Announced today on the LensRentals blog, is their intentions to begin testing lens variance - something never formally done in photography before. Lens variance is simply the difference from one copy of a lens to the other. By testing this, they'll be able to debunk any misinformation found in reviews that test only a single copy of lenses. This will result in higher accuracies, and more information prior to buying.
The article itself is incredibly nerdy - or as article author Roger Cicala calls it - a Geek Level 3 article. That said, the information provided could bring a whole new level of data to lens testing, showing us which lenses and brands have the sharpest image and lowest variance across the board.
I remember when I first purchased my Sigma 50mm f/1.4 (non-Art). I purchased locally because I had heard issues of inconsistency in sharpness from one copy to another. The result was me having the local shop's employee bring out 5 different copies, as I sat there for 45 minutes with my camera, diligently testing the sharpness by my own eye. I wanted to make sure I had the sharpest copy available, which meant I had to jump over some hurdles. The reality is that they all were likely close to identical in sharpness, and I was just perpetuating those rumors, as I had no real way to test these lenses, other than by hand.
Finally, there soon will be a scientific measurement used to see the variance of each lens in the coming weeks. While the news has only announced so far, in the coming weeks and months, they'll have even more information to look over prior to buying our next new lens.
For an incredibly detailed and nerdy explanation on how they're finding the lens variance, be sure to head over to LensRentals' blog.