Here’s a brief look at the Tamron 17-50mm F/2.8 XR Di II zoom lens. Scroll down for the main review.
|
Lens
|
Tamron SP AF 17-50mm F/2.8 XR Di II LD Aspherical [IF] Model A16
|
|
Box contents
|
Front and rear caps, users manual, and plastic petal type hood.
|
|
Cost
|
|
|
Build quality
|
Good
|
|
Additional information
|
|
Specifications below
|
|
|
Optical configuration
|
16 elements in 13 groups
|
|
Angle of view
|
78°-45° to 31°-11°
|
|
Aperture
|
7 blades, curved
|
|
Full frame and APS-C
|
APS-C only, with a 35mm equivalent of 26-75mm. Works on full frame camera, but vignettes hard at all focal lengths. Use APS-C size capt. for correct pictures on Sony FF cameras.
|
|
Depth of field and focus scales?
|
Focus scale and focal length index marks at 17mm, 24mm, 35mm, and 50mm.
|
|
Minimum focus, image plane to subject
|
10.6″ (269mm)
|
|
Minimum focus, end of lens barrel to subject
|
4.1″ (104mm)
|
|
Hard stop at infinity focus?
|
No
|
|
Length changes when focusing?
|
No |
|
Focus ring turns in AF?
|
Yes
|
|
Filter size
|
67mm
|
|
Filter ring rotates?
|
No
|
|
Distance encoder?
|
Tamron claims “D” compatible on the box.
|
|
Max magnification
|
0.22x, or 1:4.5
|
|
Min. F/stop
|
F/32
|
|
Sony teleconverter compatible?
|
No
|
|
Length changes when zooming?
|
Yes
|
|
Dimensions WxL (my measurements)
|
2.9″ x 3.4″ 73mm x 85mm. Add 2mm in width for zoom lock switch.
|
|
Maximum extended length (my measurements)
|
4.5″ (114mm)
|
|
Weight bare (my scale)
|
15.4oz (438g) 16.05oz (467g) with caps
|
Requisite product shots.
 |
| Box and contents |
 |
| Side shot fully drawn in, with zoom lock engaged |
 |
| Side shot fully extended |
 |
| Backside mount. |
 |
| Front element |
The
Sony A700 was used for this review. For a better understanding of terms and methods used in this review,
go here.
The Tamron 17-50mm F/2.8 is a fast, medium length zoom lens that covers about the same area as the standard issue “kit” lenses from various manufacturers. It has a large constant aperture of F/2.8 and is built reasonably well. It’s also similar in design and performance to the
Sigma 18-50mm F/2.8 lens reviewed here. This lens, like the Sigma, has been designed for exclusive use on APS-C cameras. The 35mm equivalent focal length is a pretty handy 26mm-75mm. Check out the excellent
Sony DT 16-50mm F/2.8 SSM also reviewed here.
The made in Japan Tamron 17-50mm F/2.8 appears to be made mostly of plastic around the barrel, but metal has been used for the mount. There’s a nice wide rubber grip area around the zoom ring, reminiscent of Konica Minolta. Fit and finish are good, with smooth actions for zooming and focus. I didn’t notice any zoom creep during my limited use, but should you encounter that problem, there’s a zoom lock button to keep the zoom locked at the 17mm length only, it can’t be set at any other length. You can tell you have the zoom lock set as it shows red on the slide to let you know, see product shot above. Zooming adds just over 1″ (29mm) of length to the lens. Additionally, the EXIF data is correct at all focal length index marks.
In the box comes the lens, a petal-type plastic hood, an owner’s manual in several languages, and front and rear caps.
This lens has an internal focus design, so there is no additional extension of the lens. The focusing ring has a rubber insert for easy gripping, and the ring rotates when using AF, so watch your fingers. Auto-focusing speed is average to slow, but appears mostly accurate in my limited use. I noticed it wants to roll all the way over past the infinity index mark when focusing to infinity near the wide end, fortunately, that seems to be the sharpest position. I think the index mark might be off or possibly the calibration, but again, it’s nice and sharp at that setting. At the long end, AF goes to the infinity index mark when focused at infinity, and is good and sharp there. When focusing manually, you only need about 1/6 of a turn, which is pretty fast, and doesn’t allow any attempt at fine tuning, especially for people with jittery hands. Also note this lens has a reversed focus direction as compared to most of Sony’s lenses, meaning the focus ring turns clockwise to go to infinity, as you are holding the camera properly.
Internally, Tamron boasts of “XR” or extra refractive index glass; at least one aspherical element, and “LD” glass, no specifics are given. The lens is multi-coated, and shows green/magenta, with emphasis on green.
