS sensor. The room for improvement is astronomical.

      2. I agree- but I feel your statement on the basic need for performance is a secondary case in terms of innovation. We don’t need V8 sports cars either- any humble eco car will do. It still doesn’t negate the want consumers have for better things- there’s nothing wrong with that either. I’m strictly speaking about serious advances that offers us something worthwhile. At the minimum, at least something we haven’t seen. I’m all for that. You could have said the same for film, it was good enough to meet our basic performance needs as well.

      3. I agree I shoot an A900, which continues to take beautiful images and still surprises me, sure it has its weaknesses, but I know what they are and bought knowing them and they have never been an issue. My clients never complain about the quality of the files, I shoot a lot of artwork for book reproduction and although an A99 produces better colour in side by side testing my A900 files have never not been wanted when I hand them over. As a camera I love it faults and all, I find looking at possibility instead of desire and being content with what I have instead of what owns me a richer experience. I enjoy the new cameras I get to try them all but when my A900 finally stops working I will get one then and continue growing with that camera.

  18. canon will be the old and new leader!!…..majority of pros using them beside hassy, phaseone…..

    it will never change…..;) too big to fail/fall

        1. A company doesn’t need to lose market share to start failing, just profits. Microsoft are not doing very well despite being the biggest player in forever. They haven’t ‘failed’ yet, but they are failing.

          1. so, you either have no job, working for yourself, or working in a place that does not required a pc lol….

            keep in mind that most of company in this world are using pc… duh… if you have friends working at some level of management in a company, ask him/her to see how much does it cost to migrate into a new system, not even mention about broadly used…

          2. Lol, a personal attack. The best ‘argument’ of those who can’t make one.
            The facts that many companies are still using PCs and that Microsoft is failing aren’t mutually exclusive.
            Try harder.

          3. lol… you have facts to claim? i believe in data, facts… not media… for example, data, 10k documents. i already gave you my fact that was number of pc sales always more than number of whatever wanted your system.

            note: you already agree that many company, if not most, are using PCs. now your turn to prove that microsoft is falling. i do hope that you know how to read 10k documents… if not, oops…

            try harder… kakakaka

      1. They have forever been the most innovative camera company unlike Nikon which always does bring a good product but too late to the market. That’s why I ditched Nikon. I really really admire the build quality of Canon. Hold a 5D, a D800 and a A99. i have always marvelled at the solidity and feel of the 5D and 7D, while the other two feel cheap and hollow. But i think I am going to ditch Canon for Sony. Canon is too sluggish now.

        1. That’s funny to me the 5D (mk2 never spent much time with a mk3) feels plastic-y and cheap compared to the D800—I shoot with neither so I’ve got no real bias.

        2. I had the 5D Mark III for 9 Months and sold it to an a99.

          Best handling doesnt help if image quality is shit and you only get extremely overpriced lenses which are far from outresolving 22 Megapixel.

          Especially the very high priced primes like 35mm 50mm 85mm and so on are far from outresolving the 22 Megapixel of the 5D3.

        3. When I went digital I ditched my Nikon & 503CM kits for the 5D & the original 4/3 Olympus E1. Now if we’re talking about the nicest camera I’ve ever held it would be the E1, the 5D felt like a lump of wood by comparison. But the 5D made a better looking image by far. Problem with the 5D, nothing locked in place, whereas the Olympus E1 was beautifully designed — everything was tight & secure — pity the file it produced was absolutely crap viewed alongside the 5D. There’s no perfect choice no matter how hard you try. (:-)

  19. I am waiting for the A6000s by next year in addition to my A6000. Then I can safely forget these wars and shoot in peace.

