In-Store Lens Testing

[AndrewKulin]

I will be traveling to Calgary in a couple of weeks time on vacation and plan to upgrade my Canon EF 75-300 IS with a Canon  EF 70-200mm f/2.8L IS USM Telephoto Zoom.  My camera is a Canon 40D and this lens would complement my 10-22 EFS and 17-55 2.8 EFS lenses.

I have seen posted statements here and there that there can be significant differences in image quality between different individual units of the same lens model, even with L-Series lenses.  And enough so in some cases that purchasers have had to buy/return two or more copies of the same lens before they found a keeper.

Since:

a - I am on vacation and will rather find myself located about an hour or more further west of Calgary; and,
b - I hate having to go through the return process for anything.

What in-store tests would you recommend doing to test out different units of the same lens (assuming the store in question has more than one sample when I get there)?  I would have my tripod, laptop, card reader handy if needed.

For starters I would think that:

Tripod Mounted and combination of:
- 70, 100 200 mm zoom
- 2.8, 5.6, 11 and 32 fstops
- don't think shutter speed or ISO would matter, other than perhaps using mirror lock-up

Hand held to compare IS vs. non-IS.

Compare images on my laptop for sharpness, vignetting, etc.

I also want to keep it short and to the point.  The above list I put together could be quite time consuming to do with 2 - 3 lens samples.  I wouldn't imagine the store would mind accommodating my request (I'd be shopping Vistek, in part because they also have stores in Toronto so if for whatever reason I had to bring back the lens after vacation I could go to the Toronto store.

It may also be overkill doing all this for this particular lens but let's just say I got a bit spooked by some things I have read, and it is going to be my most expensive lens purchase to date.  $2K is a chunk of change

[SecondFocus]

The first thing you should do is to ignore most of that junk you read online. Next, the 70-200 2.8 is a reliable lens and I don't think you are going to find much difference between any of them. Buy your lens, shoot some and make sure it works before you go, that is just a good habit anyway, and then go have fun.

[Ray]

It might well be the case that the 70-200/2.8 is reliably consistent. One would hope so. It's a lens that's been around for a long time and Canon should therefore have sorted out any QC issues.

Nevertheless, it is a fact that QC variations do exist with many lenses. I've experienced it myself and it can be a big time waster confirming that a particular lens is substandard, which is why I would prefer to pay a bit extra for any lens I buy if it could be confirmed with a reliable, real-world MTF test, that the lens meets a certain standard.

One may change camera bodies frequently, but a good lens is priceless.

Lenses that I've returned to the store after rigorous testing have been;

(1) The Canon EF-S 10-22mm due to inaccurate autofocussing and a general lack of sharpness even when manually focussed.

(2) The Canon 400/F5.6 prime. Not as sharp as my 100-400/F5.6 zoom at 400mm.

Lenses which I didn't test thoroughly before buying and which I'm not satisfied with:

(1) Canon 50/1.4. Not quite as sharp as my much cheaper 50/1.8 and also has some difficulty in accurate autofocussing, which is apparent when used at wide apertures.

(2) Canon EF-S 17-55/2.8. Quite sharp at full aperture when manually focussed, but autofocussing is poor.

(3) Canon TS-E 24mm. This is a lens which has been around for a long time and has apparently been improved in recent times. I must have got one of the older models because its sharpness is on the level of a low cost zoom.

Probably more important than comparing different models of the 70-200/2.8 would be first to compare the lens with your 75-300, at the same focal lengths. If the 70-200/2.8 is not significantly better in a worthwhile sense, then you should look at other models of the 70-200.

I don't think there's much point in comparing lenses at F32. Good luck!  

[thompsonkirk]

To begin with, a 100% reliable recommendation - don't buy new equipment for a vacation without having time to get used to it first!  The alternative is to order over the Internet from a reliable dealer & try it for the week that the dealer usually allows, or the 3 days that your credit card presumably allows.

The problems with Canon zoom lenses have in my experience proved to be matters of adjustment.  Just make a few tripod shots to be sure corners are equally sharp.  Once you're past that hurdle, just use it a while.  If necessary Canon service, which in my experience is quite prompt, can adjust the 'ideal focus point' more accurately than they bother to do at the factory.  I've sent 3 zooms in for this, & so have colleagues.  They've always come back performing a little better than new.  

Kirk

PS, as above: diffraction will increase below f8; not much point testing beyond that.

[stewarthemley]

The first thing you should do is to ignore most of that junk you read online. Next, the 70-200 2.8 is a reliable lens and I don't think you are going to find much difference between any of them. Buy your lens, shoot some and make sure it works before you go, that is just a good habit anyway, and then go have fun.
[a href=\"index.php?act=findpost&pid=213572\"][{POST_SNAPBACK}][/a]

I have tested quite a few "pro" Canon and Nikon lenses and there are too often clear differences. Not just in overall sharpness but also in fall off at the edges and corners and more alarmingly in unsharp areas of individual lenses. Considering the cost, Canon and Nikon should offer better quality control. My dslr at the moment is Canon and I tried three 24-70 L zooms and four 70-200 L IS zooms and the differences would have embarrased Canon directors (hopefully).

I would expect the 40D (haven't used one) and a laptop with a reasoanble screen to show any differences quite clearly. I think your ideas re testing are sound (noting the good pint about refraction above f8/f11) and similar to what I did. I went in to a pro dealer (in the UK), said quite clearly what I wanted to do, was given a quiet corner and told enjoy myself. I went to Vistek a few years ago when I was in Toronto and my guess is that they also would be happy to help.

An interesting point that arises from this is, do lens reviewers check for sample variance? I would imagine that if the manufacturer supplies a lens it will be the best they have, but I think it would be a great addition to any lens test to borrow three or four from a friendly dealer and see how consistent they are.

[Geoff Wittig]

Testing the 70-200 f:2.8 L zoom on a Canon 40D won't tell you a thing about vignetting or corner sharpness because the image is cropped out of the center of the lens's field of view. So you could have a mediocre or poor copy and you'll never know. You may want to beg or borrow a full frame camera like a 5D or a 1Ds series, or even a film body, and check out the corners. You really do need to lock the whole rig down on a solid tripod and shoot at f:8 or wider to eliminate camera shake and diffraction losses from the sharpness equation.

[stewarthemley]

Geoff, good point about the edges and corners on the 40D, I'd forgotten it wasn't full frame. Though I'd still expect to see a sample difference wide open up to maybe 5.6. And with the L lenses much of the attraction is in the wider apertures.