brainiac wrote:
A cacophony of confusion, in my opinion. The enlargement factors you cite as advantageous are a feature of the monitor. Resizing images using bicubic interpolation removes monitor-related resolution issues.
Any image viewing software will be interpolating the image before display. It is never an issue of the monitor. Yes, bicubic in photoshop is better (and slower) than what many image viewers will use (some do it on the gfx card side and do quite well).
Bottom line though is that whichever image is interpolated will suffer.
Almost the only way to be fair would be to pick a size paper and then interpolate/scale both to 300 ppi on that size paper. Even then you have to worry about which algorithm to use for each one.
Edit, here is a link to a test run by smugmug comparing different levels of sharpening with lanczos downsampling.
+1 on the 1.4 Tele. I think the 5d Mk II w/ 1.4x is more capable in a way than the 7d w/o a tele because you can slap the 17-40mm and get a nice wide-angle FoV.
I just got a 5d Mk II this week after going back and forth between that and the 7d. The full-frame ended up winning (plus I got a reasonable deal on a low-mileage used one that turned out to only be $400 more for the Mk II.
I would have to say the resolution captured by the sensor is intense...
My 2 cents would be to crop a 5d Mk II image or use a TC rather than be restricted with the 1.6 crop sensor.
Also I don't beleive it to be fair to look at anything over 100%. I would as @brainiac said, look at the 7d at 62.5 while keeping 5d @ 100.
jxsq wrote:
3. 5d2's center AF point is under-stated and 7d's AF is over-stated.
I agree the 7D AF is over-stated.
Have you compared AF tracking in low light of 5D2 vs 7D using the center AF lone? I suspect they are nearly the same but some fellow in another forum disagree.
Talk about limiting your test group! Most photographers who select one format or the other will select lens most appropriate for use with that system. (An exception perhaps being those who shoot certain subjects with the very longest focal lengths.)
I guess that the question you are trying to "answer" is one that has been asked and (maybe or maybe not) answered many times, namely: If you use the lens that is right on a cropped sensor body, what is the image quality going to be if you use the wrong focal length on a full frame camera and try to compensate by cropping....Show more →
that's kind of a poor way to frame it
1st he said exactly that he was going after a limited target audience
2nd it is pixel density not sensor size
3rd it is kinda of ridiculous to call it the 'wrong' lens choice if that is all the lens one has or perhaps wants to carry, so unless someone is willing to buy and carry 800mm f/.56 and 400 2.8 for everyone.... and then what if even those two are limited? there are no alternatives!
thedigitalbean wrote:
Yep looks like the 7D isn't getting you a lot with that particular lens.
i've gotten noticeably more reach from a 300 2.8 + TC wide open from a 7D than a 5D2
the exact focus plane is very narrow for this sort of thing and you can't pick spots all over the frame to compare, you can't shoot through bad atmosphere, this subject is less than ideal for comparison
Have you compared AF tracking in low light of 5D2 vs 7D using the center AF lone? I suspect they are nearly the same but some fellow in another forum disagree.
I didn't get much chances to shot at low lights - not a sports person myself.
I did shot two gymnastic events, iso 6400, f2, 1/1000 for smaller gym, and iso3200, 1/1000. f2 for bigger one. I guess the light was decent and is not really low lights.
Was using 200/2, 135/2 and 85/1.8, I think both bodies can focus easily without any problem. I don't have any statistics on keeper rate based on only in-focus shots but i'd say it is about same - both performs well.
for the kind of lights in those two events, I'd say both are about same.
The iso performance is more apparent. And of course the frame rate. If i had to pick one body for this, that'd be a tough call - really iso vs. fps.
brainiac wrote:
A cacophony of confusion, in my opinion. The enlargement factors you cite as advantageous are a feature of the monitor. Resizing images using bicubic interpolation removes monitor-related resolution issues.
With CS4 there is no penalty at any size now. 33% and 66.7% look just as good as 25% and 50%. This is as long as you have graphics card acceleration enabled.
(And over-simplified...)