      1. is it difficult to get a 12MP APS-C? I would be so sorted. The only other camera i wanr is a RX2-50. Done, no more SAR until the A7000 and A7000s or the A9 and A9s

          1. But is it new sensor tech like the A7s? I just need a higher ISO(50,000) and 4K just like the A7s

          2. Right now my A system is APS-C. And I don’t want to expand my expenditure into FF until may be end of next year, by which time i hope there will be better FF systems for video capture(internal) for under 3500$

          3. Yes, but you said you wanted a better low light, 12MP APS-C, so why not get an A7 with 10MP APS-C crop mode? They aren’t super expensive and you can start expanding into FF as you go? I’m just curious as quite a few people have said they want a crop A7S…

          4. Right now, the 4500$(+ atomos) bill is out of reach for me. But next year if my finances allow i will try to get 4K video.
            i shoot events. The smaller and unobtrusive the equipment the easier is my job and less frequently I would be identified as a photographer at the event. APS-C is perfect for events and the client is perfectly happy. And APS-C 4k video is perfectly enough for the clients, for a long time to come One needs FF only for high end productions. May be that’s the reason other photographers want APS-c A7s. that would complete the Sony APS-C system for the event photo and videographers.

          1. The rx2..curved sensor ff..I would be inerested in would be with a 90mm macro lens and very high resolution.

  20. Canon Rumors says that their sources say that what Canon will introduce in the new 7DII at Photokina will be more than evolutionary, but not Foveon-like. Otherwise their sources have made no indication what it will be, and they have have given the rumor C2 status.

    1. We do know Canon and Sony are both working on a stacked sensors.
      If Canon is anything like Sony they probably spread deliberate missinformation to help keep thier secrets.
      Therefore this could very well be a stacked sensor despite the claims to the contrary.

      1. Sony also working on a curved sensor

        The super lowlight sensor from canon from 2013 is, so I interpret the press infos from canon, not a stacked sensor:

        ,,The newly developed CMOS sensor features pixels measuring 19 microns square in size, which is more than 7.5-times the surface area of the pixels on the CMOS sensor incorporated in Canon’s top-of-the-line EOS-1D X and other digital SLR cameras. In addition, the sensor’s pixels and readout circuitry employ new technologies that reduce noise, which tends to increase as pixel size increases.”

        1. I wpuld like to see Sony release the new 54mp Barry Senser asap..even if they are not forced to do so by Canon.

  21. The problem is that it will still be a Canon and (while I like Canon) they will still charge a Canon premium.

  22. >At least at Low-ISO the 3 layer sensor from Sigma already shows an impressive performance (see here at MirrorlessRumors). But there are many issues on noise handling and so on. I am sure Sony has the knowledge to fix them and also to match if not beat any new Canon sensor tech.

    Issues on Noise Handling?

    The 46 Megapixel(!) SD1 have the same amount of noise then the Sony a77 with 24 Megapixel Bayer Sensor!

    I cant see a noise problem there, also Sigmas RAW-Converter is much more competent in removing noise than IDC/LR/ACR are.

    Can not see why Sigma is struggling with noise problems, the opposite.

    With the new Quattro Sensor they was able to increase the resolution in 30% and reduced the noise by 20%

    Show me when Sony increases from 24 to 32 Megapixel and have 1 stop less noise than before.

    It took them 5 years to reduce the noise by ~1 Stop with the same Megapixel!

    The new APS-C Sensors with an Bayer Equivalent Resolution of 39 Megapixel have about the same amount of nosie then the Sony a99

    An 39 Megapixel APS-C Sensor is beating an 24 Megapixel Full-Frame Sensor, cant see where Sigma have noise problems and i doubt by far that Sony would able to resolve them (if they would have any).

    > All I can tell you know is that I heard about a 54 Full Frame Megapixel sensor from Sony (with 2460 focusing points (no joke!) and the focusing area covers 78% of the entire sensor). It was actually planed for a 2015 release but maybe Canon will make them change those plans…

    I heard that too. I also heard tons of times that aliens landed somewhere in USA.

    Couldnt this be at least a bit realistic?

    1. The aliens are actually reported to be russian children with surgical modifications and the flying saucers were desighned by the Horton brothers who were captured from Germany by russia following WWII. Since my comments arealready rather controversial I refer you to the book Secrets of Area 51 by Annie Jacobson for more details.

      1. I did, i own an SD1 beside my a99.

        But as my image management software tells me, less than 3% of my images are taken over ISO 1600 so i dont really care.

        But the Images are still usable and produce much more sharpness than the a99 at ISO 3200, just more color noise.

