Khalid Sayood's book on data compression does a good job of explaining
the underlying science, partly in terms that I can understand. The
bottom line is that nothing is free. Log can do a good job of
protecting shadows and highlights by placing them in the codec's "sweet
spot" but if the codec is also being told to apply lots of compression,
it will be applying more to the midtones. OTOH, if high compression
isn't needed, log might be helpful. --J.B.
bigfish@pacbell.net [Avid-L2] wrote:
>
> I've heard it said that shooting log puts things into the sensors
> sweet spot, as been mentioned in this thread, but I don't understand
> that. My understanding is the intention is to sacrifice detail in the
> highlights to afford more detail in the low lights where it is more
> critical. Log is a math transform to compress too much contrast range
> into a smaller data range that the codec can handle. I don't see how
> any of this puts things in the sensors "sweet spot." The sensor is
> hit with whatever light comes through the lens and responds
> accordingly. The Log C or whatever transform is applied after this so
> it has nothing to do with a sensor sweet spot as far as my limited
> knowledge tells me. I'd welcome an explanation as to how the math
> transform has anything to do with the sensors sweet spot.
>
>
Posted by: john beck <jb30343@windstream.net>
| Reply via web post | • | Reply to sender | • | Reply to group | • | Start a New Topic | • | Messages in this topic (10) |
No comments:
Post a Comment