[Nesster]

I tend to use the average rating as a way tell the dogs from the rest.

Otherwise, I have my own values where my esteem of a lens may not be that of the general consensus. For example, the 28mm lenses don't get a lot of love, from Takumar 3.5 to SMC-A 2.8, and they do tend to render a bit differently than other lenses... but I like them.

Rarity and cool factor tend to influence ratings as well - the rarer and cooler the higher the ratings tend to be. In the case of the 28mm:

SMC/Super Takumar 28/3.5 = 8.14
K 28/3.5 = 9.05 (AFAIK this is the same as the SMC Takumar, but with K mount, and rare/cool)
K 28/2 = 9.83 (performance may be super excellent, but do rarity, cool, cost, speed factor into this?)
M 28/2.8 = 8.25
M 28/3.5 = 8.85
A 28/2.8 = 8.5
A 28/2 = 9

Now, maybe these are all very different, I haven't tried them all, but I do know this: the A 28/2.8 renders just about the same as the Takumar 3.5, with similar color, but with a smidge more resolution and is faster. In this the two ratings would seem to be relatively valid. The stuff in between? Who knows. Are the K lenses that much better than the other 28's? I would tend to doubt that.

Another way to use the ratings - one tends to make friends and see people with similar interests. And one tends to see people whose interests and direction seem to be contrary to one's own. I would pay more attention to the first and less to the latter when reading reviews.

[Zav]

Numerical rating is pretty useless to me. You sometimes see new owners giving a 10 to a good but not excellent lens as they are overjoyed by their new toy. Too many personal factors are involved I think. I prefer to rely on the text and on sample shots.

[jac]

Just a personal little rant. I'd like to see all reviews dispense with that nebulous little grading of price/performance in the main ratings. Sure, include this opinion as a second ranking if somehow it's necessary but this should not influence rankings for lens quality.

[paperbag846]

The numbers are useless, read the words.

[geezer52]

Regards the difference between the 18-55 II and the 18-55WR. The "II" is the standard kit lens, generally well regarded for a kit lens, and produced in mass quantities. For many it may be the first and only DSLR lens they've used and they're impressed by the improvement over whatever camera they were using before. The "WR" on the other hand, while offered in some kits, is a bit more of a special needs lens more likely to be bought separately by more experienced and discriminating shooters. That's one thought, while the two are supposedly optically the same, I suspect there may be differences in manufacturing that makes sample variation more common in the "WR" version.



First I agree when a lens has 10 or more ratings, the aggregate rating is a good indicator of the lens's worth in both quality and value.

When reading other reviews I look first for any from individuals whose opinion I've come to respect. Then I go thru the rest and try to put the individual ratings in context with what the individuals said - It's usually pretty easy to tell the ones that are isolated opinions, or over the top for good or ill. And while I think rating a lens for value or "bang for the buck" is perfectly valid, the reader needs to use some common sense in forming expectations about the len's absolute quality. For such lenses it may be best to simply consider the ratings as a "happy factor".

Expensive lenses imply quality and deserve to be evaluated critically. I also read these reviews most critically. These lenses are an investment that I buy for the long haul and my expectations are quite different from what i expect from any knock-around lens.

Review readers need to clear in their own mind what they need the lens to do for them and for what purpose they are going to use the lens for. I've sprung for what I thought might be hidden gems a few times with varying success but never when i needed a lens for a specific purpose. Those "gems" are for me to play and have fun with and maybe later using seriously.

[RioRico]

Even aggregates can be misleading. Case in point: Takumar 55mm's f/1.8 vs f/2, both with sufficiently-many reviews (36 vs 14 now). EXACTLY the same lens, EXACTLY the same optics; but one is jiggered so it's less than 1/3 stop slower. The current ratings are: f/1.8 = 9.25 vs f/2 = 8.15. That <1/3 f-stop makes the most minimal difference in performance (I know, I have both), yet a tremendous difference in ratings, and prices.

What the overall lens ratings show isn't actual performance, but perceived price/performance. Many of the comments on the FA100-300 (silver) admit that it's optically and functionally fine, but it looks cheezy on a black camera, so it's down-rated. Perception, not performance. So I have only moderate confidence in these ratings. Yet I still read them, and base decisions on them. Arrrgh...