        1. oh dude, just because u dont take pics over iso 1600 that often does make it nonsense.

          and yes, oversampling in iso does help to improve the sharpness, but at the cost of color. the red layer of the foveon x3 sensor almost receives no light in high iso, but taken that u dont take low light pic that often, well, u should enjoy your sd1 😉

    2. The Noise Problem comes, when Sony want to offer a Big MP Camera with 3 Layer Sensor and ISO 102000 and above 🙂

  23. I owned an a700 and an a850 when I bought an a55. I enjoyed the shooting experience with an EVF so much that I did not to use the a700 of a850 except in the studio. I then sold the a850 and got an a99. Now over three years later, an EVF feature in a camera has become the most important feature for me. That is the foundation feature. If the camera doesn’t offer than feature then I move on. So unless Canon or Nikon provide an EVF in an DSLR body I will remain disinterested. I cannot adjust, compose and review a picture in an OVF. I do not use LCD much. I started shooting track meets this year and I chuckle to myself when I watch the OVF shooters trying to manage their exposure with the LCD at an outdoor event. The next important feature for me is focus peaking. I shoot with manual focusing most of the time. Canon and Nikon have yet to offer a DSLR with focus peaking. I have enjoyed the a77 very much. The only adjustment I had to make with it was that I did not normally go over iso 1600 with it. I shoot a lot of stage events and the iq of the a77 and the Sony 70-200mm SSM has been great. I will upgrade to the a77ii in about a month so I have a better focused system with the 70-400mm G. So far Sony has made the adjustments in their lineup to keep me happy.. I am happy with the upgrades of my a6000 over the nex 6 except that the digital level is missing. Many people seen so focused on what Canon or Nikon is doing. I do not care about them at all. So consider taking your preoccupation with Canon and Nikon products to those forums. So for now Sony and Panasonic has my attention. They are the ones who are pushing camera tech development. I hope with the releases of the Panasonic GH4, LX8, and FZ1000 that Sony will have some competition and lower their prices on the A7s, RX100iii, and RX10. They are a company that overprices their products.

  24. Admin,
    Just read the article and did a search, I could NOT find the word WAR.
    Can you please highlight the part where Canon says “war”?
    As far as I am concerned there is no war from Canon or Nikon. When you have the market share they have its not war.
    Why can’t you just be happy about Sony innovation and stop using words/phrases like “destroy”, “war”, etc? All it does it create fanboys, anger, etc, and does not do anything to help the Sony community become better photographers.

  25. A gear-heads wet-dream, a photographers nightmare.

    A 54mp sensor would be complete overkill for any sane user. I’m guessing it’d be expensive, and probably cost 4 or 5K.

    The camera using it would have to be large in order to house the larger processor, memory, and all the other bits and pieces needed to operate it. Heat would be a big issue. It’d probably gobble up more power than the current A99, with its less than stellar battery life. The burst rate wouldn’t be anything to write home about.

    The pixel count would be outstanding, but I doubt a 54mp sensor would be an actually be an improvement over the current 36mp sensor.

    I wouldn’t buy it.

    1. I don’t need a 30 MP Sensor and much less a 54 MP Sensor. I Hope Sony and Canon give us Cameras for difficult user. Big MP Cameras for photographer who need this. Lowlight Kings for photographer who need this.

    2. I, like you, doubt that 54MP sensor will be a good thing, but I dissagree with you about camerasize and battery ussage of the 54MP sensor. As you know parts are getting smaller and smaller. You can buy 64GB of memory in one small chip, not larger then a chip with 4GB. Processors are getting better and better, faster with less energy consumption. So in the end the camera can be as large as any 24 MP camera now. And for the A99 the low battery life is due to the EVF/LCD combination, a “real” DSLR is using an OVF that uses no power at all, the EVF is using a lot of power…

  26. Sonyalpharumors post news from Canon 😉
    As I see it- In the Next month (and 2015) we become real new innovations! Great 🙂

  27. I just want to say that a post like this, with a rather sensational tagline that includes the word “war”, doesn’t really make the best case for a Sony community.

    I still see Sony as an underdog in photography, yes. I love their innovation and for trying new things. I love it, and love their cameras for it.

    However, I do not see other brands as “the enemy” or attempt to stir emotions by saying that Canon, effectively, will unleash the sensor-dogs of war and “God help us all for whatever may come…”

    Canon is struggling to keep its image as a force beyond their current user base and that is something that has been happening for a while.

