Re: [情報] Nikon D800 tested at DxOMark, gets the
我一直在查 D800 的高感數值是怎麼測的,為什麼可以跟 D4 不相上下?
甚至比 D700 還高那麼多?
這樣 D4 為了高感光低雜訊把畫素降到 16mp 的優勢在哪裡?
我目前看到兩種說法:
1. DxoMark 是以純淨度為目標測的。所以這數值說明 D800 和 D4 在 iso 3000 的時候
都是非常純淨的(事實上這對 D800 還是非常了不起了),但在 is 6400 尤其是
12800 的時候,D4 的雜訊抑制就明顯甩過 D800 了,而這部分剛好是 DxoMark 的測
試方式表現不出的。
2. 其實同片幅下畫素愈少、高感光的畫面愈乾淨是錯誤的,或說只有一半對。
(即 pixel density 和高感光能力的正相關理論。)
原因是,如果你將一張 36mp 高感照片降噪後製,再縮成 16mp 的大小,它的畫面和
16mp 的同片幅相機下的高感照片基本上就會是一樣的。(假設我們忽略電子運算的
差異。)
該理論的測試照片在此:http://tinyurl.com/7hxfvu6
但現在我覺得第二個理論用在解釋 DxoMark 給 D800 的數值是說不通的。
因為它在之前評測 D3X (24mp)和 D3s (12mp)時顯然不是這麼做的。
--
※ 發信站: 批踢踢實業坊(ptt.cc)
◆ From: 24.15.67.209
推
04/01 10:49, , 1F
04/01 10:49, 1F
推
04/01 10:58, , 2F
04/01 10:58, 2F
http://www.dxomark.com/index.php/About/Sensor-scores/Use-Case-Scores
Sports & action photography: Low-Light ISO
Unlike the two previous scenarios in which light is either generous (studio)
or stability is assured (landscape), photojournalists and action
photographers often struggle with low available light and high motion.
Achieving usable image quality is often difficult when pushing ISO.
When shooting a moving scene such as a sports event, action photographers’
primary objective is to freeze the motion, giving priority to short exposure
time. To compensate for the lack of exposure, they have to increase the ISO
setting, which means the SNR will decrease. How far can they go while keeping
decent quality? Our low-light ISO metric will tell them.
The SNR indicates how much noise is present in an image compared to the
actual information (signal). The higher the SNR value, the better the image
looks, because details aren't drowned by noise. SNR strength is given in dB,
which is a logarithmic scale: an increase of 6 dB corresponds to doubling the
SNR, which equates to half the noise for the same signal.
An SNR value of 30dB means excellent image quality. Thus low-light ISO is the
highest ISO setting for a camera that allows it to achieve an SNR of 30dB
while keeping a good dynamic range of 9 EVs and a color depth of 18bits.
A difference in low-light ISO of 25% represents 1/3 EV and is only slightly
noticeable.
As cameras improve, low-light ISO will continuously increase, making this
scale open.
推
04/01 11:02, , 3F
04/01 11:02, 3F
筆誤 多謝指出
※ 編輯: breath35 來自: 24.15.67.209 (04/01 11:11)
→
04/01 11:46, , 4F
04/01 11:46, 4F
→
04/01 11:47, , 5F
04/01 11:47, 5F
→
04/01 11:47, , 6F
04/01 11:47, 6F
→
04/01 11:48, , 7F
04/01 11:48, 7F
推
04/01 11:53, , 8F
04/01 11:53, 8F
推
04/01 12:03, , 9F
04/01 12:03, 9F
→
04/01 12:03, , 10F
04/01 12:03, 10F
→
04/01 12:04, , 11F
04/01 12:04, 11F
→
04/01 12:05, , 12F
04/01 12:05, 12F
→
04/01 12:05, , 13F
04/01 12:05, 13F
推
04/01 13:04, , 14F
04/01 13:04, 14F
推
04/01 13:11, , 15F
04/01 13:11, 15F
→
04/01 13:11, , 16F
04/01 13:11, 16F
推
04/01 13:21, , 17F
04/01 13:21, 17F
推
04/01 13:39, , 18F
04/01 13:39, 18F
推
04/01 16:14, , 19F
04/01 16:14, 19F
推
04/01 17:00, , 20F
04/01 17:00, 20F
→
04/01 17:01, , 21F
04/01 17:01, 21F
→
04/01 17:02, , 22F
04/01 17:02, 22F
→
04/01 17:02, , 23F
04/01 17:02, 23F
推
04/01 20:39, , 24F
04/01 20:39, 24F
→
04/01 20:39, , 25F
04/01 20:39, 25F
→
04/01 20:40, , 26F
04/01 20:40, 26F
→
04/01 20:43, , 27F
04/01 20:43, 27F
推
04/01 20:57, , 28F
04/01 20:57, 28F
推
04/01 22:27, , 29F
04/01 22:27, 29F
是的,如果是normalize到一樣大小後,並考慮雜訊/細節的比例,那就是理論二。
這樣 D800 測出來的 95 就有作弊灌水嫌疑。
好像之前他們給 D700(12mp) 高感光 2300 的分數,5D2 (21mp) 1800,在我看來也
是不合理的。
※ 編輯: breath35 來自: 24.15.67.209 (04/02 01:22)
推
04/02 10:00, , 30F
04/02 10:00, 30F
推
04/02 10:44, , 31F
04/02 10:44, 31F
→
04/02 10:44, , 32F
04/02 10:44, 32F
→
04/02 10:47, , 33F
04/02 10:47, 33F
→
04/02 10:48, , 34F
04/02 10:48, 34F
→
04/02 10:52, , 35F
04/02 10:52, 35F
→
04/02 10:53, , 36F
04/02 10:53, 36F
推
04/02 12:08, , 37F
04/02 12:08, 37F
→
04/02 12:09, , 38F
04/02 12:09, 38F
→
04/02 12:19, , 39F
04/02 12:19, 39F
推
04/02 12:34, , 40F
04/02 12:34, 40F
討論串 (同標題文章)
完整討論串 (本文為第 2 之 2 篇):
DSLR 近期熱門文章
21
42
PTT數位生活區 即時熱門文章