new boyz: So.. Canon can finally buy Sony sensor from a non-competitor company? I hope so. I would like to see more flavors of Sony sensor. Pentax flavor is the best for now.
ADC's aren't part of the sensor in the Canon implementation. Anyone who thinks they are hasn't bothered to learn how the cameras work. The fact that Sony has been able to maintain a consistently low front end noise figure across all gain settings is the key to their DR performance. They achieve that by taking a sensor design that is really no better than Canon's and integrating a low noise distributed ADC on chip. The fact that Canon has not chosen to do that does not make their sensors bad. If you are going to compare, please compare apples to apples. Not doing so just adds to the confusion.
As for the QE, the difference is measurable but probably mot disenable in real use. Most of the cameras that get routinely discussed in these forums are around 50% (D800, D810, A7r, etc.). Yes, not exactly the same but hardly worth arguing about.
T3, it is a matter of economics. My guess is that Canon's semiconductor group makes more than just sensors for cameras. Canon makes a whole bunch of products that rely on semiconductor technologies.
Not only that but Canon's sensors aren't that bad, in fact they perform just as well as Sony's -- this is a documented fact, even by DxO's own data.
Gesture: We can't make both gum drops and marshmallows.
Exactly. This reminds me of when I worked for Rockwell Semiconductor who also owned Collins. They decided that it made sense to separate the two so they spun off the semiconductor division as Conexant and kept the rest.
SingerNick: Yes, Canon has been hitting it out of the park with these premium optics. Hats off to them. But how about they stop shipping 6-year-old sensors in their flagship compact?
An "uproar" alone is meaningless. What needs to happen is for people to move to some competitive product. To paraphrase a common cliché "Money talks, whining walks". Apparently, the competition is not sufficiently compelling, or the issue is not significant enough to cause a big enough churn.
People keep raising the canard: "too many Canon users just bow down and kiss the feet of Canon no matter what they do" the reality is people use the gear because it does the job they need done. If that isn't the case for you, there are a boat load of alternatives that you can choose. However, apparently, these alternatives apparently aren't sufficiently compelling to get a large enough number of people to choose.
So have you dumped your Canon stuff and moved to something else yet? If not, then it is you that is the problem, not me. Once people stop just bitching and start taking action, then you will probably see your change.
"And what does the user care about ROI? Are you a photographer/videographer or a bean counter?" I am an engineer working in the semiconductor industry. I am faced with these sorts of decisions every day.
I am fully aware of what it costs to make the changes that some people on these boards just brush off as trivial. Even the simplest solution of dropping in a Sony sensor represents a considerable investment (SW, HW, docs, etc.). This must be traded off against the return it will bring. From their perspective it is a simple matter of if we do this, how many more cameras will it sell. All of these companies are in business to make money and that is how business works.
FWIW: I have never denied that this is an issue for some but so far, it doesn't seem to be an issue for enough to have become an issue for Canon.
I am still amazed that they can make this "kludge" and compete in terms of recurring cost against their competitors though. Their fab costs must be very low.
I suspect that if Canon made whatever investment is needed to fix this so-called "Issue", they feel that it would not make much impact to their sales (or not enough ROI). There has to be a financial reason why they aren't doing this -- particularly when their solution appears to be significantly more expensive to produce. Search for that and you will have the answer. And when you find it, please tell the rest of us because many others are at a loss as well (I certainly cannot figure it out).
Wrong again. I don't own any kit glass, never did. FWIW: Your comment makes no real sense. All the best cameras and glass are available to you now so exactly what the heck are you bellyaching about? Figure out what you want and go buy it. If more people do that, Canon will be forced to improve their technology.
And, BTW, once you do that, let's see the great art you make with it. The simple fact is that there is a great deal of excellent art out there generated with these supposedly sub-par sensors. Similarly, there are quite a few here who continue to grouse about the Canon technology but have nothing to show for whatever it is that they are using -- it's laughable, really.