    I think we ought to grow up a bit and stop pretending brand support immediately draws the line between brand X and brand Y and therefore we live in imaginary trenches for other photographers.

    My colleagues still don’t see why I choose Sony. They admit the quality of the images but they don’t like the rather sterile design and user interface. However, I don’t go out and berate their choice to shoot whatever they want to shoot. I know I have what I have because I like what I get. If Canon was to bring a super amazing sensor next week…I may go that way. After all, Zeiss ZE lenses are still available and without that, I would be more hesitant.

    1. Who shoots pictures with the user interface. I access the menu twice a year to switch the video button on and off, for the rest, the camera has all buttons in the right place. A6000 is here different since some buttons like the program wheel is bullshit, and it is there where the settings reel should be. Some cameras are poor in layout and disposition of buttons and commands, like Pana’s GM for example that is a terrible camera to use in this matter, but generally, at Sony cameras I found all to be in the right place on most of their models. The menu of Nex models is somewhat old Fuji S5 style, split access, but since Fuji writes the software for Sony, this is not surprising. Even Fuji has some of this crappy menu layouts still today. They look like japanese translated to english by Google. I hate the menu system of Fuji’s S5, but the camera is just a hammer all over. A fortune that the body was made by Nikon.

    2. More to cameras than just the sensor. Sony still lacks the system, the support and the dedication that they’ll stick with something longer than 5 years.

      however this blog post is pretty fanboyish. canon? war? really?

      “Canon is struggling to keep its image as a force beyond their current user base .. ” meaning they are struggling at being the #1 ILC company in the world? they sell more rebel bodies then all MILC companies sell cameras as totally depressing as that may sound.

      1. I do not believe Sony intends to give exclusivity this time…I have heard that all the manufacturers will be ontroducing stacked sensors at the same yime…many will use the Sony senor…Canon will have thier own.
        I suspect Canon may have both a ff and C size stacked sensor…Sony possibly only ff.

  28. Perfectly said, anger and use of aggressive wording invites annoying fanboys and deteriorates the quality of site in general. So stop using words like war destroy, etc.

  29. CANON NIKON SONNY PANASONIC!!!

    The only war you can start is the PRICE-WAR.
    means aps-c and 4/3 BODYS for 249,- ( oh my good it has no 4k)
    aps-c and 4/3bodys-4k for 399,-
    Fulsensor maximal 699,-

    If you think you can not go this way…
    for filming i have now 2 KINERAW mini 4k
    and for normal Photography ( also textures to use it in 3d and nuke …)
    old normal canon 600… because I KNOW waht do DO!
    YOu al made much failors since 2010 and made thwe blinki-blinki show for the masses.
    the masse is told waht they have to do.
    the young masses( in asia sorry- do not htink taht it is told-and if…)
    and more People have not the money ore the SMARTPHONES kill this all.
    / if you get a smartphonen-4k-2015gen for 15,90,-/month …..what will bee?

    My old good C300 is in ebay
    the good SONY F3 is gone
    the faboulouse Red Epic lost against KINERAW MINI..

    I spend my money in some Lenses… and think taht most in her also do not
    want to go this ” KASCHPERLTHEATER” carnenevall.. no

    and on the ebay-piles we see trillions of good cameras from all this super manufactures..
    so what? FUKUTJIMA??

    we had Chernobyl first.. waht? MARKETING? FUCK OFF..

  30. Canon already probably have the best sensors, so they could quite easily catch/pass Sony there, the question is whether they have something all new (non-Bayer perhaps) or have just done the obvious (on-chip ADCs)?

    (Long version)
    The Canon 6D sensor (IIRC their more recent FF sensor) has over 15.5 stops of Dynamic Range and class-leading noise banding. The issue is how they digitise the data, which limits them to 12-something stops of DR at low ISO (as it adds noise, that effect reduces with increasing ISO which is why they are competitive at higher ISOs).
    Adding on-chip analogue-to-digital converters (ADCs) would enable them to use more of the sensor’s DR and so show a big leap in performance (although would need 16-bit ADCs to take full advantage).

