 |
|
Photodo stats
1102 lenses 428 MTF tests 74 in-depth photodo reviews 100+ users join each day Help the lens community by reviewing or rating a lens today via our lens search | Sigma 18-50 f/2.8 EX DCPage 1 of 1: 1
Specification
Build and handling Sigma’s well proven AF/Manual focus switch is located in the usual place, left side of the barrel just in front of the mount. Above, and in front of this, is a zoom lock that locks the lens at 18mm, but I found this completely redundant as the mechanism was so well torqued that it did not creep, but was still easy to adjust. The zoom ring is marked at 18, 24, 28, 35 and 50mm and occupies 2/3rds of the barrel, ample wide enough for even the biggest of hands. It's mainly covered in a large pattern ribbed rubber surface. The remaining third of the barrel is taken up by the focus ring, which has a smaller pattern ribbing and a basic distance scale in metres and feet. The front element, which extends some 21mm on zooming to the 50mm focal length, does not rotate during use, helpful when using filters. Focusing is fast. In fact it is almost instant, one of the advantages of the large aperture of this lens. Despite using it in some very low light situations the lens never once hunted, picking up the focus first time, every time. The supplied petal hood is a bayonet fit that is reversible for storage, and does its job well with no signs of vignetting.
Below is our lens test data. To find out how to use these graphs look at this article: How we test lenses
Verdict In summary the main positive points of the Sigma 18-50mm f/2.8 EX DC are: Negative points are: Check the latest price of the Sigma 18-50mm here Test by Ian Andrews www.wildaboutkent.co.uk This review is a perfect example why you shouldn't post reviews with no regard to what camera was used.
Why? 1) use of antialiasing filters 2) pixel size/sensor resolution 1) I believe the camera used was either sigma SD10 or sigma SD9. With no antialiasing filter used these cameras indeed produce very sharp details, but with a lot of aliasing artefacts. If my understanding is correct a perfect MTF value with no aliasing would NEVER be greater than 0.5 cyl/pxl. Values of up to 0.7 clearly reveal excessive aliasing. 2) High MTF values might lead people to believe the lens will work equaly well on different cameras. That is misleading. Value of MTF in cyl/pxl is directly linked to the pixel size. A Sigma SD10 pixel is almost 2 times larger than a Nikon D300 pixel! Does this sigma still get 0.7 cyl/pix on Nikon D300? Well, I use one on a D200/D300 and it is one of the sharpest lenses I have.
As stated, the test was done on an SD10. Ian I Use this lens on my Pentax K10D and I am verry happy with the lens, very sharp pictures
I'm sure the lens is perfectly fine.
The point is, this simply can't be the "Sharpest lens tested to date" judging by its measured cyl/pix. This methodology is flawed and misleading. 1) use of antialiasing filters Take a close look at some of the cables hanging from the masts in "18-50mm at 18mm f/8" image. Sometimes it looks as if the cables have gaps along the length. This is where the high cyl/pix values come from - aliasing and less than 100% pixel coverage. This lens/camera combination produces details, that are not there. This is the reason why most cameras use anti aliasing filter. 2) pixel size/sensor resolution If you take the highest measured value of 0.7 and take into account pixel size of d300, you get cyl/pix value of 0.35 That's not so impressive any more, is it? The use of cyl/pix as if it's some universaly applicable property of the lens is flawed and misleading. The lens is perfectly fine. Not the best F2.8 but clearly you need to pay twice as much to beat it.
