Poll: Which of the following is the best?
Poll Options
Which of the following is the best?

Reply
Thread Tools
schettj's Avatar
Posts: 501 | Thanked: 292 times | Joined on Nov 2009
#11
I would need to see the original non-resampled images. Two photos taken at the exact same time from side-by-side positions.

Otherwise judging these are irrelevant to determining anything at all.
 
ndi's Avatar
Posts: 2,050 | Thanked: 1,425 times | Joined on Dec 2009 @ Bucharest
#12
FLOWER SHOT

F is pretty well balanced. No obvious corrections and, even though I wasn't at the shooting it seems to be the closest to the original. No odd colors, no signs of weird tampering. Has yellowish tint. Can be corrected though, so it's better like that. 4/5

E tried to do brightening and failed, by nuking the sky. Usually, not sensitive enough sensor, so shot at whatever ISO it could and corrected by burning the image up. Pretty low. Has blueish tint, attempting to correct sensor. Also, too compressed or poorly compressed, I can see JPEG artifacts better than the flowers. 2/5

D is burned to hell. Colors are at 200% to make it stand out, brightness way up. Sky is gone, and even though the lake(?) nehind the trees no is way visible I think it wasn't before. There isn't s spot of dull gray in the image, a sign that you either live in Wonderland or the colors have been turnes way, way up via vibrancy. The flowers have no detail left, they are color patches. It's so bright the white has bled into the trees. Bad, 1/5.

C is blueish, sky is neigh gone. Not as bad as E, though. It's way too blue, some detail lost. Also, oversharpened. I can see pixels turning white at edge intersection from all the sharpening and still it's dull and lacks detail. Usually corrects for bad optics. 2/5.

B has been brightened, sky is gone, a good sign of edge dynamic range processing. Details exist, but faded. Also, lake is up, meaning blacks have been raised to enhance too dark images. Again, tricking user brightness instead of exposure. Color flowers lack detail. 2/5

A Sky is gone, again a sign of post-processing, but not by much. The rest of detail is pretty nice, ground is visible, which means good mid-range depth, but lake is altered, a sign of black raising and the sky is gone, a sign of white rising. Most likely, an adaptive shift. Colors have been altered by curve/vibrance, meaning less color artifacts but still no color saving. Because it's on a curve, darker sports like ground have good detail, while white areas and red areas tend to become washed. 3/5

So, F is my best, even it looks unimpressive, because it does a full capture and details still exist both in color and in darkness/light.

Followed by A (not bad, but artificially brightened and colored)

Followed by B and C, and finally D, which is downright frightful. Seems the only thing that does is make images look nice for a presentation shot, like a flower.


=============================

SUNSET

F is nice. I wasn't there but colors look natural, no dark or bright sport, no missing detail in the shadows. No blur. By the plane silhouette, I'm guessing twice what the pic is resized to. Pretty good shot, it could pass for a camera. Unlikely to pass for a DSLR, but a compact camera could be. 4/5. Very good shot. If it really is a phone and it's not a trick, then 5/5.

E blurry, no details in dark areas, colors have shifted. 2/5

D tree detail is almost gone. Exposure software failed bug time, exposing the sun and ignoring the rest. All dark detail is gone, even white suffers. 2/5

C exact same problem as D. It's even worse. Also, over-compression gives artifacts in the gray area of the sky. 1/5

B is correctly exposed, colors shift but only slightly. It's a bad picture, but compared to previous disasters it looks pretty good. Would be better, but I see JPEG artifacts in several places. Most visible on the roof. Typically, these color artifacts appear after brightening an already-compressed JPEG, where the actual color shifts are subtle. It's either over-compressed or re-compressed. 3/5 because it actually has trees in it.

A building windows are dark, trees are dark but not invisible, some exposure issues, but not much. Same yellow color shift we see in most small CMOS in low light, uncorrected. Not bad, but still, color shifting, exposure bad. 2/5

Again, F seems to be best by far. It's the only one that can actually take a picture under these conditions.

Followed by B, then A, C and D. Throw E wherever.

==========

By both accounts F is the best out there. I'll call it 100%.

