Introduction

Canon EF 35mm f/1.4L USM, which was released in late 1998, heralded a new era for Canon's lineup of fast wide angle lenses (this lineup now includes EF 14mm f/2.8L USM, EF 14mm f/2.8L II USM and EF 24mm f/1.4L USM). Priced at about US$1,100 (as of September 2007), the lens is targeted towards professional and enthusiast photographers rather then mainstream consumers.

The optical construction of the lens consists of 11 elements in 9 groups, including a single aspherical element intended to correct spherical aberrations. Like many wide angle lenses, Canon EF 35mm f/1.4L USM features floating element construction which improves picture quality throughout the aperture range. The build quality of the lens is superb, as expected from a professional grade L glass - there's no wobbling inside or out, rubberized focusing ring is smooth and the lens looks and feels very sturdy. The lens offers fast and near silent ring-type USM AF as well as full-time manual focusing system, which can be controlled with an AF/MF switch located on the side of the barrel. Like other modern Canon lenses, EF 35mm f/1.4L USM does not have a dedicated aperture ring and aperture levels have to be set directly from the camera.

Like its other wide angle cousins (EF 14mm f/2.8L USM and EF 24mm f/1.4L USM), Canon EF 35mm f/1.4L USM is a bit 'bulky', measuring 79 x 86mm (3.1 x 3.4in) and weighing 580g (20.5oz). The minimum focusing distance is 30cm, resulting in 1:5.5 maximum object magnification at this distance, the minimum aperture level is f/22 and the filter size is 72mm. The lens sports internal focusing mechanism, so the lens cams do not extend during focusing.

 

Image

 

The lens is designed to fit Canon's full-frame EF mount and on APS-C type cameras with 1.6x crop factor the field of view of the lens will resemble that of a 56mm lens on a full-frame body. The lens is compatible with Canon's Extension Tubes EF 12 II and EF 25 II and accepts Gelatin Holder Adapters III and IV. The factory box includes Canon EF 24mm f/1.4L USM lens, front and rear caps, EW-78C petal shaped lens hood, LP1214 soft case, manual and registration card.

 

Summary
Lens Composition 11 elements in 9 groups
Angular Field 63 degrees
Minimum Focus 30cm/1ft
Focusing Action AF/MF, USM
f-stop Scale f/1.4-f/22, camera-controlled
Filter Size 72mm
Lens Hood EW-78C (included)
Weight 580g/20.5oz
Dimensions 79x86mm/3.1x3.4"
Lens Case LP1214 (included)

 

Field Tests

The lens produced rather sharp results in the center across all aperture ranges. Borders were a little bit soft at wide apertures, continually improving with stopped down aperture. The softness around borders is more pronounced on a full-frame camera, which is expected.

 

ISO 400, 1/320, f/1.4, 35mm (Canon 5D)
ISO 400, 1/320, f/1.4, 35mm (Canon 5D)

When shot with wide open aperture, Canon EF 35mm f/1.4L USM produced well rounded and uniformly lit out-of-focus highlights, that carried pretty neutral boundary transitions (once in a while you'd see a slightly harsher edge highlighting, but that was quite uncommon). Contrast transitions in the OOF areas were pretty smooth and pleasing, and there was no sign of double-edging around OOF back/fore-ground objects.

 

Vignetting @ f/1.4 - full frame vs 1.6x crop
Vignetting @ f/1.4 - full frame vs 1.6x crop

The lens showed moderate amount of vignetting with wide open aperture on a full frame Canon 5D. Vignetting is reduced significantly once stopped down to f/2 and by f/2.8 it becomes pretty minimal. On APS-C camera the lens did not exhibit any major vignetting throughout the tested aperture range. The lens showed minor levels of color fringing across the frame, but managed to hold its ground against flare. The lens also showed no visible barrel distortion.

 

ISO 100, 1/4000, f/1.4, 35mm (100% crop)
ISO 100, 1/4000, f/1.4, 35mm (100% crop)
Canon EF 35mm f/1.4L USM image gallery...

 

Lab Tests

Please note that MTF50 results for APS-C and Full-Frame cameras are not cross-comparable despite the same normalized [0:1] range used to report results for both types of cameras.

