How Does The Canon 10-18mm Compare To The Sigma 10-20mm?

February 14th, 2016

Why10-18James recently asked the following question:

I’m going into real estate photography I have a Canon 1200D. I’m just wondering if there is a difference between the Canon 10-18 lens and the Sigma 10-20. I see the Canon lens is slightly cheaper will I be OK with the Canon lens?

For years, Sigma 10-20mm f/4-5.6 lens has been one of the go-to lenses for real estate photography but in the summer of 2014, Canon introduced the Canon 10-18mm f/4-5.6 wide-angle lens for $299. At first, it wasn’t clear how the Canon 10-18mm stacked up against the Sigma.

Now, it’s pretty clear that the Canon 10-18 f/4-5.6 and Sigma 10-20 f/4-5.6 are very close in quality. Check the DxOmark comparison here. The Sigma tests slightly better than the Canon. For a long time, Sigma kept all versions of the 10-20mm f/4-5.6 priced around $429 as it had been for years but now the Sigma has dropped their price to $340 for only the Canon mount lens but left the Nikon, Sony, Pentax versions (same lens with different mounts) of the lens at $479. This clearly indicates the competition Sigma feels from the Canon 10-18mm.

The Canon 10-18 has some other distinct advantages:

  • Image Stabilization
  • Around half the weight of the Sigma
  • Shorter and smaller in diameter

For Nikon, Sony and Pentax cropped sensor bodies the Sigma 10-20mm is still probably one of the best choices of a wide-angle lens for real estate shooting.

Share this

4 Responses to “How Does The Canon 10-18mm Compare To The Sigma 10-20mm?”

  • Photozone’s ( review indicates that, while the Sigma 10-20 F:3.5 performs very well in many respects, it does not do so well with sharpness in the extreme corners, especially at the wide end, whereas the Canon does significantly better in this respect. Granted, the only Photozone test with the Sigma was done on a Nikon aspc body, but I think the comparison is still fairly relevant, when considering performance variations across the field of the sensor.

    It is possible that the performance difference is attributable to variations from one copy of the lens to another. From what I have seen, these kinds of tests do not involve multiple copies of a particular lens unless the first copy or copies tested exhibit performance issues that suggest that the copy of the lens may be outside of the manufacturer’s specifications.

    Assuming optical performance is the main criteria, based on the Photozone and DxO reviews, it seems to me that trying both models out for yourself is the only real way to decide which one will work best for you, bearing in mind that you might need to try several copies of each lens, since manufacturing quality control is, unfortunately, not all that we might wish it to be.

  • I owned the Sigma 10-20, f4-5.6 for several years and was very happy with it. The build quality is like a tank, very solid, and heavy. It was very sharp in the middle, going soft to the edges. It also seemed to have some irregular distortion patterns towards the edges, that you couldn’t fix with PT Lens. (LR 3.0 didn’t have it in it’s database at the time for lens correction.)
    I recently sold mine and got the Canon 10-18 when it was on sale for $250. I am very happy with my purchase as the optics seem sharper with less distortion, and as Larry said, IS, less weight and smaller.
    In my opinion, you get twice as much for half as much.

  • I just realized that the DxO comparison to which Larry linked involves a different Sigma lens than the one to which he refers in his blog post. My comment relates to the Sigma 10-20 f:3.5, whereas Larry refers to the 10-20mm f:4-5.6. Neither DxO nor Photozone have reviewed the Canon mount version of either model; however, I think their evaluations of the Nikon mount version have some relevance and indicate that both are good lenses that should serve well for most real estate photography. Between these and the Canon 10-18mm, for me, I think it would be a close choice between the Canon and the Sigma 10-20 F:4-5.6.

  • I know they are not tested on Canon bodies, but it could be useful.

Comments RSS

Leave a Reply