Nikon D5100 Review – Page 4
Welcome back to the next part of this story. As written earlier, I have a different dayjob, so I won’t be able to continously crank out posts and pages on my blog, neither should it be just for the sake of writing. Anyway, I haven’t concluded part 3 as intended last weekend, it grew too long to reasonably read it. Will do so in this part.
I’ll start with battery life as it happened that my D5100 ran out of juice just after about 1.000 photos. About 10% of these images were shot with LiveView, I did no video taping and basically didn’t use the internal flash. All in all, its ok, not outstanding like the D7000, which – having a bigger battery – has an amazing battery life for photographers. During last New Year night, I was out there at –25 degree celcius and the D7000 was way better than the D300 in handling this condition (Both have similar battery capacity, so it looks like the savings came from the consumption side.)
You might be interested which components in a D-SLR use most of the scare energy resources, so let’s spend some time in this department. In my humble view, the D2/D3 series are leading the Nikon cameras in battery reach. it is not unusal to get between 3.000 – 5.000 images with one charge. The D7000 is from my point of view now the second best “class”. I don’t have any real (measured) data from the D5100, but we can draw some comparison from other classes to better understand the balance between energy capacity in the battery and camera consumption in different scenarios. Let’s compare 3 bodies – the D5100, the D7000 and the D3.
- D5100 (EN-EL14): 1030 mAh & 7.4 V = 7.7 Wh
- D7000 (EN-EL15): 1900 mAh & 7V = 14 Wh ( +81% vs. D5100)
- D3 (EN-EL4): 1900 mAh & 11.1V = 21 Wh ( +174% vs. D5100)
- there is a newer battery for the D3 series available, the EN-EL4a
- D3s (EN-EL4a): 2500 mAh & 11.1V = 28 Wh ( +260% vs. D5100)
- Due to lack of D5100 data, I’ll take D3 values as baseline and leave it up to the reader to guesstimate it for the D5100 appropriately.
Basically, modern D-SLR cameras consume in OFF and STANDBY mode (no button pressed) about the same energy. We are talking low levels, about 2-300 micro ampere. Press any button (turns the electronics, AF and metering on) and energy consumption increases roughly 1.000 fold to 300 milli ampere. That’s quite a lot! Keep the metering in OFF mode and turn on the display in the back of the camera for image viewing gets you into the same territory. The most rapid way to drain any battery is via the often requested LiveView feature, which gets you into the 1.000 milli ampere range.
Looking the other way around:
- LiveView uses about 3-4 times more energy than a classical D-SLR shooting mode.
- As metering uses approximately 1.000 times more energy vs. standby, try to avoid any long setting in this regard (Option C2 in the custom setting, page 160 in the D5100 manual)
- Playback takes also 1.000 more energy vs the camera just being ready for the next shoot. People with extensive chimping habits might be the first to get a second battery.
- LiveView takes about 3 to 4 times more energy as normal shooting. This wouldn’t be too bad, but the average time those things are turned on varies significantly, taking the LiveView mode even deeper into the energy negative territory. Normal setting for metering (remember it’s 300 milli ampere) is set to turn metering off after 8 sec. Typical shots with LiveView are signifcantly longer. So you have to multiply your higher engery consumption (3-4 fold) with the longer time vs. normal shooting.
- There is almost no benefit to turn the DSLR camera OFF vs. keeping it in standby.
- Ok, this are roughly the numbers for the D3/D3s and Nikon might have used the time to find some magic ways to get the efficiency of individual components significantly up. But the basic pattern will be very similar. As the D3s, D7000 and D5100 have now the same resolution displays in the back of the camera, I was interested if there was any significant brightness difference in this component. To force the displays to maximum brightness, I shot with all 3 cameras an overexposed photo of a softbox and turned the highlight mode off to avoid flashing displays.
- Here is a shot of the D3s, D5100 and D7000 displays
Checking in CNX2 the brightness, the D7000 was by a small margin brighter (220), vs 215 for the D5100 and 210 for the D3s. Basically, the brightness of all 3 displays are by any normal means identical. If (and only if) the display in the D5100 has a similar energy profile than the one in the D3s, it will drain the smaller battery of the D5100 very fast.
