selmslie wrote:
It's only yada yada if you don't understand the difference.
Precisely my point. Many here do not really understand the difference. Because if they do, they wont be polarising one from the other.
selmslie wrote:
Below are the results from testing six cameras from three different manufacturers. They produce almost identical results for the raw file from about EC-5 through EC+4. The JPEG results vary a little in the middle brightness range from one model to the next. The only outliers are the two 16MP cameras. They all have the same curve shape.
The raw file captures over 8 stops of brightness using integers from about 32 through 16000. These can be converted to evenly spaced log values from about 5½ to 14 but the log conversions are not rounded.
The camera's JPEG uses the converted log values, applies a curve to them and rounds them off as integers from 0 to 255.
Below are the results from testing six cameras fro... (
show quote)
Youve given proof that JPEG is basically the same banana. It's a compression software, yet there are those who would insist that a camera jpeg is worse than a jpeg produced in a pc.
selmslie wrote:
We all know what happens if you overexpose the image. The highlights stop recording at about 16000 and the JPEG can't go higher than 255. Blown raw highlights cannot be recovered.
It is correct that JPEG has a narrower dynamic range, that is why one must be even more vigilant and precise in his/her decisions when shooting directly to jpeg. It takes more skill to create a good jpeg image straight out of the camera yet many hoggers look down on guys who do it.
selmslie wrote:
But two things happen if you underexpose. First, the JPEG does not get close to 255 and it and it moves down the curve toward the left. You will end up with a little more highlight contrast.
The second problem is that the deep shadows move to the left at the same time. This is something we can fix if we start over from the raw file but not if all we have is the camera's JPEG. There may be only 8 possible values left.
Possible, but not all scenes has a wide a stop requirement. In some scenes, JPEG can provide all the dynamic range that is required. If a little bit more is needed then a one can push the D-lighting or any other wide dynamic range settings like "flat" or use other techniques such as multiple exposures. Another is that it wont matter if one shoots RAW if they are pushing the ISO hard, because such raising of ISO narrows the dynamic range of the sensor itself.
selmslie wrote:
The third problem is that any attempt to correct the contrast or brightness in the JPEG can result in banding. That not likely to happen when you are working with the raw data.
Push the RAW hard, the same result happens. Everything has its a limit. What flew in many peoples head is what the question was about.
When there is no requirement to edit, is it ok to use JPEG? and the answer is Yes, yes yes. The question was never about editing. Yet that is what we keep seeing as the answer. Besides, with any single color gradient, the RAW file will still band upon its final convertion to high compression JPEG.
selmslie wrote:
It may take some effort to absorb this concept but once you see the ramifications you will discover why raw is worth the effort.
Tell that to the polarizing group. They have a hammer and everything is a nail.
I will use Jpeg when it suffice and I will use RAW if that is the better solution for the scene. I dont need to compare and polarize. They are just image files from a myriad of other file formats, that has their own field of use.