Does anybody know why zoom lense ranges are watershed at 70mm (in 35mm)?

[xpatUSA]

I assume your 17-55mm is a crop-sensor lens? Than 55mm becomes equivalent to 70mm (more or less) on full frame. Still variation on the 70mm theme. 12-24 on the long end, obviously, borders the short end of 24-70mm.

But using the so-called "crop factor" doesn't work for MFT, as I mentioned earlier.

http://www.four-thirds.org/en/microft/lens_chart.html

I'm not trying to create something that doesn't exist. If you believe there isn't enough evidence of 70mm serving as the dividing point for many systems, then fine. But I thought there is and wondered why. The reason may have been purely arbitrary by the frontier, mindlessly replicated by the followers, as Colin suggested . . .

. . . and with which I agree. I even provided a table showing conclusively that 70mm is exactly half-way between the historically 'normal' and 'short telephoto' focal lengths, in old 35mm film terminology. However, that does seem too simple a reason to be acceptable - not enough PO (photographic obfuscation) perhaps.

Could it be that lens designers think more in terms of angle of view than they do in focal length? So, could the watershed be an AOV?

Perhaps we should go to somewhere like slrGear.com and do some statistics on all their listed lenses, wouldn't take more than about 70 seconds or so . How about a histogram of the number of occurrences of 70mm as the long end and another for 70mm as the short end? Perhaps repeat for FF, APS-C, MFT . . . . and Foveon but not Canon (just kidding).

Rick, OT but I had a D50 for a long time and bought the 18-55mm kit lens for it. Boy, talk about flare . . .

Cheers,

[shreds]

Manufacturers will produce lenses to suit pro use first and foremost. Therefore it is hardly surprising that the historic precedents have followed through as most pros still use FF cameras.

You also have to consider carefully what part of the market the photographers are in, and what they shoot. This has a major influence.

Not that other sectors of the market are unimportant, far from it, and as mentioned the cheaper 500mm market is largely left to the likes of Sigma and Tamron, because the volume sales with sufficiently high pro quality were presumably not seen as sufficiently profitable by the big boys. (Having said that they may have missed a trick there).

I personally, rarely shoot at 70mm, despite having good glass at that length. Generally a slightly wider for general or indoor use is required. Alternatively I go for longer lenses for portraits as appropriate for my use, but each to their own. Interestingly there are few prime lenses at this length from the big boys (correct me if I am wrong). Clearly not in demand (?) and hence the reason why there may be a break at this point?

Does it really matter? I change lenses all the time to suit the situation/the shoot and it is probably as much about convenience as anything. (Or being cynical, why only sell one lens when you can sell two, if I were in Canon or Nikon's shoes)

[ajohnw]

But no penalty in sharpness on mine anyway:



http://slrgear.com/reviews/showprodu...uct/349/cat/31

One of my better eBay gambles

Compared with what - I was just referring to a comparison with high quality primes. Optical designers are capable of producing optics with the same levels of quality over various ranges. I should also have said at some setting too but as things get longer it's usually the long end that suffers.

The test result you have posted doesn't relate to any figures.

A better test is this one which shows it as it is - or at least the sample tested.

http://www.photozone.de/canon-eos/51...osapsc?start=1

You will find the Canon equivalent on the same site. Also their 17 to ?? more or less 3:1 range zoom. Unfortunately a cheaper range lens as they also do an F2.8 but it's noticeable that the performance throughout the range is more even.

These tests don't tell the entire story though as they are based on contrast. A lens is perfectly capable of putting most of the light it collects into a small spot and spreading some of it over a larger area. That has an effect on it's resolution in tests. I posted my own test of an olympus 40-150mm zoom which showed that photographically it's more or less limited by the pixel size in the camera. The test results see it as a good but not spectacular lens - contrast is reduced a bit because some light spreads outside of it's best minimum circle of confusion. Actually tested this way it's markedly better than the Pansonic at 150mm which goes to 200mm. The Panasonic generates more contrast but the resolution is woolly - blurred edges. Primes can have the same differences as well. The designers can do within limits what they decide to do. One thing they are unlikely to do is produce a slower lens which is far better than a more expensive faster one even though they probably could - they might just do that on a dedicated macro lens as there is a reason but they are bound to up the cost too.

No comment on m 4/3 lenses because as I see it they are broadly similar in the from toos as other crop factors with more inclination to offer nearer 3:1's in their ranges. Getting the most out of little pixels at a price and taking advantage of the fact that things get easier as crop factors increase - up to a point,. Early lenses on 35mm tended to go 28,35,50,100,200,300,400,500,1000mm some of which are unusual as doubling each time is more general plus odd debates about the best portrait length etc and specialist lenses. 135mm too as it was reckoned to be a max hand hold under the normal range of lighting conditions. Easy to handle too even at F2.8. 24mm etc came but was expensive. Glass improvements have had a big effect at that end.