The fact is that there does not appear to be a huge advance in the state of "ART" as a result of these supposed "state of the art", game changing sensors. My conclusion has always been that there must be more to a camera than the sensor alone, otherwise people wouldn't still be investing in the system.
Excellence for excellence sake is a waste of time and money, two commodities that successful companies generally endeavor to conserve. Spending a boat load of money to solve a (mostly perceived) problem that involves one narrow use case is probably a difficult pill to swallow.
Maybe because they work -- and there is no need to fix something that ain't broke?
Can't wait to see the DxO results.
Did they post its DR curve :-).
Detail Man: Richard Butler and Rishi Sanyal:
These two articles about sources of image-noise are well presented and explained in your statements of clarification made in comments sections.
I like the way that DPReview writers appear to have the adopted the terminology "ISO-invariant" / "ISO-variant" in lieu of the (IMO, less descriptive) terms "ISO-less" / "ISO-full", and recall suggesting the alternate use of those particular terms in this post replying to "gollywop":
... when he was in the course of editing his to be published DPReview article here:
... and which he continued to use in his subsequent DPReview published article here:
The more information regarding these subjects that is made accessible to readers within articles published on the DPReview site (in addition to appearing in forum posts), the better !
Yea, doesn't make a whole lot of difference. people who want to know will figure it out anyway. Maybe Canon will bold down a Sony sensor and we can all stop talking about eventually ;-).
Everyone here wants to work in the term "sensor" somewhere but that is not valid in the case of Canon. I think I would just say "camera" and forget about “sensor” particularly in the case of Canon since in the "downstream" read noise comes in somewhere long after the signal has left the sensor.
I think your terms (LDRN/HDRN) are fine but the "downstream" concept seems to be unique to Canon from what I can tell. I get what DPR is trying to convey but I am not even sure that the concept applies to the Sony architecture since the digitization appears to be right at the detectors themselves (I suspect there is some buffering there but it is quiet). Sony really doesn’t have a “downstream” while Canon has a whole analog signal path that doesn’t exist with the other guys. But as for your question, Yes, I do think your terms are better since they don’t use the term ISO and are more descriptive of what is actually going on – good point.
"ISO-Invariant" and "ISO-variant" make more sense to me. Never did like the term "ISO-Less".
Jonathan Brady: My first thought? MAN... DPR is REALLY going after Canon and just burying them! They must be in cahoots with Nikon/Sony. My second thought? Instead... Maybe they're just trying to educate the masses who will then demand a superior product from Canon which will (hopefully) give them the swift kick in the *** that they obviously need and the whole market can move forward with better products. Yes... Image sensor noise (and thus, exposure latitude and DR) is simply one aspect of a cameras performance, but when one company is so obviously outclassed, it's hard not to beat that dead horse. PS, I'm a Canon user. And yes I recognize they're behind in low ISO noise performance due to the DOWNSTREAM noise added by their imaging pipeline.
No mention of any particular manufacturer that I saw here. All these noise sources are present to varying degrees in any camera. FWIW: there is no "problem" different system architectures perform differently. Both these systems are performing exactly as they should perform.
David Hull: What is with that fold out screen? At least that is one thing that Canon does right.
Hmmm... Maybe that is why they did it that way then. Good point.
There isn't one (my comment wasn't about the FF body, but about the kludge fold out screen implementation) but the fold out screen methodology that Canon did on the G11 and the 60D was far superior to this IMO. Maybe Canon has a patent or something?
What is with that fold out screen? At least that is one thing that Canon does right.
Looks a bit like the iPhone. Interesting. Rumors are that the iPhone 6 may have a more beveled, bezel-less appearance. I like the look, actually. I could care less about the camera. More MP, so what.
Smeggypants: "So you find that in America your customers equate physical size with quality?"
LOL :) - They got that right. :) :)
Size matters... that's what she said anyway :-)