06-28-2013, 11:18 PM
(This post was last modified: 06-28-2013, 11:19 PM by Klaus.)
The EOS 50D is a bit old by now but technically there is not a huge difference between the 15mp and 20mp sensor of the upcoming EOS 70D -> 15% in terms of picture height (PH).
Now if we moved to the 70D, we wouldn't be able to do a lot of retesting thus you would loose the framework for comparison to some degree.
Thus what is your preference ?
No strong preference either way for me. I'd suggest delaying the decision 6 months and see if there's any hint of the 7D mk2 and what sensor that has. While I suspect they will also go for the same sensor in that, there remains some chance they may go higher MP and thus may be a more interesting choice if so.
<a class="bbc_url" href="http://snowporing.deviantart.com/">dA</a> Canon 7D2, 7D, 5D2, 600D, 450D, 300D IR modified, 1D, EF-S 10-18, 15-85, EF 35/2, 85/1.8, 135/2, 70-300L, 100-400L, MP-E65, Zeiss 2/50, Sigma 150 macro, 120-300/2.8, Samyang 8mm fisheye, Olympus E-P1, Panasonic 20/1.7, Sony HX9V, Fuji X100.
I don't think that the 7D II will be that different.
In any case we would prefer to test on the more mainstream model.
Unless the plan is to retest all lenses again, it seems better to stay with the 50D.
Otherwise, previous tests cannot be used as a reference.
I also agree on staying on the 50D in order to use the same frame of reference.
Idea for a compromise: get the 70D, but continue to test on the 50D, too. At least for a while. This way we'd keep the reference on the 50D while we start to build a new reference system with the 70D. Maybe a few retests can be done in that time, too (but certainly only of important lenses).
Tests on the 50D would however be downgraded to "lab reports", thus no sample images. Parts of the reviews which are valid for both cameras (distortion, Locas, bokeh) would be made on the newer camera only (but published in both reviews, of course).
We could use the same "slow shift" procedure in the Nikon section, too, where the D7100 is about to replace the D7000.
If a 50d supported resolution is y, 70 x and a lens is z (somewhere between x and y) are you likely to notice a big difference in the behavior if the 70d is used instead of the 50d (x < z < y) ?
I.,e, is the drop off as the camera resolution increase (relative to a lens that is borderline vs one that is >>> camera resolving power) obviously more visible with the slight in crease in resolving power of hte 70d over the 50d ?
If the border LW/PH of a lens is 1000 there'll be no significant difference (<5%) between the 50D and 70D.
If the border LW/PH of a lens is 1500 there'll be a marginal difference (~5%) between the 50D and 70D.
If the border LW/PH of a lens is 2000 there'll be a slight difference (5-10%) between the 50D and 70D.
If the border LW/PH of a lens is beyond 2500 there'll be a medium difference (~15%) between the 50D and 70D.
Now this assumes that I would take the same RAW converter with the same profile.
I suspect that Canon will follow the trend and use a weaker AA filter in the 70D. In this case you can double those figures - AA filters have a linear effect.
The point is that there is always a scaling effect with the bigger sensors. This multiplier is similar to a USM. If there is no sharpness at all it will not help. However, the clearer the edge the more accelerated is the effect.
No reason to change it. When pixels number grow twice is would be fine to thinks about changing of camera.
E.g Canon 350d-8MP -> 50d 15Mp -> canon xx/xxx 30MP+