Personal View site logo
Make sure to join PV on Telegram or Facebook! Perfect to keep up with community on your smartphone.
Expose To The Right (ETTR)
  • 55 Replies sorted by
  • @LPowell, it seems that there are two different expose to the right understandings. To me expose to the right was always about perhaps overexposing visually (using a histogram as a reference so not to clip the highlights) but recording more info and ten correcting in post. The article you are unhappy with seems to concentrate on the exposing to the right as the right being an absolute limit for the highlights, which for the high contrast (as the scene they use in the article) scenes will of course cause an extreme underexposure. This however does not change the fact that it is a good practice for digital just like it was for the slide film to expose for the highlights (do not clip them) and then use all the fill and bounce you can afford to equalize the exposure. I fully agree that for the low contrast scenes expose to the right is the best way to go.
  • To me ETTR only makes sense with a recording at 10 bit (or more), like RAW, S-log, Log-C or the like plus if a serious color grading session is in your budget and time.

    If you recording in 8 bit you need to get exposure and contrast right in camera, no second chance. You may say that working with a cheaper camera is more demanding than with an expensive one.
  • @dkitsov
    I don't think I've ever seen anyone advocate overexposing highlights on a digital camera, and I think that's where the article indulges in a straw-man argument. In addition to blowing out highlight detail, overexposure will shift the color balance of any non-white highlight. A reddish-orange car taillight, for example, can turn yellowish as the red channel saturates before the green does.

    In my view, the author's most misleading claim was "These days, noise is really not a big source of image quality loss..." In my work, I hit the noise floor every time I deal with GH2 shadow detail, especially with 8-bit AVCHD frames whose noise bits are much more granular than 12 or 14-bit RAW images. It's always preferable to digitize shadows with as many bits as you can squeeze out of the camera, so long as you don't blow the highlights. In post, the more bits of color depth you have, the easier it is to grade the shadow details.
  • @LPowell
    The author is talking about digital raw stills and I think he's making a good point.
    Digital noise is not as much of a deal breaker as it used to be. And even more so if you compare to film.
    I'd like to see some ISO 51000 film that's as clean as what is possible from some DSLRs today.. oh wait.. not possible.
    However, abrupt blown highlights are still a problem even with the best DSLRs out there. It's a dead giveaway that it's digital.
    But to each their own I guess. I'd personally have some shadow noise and avoid blunt highlight clips.
    Again, this article is written completely about RAW stills and not video.
  • @hishimaru - "The author is talking about digital raw stills and I think he's making a good point."

    As long as you've got good lighting along with 12 or 14-bit RAW color depth. Only then are highlights more of an issue than noise. In nighttime available light, however, pulling the shadow details off the noise floor is still crucial, along with keeping ISO as low as possible.