Canon L's - Old vs. New - Any build diffenences?

I'm looking to purchase a 70-200 2.8 (non-IS) and I have seen a few lately for reasonable prices but they are older (by date code one was built in 1995). So my question is - Are there any differences in this model of lens from say 15yrs ago until now? Have they updated and upgraded they way they are built or the components they are made with? Or, are they virtually the same?

Logic tells me something this complex with electronics, glass, motors, etc. would surely have been updated in the last 12yrs - but maybe not. By the prices they are fetching, the shape the lens is in seems to be more important than how old it is, which makes me think maybe Canon takes the old "If it aint broke, don't fix it" approach and they are are all the same.

Anyone know about this stuff?

Chris

I have two of the older L's (28-70 f2.8L and the 70-200 f2.8L) and several of the later L's - there's no significant difference in the build and I've never felt the need to upgrade to newer L's.

Mark

Canon does not typically make changes to their lenses without giving them a new designation - i.e. EF 50mm f/1.8 followed up by the EF 50mm f/1.8 II. That was a drastic change, but that is how they introduce change. The exception is when there is a problem like the flare issue when the EF 24-105 f/4L IS was introduced. Canon recalled them and fixed the issue.

Over the years the L lenses have acquired water resistance, and of course better and faster USM, and better IS. But they do not just put those things on the current models.

Have Fun,

Hi Chris,

There certainly has been at least a mechanical change in the 100-400 L. Up until about 3 years ago, the friction ring required 4 inches of turning or so from locked to fully unlocked; since then it only requires about 1 - 1 1/2 inch. It seems that QC has improved as well. People report fewer initial bad copies these days, if any.

To me it seems they have done something to the IS as well; I get about 50% sharp shots at 1/45s at 400 mm; with IS off I have to use at least 1/500s to get the same results...

Kind regards, Wim