On a second note: The 3 displays reproduced color in a slightly different way. When using the white balance tool in CNX2, the WB settings per display are different. It might be influenced by the different viewing angles, so I will not go deeper into this. As said, just a simple note that the displays might not be completely identical beyond its identical resolution.
- WB correction for the D3s display (red channel 1,13 / blue channel 0,97)
- WB correction for the D5100 display (red channel 1,18 / blue channel 1,11)
- WB correction for the D7000 display (red channel 1,09 / blue channel 1,05)
- Where does this leads to?
- The kind of exercise above was meant to increase awareness about the impact of energy consumption in general and LiveView in particular. D-SLRs had a tradition that their efficiency came from keeping all the energy hungry components most of the time OFF, only waking up for seconds to do their job. The market demand for LiveView and Video recording will bring D-SLR closer to consumer cameras from a battery life perspective. Don’t have an issue with this, as long the new features don’t disable the old ones.
- Energy aware shooters might choose to stick with the old habits. If you are an environmentalist you should do so as well
Autofocus performance in low light
The latest generations of cameras cranked up ISO to levels to a level which had been unimaginable just a couple of years ago. While most attention in communities and internet fora centered around the ability of a new sensor to deliver ultrasharp photos at nightvision light levels, this discussion and the “tools” to prove it was flawed for many reasons. (I am guilty as well)
The D3s and the latest generation of fast lenses like the AFS 24mm/1.4G, AFS 35mm/1.4G or AFS 85mm/1.4G exhibited a new boundary condition. It wasn’t the ultra sensitive sensor for “normal” shooting conditions, which limited the versatilty of the camera, but other components. In my opinion the CAM modules will need a major overhaul in the future, if this trend of increased sensor sensitivity continues.
Let’s get deeper into this technical point.
Most AF modules in contemporary D-SLRs have an operational range of EV –1 until 19. An exposure value (EV) of 0 is defined as an exposure time of 1 second, taken with a relative aperture of f 1.0 and at ISO 100.
Here is an EV table for ISO 100. The shaded fields (above EV 19) are outside the specification of current Nikon CAM modules.
Ok, so far so good. How’s about the latest generation consumer cameras? Nikon marketing claims ISO 6400 for the D5100 and D7000. Fortunately they are correct, ISO 6400 is from a sensor perspective still usable. But what is the impact for the AF system at ISO 6400?
As in the ISO 100 diagramm above, the shaded fields in this ISO 6400 table are EV values outside the specs of 3 cameras. The 3 yellow fields are quite often used in handheld available light photography. See how close they are to the edge of the specification of the AF modules?
One more caveat: AF fields of cross sensor type have better phase detection in low light than simple line AF fields. Why does this matter? Because the usual way of focussing with f1.4 lenses is a bit different from kit lenses. Imagine you would like to do a portrait with an f-stop of 1.4.
With most slower lenses it is perfectly fine to use the center AF field to focus on the closer eye, keep the shutter half pressed and recompose for your photo. With f1.4, the DOF is so shallow, that this mode of operation will lead in many cases to undesirable outcomes as the original intent (to have the closer eye in focus) did not materialize.
Here is a rough diagram of the problem. The red camera shows the position and direction during the focussing phase. The blue camera displays the direction after recomposing. Combine this shift in the focusplane with a VERY narrow DOF, and you are immediately off focus with the eye.
How can this be solved? Quite easily. Select a different AF field, closer to the eye you’d like to get in focus and which fits the intended composure and you are done. If there is enough light, then you wouldn’t notice in most cases if the AF field is of a cross type or not.
In low light, you get faster to the limits of the CAM system than indicated above in the EV tables. The D5100 has only the center AF field as a cross type, the D7000 has 9 cross sensor AF fields and the D3s has 15. We will see later, what difference it makes.
BTW, a side note: Nikon has 2 different versions of one CAM module. The CAM 3500DX for the D300/D300s series and the CAM 3500FX for the D700 and D3 series. With the D7000, Nikon introduced the new CAM 4800DX module. Think, we can expect the CAM 4800FX in the next generation of FX bodies.
With that, lets close this technical chapter on AF sensitivity and its role in the current cameras.