Lens flare/ghosting. Average to above average control for a wide-angle zoom. I see mostly green colored ghosts if the sun is at the edge, or inside the image. The ghosts change shape and color with focal length and aperture. Look below for examples. Veiling glare control seems average to above average as you zoom out, though as always, use your hand to block the sun or other super-bright light source with your hand to keep the contrast up in your shots, if the sun isn’t actually in the shot.
Color fringing (CA). Above average control. Look for mild amounts of lateral color fringing at and near 17mm, diminishing to almost nothing at 50mm. Moderate amounts of lateral color fringing can be seen at close focus, but that’s fairly normal. I see no axial color fringing at any aperture.
Bokeh. Harsh at all focal lengths, similar to the Sigma 18-50mm F/2.8 and the kit zoom lenses. Look below for sample crops.
Color. Neutral to slightly warm.
Close up filter. N/A
Coma. Mild at F/2.8 through most of the focal lengths, almost gone by F/4.
Regular filters cause no additional light fall-off problems.
Filter size is 67mm. This is an oddball size, the only Sony lens to use it is the new 28-75mm F/2.8.
Distortion. You’ll notice moderate barrel distortion at the wide end, becoming flat around 20mm, then very light pincushion distortion starts at 24mm, turning moderate by 50mm. Check out the cropped samples below.
Distortion examples directly below.
 |
| 17mm, moderate barrel distortion. |
 |
| Sort of flat at 20mm. |
 |
| 50mm, moderate pincushion distortion. |
Distortion is complex wave-type at the wide end, where I see moderate bulging distortion. At 20mm, you’ll see a nearly straight line, although you can never get it completely straight. Light pincushion distortion starts at 24mm, then gets stronger as you zoom out to 50mm. The results here are similar to the Sigma 18-50mm F/2.8.
Bokeh crops next.
|
17mm F/2.8
|
17mm F/4
|
|
|
|
|
50mm F/2.8
|
50mm F/4
|
|
|
|
|
17mm F/5.6
|
50mm F/5.6
|
|
|
|
Bokeh is harsh at all focal lengths using wide apertures, but looks ok at F/5.6. Again, similar performance to the Sigma 18-50mm F/2.8. The bottom two crops were taken at a different time and location from the other crops.
Lens flare/ghosting examples
|
17mm F/5.6, ghosting.
|
17mm F/5.6 sun centered.
|
|
|
|
|
17mm F/5.6, sun just out of picture.
|
50mm F/5.6 sun in shot
|
|
|
|
|
50mm F/5.6 sun just out of shot.
|
50mm F/5.6, hand used to block sun.
|
|
|
|
I see mostly green blobs when the sun is near, or inside the image at 17mm. When the sun is in the center of the shot, there is a red ring just visible in the corners, and it becomes a little more noticeable as you stop down. There’s also a red arc visible opposite the sun when the sun is just out of the image, see the first image in the second row. This lens produces an average amount of ghosting and veiling glare at the widest lengths.
Towards the long end I see average, to slightly above average control of glare, with some red and green blobs thrown in. When the sun is in the center, look for a green halo around the sun. I show in the last row what you can do to prevent veiling glare when the sun is just out of the image, use your hand to block light from hitting the front element, the hood really doesn’t do much. I’m showing the worst of glare and ghosting here, it may not show nearly this much depending on what your background is composed of, and the exposure values used.
Light fall-off.
See the crops below. Light fall-off or corner shading is light to moderate at the extreme ends, one stop down and most of it is gone.
|
17mm F/2.8
|
17mm F/4
|
|
|
|
|
50mm F/2.8
|
50mm F/4
|
|
|
|
Light fall-off shot below.
The full image directly above shows real world light fall-off at F/2.8, 17mm with a shutter speed of 1/4000, -0.30ev. The slightly dark corners don’t bother me, but you can stop down to a more reasonable aperture and do away with this issue in bright conditions as above.
Center and corner sharpness.
Below are crops from the
image centers and corners at 17mm.
|
F/2.8 center
|
F/2.8 corner
|
|
|
|
|
F/4 center
|
F/4 corner
|
|
|
|
|
F/5.6 center
|
F/5.6 corner
|
|
|
|
|
F/8 center
|
F/8 corner
|
|
|
|
|
F/11 center
|
F/11 corner
|
|
|
|
At 17mm, F/2.8 shows a little soft from veiling haze, but still has good detail. One stop down to F/4 gets rid of the veiling haze, and the centers look sharp, peaking at F/4-5.6, with diffraction softening starting to show around F/8. The corners are fairly sharp also, looking almost as sharp as the centers at F/4-5.6, and don’t seem to sharpen up any more by stopping down further, which is good performance, especially at the wide end. All crops above and below were taken at infinity focus.
