Sony has announced a high-end ‘normal’ prime for its Alpha E-mount line of cameras: the Planar T* 50mm F1.4 ZA lens. As the third normal prime for the system, we wanted to know what it offers over the already excellent FE 55mm F1.8 ZA, so we set about performing some benchmark tests.
We’ll take a look at sharpness on this page, and bokeh, coma, and longitudinal chromatic aberration on the next.
Below, you’ll see a series of aperture progressions for the 50mm F1.4 ZA and 55mm F1.8 ZA. Have a look around the scene at various apertures to get an idea of the capabilities of these two lenses – with the caveat that this performance is only representative of our single copy of each lens.
Sony Planar T* FE 50mm F1.4 ZA
Comparing both lenses at F1.8 (a more level playing field), the 50/1.4 catches up to the 55/1.8 in terms of, but still lags in peripheral sharpness on the and sides of the frame. By F2, though, the 50/1.4 the 55/1.8 in central sharpness, though off-center it still . By F2.8 though, the 50/1.4 of the 55/1.8 even here off-center, and particularly where it pulls andstays ahead at higher F- numbers. Peripherally, though, the 50/1.4 never quite catches up to the 55/1.8, not at , and not even by (the lenses are a bit more even on the left side at and due to the weaker performance of our 55/1.8 on the left but, technically, the F1.4 is still a little bit behind).
What does this mean?
The new 50/1.4 ZA displays impressive sharpness and contrast at F1.4. Our copy didn’t hold up as well as the 55/1.8 wide open, but displayed particularly respectable performance considering the 2/3 stop light and depth-of-field advantage. These new lens designs deliver sharp and punchy images wide open, instead of the soft and hazy images you may be used to getting if you slap on old F1.4 designs on such high resolution sensors (remember that we’re using the unforgiving 42MP a7R II for this test).
That said, the new 50/1.4 does not retain this sharpness across the field as well as the 55/1.8, which offers better field uniformity at all apertures. By F2, though, the new 50/1.4 ZA matches the 55/1.8 in central sharpness, and surpasses it at all smaller apertures. Considering the high bar set by the 55/1.8 ZA, this is very impressive. However, you give up off-center sharpness at the widest apertures. If we were forced to pick an overall winner here in terms of sharpness, we’d probably go with the 55/1.8, but really there isn’t a huge difference between the two, particularly when you factor in the realities of copy variation.
|Roger Cicala over at LensRentals found the Sony 50mm F1.4 lens to be the sharpest centrally of any 50mm prime, outperforming the 55mm F1.8 ZA. However, peripherally, the 50/1.4 takes a plunge in terms of resolution, and the 55mm F1.8 pulls ahead. In fact, just 4mm out from center in the image circle, tangential resolution (which we assess by considering the highest frequency MTF trace: 50 lp/mm) drops below that of the 55/1.8 (solid purple line). Source: LensRentals Blog|
It’s worth noting that Roger Cicala at LensRentals found the central sharpness wide open of the 50/1.4 to exceed the 55/1.8 (see MTF traces above), while our visual results don’t show the 50/1.4 to exceed the 55/1.8 until F2.8. We can’t rule out the possibility that our copy of the 50/1.4 slightly under-performed relative to the average, perhaps due to decentering; however, it’s reassuring that he found the 55/1.8 to offer greater uniformity. This difference in peripheral sharpness may be the reason for the apparent discrepancy in our results, though copy variation is also highly likely to play a role (note that the worst performing 50/1.4 Roger tested was significantly worse than the best 55/1.8).
To elaborate, below we show our infinity scene overlaid with red and blue rings representing image heights of 4mm and 16mm, respectively: the two points where the 50 lp/mm MTF traces of the two lenses intersect. Between these rings, Roger’s 50/1.4 tangential 50 lp/mm trace falls below the 55/1.8. Hence, our visual shootout seems to agree with Roger’s results: aside from a very small region in the center, the 55/1.8 doesoutperform the 50/1.4 wide open (though we don’t see the improvement towards the edges of the 50/1.4 that Roger sees: our 55/1.8 remains ahead peripherally). And, again, we may have tested a copy that under-performs relative to the average, which Roger’s data is more representative of.
|Our infinity scene with red and blue rings that represent 4mm and 16mm image heights (distance from center in the image circle). Between these regions, Roger’s own MTF data indicate a dip in tangential resolution. It may only be a small region in the center (within the red circle) where the 50/1.4 significantly out-resolves the 55/1.8 wide open, which would explain the apparent discrepancy between Roger’s results and ours. Copy variation is also likely to play a role in the differences – we only tested a single copy.|
All that said, sharpness isn’t everything. How does the new 50/1.4 fare in terms of bokeh, coma, and purple/green fringing?
Bokeh is hard to quantify or even test in a controlled manner. But taking a look at the rendition of out-of-focus highlights near and far from the focal plane can give us a good indication of what to expect. Here’s we’re comparing how the two lenses render defocused city lights as we focus beyond, and in front of, the plane of focus.
Immediately, it’s obvious that the 50/1.4 has(a tell-tale sign of aspherics) compared to the 55/1.8. This helps the 50/1.4 render more natural out-of-focus highlights, which gives it a leg up in portraits like this.
Sony Planar T* FE 50mm F1.4 ZA
Interestingly, for objects in front of the focus plane (foreground objects), the differences appear, and it appears this is due to the 55/1.8 in addition to its onion-ring bokeh. In fact, the 50/1.4 has the edge here, because at least it lacks the severe onion-ring effect of the 55/1.8.
