Sony FE 24mm f/2.8 G Review

Focus Speed and Performance

The Sony 24mm f/2.8 G autofocuses quickly and quietly, with impressive accuracy even in low light. I didn’t notice any significant focusing errors whether in the field or in a strict lab environment.

The close-focusing capabilities of the 24mm f/2.8 G are nothing to write home about, with a maximum magnification of 0.13x (or 1:7.7 magnification). Yes, you can focus on some nearby details, but nothing approaching macro territory. If you want a wide-angle lens for close-up photography as a Sony shooter, your best option is a third-party lens like the Venus Laowa 15mm f/4 or Sigma 17mm f/4 C.

Distortion

The Sony 24mm f/2.8 G has extremely high levels of distortion for a prime lens at -8.77% barrel distortion. This is nearly fisheye territory, as you can see from the photo of our test chart below:

By comparison, the Nikon Z 26mm f/2.8 has “just” -4.98% barrel distortion, and the Nikon Z 28mm f/2.8 has -2.39%, which I called “on the high side for a prime” in my Z 28mm f/2.8 review. However, Sony still doesn’t set our all-time record, which is held by Canon’s RF 16mm f/2.8 at a whopping -11.71% barrel distortion.

For years now, mirrorless lenses have been trending toward optical designs with high levels of distortion, yet better qualities elsewhere (like high sharpness or light weight). The theory is that distortion is easy to correct in post-production without compromising image quality. Although this is true, major distortion corrections can still contribute to softness in the image, especially in the corners, since the photo must be stretched significantly in order to look rectilinear. It should go without saying, but the -8.77% barrel distortion found on the Sony 24mm f/2.8 G is high enough that this possibility comes into play.

Vignetting

In uncorrected images, the Sony FE 24mm f/2.8 G has relatively high levels of vignetting wide open, especially if focusing on distant objects. Stopping down improves the situation a bit, but even at narrow apertures, there is still noticeable vignetting on this lens. Here’s what we measured in the lab:

The maximum of 2.39 stops of vignetting is on the high side. In fact, Sony’s own 24mm f/1.4 GM maxes out at 2.07 stops of vignetting despite the much wider maximum aperture of f/1.4!

This level of vignetting is part of the compromise that Sony made when designing such a small and light lens. It’s not out of line relative to other ultra-portable lenses, but it’s still more than I would have liked to see.

Note that the chart above measures vignetting in the raw image before distortion was corrected. Correcting distortion will crop out the darkest portions of the corners and lessen the vignetting slightly. However, it is not a particularly drastic effect on this particular lens:

Lateral Chromatic Aberration

There is a fairly low amount of lateral chromatic aberration on the Sony FE 24mm f/2.8 G. Here’s the chart:

Anything under about one pixel is almost impossible to notice in real-world images, even with chromatic aberration corrections turned off. By comparison, the Sony 24mm f/2.8 G maxes out at a 1.42 pixels of chromatic aberration, which is still on the low side. You probably won’t see lateral CA on this lens very often, and even if you do, automatic software corrections will fix it without any meaningful side effects.

Sharpness

The Sony FE 24mm f/2.8 G is sharp in the center, decent in the midframes (APS-C corners), and so-so in the corners. Here’s how it performs as measured in our lab:

Given the somewhat weak corners, I requested a second copy of the Sony FE 24mm f/2.8 G to test for sample variation and decentering. It performed extremely similarly to the first copy, with slightly weaker central and corner sharpness overall, but slightly better midframes at f/2.8 and f/4:

The differences between both copies were small enough that, outside of the lab environment, I could not reliably tell which copy was which. That’s actually a good sign for Sony’s sample variation, even though the pure sharpness numbers on the 24mm f/2.8 G aren’t best-in-class. (The next page of this review will put these results into context against other wide-angle prime lenses available today.)

Finally, there was minimal focus shift or field curvature on either of my copies of the Sony 24mm f/2.8 G.

Coma

Related to sharpness is coma, a lens aberration that can make dots of light in the corner of a photo look like smears. Coma isn’t usually visible in everyday photography, but for something like Milky Way photography, it can be a factor. The Sony 24mm f/2.8 would not be my first choice for Milky Way photography (it’s a little long at 24mm, and a little dim at f/2.8 + high vignetting) – but it’s still within the realm of possibility. So, I wanted to put its coma performance to the test.

The crop shown after the image below is an extreme crop from the top-right corner of the Sony a7R V with the 24mm f/2.8 G. I cropped the a7R V’s 61-megapixel sensor down to 1000 × 1500 pixels and didn’t do any resizing; the second image below is a direct excerpt with Lightroom’s default sharpening and noise reduction applied:

This is impressive coma performance, with just a bit of smearing in the far corner. I have no real complaints with it at all.

Flare and Sunstars

I was happy to find that the Sony 24mm f/2.8 has good flare performance, with just the occasional dot of flare appearing when the sun is in the frame, and high levels of contrast retained throughout the aperture range.

If the sun is totally bare, you do get a bit of red dot flare near the sun (more like rainbow flare with this lens, but the cause is the same); however, this is more mild than on most wide-angle mirrorless lenses.

As you can see from the samples below, both flare and sunstars look good on this lens. You get reasonably well-defined sunstars so long as you shoot at narrow apertures like f/16.

The next page of this review dives into the sharpness numbers a bit more, with some comparisons against other lenses that Sony users may be considering. So, click the menu below to go to “Sharpness Comparisons”: