I've been using the Tamron 28-75mm f/2.8 on my Pentax K-x with fantastic results, but I sometimes miss the wide angle. I planned on replacing my Tamron 28-75mm f/2.8 with its little brother, the Tamron 17-50mm f/2.8. Assuming I was happy with the 17-50mm, I would then sell the 28-75mm and buy a Sigma 50-150mm f/2.8, so I would then have 17mm to 150mm at f/2.8.
I just received the 17-50mm, so phase I is complete. After doing back-to-back comparisons of the 17-50mm vs the 28-75mm though, I'm starting to have second thoughts.
I haven't used the 17-50mm very much yet, but it seems to take nice sharp pictures with great colors and contrast, but it's not as bright as my 28-75mm at equivalent f-stops.
Comparing both lenses at f/2.8, they are very similar, but the 28-75mm is a little brighter. The difference becomes more noticeable as the aperture decreases, becoming easily visible by f/5.6 and strikingly different by f/10.
What this all means, is that I have to use a slower shutter speed on the 17-50mm than I do on the 28-75mm, which will be a limitation in some situations.
I first noticed this behavior when shooting a portrait of a friend in my home studio. I started out using the 28-75mm, and near the end of the shoot I put the 17-50mm lens on the camera to see how it compared under controlled lighting conditions. I had just received the lens in the mail a few hours earlier and had not yet used it. Both lenses were set at f/10 and 1/125 sec throughout the shoot, and the strobe settings remained constant.
I was surprised to find that the shots taken with the 17-50mm were significantly underexposed, so that my subject was very dark and my white backdrop was no longer blown out (as it should be for high key photography).
Here are a couple test shots from the shoot (BTW, it's just a coincidence that he is smiling in the first picture and looks less enthusiastic in the second):
I then took some test pictures at various apertures using both lenses. The subject was a toy car that was just lit by the light coming from my LCD monitor and dim overhead lighting. (note that I horizontally flipped the second image so that the images would mirror each other, since lighting is not even across the image).
At f/2.8 you can see that brightness seems to be similar, but on my monitor I can see that the 28-75mm picture is just a little brighter. As the aperture increases, however, the difference becomes much more pronounced.
And by f/11 there is a huge difference in brightness.
I suppose that the most important thing is that the 17-50mm still performs well at f/2.8, producing an image almost as bright as the 28-75mm. But I find it disturbing that at larger apertures the lens is clearly not capturing as much light as the 28-75mm.
Since the 17-50mm is supposed to be based on the 28-75mm, with a similar optical formula, I am surprised to see such a difference. Could this be because the 28-75mm is a full frame lens (35mm), while the 17-50mm is designed to only cover a smaller, APS-C sensor?
The way I understand it, f-stops are not fixed dimensions, but are based on some type of ratio involving the length of the lens and the width of the pupil (ie aperture) of the lens. So with the 28-75mm being a large lens (length and width), is f/10 larger on the 28-75mm than f/10 is on the 17-50mm, therefore letting in more light on the 28-75mm?
Is there any way, other than f-stop ratings, to rate the actual brightness of a lens?
Anyway, what it all boils down to is that the 17-50mm Tamron is not as fast/bright as its 28-75mm big brother. So now I'm rethinking my plan, and I may decide to return the 17-50mm, keep the 28-75mm, and still buy the Sigma 50-150mm. When I need a wide angle, I could always throw my 18-55mm kit lens back on.
This decision is made harder by the fact that I'm really happy with the results I've gotten from the 17-50mm. Image quality seems to be very close to the 28-75mm. I took some shots of my girls this morning at f/2.8 with both lenses using just natural lighting from the windows, and it looks like the 17-50mm may have a very slight edge in contrast and color, while the 28-75mm seems to be a little sharper. But the differences were very slight, and two happy little girls bouncing around on a bed is not exactly the most controlled of circumstances.
Here are some of the shots from this morning with both lenses. These pictures are straight out of the camera, with absolutely no color or exposure adjustment, and no sharpening. The only thing I did is crop the two vertical images from 3:2 aspect to 4:3 aspect (the native 4:3 aspect is something I seriously miss from my old Olympus Four-Thirds DSLR).
Click on the image for the full-size original, but be prepared to wait if you don't have a fast broadband connection.
Last edited by Edgar_in_Indy; 11-26-2010 at 11:10 AM.