The exposure differences in the corner crops are from light fall-off. These right side corner crops are about 400 pixels in from the very corner. The corner crops were taken from the same spot as the centers, they look different because of the way the corners are stretched towards the centers with a wide angle lens, this is normal, and you wouldn’t notice this unless viewed side-by-side as I’ve done here.
Below, crops from the 28mm centers and corners.
|
F/2.8 center
|
F/2.8 corner
|
|
|
|
|
F/4 center
|
F/4 corner
|
|
|
|
|
F/5.6 center
|
F/5.6 corner
|
|
|
|
|
F/8 center
|
F/8 corner
|
|
|
|
|
F/11 center
|
F/11 corner
|
|
|
|
At mid-zoom, the center performance is similar to the 17mm crops, but the corners don’t seem to sharpen up to their very best until F/11, though F/4-8 are not too much different.
Below are centers and corners from 50mm.
|
F/2.8 center
|
F/2.8 corner
|
|
|
|
|
F/4 center
|
F/4 corner
|
|
|
|
|
F/5.6 center
|
F/5.6 corner
|
|
|
|
|
F/8 center
|
F/8 corner
|
|
|
|
|
F/11 center
|
F/11 corner
|
|
|
|
The 50mm center crops are slightly soft at F/2.8, but sharpen up nicely at F/4. It looks as though the sharpest centers (and corners) come at F/8, you notice this by looking at the tan colored blocks along the roof, and the way the mortar joints show up, it’s hard to see unless you inspect closely. There’s very little difference in sharpness between the corners and centers at F/8. Notice the near absence of lateral color fringing, which is a pleasant surprise since these crops were taken just 200 pixels from the very corner.
Let’s check out the macro capabilities of this lens.
Below, check out the 100% cropped portion (284kb) of the full image. The sample shot was taken with the
Sony A 700 12.2MP camera. The subject is a standard US stamp, 0.87″x 1.0″ or 22mm x 25mm. Also, note the macro shot was taken as close to the subject as focusing allowed; in this case a very short 4.1″ (104mm), measured from the front of the lens barrel to the subject.
This lens has a reproduction size of 0.22x which is a little small compared to other lenses like this, but it produced a nice sharp shot, the absolute best coming at F/8, but F/5.6 looked good also. Shooting close focus subjects at F/2.8-4 shows very soft, and there is a big leap in sharpness from F/4 to F/5.6. While taking the macro shots, I noticed this lens produces barrel distortion at 50mm, focused close, instead of pincushion distortion when focused at longer distances and infinity, see the distortion samples to see what I’m talking about. As a side note; the “1996” on the bottom left of the stamp measures a mere 1mm wide.
 |
| As close as you can get. F/8. Click for larger image. |
Coma shows up in the corners at F/2.8 at all focal lengths, but less so as you zoom to 50mm. It only takes one stop down to get rid of this. These 100% crops show the lower left corner, and the whole image would print out about 45″ (1.14m) wide as you see them on your computer screen, so there is no real problem.
This crop was taken from the last 700 pixels on the middle/left side. Magenta and green are conspicuous beside the upper tree branch, other than that, things look pretty good. Shot is at 17mm, F/5.6.
Conclusion.
The Tamron 17-50mm F/2.8 XR Di II turned in a very good all-around performance. Things I like are; sharp images across the entire frame when the aperture is closed a couple of stops, at all focal lengths; very low color fringing, especially towards the long end; light fall-off is almost eliminated one stop down, and coma is almost non-existant. There are a few characteristics that are typical for this type of lens, and include poor bokeh, moderate distortion at both ends of the zoom; and manual focusing which is geared too quick for accurate results.
The sharpness at F/2.8 is perfectly acceptable in normal picture taking, especially in low-light conditions, but sharpness jumps up noticeably at F/4. Auto-focusing is mostly accurate, though you may want to check your screen when using F/2.8, to get the best possible sharpness. I also notice the AF is a little slow, and seems to want to focus in “steps” before locking.
The Tamron 17-50mm F/2.8 XR Di II is a very good buy at the current street price, and performs similarly to theSigma 18-50mm F/2.8 EX DC macro, also reviewed here. I don’t have the Sigma available for comparison, but if memory serves me correct, I think the Sigma was a little sharper in the centers at F/2.8, but the corners on the Tamron are sharper from F/4 and beyond at all focal lengths. I might add I had to send back my Sigma to have the rear group re-aligned, it was unusable as I received it brand new(!) I’ve had several Sigma lenses, and only had one I didn’t have to send back. Tamron may have better quality control.
Don’t forget to browse the Sony DT 16-50mm F/2.8 SSM review, it’s a great lens for more money.
Check out the quick comparison between the Tamron, and the Sony CZ 16-80mm here, at bottom of page.
Another comp between the Sony CZ 16-80mm and Sony 16-50mm at the bottom of the page.