Fast night shots with the D5100
Back to the camera this review is about, the Nikon D5100.
My curiosity centered around the “partnership” between the CAM1000 AF module and the stellar 16,2 MP sensor’s ISO abilities (at least for a DX camera). To get a base line I compared the D5100 with the D7000 and the reference body for low light photography – the D3s. To get some insight into the differences between the 3 cameras I shot in AF-C (with all AF fields enabled) cars passing by on the nightly light city streets my hown town. ISO settings were between 3200 and 6400, aperture stops were either wide open or max at f2 to keep shutter speed above 1/125 to freeze the cars passing by with between 30 and 40 mph (50-65 km/h). As this setting was somehow unpredictable/unrepeatable I did 10 runs each with each camera. The first scenario was panning cars passing by, the second the position was more heads on to the approaching cars.
An AF sensor needs light to do his job. Where does light come from? Especially important for my not-so-serious night shooting exercise in continous mode.
1) A fast lens. In this regards the AFS 35mm/1.8 differentiates to the kit lenses. What apterture do you get at 35mm?
- AFS 35mm/1.8G = f 1.8
- AFS 18-55mm/3.5-5.6G = f 4.5
- AFS 18-105mm/3.5-5.6 VR = f 4.5
- AFS 18-200mm/3.5-5.6G VR II = f 4.8
- Roughly speaking, the 35mm prime lens gathers 4-5x more light than the kit lenses. While the sensor sensitivity progressed significantly, the AF module’s sensitivity didn’t. See the pattern?
2) A fast blackout time. The faster the shutter/mirror combination can do its job of getting the mirror up, open the shutter, expose, get the shutter and the mirror down, the more time is available for the AF system during continous shooting. (The AF gets only light when the mirror is in its downward position). The blackout time for the D3 is known to be 55ms, it is not available for the D5100 – save to assume it is slower.
Not suprisingly, the D3s is a class in itself. Not only with the cleanest image, but the AF system captured 10 out of 10 scenes in both cases. The D3s didn’t differentiate the colors of the cars, focussing on black cars as well as white and other colors. A joy to use in these conditions.
The D7000 got 7out of 10 and 6 out of 10. It had mostly challenges with dark cars or cars with less edges to “help” the AF. I’m not sure, but the longer blackout time might have impacted the CAM 4800DX performance as well. I am really curious what this camera with “semipro” components like faster blackout time and better filter technology will be able to deliver. Really interested in the capabilities of the upcoming D400 in low light (or what ever the name will be).
The D5100 is a great low light camera, but the chosen setup was kind of mean to the camera. With only one cross sensor AF focus field, the oldest CAM module, a quite long blackout time, it clearly showed the difference to the higher priced Nikon cameras. Only 2 or respectively 1 runs got properly focussed (out of 10 each). To express it positively (which is my default point of view): The D5100 is a great camera in low light, based on a fantastic value. Full stop. If the photographer has additional requirements, Nikon has more advanced models in its lineup with increasingly more sophisticated capabilities. Its up to each one interested to justify the additional expenses vs. these additional features.
Scenario 1: Panning cars passing by
The D5100 had issues when cars were in the lane closest to me (or was it me?), with dark cars and got most easily confused when more than one car was in the viewfinder. Here is a run with a single red car (the best run).
While red and other colorful cars were a kind “prerequisite” for the D5100 AF, the D3s had no issues whatsoever with different car colors, speed of the cars or distance passing by.
D3s and black cars
Scenario 2: Cars coming more heads on to the camera
The D5100 had initial acquisition issues at distance and when the AF locked on, it usually lost the car within the next one or two frames. Quite often it recovered later on, but the frames in between were not focussed properly. Here is one of the better runs.
The D7000 with its more advanced CAM module fared better. Initial lock-on was consistently good, following the cars getting closer seemed to be better suited to the 39 AF fields covering a bigger part of the frame with more AF density. The higher frame rate in low light vs. the D5100 is also visible (more images)
Last in this part 4 is the D3s. As written earlier, the camera shines not only with its sensor, but also the AF system. Pick and choose your car color – how’s about a black one?
With that, this concludes now the lowlight part of the D5100 review. The next part will potentially cover image resolution and sharpness with some lenses.