John
-

[xpatUSA]

Compared with what [?] - I was just referring to a comparison with high quality primes.

Hello John, didn't mean to irritate,

I was only responding to the one statement, which I took as talking just about 4:1 zooms. It said:

"There have been some [zooms] that hold up over a 4:1 range but there is usually a significant penalty at the long end. I would be inclined to say always really."

You were inclined to say that there is always a significant penalty at the long end. I must have missed the part were you said "compared to high quality primes"

Doesn't really matter, I was just showing off the fact that I have a Sigma 4:1 "macro" zoom that, at least, is pretty sharp wide open. Probably has terrible CA, poor contrast, severe coloration, awful distortion, seriously bad vignetting but it's sharp, by golly.

Thanks for the link to PZ but it refers the latest version, mine is an earlier model with no OS, no HSM and different optics.

Later,

[ajohnw]

No irritation here, I just sound like that sometimes. I may be irritated by something else at times too but not in this case. I do have a bit of a thing about some styles of testing in reviews. Why not just publish line pair per mm and then point out the implications on various formats the lens can be used on. The test I linked to works on the basis of sensor height and how many lines can be projected onto it I suppose we could maliciously work out how finely it uses the pixels.

No my fault anyway I should have mentioned primes but in general a 3:1 will be found to be better than a 4: 1 and that better than a 5:1 etc. At the short end things do get worse and by the look of many at the long end too. But as pointed out there are all sorts of complications even what's in front of the camera when the lens is used.

Tests are a load of ############# in some ways but me for one will always look at them as they are a sort of guide but I feel some are better than others.

John
-

[Glenn NK]

I was just showing off the fact that I have a Sigma 4:1 "macro" zoom that, at least, is pretty sharp wide open. Probably has terrible CA, poor contrast, severe coloration, awful distortion, seriously bad vignetting but it's sharp, by golly.

It reminds me of the quote, "aside from that Mrs. Lincoln, how was the play?"

[Glenn NK]

I assume your 17-55mm is a crop-sensor lens? Than 55mm becomes equivalent to 70mm (more or less) on full frame. Still variation on the 70mm theme. 12-24 on the long end, obviously, borders the short end of 24-70mm.

I'm not trying to create something that doesn't exist. If you believe there isn't enough evidence of 70mm serving as the dividing point for many systems, then fine. But I thought there is and wondered why. The reason may have been purely arbitrary by the frontier, mindlessly replicated by the followers, as Colin suggested.

AFAIK, the focal length of the lens doesn't change when it's on an APS-C body.

If the 12-24 borders on the short end of 24-70, then there must be two divides - jeez this is getting complex.

One of Canon's most popular lenses (a kit lens with the 5 series) is the 24-105 (which I have). Maybe 24 is the magic number.

[Scott Stephen]

I find that a 24-70 is frustratingly short on a FF body. First, it is just too short to frame a lot of the things I shoot outdoors, and second, it does not reach up in into the "classic portrait range" of 85-135 for posed shots.

I like the 24-105 range much better, but the quality does suffer in the existing models, though these are old designs and cheap "kit" models. Perhaps they could be improved?

Not to be an iconoclast or heretic, but I wonder if, absent an improved 24-105, someone could do something in-between, like 24-90, or 20-85, or something like that?

[Colin Southern]

I find that a 24-70 is frustratingly short on a FF body. First, it is just too short to frame a lot of the things I shoot outdoors, and second, it does not reach up in into the "classic portrait range" of 85-135 for posed shots.

It's just horses for courses - I LOVE my 24-70 on a FF camera - it's my go-to lens for 3/4 -> full length portraiture. Once I go above 70, it's the realm of the 70-200 (where 105 would often be too short).

[ajohnw]

The Canon 24-105mm is a wedding photographers lens in my view and has been designed for that sort of use. Not a lot asked of it really. Central resolution is decent though out the range Your lucky it's not the Nikon 24-85mm, spectacular in the centre but a bit oh dear elsewhere. The Nikon 28-70mm is much better but still shows the problems with including a wide angle in the range.

Curiously the best lens in this sort of range I am aware of is the Olympus 14-42 on straight 4/3 it's ok on micro too but falls a bit short. That's 28-84 in full frame. Resolution is ok all over and through the range. Always good plus on 4/3. The 12-60 F2.8 is the same too but came at a heavy price. Good example of the effect of smaller lenses due to smaller formats. Scaling lenses up generally never works out. The problems go up related to the area of the lenses. Not that compact lenses are that good.

John
-