The 7D outperforms the 5Dmk2 at high ISO image quality per area sensor.
A 7D sensor scaled to FF at 46MP would have better image performance than 5Dmk2 in every aspect.
You keep saying that, where's your proof? I think it's clear that the 7D gave up ISO performance to hit the DR numbers it did. DxO labs is probably the most respected independent lab out there, and their tests show that the 7D is not even close to the 5DII in ISO quality.
Again, the 5DII hits 1815 ISO before falling below acceptable quality.
The 7D is only able to hit 854 ISO before falling below acceptable quality.
The 7D does a decent job, but it in no way "outperforms" the 5DII in ISO quality.
So show me the proof, or I write you off as a troll.
And, no, at best a 46mp FF sensor based on the 7D would perform at the level of the 7D, possibly more noise issues cropping up from the larger size. It would still be subpar to the image quality of the 5DII in every category. The trade off might be ok for you, but some of us actually want better quality at high ISO, and a 7D equivalent FF sensor doesn't cut it.
Sensor Scores 5DII 7D Bold type denotes winner in each category
Over all Score 79 66
Color Depth (bits) 23.7 22
Dynamic Range 11.9 11.7
Low Light ISO 1815 854
Edit - to include other DxOMark scores for the 5DII and 7D
Please educate yourself before you write about things you dont understand. Look at the nosie dxomark SNR graph and you will se that the 7D sensor outperforms the 5Dmk2 sensor per area unit.
Photography isn't just about technical specifications. I just read an article on Luminous Landscape that someone linked to in one of these threads and it reaffirmed some of my thoughts over the years. Lens reviews often don't show the real world, they may comment on how sharp lens x is compared to lens y, but they don't always look at real world images and other equally (sometimes moreso) important aspects of lens characteristics, such as bokeh and contrast (especially where it counts). To paraphrase the article, why have a sharp lens, if it makes the out of focus areas look even worse than they are already. Ok, it isn't the sharpest lens in the Canon arsenal, but the bokeh on the 100-400 makes me feel ill when I look at certain backgrounds (green woodland for example), that is rarely mentioned in the reviews and you find out when you try it for yourself. That is the important thing. Taking that onto the differences between the 7d and the 5d MkII, the facts and figures don't always match what I see and I think even the DxO stats don't do the 5D justice. To my eye, the usable dynamic range is significantly greater in the 5D MkII than the 7D. What the stats don't show, is that when the highlights on the 7D are blown, they start to get a colour cast, granted, it was improved by an early firmware update, but it isn't completely fixed. Likewise, the shadow detail might show lower noise in the stats, but the detail is greater on the 5D MkII. Conversely, the DxO technical data shows that the 5D MkII has a greater usable ISO, however, it doesn't show the quality of the noise. Yes, the 7D is noiser to my eye, but up to a point (i.e. when there is a lot of it or the detail is being lost), that noise is more pleasing to the eye. Many have described the noise on the 7D as more film like and while not completely accurate, it does have some truth. On a related note, when the noise gets too high on the 5D MkII, you get banding, which has been pretty much eliminated on the 7D. Banding looks pretty awful and is unusable, it was something I used to hate about the 40D, which suffered from the same problem. Again, when compared to the 1D MkIV, this is missed on the technical data. DxO shows that the 5D MkII has less noise than the 1D MkIV, yet many reviews when the 1D MkIV was released indicated there was little in it. Again, while I haven't used the 1D MkIV, so can't comment personally, the technical data doesn't seem to indicate what is seen in real life, based on those reviews.
In summary, what I'm basically saying is, don't always believe the technical data that is available, it may not tell the whole story or it may be of little relevance in the real world. Look at the technical data by all means as a guide, but then see things for yourself to see if a) it matters to you and b) your eyes see the same problem (or advantage) that the technical data might show.
It is true that despite all the banding naysayers things like banding differences are often real world noticeable before the often minor SNR and DR differences between models of reasonably similar technology.
Even taking entire sensor area into account the 7D will sometimes produce more pleasing photos at ISO3200 than the 5D2, when most of the image consists of very, very, very dark areas because then most of the image lies in teh very depths and the fact that the 5D2 bands more at high iso than the 7D, can in a few cases, make the 7D do better even taking the entire identically framed images into account.
The 7D often has vertical banding though which hurts it's SNR ratings a bit and kinks the SNR curve oddly. THis does vary quite a bit copy to copy some have this a lot more than others and it tends to vary in degree in different parts of the frame on any given body. I've never seen a bigger copy to copy difference with the CAnon bodies than with the 7D and vertical banding. It's gain rather than offset banding.
The 5D2 has more offset banding in deep low iso shadows than 1Ds3. Converters tend to handle it muchbetter than when the camera was first introduced though.
SOme of the recent sony/nikon sensor has less read noise and banding and not just a lot more low iso DR but a lot more usable low ISO DR. You can also see that it depended not at all on teh MP count but only on whether the new column ADC system was used or not.