A and B have their own issues. Some are tolerable, some are not. Depends on conditions, really. However, visibly below F. I'll put both at 70%

E overcompresses in both shots. Too bad, it could have been a decent camera. I have no idea why it compresses like that. Either it's an older phone with little space and CPU, or the camera is upscaled. I'll say 60%.

D is an odd ball. Nice sensor, has some life in it, but completely ruined is post-processing. Someone really, rally tries to enhance color and contrast and ens up ruining contrast, color and introducing artifacts. Because of the horrendous color shot, 50%.

C exposes bad optics and compression and artifacts in the light and is among the worst in dark. Barely worth taking the shot. 30%

Finally, I know what you took the pictures with. However, I won't say anything, you worked too much for this. Secondly, I reviewed the images before looking, so I'd be fair. Curiosity got the better of me just at this paragraph. I would have never guessed.
__________________
N900 dead and Nokia no longer replaces them. Thanks for all the fish.

Keep the forums clean: use "Thanks" button instead of the thank you post.
 

The Following User Says Thank You to ndi For This Useful Post:
Posts: 3,664 | Thanked: 1,530 times | Joined on Sep 2009 @ Hamilton, New Zealand
#13
Originally Posted by schettj View Post
I would need to see the original non-resampled images. Two photos taken at the exact same time from side-by-side positions.

Otherwise judging these are irrelevant to determining anything at all.
They all are not resampled photos. It's as the original came out of the camera itself. The only thing I change is the resaving method and add the letter A-F to make sure i don't mix it up while uploading If you want none sample then give me your email I'll send it to you directly


NOTE: All pictures uploaded hasn't been enhanced at all. It's how it look like from the camera.

The only thing I did is a little crop to confuse you people. Only one is cropped out of the 6

I've uploaded the RAW image with only information stripped out for you to look at. This would make you shuttup!

http://www.filefactory.com/file/b26c28g/n/original.7z

Last edited by maxximuscool; 2010-07-06 at 03:16.
 
Posts: 3,664 | Thanked: 1,530 times | Joined on Sep 2009 @ Hamilton, New Zealand
#14
All I can say is I'm pretty shock at the comments and result from this

Very surprise!! I'll revealed the answer soon! Keep the comment and vote coming guys!
 
Posts: 842 | Thanked: 1,197 times | Joined on May 2010
#15
F is definitely the best.
 
Posts: 193 | Thanked: 92 times | Joined on May 2010 @ galveston, tx
#16
F is the most well-balanced picture of all.
I am using Fedora 13, Linux OS, with sRGB color scheme on my DELL 2408WFP panel.
 
Posts: 3,664 | Thanked: 1,530 times | Joined on Sep 2009 @ Hamilton, New Zealand
#17
I like F also but i'm not Voting since i'm already know which is which. I'll stay fair.

Remember these are the Mobile Camera showdown! Same spec camera! May the best 5MP camera phone win.

Seriously people being so mean toward "D" and "E" lol.
 
Posts: 1,096 | Thanked: 760 times | Joined on Dec 2008
#18
Originally Posted by maxximuscool View Post
All dimonsion are the same, some a little smaller due to i cropped it to make it a fair test. I do this to stop all the fanboy who would only vote for their device that is why i'm not voting because i know which device is which picture.

But yes all camera here are 5Mpx, should be about the same size :P

I'm not revealing yet.

The outcome is going to a surprise
oh sorry, the first time A must not have loaded all the way and it looked like the wide format like n900 pics.
 
ossipena's Avatar
Posts: 3,159 | Thanked: 2,023 times | Joined on Feb 2008 @ Finland
#19
is this a review of each cameras automatic functions? or why exposure differ approximately 1-2EV at most?
__________________
Want to know something?
K.I.S.S. approach:
wiki category:beginners. Browse it through and you'll be much wiser!
If the link doesn't help, just use
Google Custom Search
 
Posts: 3,664 | Thanked: 1,530 times | Joined on Sep 2009 @ Hamilton, New Zealand
#20
Should I reveal one device? or should I leave this going for a while? Any suggestion? Which one do you want to be revealed?
 
Reply


 
Forum Jump


All times are GMT. The time now is 16:53.