Sony 24-105 vs prime. Overhyped zoom? Bad copy or normal?

Started 3 months ago | Discussions thread
Flat view
borgein Contributing Member • Posts: 680
Sony 24-105 vs prime. Overhyped zoom? Bad copy or normal?

Hi guys

In the last 7-8 years I've mostly only used primes for my photography, from various vendors: Leica, Fujifilm, Canon, Sony and Zeiss.

I decided that I wanted to have some more flexibility with my A7rIII, and purchased the FE 24-105/4 that scores so well in every review everywhere. The hype is real with this lens.

But I can't help but feel disappointed with the zooms performance after running it head-to-head on an A7rIII versus the Leica M10 and a Carl Zeiss Tele-Tessar T* 4/85 ZM lens. The Tele-Tessar is an older ZM prime that I purchased used for about $500 in mint condition, and it weighs in at about 260 grams. The lens is coded as a Leica Macro-Elmar-M 90/4, and so Lightroom shows it as a 90mm, but it is in fact a 85mm lens.

I've made several landscape shots on each lens on tripods with matching settings, and I noticed a pattern when I started comparing my shots. The M10 + Zeiss combo was simply breathtakingly sharp edge-to-edge in all my landscape photos. The FE 24-105 was always reasonably sharp at the point of focus, but, at the mid zones and outer zones of the image things get blurry. Also, details tend to look stretched and proportionally incorrect compared to images made with primes.

So I decided to do a quick test at home. I made two images at the same focal length (86mm on the FE), from the same distance, and within 10-15 seconds of each other. Both cameras and lenses were shot at f/5.6. The Zeiss ZM lens was set to infinity focus, while the Sony FE was focused on the illuminated building.

Then I took the image from the M10 and upscaled to the exact same frame dimensions of the 42mp files from the A7rIII using the "Preserve Details 2.0" resampling method, which is by far the best upscaling method available. I was kind of shocked by the results. Even upscaled from 24mp to 42mp, the images from the M10 + older prime was much clearer, sharper, and had a much flatter field of focus than the A7rIII and the FE 24-105.

After performing this test I simply cannot use the FE 24-105/4 for landscapes any more. I didn't expect it to be at prime level performance, but I expected better than this. The 42 megapixels is literally wasted while using this lens.

I've made the raw files available for download here:

Leica DNG:
Sony ARW:

These are comparison shots after upscaling the Leica image to 42mp (100% crops):

Leica left

Sony left. Notice the stretched roof-top.

Leica right

Sony right

Upscaling settings applied to the M10 files

As you can see, the trees and other details on the sides basically look smudged on the Sony, and crisp on the Leica, even after upscaling from 24mp to 42mp. Quite a dramatic difference that I did not expect.

This is not a Sony VS Leica thing. It's more of an opinion and question about the FE 24-105/4 that everyone praises and says gives "prime-like performance". If I attach my Sonnar FE 55/1.8 to the A7rIII and compare that in the same way with my M10 + Summilux 50 ASPH, then the A7rIII is significantly better after the same upscaling methods have been used on the M10 files.

Is this normal for the FE 24-105? Or did I get a bad copy?

All the files attached here + more are available on my website folder:

 borgein's gear list:borgein's gear list
Fujifilm X100F Leica M10 Sony Alpha a7R III Leica Elmar-M 24mm f/3.8 ASPH Leica Summilux-M 35mm f/1.4 ASPH +10 more
Carl Zeiss Tele-Tessar T* 4/85 ZM Leica M10 Sony Alpha a7R III Sony FE 24-105mm F4
If you believe there are incorrect tags, please send us this post using our feedback form.
Flat view
Post (hide subjects) Posted by
Keyboard shortcuts:
FForum PPrevious NNext WNext unread UUpvote SSubscribe RReply QQuote BBookmark MMy threads
Color scheme? Blue / Yellow