近日收到不少網友意見,指 DCFever.com「吹水閒聊」充斥愈來愈多粗言穢語、人身攻擊、誹謗等不理性討論。作為 一個以攝影興趣為主題的網站,此等漫罵式討論已超出 DCFever.com 設立討論區之原意。DCFever.com 呼籲大家以 攝影交流為目的作理性討論,並正密切留意事態的發展。
上個主題 :: 下個主題 |
作者 |
訊息 |
kkyy
註冊: 2003-01-14
上載我的肖像
|
CANON & NIKON :銳利對柔 (Part 2) Post time: 6 月 17 日 |
|
|
There was a previous thread (again!) on the same old story, initially didn't want to go into the debate because of mostly fruitless outcome. However, after replying in another thread of "DC fever 今非昔比", I think I should put my suggestion into action: posting to contribute.
________________________________
When will this local-grown myth of "CANON & NIKON : 柔對銳利" be corrected?
How about some objective tests on camera sharpness?
http://www.imatest.com/docs/sharpness_comparisons.html
For those who can't read technical English or too busy to read through, let me summarize a few points that I think should stay, rather than "XY Company cameras are sharp/soft":
1. It is not the lens that is the bottle neck of the perceived difference in sharpness between the 2 brands.
2. It is not CMOS or CCD that is the bottle neck either.
3. The different anti-aliasing filter used has some effects on final sharpness.
4. Nikon D70 choose a combination that produce an outstanding performance in sharpness, but with moiré in return. The Canon 1Ds, however, is even sharper. The EOS 10D + 300D is considerably softer. Do you know how many of these models sold from these 2 companies? Obviously the 1Ds is not the best selling, so this perpetuated the generalization of the fact, or myth, that Nikon is sharper than Canon. Look at Nikon D100 from the test, it is much softer than D70.
5. The Canon 20D is tuned to be sharper after the complains from the previous 1D2 users, so the sharpness now is comparable to D70. In fact from dpreview's test, with MTF comparable lenses used (50mm 1.4 on both cameras), the 20D is found in that test to be sharper than D70.
http://www.dpreview.com/reviews/canoneos20d/page25.asp
6. Even for the same camera, say my 1D2, shooting RAW or JPG has significant effect on perceived sharpness.
Look here:
http://www.pbase.com/feharmat/image/29893590/original
This is a single photo taken as RAW+JPG. The upper part clearly shows what the RAW capture should resolve but the lower part shows that the on-board algorithms smooth-out a lot of detail at or beyond Nyquist spatial frequencies.(claimed by many as a way to avoid moiré/colour artifact)
I don't have solid data but I suspect as a whole people shoot JPG much more than RAW+post processing. That will also contribute to the perceived softness of Canon JPG in some previous models.
--------------------
Too boring right? One simple advice, drop the firmly held belief that XX brand is sharp/soft, instead, compare individual camera, in a scientific way.
Happy shooting.
kkyy
|
|
|
|
 |
雞能
註冊: 2004-01-20
上載我的肖像
|
Re: CANON & NIKON :銳利對柔 (Part 2) Post time: 6 月 17 日 |
|
|
addition,
Canon use CMOS for very huge noise level than any brand CCD(DSLR)
at original image data directly from CMOS.
Each pixel in Canon CMOS have internal circuit to do noisereduction, noise reduction process will make pictures softer than original, so Canon user can feel happy to use ISO 1600 and get softer picture... and this process make image look perfect smooth that peoples like.. that's why Canon say
"IMAGE PERFECTION" solgan
...and CMOS is the cheapest sensor in the world,
Canon make high profit on that @!#$...
just like magic, it turn rubbish to the best...
|
|
|
|
 |
|