pixel_junkie wrote in post #6930110
That is right Tony. It is a random poll with a random group of people taken at a random time. That is all it is, you chose if it is worth anything to you or not.
No, it isn't random. It is fraught with sample bias. When we want to check prevalence of a virus infection in a community, we don't ask people to come in and give us a sample because it only attracts those with strong feeling about the disease (i.e., "Oh, I think I had that disease. I should go get checked."). We take samples from the blood bank and they are randomized. This is how it was determined that four of five people infected with West Nile virus didn't even know it - it's asymptomatic in 80% of those infected.
Here you have the same issue. On top of that, your sample only includes Canon users. It doesn't take into account lens lot numbers. It doesn't take into account camera model numbers. There are lots of issues with "polls" like these. They're fun, and they can be enlightening sometimes, but they generally have little scientific merit.
But I bet you anything that if I take the same poll a month from now and then 3 months from now, the results are going to be very similar to the one I did. The lens has a problem, and that is a fact so it should be mentioned.
All lenses have problems. If you did the same kind of poll asking if the EF 50mm f/1.4 is sharp wide open, you'd find 80% of people saying "yes". But if you look at the images in archive that are shot at f/1.4 you'll see that the great majority are not sharp at f/1.4.
"Raw" is not an acronym, abbreviation, nor a proper noun; thus, it should not be in capital letters.