Krosavcheg wrote:
Tried 7D yesterday.
Not really something I would consider buying.
Of course, I didn't have chance to study it's ISO performance using a decent screen, but compared to 5D MkII side by side it wasn't impressive at all.
Holding it in your hand feels like slightly larger Rebel and it does feel a bit like a toy.
One good thing though is the battery status indicator can be seen when shooting (through viewfinder).
well it would be tough to match a modern FF
as for a toy, it weighs more than the 5D2 and has at least as good sealing, if not better, and more advanced everything other than sensor size....
sure it is not a giant 1 series brick, but being a brick doesn't automatically make something pro, being a brick has nothing to do with it
durandal2 wrote:
and for the pleasure i just dit the whole thing with the 7D raw files, left row is plain 7D raw detected with the correct iso levels by ACR, and on the right, the raw detected only as 100 iso raws.
alundeb wrote:
Optical illusion? I got curious and tried rearranging the columns. The one to the right always looks brightest because of the gradient across.
+1. In fact you can even claim the opposite as the left column shows better shadow detail making the right column seem darker.
I appreciate the effort with the colored cubes, but it's not the best test for signal to noise ratios. Not much detail in the cubes to base judgments from.
Because Nikon is apparently doing some kind of in-camera smoothing even with RAW images, I'm starting to think that the best test would be for unbiased individuals to take the exact same shot with several different cameras, preferably with a detailed subject, and apply their own post-processing techniques to all shots to maximize IQ.
I know it would introduce variability from the techniques used by the individual doing the experiments but, as mentioned earlier, it looks like Nikon is already adding an uncontrolled variable into the equation.
durandal2 wrote:
a simple hexadecimal editor , the hardest is to find the iso value, it's not always at the same place but on canon files, the bytes orders are inversed, so instead of 0C 80 for 3200 iso, it's writen 80 0C in the file.
HAH!! Haven't done that since back in DOS days hacking games save files value to cheat on games I couldn't finish.
alundeb wrote:
Optical illusion? I got curious and tried rearranging the columns. The one to the right always looks brightest because of the gradient across.
haha, damn, I think you are right, tricked by an optical illusion
durandal2 wrote:
for noise lovers, guess where is the 7D in this picture:
I removed the ACR noise reduction by changing the exif of all the raw files to 100 iso instead 3200 so every one is at the same level, and I reduced the original picture to the size of the smallest one of the group.
I fail to see how showing speckled swatches of pure color can give any idea on the relationship between the noise level and preservation of sharpness and details. Any perceived conclusion on the sharpness level is false because the image lacks of any spatial scale.
gfiksel wrote:
I fail to see how showing speckled swatches of pure color can give any idea on the relationship between the noise level and preservation of sharpness and details. Any perceived conclusion on the sharpness level is false because the image lacks of any spatial scale.
it wasn't about detail preservation, but only noise level.
fraga wrote:
The 7D body appeards to be, at least from what I've seen, basically the same size as the 50d.
And while there is a definitely a difference to the rebel line, the difference to a 1 series is much, much bigger.
rami.halim wrote:
the camera specs are awesome.. but the sample pictures lack the crispness compared to 40D or 5D ... noise at iso 1600 looks terrible as well and not that great at iso 200 either ...
M Vers wrote:
Right now it's just a high fly ball to center field...waiting for the wind to blow any which way. We'll see in a few weeks whether or not it makes it out.
____________________
gfiksel wrote:
I fail to see how showing speckled swatches of pure color can give any idea on the relationship between the noise level and preservation of sharpness and details. Any perceived conclusion on the sharpness level is false because the image lacks of any spatial scale.
It's rather obvious to me the noise characteristics from both sets are vastly different especially at high ISO.
Without ACR's dabbling (i.e., when EXIF ISO values are edited to appear as 100), the noise characteristics look grainy and pleasant looking. This is why ACR treated high ISO images always appear soft even when NR settings for both luminance and color are reduced to zero.
This series of photos is akin to comparing ACR against DPP, all with NR = 0.
A 'neutral' program like DCRAW is probably more suitable when comparing camera RAW output.
It is probably down to weight. My hands are not huge but fairly large.
I held the 7D in one hand and 5D MarkII in the other, and I must say 7D felt much lighter.
This of course can also be due to the fact that my right arm is stronger than the left.
One thing which make me compare it to Rebel series was that 7D's finish felt somewhat plasticky. Another thing to consider is also lenses attached. 5D MkII had 24-105mm L kit and 7D had 15-85mm EF-S lense..so obviously diffference in weight there already..
Shutter noise was also much more subdued than XX series and reminded me of a click of my Rebel. Good or bad? Well...I personally really love the "bolt action" sound of a shutter..
Not really conclusive verdict so to say, but more my associations. Maybe in a way me having negative expectations, hoping for 7D to be atleast APS-H camera...
thw2 wrote:
Without ACR's dabbling (i.e., when EXIF ISO values are edited to appear as 100), the noise characteristics look grainy and pleasant looking. This is why ACR treated high ISO images always appear soft even when NR settings for both luminance and color are reduced to zero.
This series of photos is akin to comparing ACR against DPP, all with NR = 0.
A 'neutral' program like DCRAW is probably more suitable when comparing camera RAW output.
I wonder if it is really NR at play here, or whether ACR is instead toning down its internal sharpening algorithm for the high ISO files, in order not to amplify the noise footprint? I'm just speculating, because I haven't done the testing others have done, but from the comparison posted it seems like different sharpening might explain the results too?
In either case, it's useful information to be aware of.
There we go, I was wrong indeed. Well, remains to see it perform in action.
Aside from weight, I didn't really like Focus Points - outlined bold in brownish red with dots in the middle. Couldn't really see clearly which point was focusing when all points were active. Tried selecting individual focusing points, but same story there. Focusing system is truly different from 1D/5D series.
Saw a magazine doing a 7D exclusive today in a local book store. Photos look great. However, I was not really convinced by the action shots - there were some football photos looking somewhat fuzzy...would love to see some stories on RL AF performance.
Maybe predictive AF suffered from same syndrome as 1D MkIII??
Krosavcheg wrote:
Saw a magazine doing a 7D exclusive today in a local book store. Photos look great. However, I was not really convinced by the action shots - there were some football photos looking somewhat fuzzy...would love to see some stories on RL AF performance.
Maybe predictive AF suffered from same syndrome as 1D MkIII??
Did you see it with your left or right eye this time ?
, the hardest is to find the iso value, it's not always at the same place but on canon files, the bytes orders are inversed, so instead of 0C 80 for 3200 iso, it's writen 80 0C in the file.