[Just1MoreDave]

K 28/3.5 = 9.05 (AFAIK this is the same as the SMC Takumar, but with K mount, and rare/cool)

No - 8 elements vs. 7, and a different minimum focus distance. If I remember right, they look a lot different side by side, but it's been a while.

I think about the K28/3.5 in particular when user opinions are mentioned. My copy was maybe a 5. I've used maybe ten different 28mm lenses, have six now, done some comparative testing, lots of buying and selling, taken actual photos, and someone might conclude that my opinion is useful. In this case, it's not. I am pretty sure either my copy was bad or I was just totally wrong about this lens, and others must be right. If I reviewed this lens, mine would be the one to chuck out.

Overall, the reviews aren't perfect but they are more useful than any other user review on the internet.

[newarts]

The numbers are not totally meaningless but can vary a lot compared to truly objective measurements. The numbers give a false implication of objectivity given the lack of quantitative measurements.

Comments contain much more usable info; for example when someone downgrades a lens due to a feature you don't care much about. Also with long zooms I pay attention to wide open center sharpness at the long end (that's where I use long zooms.) On the other hand, wide angle edge performance rules (if you don't care about the edges why use a wide angle?)

[stevebrot]

The numbers are useless, read the words.

User ratings are very useful, read the words...


As with all numbers, they should be taken in context. Look at the average and consider the outlyers. Look at the pros/cons. Remember that value (price/performance) has a great influence on rating. Consider too, the experience of the reviewer and the date of the review. High post count and forum participation is not a bad measure of review validity. And probably most important of all is the notion that almost all lenses are generally pretty good. Only the true stinkers get rated below 5.


Steve

(...has contributed reviews for all of the M42 and K-mount lenses in my collection...none rated above 9)

[Lowell Goudge]

Just a personal little rant. I'd like to see all reviews dispense with that nebulous little grading of price/performance in the main ratings. Sure, include this opinion as a second ranking if somehow it's necessary but this should not influence rankings for lens quality.

I disagree. For me, the price of a lens determines an expectation of performance if the lens performs as you expected for the price then I would say it is an 8. I would only rate a lens higher if it surpassed the expectation I had based on the price I paid.

[rparmar]

The numbers are useless, read the words.

To add insult to injury, many of the words are also useless.

[SpecialK]

I think they are about as accurate as Amazon reviews.

[Digitalis]

I think they are about as accurate as Amazon reviews.

that is a little harsh.

[reeftool]

Another point that is relevant (to me at least ) - which camera you are using. I gave up on the 55-300 because I could not get satisfactory images in fairly low light - my fault of course, not the lens', and note that I did not enter a review for it. That was with my K-7, which is a fine but iso-constrained camera. Now I own a k-x and can bump iso up another stop or more without discomfort, and the 55-300 regains some appeal.

Funny to look back - I really liked the 16-45 and 55-300 lens combo, but with the K-7 became disillusioned and sent them on while trying other lenses (especially WR types). A year and a camera later, I've gathered both those lenses back into my fold. Thankfully they were regained with exceptional deals, so other than time & sanity I lost very little..

An important point made. Far too often, someone purchases a lens and tries to use it for shots beyond it's capability, your low light shots for example. So they try to shoot a night ball game and can't get the fast enough shutter speeds needed and then rate the lens low. I bought my DA 10-17 Fisheye in the Marketplace at a very good price because it's previous owner somehow believed he could remove all the distortion in PP. The differences in the rating of the various 18-55 models would be expected as a lot are probably from beginners just learning the craft. I like my 18-55 and while I have a few lenses now that overlap in that focal range, the kit lens is still in my bag. As I look back through my photos, I have a huge amount of wonderful shots with that lens.

I do have a problem with reviews who list dumb things as "cons". I have seen barrel distortion listed as a con for fisheye lenses. Well, without it, the lens wouldn't be a fisheye. I have also run across reviews that said too short. What? Why did you buy it? It's not a con if a 200mm lens isn't as long a telephoto as a 300mm.

[Adam]

Since the ratings consist of just a single number, you have to read the reviews to get an accurate sense of the optical quality of a lens. It would be better if we had multiple scores, i.e. one for each parameter as we have in the in-depth reviews.