This is a thing that’s nice to know but that I personally don’t care about that much. Because: You can correct it rather easily, mostly even in-camera when shooting JPEG. Plus, in most scenes where you shoot wide open, it’s usually something that looks pleasing anyway. Which you can see from the fact that it’s one of the most frequently used effects in post and in cell phone cameras. But yes, it can get in your way if you don’t need it.
Here, I’m giving you two results. One from the RAW files, where you can see the actual vignetting of the lenses, and one from JPEG files, meaning with in-camera correction applied, so you can see how well this is corrected for. The following chart is made from shots against the evenly illuminated (two big softboxes on modelling light) white studio wall. Spot metering in the middle of the image frame, and no further corrections of exposure etc. Here’s the RAW chart:
What you can see:
- Wide open, they all have very strong vignetting, with the Zeiss Otus, the Nikkor 50/1.4G and 50/1.8G being even more prone to it.
- The Zeiss is always overexposing a bit, so it looks “better” than it actually is. Vignetting is as strong as the two Nikkor 50s, and more importantly, it gets really dark in the corners.
- The three lenses (Zeiss and the Nikkors) are also the heavyweights at f/2.
- The Sigma and the Nikkor 58/1.4G are noticably softer with their vignetting.
- The Sigma vignetting is already at f/2.8 improving to a level that’s hardly recognizable in practice.
- At f/4, vignetting with all lenses is virtually gone.
And here’s the JPEG chart, with in-camera vignetting correction set to “normal”:
What you can see:
- The Zeiss isn’t corrected for.
- For some reason, the Zeiss is very prone to over-exposing, most strongly at f/2.8. I don’t know why, but that’s consistent for all shots I did, no matter what metering pattern I chose.
- For some magic reason, the Sigma is almost completely corrected. I always thought it’s the camera that has to know the lens, but apparently, this is not how it works. Can someone enlighten me, please?
- The rest, well: Looks ok to me, but nothing spectacular. I’m wondering why the Nikon lenses on a Nikon camera don’t have the same perfect correction as the Sigma.
Coma are those “blurb” effects on bright points of light towards the corners of the image frame when shooting wide open. Again, nothing that I’d care about, but I was curious.
Why don’t I care? Firstly, I’ve never used a 50mm for motives where this would matter (architecture and the like). More importantly: Coma is something that affects only large apertures, and whenever I’d shoot a night scene where this matters, I have a tripod, and can stop down. Yes, yes, there might be daylight scenes, but you’d have to construct them and they’d be very unrealistic. Or when was the last time you shot the drops of a sprinkler against the sun at f/1.4 with a 50mm lens? Right, I thought so.
Anyway, the result is that the 50/1.4G is prone to it, the rest isn’t so much, and the Zeiss is completely free from it. Here’s one of the motives to check it out:
See those small lights on the bottom right of the frame? Let’s look at 100% crops, shall we? If you don’t know what coma looks like, the first comparison shows the Zeiss crop for you to see how the lights looked in reality, slide to view to the 50/1.4G crop to see creative re-shaping. Click through the gallery to see the other lenses, too (non-high-dpi 100%)
Just like with the coma issue, the sunstar quality of a 50mm is not something that I’ve ever used in practice: Motives that I take where I would want sunstars are usually shot with a wide-angle lens like 35 mm and below. Anyway, it never hurts to know what if, so here you go.
The number of rays of sunstars are defined by the amount of blades of the lenses’ aperture diaphragm: Effectively, an even amount will produce that even number of rays (e.g. 8 blades = 8 rays), an uneven number will produce twice the amount (e.g. 9 blades = 18 rays), usually interchanging stronger and softer rays. How “strong” these rays are and how clear the sunstar is depends mostly on the shape of the blades, round blades have a harder time than straight, and will need more stopping down, which is something you need to do anyway. Results at f/16:
As you can see:
- The Nikkor 50/1.8G produces the clearest shape, followed by the Sigma. Both lenses produce usable sunstar results already at f/8.
- The Zeiss, the Nikkor 50/1.4G and 58/1.4G don’t produce such a clear star shape.
- In fact, their shapes above at f/16 are achieved with the Sigma and the Nikkor 50/1.8G already a stop or two lower.
8. Color Rendition
The color rendition only matters whenever you cannot use a gray card, which, depending on what you take photos of, may be very often. Direct out-of-camera shots always have a slightly different cast to them from the different lenses. They do. And no, using Nikon-only lenses does not give you one certain rendition, each lens is different.
You can see how the tints look in the bokeh comparison, where you get an overall impression. Here’s a chart:
I used the X-Rite Color Checker to see if these impressions are valid also if you measure them. Keep in mind that this is all only relative to eachother, as the camera’s sensor already has its own color rendition. These are the trends:
- The Zeiss, the Sigma and the Nikkor 50/1.4G all have a rather warm color rendition, with the Sigma being the warmest.
- The Zeiss and the Nikkor 50/1.4G have a slight green tint. Which I didn’t see without measuring, but I’m just sayin’. Anyway, the Sigma doesn’t.
- The 58/1.4G and the 50/1.8G have a rather cool rendition to it.
As said, all these color casts can be easily neutralized using a good grey card. Then again, you don’t always want to put a grey card into every single picture you happen to be taking.
Go back to page 1: Introduction
Go back to page 2: Perceived sharpness, focus shift
Go back to page 3: Bokeh quality
Go back to page 4: Sensitivity to bright light inside/outside the frame
Go to next page 6: Handling and Quality
Proceed to page 7: Summary