忽然想到photozone网上可以参考下各镜头的光学质量，
就进去看了下，洋文不懂，主要看图。

20饼的畸变只有0.14%，居然好到这种程度，可是下面还有一个表示畸变的图，再借助于翻译软件一看，原来下面一个图才是真正的光学畸变，达到了3.26%，上面一个是软件校正过的。

这样一来，如果拍RAW格式，是不是就没有校正这个动作？同一场景，可以拍出两个效果来？

此页最下面有一张对比图，大概就是有校正和没校正的对比，鼠标放上去或是移开就有变化，而且很明显。17饼更差，色散更大。

据说LX也是校正过的。

http://www.photozone.de/olympus--four-thirds-lens-tests/464-pana_20_17?start=1

如果装在EP系列上，那会怎么样呢？还有G1有没有机内修正呢？

这看各人的接受程度了。饼干头光圈要做大，总是要做出点牺牲的。

我在GF1上用这个镜头，用Adobe Camera Raw解raw，畸变已经经过机身处理，跟jpg没有区别。

按说小像场镜头应该更好设计才对，20饼就一标头，应该是很好设计的，为什么变形还那么大呢，比如宾得的40饼（M40是全幅的）畸变好像也没这么大。是不是松下有了软件修正的技术就不在光学上下功夫了？或是这样来节约成本？

dpreview:
As expected, the 20mm F1.7 pancake takes advantage of the Micro Four Thirds system's ability to correct geometric distortion in software. For most users this is completely transparent - the camera corrects both the viewfinder image 'on the fly', and the JPEG files it records. Also, both the software supplied with the camera (Olympus Master or SilkyPix) and industry-standard alternatives such as Adobe Camera Raw convert raw files correctly too. However users who like to experiment with more obscure raw developers which are unable to apply the requisite corrections will find themselves with distorted images.

To investigate the lens's 'native' distortion prior to correction, we shot our standard test in raw and converted the image using dcraw, which doesn't apply any corrections. The comparison below illustrates the transformation from the barrel-distorted raw image to the properly-corrected final version. Notably the 20mm has lower native distortion than the other MFT lenses we've tested so far, so the corrections required aren't as extreme. In context many superzoom lenses for DSLRs show more barrel distortion than this, albeit at wideangle settings.


- Raw/JPG 软件(SilkyPix,ACR)/机内都自动修正畸变。
- 20/1.7真实畸变并不比目前其他m4/3镜头或者DSLR大变焦镜头差。

Photozone:
It's a bit surprising but the lens has a fairly mediocre vignetting characteristic and it is not automatically corrected.

某些机器还修正镜头失光，不点名了。
不过m4/3镜头没有被自动修正失光。