Just got the 18-50mm F 2.8 Sigma with the 4/3 mount-terrific
Just a question I have:is there a diffference in DOF/Sharpness withthe 4/3 system vs. the 3/2sensor since the 4/3 uses more of the center of the lens-when using the same lens in for both sensor systems??? or am I worng?? Dr. LR larryro2,
I assume by 3/2 sensor you mean 36*24mm full frame sensor. DOF of 50mm f2.8 is exactly the same on 4/3 as it is on full frame, but the framing is tighter. The end result is 50mm f2.8 on 3/4 looks like 100mm f5.6 on full frame. For the same megapixel count 4/3 sensors utilize much smaller pixels than full frame, so it's much more demanding on the optics. Generaly, larger sensors will always be sharper than smaller sensors - medium format will always be sharper than "full frame", large format will always be sharper than medium format. About 4/3 using using more of the center. This is correct only if you use full frame lens. Sigma 18-50 is a DC lens. It works on smaller sensors, but the image circle is too small AND too soft at the edge for full frame. Off center it behaves similarly as 28-80 full frame lens would, just on a smaller scale. This lens is indeed very nice. I use it on a K100d with an APS-C 6MP sensor and am quite happily getting more and more used to it :)
I can vouch from personal use for this beast being sharp even wide open, more so than a lot of Canon-L shots I see posted on the web. Personally I scratch my head about the Canikon lens hype. They have great gear, but is pricey, and a sharp lens one can buy is better than an exotic lens one cannot buy. Kinda like dreaming of a Ferrari, when the souped up Mini (OLD style, not the luxury BMW version) or Trabant is affordable and still allows a working stiff some weekend racing :) About the critique regarding sharpness: Ok, let's say this MTF chart was made on a Sigma SD 9 or 10. Here is the Foveon sensor's spec: 13.8mm x 20.7mm 3.4 Million sensor sites (tripling to 10MP if counting the three colors on each cell site, as other manufacturers do every pixel, regardless of them only recording one color each on a Bayer array) So the sensor is smaller than an APS-C sensor, with a cropping factor of 1.7, about halfway between APS-C (15.7mm x 23.5mm) and 4/3 (17.3 x 13.0mm) Ok, lets say you want to compare this lens mounted on a 10MP APS-C to one mounted on the Foveon. IF you keep the photo sites the same sites, the Foveon with 3.4MP needs to be upscaled to 10MP, covering a surface that is about 3 times larger, right? Ok, but that means we scale in two dimensions, so in EACH dimension we are scaling up with a factor of the square root of 3, which is about 1.7. What does that mean for MTF now? Simple, you divide the MTF assumed on a Sigma Foveon camera by 1.7, so the 0.7 result divided by 1.7 becomes a wee bit more than 0.4. That is still quite respectable. BUT WAIT! The APS-C sensor is a little bigger, isn't it? 23.5 / 20.7 = 1.13 Multiply 0.4 by 1.13, you get a final MTF of roughly 0.45. Not bad at all for a sub 300 dollar lens :) And on a 6MP APS-C camera, the news is even better at an MTF of almost 0.6 - which is positively impressive (and I was wondering why my pictures turned out so nice ...) HI
I TOOK SOME PICS WITH THE OLYM 14-45 KIT LENS AND COMAPRED THEM TO THE SIGMA 18-50- BETTER CONTRAST,COLOR AND sHARPNESS WITH THE SIGMA- QUESTION- WILL THERE BE AN E-520?????? AT FIRST I THOUGHT I WAS MISSING SOMETHING BY HAVING A SMALLER SENSOR BUT I FELL IN LOVE WITH THE E-500-EXCEPT FOR THE VIEWER-A BIT SMALL-AND INFO TO THE SIDE-ILL LIVE!! LARRY Hi!!
How will this lens compare to the Tamron 17-50mm f2.8? I am using the Canon Rebel XT camera (350d) as far as construction, sharpness and focusing is concerned. Page 1 of 1: 1 Add your messageLogin required
Please login here or if you've not registered, you can register here. Registering is safe, quick and free. |
This is Sigma’s EX version of what is fast becoming a standard lens for digital SLRs with a cropped sensor. The lens carries the DC classification, signifying that it will only work on cameras with the smaller sensors. Sigma already produces a budget f/3.5-5.6 version of this lens’ focal length, so why have they tooled this lens, and is it worth the extra? We put it through our test program to find out.
Quick, accurate autofocus
CA wide open