Pop Photo Lens Testing Methodology

I am interested in purchasing either the Canon 100-400mm L, Sigma

100-300mm f/4 or Sigma 80-400 f/4.5 for use on my Canon 300D.

Readers of sites such as Fred Miranda and PhotographyReview.com give

them very high ratings in terms of sharpness.

 

However, Pop Photo was not as kind to the Sigmas especially at the

long end. They test resolution at the closest focusing distance at

various marked focal lengths. Is this the right way to test

especially for a lens that will most likely see use at the longer

focal lengths? I intend to use either lens for zoo photography and

doubt I will be using the lens at its closest marked distance (about

5.9 ft) very often. Since my subjects will be animals and not

architecture, brick walls and such, I am not overly concerned about

CA, distortion, etc.

 

Which of the reviews carries the most weight?

 

Thanks in advance for your opinions.

<I>Which of the reviews carries the most weight?</i><P>

 

The only review that really counts is if you try it yourself. There is some sample variation

in any complex lens, so if a published review is good (or bad) you could get a sample that

is different. For any of these lenses, don't expect the kind of sharpness you'll get out of a

good prime at 300-400 mm. If there are a lot of reports of good results from forum

contributors, that's a good sign -- although it's true that users can be a little self-

delusional about the quality of gear they've spent a chunk of money on.<P>

 

<I>Since my subjects will be animals and not architecture, brick walls and such, I am not

overly concerned about CA, distortion, etc.</i><P>

 

I can understand why you might not worry too much about distortion, but CA is not good

for any photographic subject. It can be corrected in software to some degree in a digital

image, but the less there is in the raw image, the better.<P>

 

I also noticed that rather unflattering Pop Photo review of the Sigma 80-400, but people

who have it seem to like it. I own the Canon 100-400, and while it got a somewhat better

PopPhoto rating at 400 mm (? IIRC?), my sample is a smidgen soft wide open at 400.

However, it's

much improved if stopped down a little bit -- very usable, although certainly not as good

as what I

can get out of my 500/4, even with converters.

.

 

You answer your own question, Louis. Since you read the review and interpreted it using your own criteria, then you decided what's important for you -- that's all any of us can do before we buy it and try it.

 

Now, if you add compatibility and resale value to your criteria, you may seriously reconsider ever buying Sigma electronic lenses for anything but Sigma cameras!

 

But, hey, as you say, long distance, non-architectural subjects is your main criteria, so you may have a budget-driven winner ... for now.

 

Let us know what you do and how it works for you!

 

Click!

 

Love and hugs,

 

Peter Blaise peterblaise@yahoo.com http://www.peterblaisephotography.com/

Reviews from users can only give you one solid information: if no one is happy with one lens,

that lens must be a complete disaster.

 

Standard tests from same source must be far more reliable. I'm very fond of photozone

http://www.photozone.de/8Reviews/index.html for that kind of data.

 

Take a look at his reviews of Canon 100-400 f/4.5-5.6 and Sigma 50-500 f/4-6.3

Although pop photo has a dubious reputation among some forum members for their testing I have found it to be very accurate. It matches my experiences with my own lenses. IMO I would give some weight to their ratings in my decision process. Bob.

In my opinion, the Canon 100-400L gives very good results on 1.6 crop bodies, such as my 20D.

 

Look at this shot, taken at f/5.6 (wide-open) at 400mm (equivalent to 640mm) at 1/160 second (thanks to IS) on my 20D:

 

http://www.pbase.com/bellavance/image/45200814/original

 

Pierre

Pierre: that's a nice picture of a hawk, but to be honest, it didn't strike me as particularly

sharp and clear. Undoubtedly that's partially due to compressing the original image to fit

Web bandwith, but I'm just not seeing the kind of detail that's possible with a good prime.

As an example, <A HREF="http://www.biology.ucr.edu/personal/MACphotos/birds2/

BBP2b.jpg">this picture of a black-bellied plover</a> has been

compressed to about 1/4 of the original pixels. However, it shows details down to the

barbules in

individual feathers (at least for those parts of the bird within the very narrow DOF). This

was shot with a 500 mm/4 plus 2X converter at 1/250 sec as the bird ran past me (hence

the blurred foot).<P>

 

Warning: although compressed, it's still a big file (~ 1.3 megabytes).

Pop Photo is often ragged on by Photo.net posters for being too kind on their lens tests.

So, if they say a lens is a bit soft at the long end, it probably is. My experience is their lens

reviews are accurate.

The problem with isolated tests, however thoroughly and independently conducted, is that they are reporting findings on one particular copy of a lens, which may be a poorer than normal copy. The much respected photodo tests (no longer conducted) significantly underrate the Canon 300 f/4L IS, for example. The problem with consumer opinions is that mostly they are formed without the benefits of the rigor of well conducted tests, and it can be difficult to evaluate the worth of individual opinions. If one particular test shows a lens to be a less good performer than other reviews, however reliable the tester reputation, it probably should be discounted.

 

So far as these particular lenses are concerned, I think the average opinion is that the Sigma 80-400 tends to be slightly sharper than the Canon 100-400; the OS works nearly as well as the IS; but the lack of HSM makes it much slower to focus than the Canon. The Sigma 100-300 stands comparison with the Canon 300 f/4 at 300, and although the prime is a hair sharper Canon offers no zoom that competes with it.

 

So far as compatibility is concerned, there are occasional reports of the Canon 100-400 causing camera lockups intermittently, sometimes resulting in the need for replacement of the IS mechanism. I have not seen reports of any such problems with the Sigmas.

 

So far as cost of ownership is concerned, people often seem to get number blind when evaluating the economics. If a $1,400 Canon resells for 80% of new, that's a loss of $280. Add the compound interest forgone for spending an extra $400 over a $1,000 Sigma during the period of ownership, and that will give the breakeven loss on owning the Sigma.

Louis, I think you may be misunderstanding the Pop Photo tests.

 

The close focusing evaluations in the text are to evaluate the close-up ability. The numbers in the charts evaluate the performance at infinity focus.

 

My opinion: If the 16x20 chart values are close to 90, this is outstanding; 80 is excellent. Note that best performance is usually at f8 or f16.