The bokeh differences become minimal when the city lights are. This is to be expected – significant defocus diffuses any patterning in the bokeh, which is why it’s hard to detect significant bokeh differences between lenses in close-up portraits with distant, defocused backgrounds. Far more telling is what happens with slightly defocused objects in front of, and behind, the focal plane, as we’ve done above.
To sum up…
Overall, bokeh is very pleasing from both lenses, as one would expect with any fast prime of this focal length. Some photographers may prefer the slight soap bubble effect of the 50/1.4, even though it may come at a slight cost in smoothness of bokeh near the focal plane. In our opinion, the lack of severe onion-ring effect gives a leg up to the newer 50/1.4.
Longitudinal Chromatic Aberration
The bokeh comparison above tells us something else about the lenses as well: the 50/1.4 has considerably less longitudinal chromatic aberration (CA) than the 55/1.8. Note less refer to widget above as you click these links). The relatively low levels of axial CA is also evident in real-world photographs: for example, purple and green fringing in the black lettering in front of, and behind, the focal plane, respectively, are very well-controlled in the image below. Usually this sort of fringing is exceedingly easy to spot on high-contrast lettering such as the ‘4’ below, yet it’s largely absent with this lens.as well as behind, and in front of, the focal plane (
In fact, you’ll generally note a lack of such fringing in all our images from the 50/1.4sample gallery, and the same cannot be said for the FE 55/1.8. This is important, as axial CA plagues fast primes with aspherical elements, and is fairly difficult to remove in post-processing. A fair number of the most compact FE lenses, like the 55/1.8 and the Batis 25/2 are notorious for their levels of longitudinal CA, and we’re quite pleased to see how well it’s controlled on the 50/1.4.
Below, we take a look at coma (the spread of point sources at edges) performance of the two lenses, at apertures ranging from wide-open to F4 (beyond F4, coma tends to minimal on most lenses). The 50/1.4 takes the lead here, showing barely any coma at all even wide open, while the 55/1.8 needs to bebefore edge highlights appear circular (which is still quite respectable performance). There’s some with our copy of the 50/1.4, but it’s . The 50/1.4 also avoids odd – but kind of cool – artifacts like .
Sony Planar T* FE 50mm F1.4 ZA
The new Sony FE 50mm F1.4 ZA is impressive. Our friend Roger Cicala over at LensRentals notes its central sharpness performance to exceed even the venerable Zeiss Otus, as well as the Sigma 50mm Art. He even found it to out-perform the 55/1.8 centrally, but fall behind it peripherally. Our results show the 55/1.8 to slightly pullahead wide open both centrally and peripherally, with the 50/1.4 pulling ahead in central sharpness by F2.8. We’ll look into nailing down the source of this discrepancy (it’s always possible our copy of the 50/1.4 was slightly decentered), but sharpness performance between the two is comparable, and the 55/1.8 was already one of the most respected normal primes with respect to sharpness.
Therefore, we feel it fair to say the FE 50/1.4 pulls ahead of most, if not all, normal prime offerings from competitors (though we’d imagine the Sigma 50mm Art to put up a good fight). Despite a severely off-center composition, take a look at the tack-sharp eye of our model below, shot at F1.4.
Sharpness isn’t everything of course, and in other respects, the 50/1.4 also impresses. Bokeh is pleasing, with very little onion ring effect, particularly compared to the 55/1.8. There is a slight soap-bubble effect to bokeh though, which may be pleasing to some, but can cost some smoothness in out-of-focus regions (but if it does, it’s certainly not easy to spot in any of the portraits in our gallery). Longitudinal CA is impressively well-controlled, to the point where you won’t notice much purple or green fringing near the focal plane even when shooting wide open. Nighttime cityscape and astro-photographers rejoice: coma is nearly non-existent even wide open. Videographers will appreciate the de-clickable aperture ring.
It’s clear that Sony is trying to cement itself as a real option for pros, and the new FE 50mm ZA helps further that goal. We have no reservations recommending this lens. That said, if you can spend the extra time processing out the axial CA in post, don’t mind onion-ring bokeh in out-of-focus highlights, aren’t bothered by coma, and don’t need the extra isolation or light-gathering capability of F1.4, the 55/1.8 is a compelling alternative that is lighter and cheaper, while offering smooth background bokeh and impressive, if not necessarily much greater, sharpness wide open. It also focuses significantly faster than the 50/1.4.* But for subject isolation that puts medium format to shame, while retaining respectable sharpness and contrast even far off-center wide open (and particularly by F2), the FE 50mm 1.4 ZA is the lens to own.
* The FE 50mm F1.4 ZA, like most recent Sony lenses, focuses stopped down at your selected aperture. This means that autofocus performance steadily drops as you stop down, since smaller apertures mean less light, and more depth-of-field (less phase difference) for the autofocus system to work with. By F9, phase-detect fails altogether, and you’ll experience significant hunting in AF-C.
While the intent is to minimize focus shift and shutter lag, ironically you may experience anincreased lag in shooting due to decreased AF performance when shooting at smaller apertures – particularly relative to the 55/1.8 which already focuses faster due to a smaller focusing group and potentially faster motor.
While we continue to express our disappointment at this focus behavior to Sony (we’d like to see the lens always remain wide open during live view, only stopping down to take the shot, with focus shift look-up tables for lenses exhibiting severe spherical aberration).