Comparing the Tamron XRs
AF28-200mm vs. AF28-300mm
I love Tamron XR lenses. I have the longer version, and some have asked whether my test pictures would be the same for the shorter version. In the interest of finding out I did a series of test shots. In this page I will refer to the AF28-200mm XR model A03 as an "200XR", and the AF28-300mm XR model A06 as an "300XR".
This series of shots were all taken from a tripod, all with the same (manual) settings of 1/1.3s, f/11, ISO 200. All with anti-shock turned on, all triggered by cable release, all auto-focused, all with a bare lens with no filter, same lighting, and in general every reasonable effort was taken to make the shots as accurate as possible. The primary variable in these shots is the zoom. I tried to get the zoom accurate just using the numbers on the zoom ring of the lenses, but it was iffy, indeed. As a result a lot of the shots are plus or minus 10mm in focal length, but they should be close enough to tell the story. The shots were all taken in RAW mode with no sharpening. Post-processing amounted to each shot being colour-corrected for 4100K, which actually made them look a bit cold, but it saved me from having to type in the number each time (4100K is standard for fluorescent in the software I used). Each RAW file was saved in the colour-corrected version as an 8-bit TIFF file and opened in Photoshop Elements 2.0, cropped to 400x320 pixels, sharpened using an Unsharp Mask with amount 310%, radius 0.3, and threshold of 1,and saved as a medium-quality JPEG with no up-sizing or down-sizing. The pixels you see on the page below are the pixels recorded by the camera within the limits of the JPEG format to reproduce them. Every image was treated just the same. The cropping was all based on the top left corner starting at a crossing of two seams on the bear's forehead.
28mm
In all cases I have the 200XR image above the 300XR image. So far no real difference.
50mm
It's getting obvious that there is colour artifacting from the JPEG compression (which is pretty high) but ignoring that the sharpness is similar... the 200XR might have a slight edge on this one, but not by much. Below the 300XR shot is a shot taken with a Nikkor AF50mm f/1.8D prime lens as a sharpness benchmark.
This is the prime shot. Less contrasty than the XRs. No great improvement in sharpness, but that is largely a perceptual thing caused by the lower contrast.
70mm
Once again, awfully close. The 300XR is a touch sharper on this shot, but so far they're basically equal.
120mm
Zooming in closer, still plenty of colour artifacting in the highlighted 'hairs' on the stuffed bear. The 300XR is a bit sharper on this one, too, but I'm not despairing for the 200XR yet.
200mm
Here we have the 200XR at maximum zoom. The 300XR still has another 100mm to go, with the extra size and weight and slight decrease in focus speed to go along with it.
The camera reported this as 185mm zoom, yet it's a tighter zoom than the 200mm reported on the 200XR. Go figure. I have to say that this one looks sharper than the 200XR shot, as well.
Bottom line - the two lenses are very close. I'd give the nod to the 300XR, but I'm biased (I own one) so that might play a role, here. Certainly neither is head-and-shoulders above the other. They are close enough that I'd be confident to say that there is not enough of a difference to say categorically that one model is better than the other... you are bound to find minor variations in different lenses of the same make, so I'm going to call them basically equal, even though the particular 300XR tested beat the particular 200XR tested.