Thread Rating:
  • 0 Vote(s) - 0 Average
  • 1
  • 2
  • 3
  • 4
  • 5
next PZ Lens Test Report: Sony E 18-55mm f/3.5-5.6 OSS
#11
[quote name='Klaus' timestamp='1282832181' post='2222']

A formal bokeh analysis is pointless for slow speed lenses. We're talking about a "27-83mm f/5-8" here. There're a few bokeh shots in the sample image section though. As far as flare is concerned - show us a really reliable method with field relevance.



As far as your correction argument is concerned - frankly, I'd guess that 80% of the users do not care. They want to have everything fixed right out-of-the-box. Panasonic is just doing that which is probably one factor for their recent success.

[/quote]

My correction remark was a response to that other post, and it should be read in that context. If one actually cares about it bit more resolution, one should also care about the lens' distortion characteristic, as correcting the distortion will lose resolution.



The 80% of the users of cameras like the Panasonics and Sony's will indeed not care about it, but also will not care about the bit more resolution in tests.
#12
Doesn't change othe fact that the lens has better build and silent AF during video. None of the other kit lenses do that. I think there is only one Pana lens optimized for video that costs three times more.
#13
[quote name='BG_Home' timestamp='1283036016' post='2287']

I wonder whether that 3.5 star optics rating is justified. Looks like fairly low border resolution throughout to me... and on a further note, this seems to be a trend for that system.

[/quote]



Have found the Photozone review, rating and subjective comments about the Nex 18-55 lens to be accurate and helpful. The Nex kit zoom is indeed capable of superb results, as Klaus hinted, even if the numeric performance charts don't look exciting compared to other theoretically more perfect lenses. See this 100% size crop.

[ATTACHMENT NOT FOUND]

True, your own results may vary if you are not as careful as I was to use the (in some ways optimum) F8 aperture, a high ISO 800 setting to give me a fast 160th/second shutter speed, snap the picture when the model was not moving, a convenient carbon fiber tripod with ball head that cost more than the camera, and a $6 dollar eBay infrared remote release (However am not having much trouble meeting those conditions for most of my paid work). And maybe the lens would not be so great for that probably small percentage of your photos where the final print size is way bigger than 11x14 inches/30x40cm, or your clients have monitors with more than 3.5 million pixels, or the really important parts of the picture involve small details way out at the corners...



The Nex-5 with kit zoom has produced a higher percentage of icy-sharp photos for some non-obvious reasons. Can use the tripod way more often with the Nex in the field because the system is so light. The lightness doesn't simply allow me to carry the tripod more often in the field without problems or getting tired. It also lets me set my tripod ball-head to a light, easily movable friction setting without the camera "drooping" or moving when I take my hands away (thus no need to lock the tripod head before each photo). And the Sony sensor noise is low enough that ISO 800 has become my standard setting for the last month, even for wedding work. Sony should advertise its fairly sweet ISO 800 capability more.



Overall the high-ISO, light-and-tripod-ball-head-friendly Nex kit lets me use tripods and higher shutter speeds/smaller apertures more often, to give me great depth of field and make up for some of the lens wide-open weakness. The Nex lens exerts maybe 8 ounce-inches of torque on the ball head under the level, tripod-mounted camera. Am ending up with a higher percentage of sharp photos in the field than would be easily achievable with a huge, optically ferocious 85mm F1.4 lens (maybe 30 ounce-inches of tripod head torque) and/or my old medium format gear.



Thanks Klaus for giving us great hints about what's out there in camera land. And thanks also to Sony for putting clear and honest MTF/resolution information about their lenses on the web:

http://www.sony.jp/ichigan-e/products/SE...html#L1_30

It's a little harder than reading a Photozone review, but the usually-adequate optical resolution of the Nex 18-55mm lens and its relative corner performance is deducible from the Sony charts (but you don't get the nice Photozone info about bokeh, various chromatic aberrations, curvature of field, flare/fringing tendencies, focus speed, etc etc). Shame on all the other lens vendors that don't post MTF(sharpness) charts the way Sony and Sigma does now.



This is not meant as a harsh criticism of your comment, but rather am pointing out the non-obvious ways that a lens and camera system might contribute to a higher final percentage of good pictures than the narrowly-focused optical charts can suggest.
#14
[quote name='thw' timestamp='1282831335' post='2220']

Klauss' optical quality scores are based entirely on various aspects of lens designs.



In real use, distortion, chromatic aberration and vignetting are most EASILY corrected in post-processing, so the only criteria worth considering are resolution, flare and bokeh. Since the latter two aspects are not presented, the only aspect left is resolution. If you miss out the details, you can never recover them.



Now, look at 18, 24-28 mm and 55 mm. Examine performance at border/extreme relative to the center. Take off your fanboyz glasses and stare at the optical results.

[/quote]



Note that the Sony has a center resolution that might exceed that of the sensor, even wide open, at wider lenghts, whereas the Canon only barely makes it into excellent in the center. But the Sony is a step behind in the borders. Different, but can't say that one is simply superior to the other.



The CA is pretty hefty on the Sony, so much so that I wonder how much it can be corrected. It would be interesting to see an analysis of this, somehow.



Note: I am not a Sony fanboy. Indeed, I avoid them like the plague for always using non-standard interfaces.



-Lars
#15
Well, to be a bit more thorough, Zeiss and Leica also publish MTF/resolution charts for their lenses, not just Sony and Sigma. And Leica shows vignetting (corner darkening) and distortion info as well. What other manufacturers publish MTF charts for their lenses? Guess I am not an expert on which manufacturers publish charts. Anybody know which charts are theoretical, and which are real test results? Am aware that the Zeiss MTF charts are not just theory.
#16
i too was somewhat disappointed with the performance of the 18-55 kitlens on the NEX 7 as far as corner and side sharpness and distortion are concerned. however, i found out that when shooting RAW images and processing them in DXO with the appropriate modules for both selective sharpening and barrel/pincushion corrections results improve quite a bit.



of course the distortion can also be removed with the appropriate module in Lightroom, but improving sharpness on a selective basis is not included. in DXO there is a possibility to correct what is called "lens softness", which in the accompanying help file is described as:



[indent]DxO Optics Modules have been created by measuring, for each camera body and each lens, the amount of blur at every point in the image area. Matching the shooting parameters contained in the EXIF data (aperture, focal length, etc.) and the map provided by the DxO Optics Module, DxO Optics Pro can apply precisely-tailored correction according to the position of each pixel in the image field. This correction is not the same over the whole image, since lenses are sharper in the centre, therefore central pixels need less correction. [/indent]



Although the above may look like a ad, the results are indeed quite good and far better then what can be obtained by using standard "sharpening" algorithms. Of course you cannot add sharpness where it isn't in the original file, but by improving local microcontrast in a subtle way things do look like quite a bit better.



Unfortunately this option is only available for native NEX lenses and when using non-NEX lenses with a adapter this option cannot be used - because of the lack of data.



If interested and you wanted to try it yourself, DXO can be downloaded for a limited free trial period.
  


Forum Jump:


Users browsing this thread:
1 Guest(s)