QuoteYour missing my point. IF <-- key word here -- Canon can reduce their ISO 100 read noise to the minimums recently seen in their cameras (1.5e- worth in the 1D IV sensor), they then could achieve an improvement in total DR at ISO 100. Sony Exmor effectively normalized read noise across all ISO settings. DXO's measurements seem a little sloppy...I've seen measurements from other reviewers that have much more consistent results, so my guess is that the D800 has a consistent 2.7e- read noise at all ISO settings (effectively the minimum read noise at all ISO settings.) IF Canon CAN DO THE SAME THING....make their minimum read noise from the highest ISO the read noise for all ISO's, like Sony did with Exmor, then Canon would have lower read noise than a D800. If that hypothetical (another key word there...my previous post was hypothesizing...might do you good to learn the difference between a hypothetical argument, which is all we can really do when speculating about future improvements Canon might add to their sensors)...if that hypothetical improvement was made, Canon could improve DR in a future 7D by 2.7 stops.
As for your presumption that Canon is incapable of developing a new fab or producing complex sensors at 180nm, there is nothing to stop Canon from innovating. Thats what competition does in a free market...it spurs innovation. Right now all we have about Canon's next cameras is rumor and speculation, but usually those rumors contain nuggets of factual, if not 100% accurate, insight. Based on the current rumors, I speculate that Canon IS innovating, and developing ways to improve their sensor technology beyond the current limitations it experiences today. That does not necessarily require a reduction in transistor size to accomplish. For that matter, does Sony even have a significant lead in transistor size over Canon? I know they use copper wiring in many of their latest CMOS sensor designs which saves them some space, but I hadn't heard that they generally had significant transistor size savings over Canon.
You are a true optimist, sony exmors advantage is the short analog signal path way to digitization= ADC in the end of every single raw at the sensor edge to hold down the read noise. Canon has nothing alike and do not have the equipment or expertise to do the same.
Yes Sony has a lead to shrink electronics from their mobile camera sensor research and manufacturing.
Time will tell what Canon can do or not, and they must do something about their sensors dynamic range when it is almost the same as in 2004
Your making wild, speculative assumptions about what Canon is or isn't capable of doing. It most likely has nothing to do with what Canon is capable of, and most likely to do with the fact that Sony owns the patents on the technology you are talking about: CP-ADC. I know exactly what CP-ADC is, how it works, and its advantages. Its only part of the equation, though. Sony has half a dozen other electronic noise mitigation patents that they also use in Exmor. The point here is that SONY OWNS THE PATENTS. That is their competitive advantage...it has nothing to do with fab capacity or capability, engineering expertise, etc. Canon is prevented from implementing technology that infringes on Sony's patents unless they come to some kind of legal or licensing agreement that allows them to do so. My guess is that Sony is going to wring every last ounce of advantage they can out of their patents before they give Canon the rights to use the technology in their own sensors. The two are epic rivals.
Canon has little option but to solve their problems in ways that are different enough from Sony's approach so they avoid infringing on Sony's patents. I am not sure what kind of leeway Canon may have in terms of innovating integrated electronics to mitigate noise. They already have their own form of CDS (Correlated Double Sampling), and they have their own patents on it. Sony also has patents on a form of CDS. Apparently the two approaches are different enough that both companies can hold patents for their specific way of implementing circuitry to solve that particular problem (the problem of dark current noise.) Who knows what other patents Canon has up their sleeve, or what patents they may be developing that they could add to their sensor tech. The issue may simply be ongoing R&D and time to market. I do believe Sony caught Canon off guard, but Sony has been filing patents for a decade, and a lot of the technology in Exmor was patented before 2004. It could be a couple more years before Canon is able to fully develop, prove, and patent their own version of the same kind of technology. In the mean time, they are apparently exploring cheaper options that can be implemented more quickly...such as some way to cool their sensors to reduce dark current noise.
The other thing about Canon is they are a very careful, meticulous company. They bide their time while they innovate, and there have been several occasions in the past where they appeared to be losing out to Nikon, only to come out with something rather radical in response. This happened when they unveiled EOS and brought AF to the market in the 1980's. That took the world by storm, and was the turning point that made Canon the supreme camera brand. The lead they gained back then still holds to today. Canon was also the first to bring full-frame DSLR's to the market. I believe they were the first to surpass the 20mp mark with the 5D II, which itself was a market changing device in its time as it brought affordable DSLR video recording capability to the table. Canon is an innovative and competitive company. It's simply that at this moment in time, Sony and Nikon have the limelight as they have introduced their own disruptive technology to the market and its making waves. That's all a good thing. It's the kind of competition the market place needs. It's the fire that spurs heightened innovation. Canon isn't going anywhere. This isn't the beginning of their long demise. It's just the beginning of a new phase of growth and innovation in the digital camera market that will probably span years, if not decades.