Two lens samples were tested. The first sample was suffering from a lens element displacement and produced extremely soft and inconsistent results across the frame with wide aperture settings. The review reports results obtained from the second lens sample.

 

Canon APS-C: The lens showed pretty solid results in the lab. Center performance was exceptionally good throughout the tested aperture range. Borders were somewhat softer at f/1.4, and the performance here was rather mediocre. However, quality improves with stopped down aperture - at f/2 border performance is already solid and by f/2.8 becomes outstanding. Results remain very solid throughout the rest of the aperture range and in the f/2.8-f/11 range the lens is capable of producing outstanding 16in and decent 24in prints. Conclusion? Even with the somewhat weak border results at f/1.4, the overall performance is very solid, even for an L quality lens.

 

MTF50 (Line Width/Inch on the Print) @ 35mm
MTF50 (Line Width/Inch on the Print) @ 35mm

 

Normalized raw MTF50 @ 35mm
Normalized raw MTF50 @ 35mm

 

The lens showed somewhat higher levels of CA around borders on an APS-C camera, where it averaged ~1.3px across the frame. The lens fared better in the center, where CA was well under control, not exceeding ~0.4px even at the widest aperture levels.

 

Image borders @ 35mm (100% crop): f/1.4 vs f/8
Image borders @ 35mm (100% crop): f/1.4 vs f/8

 

Canon FF: The lens gave up some ground around borders on a full frame camera. This is most noticeable at wider aperture levels, where unlike with an APS-C camera, quality improves with stopped down aperture but at a much slower rate. In the f/1.4-f/2 range border performance is pretty mediocre, but by f/2.8 it reaches more or less decent levels and by f/5.6 becomes exceptionally good. Center performance, on the other hand, is very solid straight from f/1.4. Conclusion? A bit disappointing, especially after such a promising start with an APS-C camera! The overall performance can still be considered pretty decent, even with somewhat softer borders at wide aperture levels.

 

Normalized raw MTF50 @ 35mm
Normalized raw MTF50 @ 35mm

 

Canon EF 35mm f/1.4L USM exhibited moderate degree of barrel distortion. At 0.95%, distortion can become noticeable in general photography, especially with architectural type shots.

 

Distortion (FF) @ 35mm
Distortion (FF) @ 35mm

 

CA was more or less under control on a full frame Canon 5D. Chromatic aberration averaged ~0.4px in the center across the tested aperture range. CA remained quite manageable even around borders, where it averaged ~0.8px across the aperture range.

 

Image borders @ 85mm (100% crop): f/1.4 vs f/8
Image borders @ 35mm (100% crop): f/1.4 vs f/8

 

Alternatives

Canon currently offers another 35mm fixed focal lens in EF mount - EF 35mm f/2, which aside from being 5x cheaper when compared to EF 35mm f/1.4L USM is also more compact as well as is lighter. The slower 35mm variant shows excellent center performance, but unfortunately like many other wide angle lenses falls short in performance around borders at wide apertures. If you're willing to consider a wider angle lens, then you might want to take a look at EF 24mm f/1.4L USM (review), which offers outstanding center performance throughout the aperture range, but somewhat softer results around borders at wide aperture levels. As I mentioned in my other reviews, the list of high quality wide angle lenses in EF mount is rather short. There are a number of wide angle prime lenses manufactured by Sigma, including 20mm f/1.8 EX DG ASPH and 24mm f/1.8 EX DG ASPH Macro, but pretty much all of them do not offer any improvement in performance, especially at wider apertures. That is probably why many photographers these days are starting to look at alternative mount lenses from Carl Zeiss, Leica and even Olympus OM and adopt them to the EF mount.

 

Recommendation

Canon EF 35mm f/1.4L USM was supposed to be one of Canon's flagship lenses. It almost reaches that status, with outstanding performance on an APS-C camera, but falls a bit short on a full frame body. So where does this leave us? If you shoot an APS-C camera, then you will be hard pressed to find a better lens (image quality-wise that is). Those of you with full frame bodies like Canon 5D might still want to opt for this lens if you can accept softer border performance at wider apertures in favor of exceptionally good border performance (quality is still outstanding once stopped down to f/5.6 and beyond).