    Sources:
    http://www.clarkvision.com/articles/evaluation-canon-6d/index.html
    Well capacity = 79800, Read noise = 1.7e, therefore DR = log10(79800/1.7)/log10(2) = 15.52
    (Also good on banding.)
    http://www.sensorgen.info/
    Well capacity = 76606, Read noise = 1.6e, therefore DR = log10(76606/1.6)/log10(2) = 15.55

  31. Count me in, in this just war. This my friends and fellow photographers is going to be the last war. Only one will be left standing. And future photographers will thank us and honor our memory for how we stood up for our sensors and lenses in this all consuming, apocaleptic armageddon between the dark and light forces of photography. Even the women and young are have their part cutout for them. The storm clouds of war are approaching fast. Rattle your sabres comrades and let’s blitzkreig into the enemy.

  32. Good news for low resolution fans!
    A 54mp sensor produces a native 18mp color file of the very highest possible quality.
    It is18mp x 3 layers perfectly alighned.
    Bayer sensors do not produce a native color image at any resolution. All bayer pixels are interpolated.
    So if you want a really really good 18mp image than you will be very happy with the 54mp stacked sensor.

  33. We do not need a sensor megapixel war, we need sharper, color fringe free lenses.
    We also need cameras that can focus in low light, so we can use the high ISO of the current sensors.

    1. Can’t speak much about Sony’s Lens line up, but isn’t what you’re talking about… what the A7s is made for? Plus, if you can get a metabones adapter, you have way more options in terms of lenses.

      1. I am using a Canon 6D and a Leica M, the 24-105 fringes from time to time and I doubt that it would be a benefit to have more pixels with that lens. I almost bought the 7r until I tried one in a low light situation (sony 24-70 f4 lens).

        I am hoping the 7s has better low light focus, as the application I use the Canon for, pixel count is not that important (general event coverage).

        I really would like to have the smaller platform of the 7 & the low light focus ability I am used to (or better).

    2. The stacked 54 mp sensor will produce perfect native fringe free pixels at 18mp
      If you like a super quality low res image the stacked sensor will do it.
      A bayer sensor CANNOT PRODUCE A NATIVE CLEAN PIXEL AT ANY RESOLUTION.

        1. Please explain your comment.
          I dont know of any current lenses that go to f65 and I do not know what you mean by q67.

          1. Acoording to your link..q67 is just the same old bayer array rotated 45° …this would simply allow a different interpolation algorithm to be used instead of current demosaicing..
            Based on the comments it is debatable whether this presents any advantage for video but clearly there is no advantage for still pictures.
            As for using a special adapter to stop down lenses to f65….why would anyone place a second diaphram in a less than ideal location to produce unmanageably long exposure times.
            I would imagine a thin flat round apperture plate can be used thereby reducing diffraction while increasing depth of field…but the exposure times would be far too long for normal photography.
            Not necessary for landscapes..possibly only studio macro of stationary objects.

          2. Q67 does better than traditional bayer array, simply because both every logical pixel has its own RGB information. So although F65 has 17.7 physical pixel, the final output is only half the physical pixel, and no longer needs low-pass filter. This is so far the most efficient way to bayer array without low-pass filter and any side-effect.

            At least F65 can use all the film lenses made by Zeiss and any lens that is a PL mount.

  34. well, this new sensor tech just might explain why nothing much has happened since the 7D in the canon world of sensors. apart from the 70D of course. But this is supposed to be something very different so should have taken a lot of engineering resources. Let’s just hope it’s as exciting as they’re making it out to be…

  35. why has no one here mentioned that sony is producing a 70mm wide CMOS sensor with new pixel array, which allows it to output oversampled 8k video?

  36. The 50L was a HUGE disappointment to me, the 85L was beautiful but the focus was annoying. I wouldn’t trade the Sony 55 f/1.8 for both the Canon’s (if offered) it’s that good!

Leave a Reply

Your email address will not be published. Required fields are marked *

Search

Search
Advertisement
SNAPCLOUDBack up everyphotoTry freeSample ad

Subscribe to Our Newsletter

*Please also check your spam folder if you don't get the confirmation mail. And mark the mail as "non-spam" to see the future newsletter.

Advertisement
SNAPCLOUDBack up everyphotoTry freeSample ad
Advertisement
Advertisement