Samsung SC-MX20 Camcorder Review
Earlier this year, Samsung blew us away with the surprisingly strong performance of their high definition SC-HMX20. Over the years, we had grown accustomed to mediocre video from the Samsung line and the HMX20 gave us reason to pause—even reason to recommend a Samsung camcorder. Unfortunately, the surprising improvements we saw in the HMX20 didn\'t trickle down to the latest standard definition camcorder: the Samsung SC-MX20 (MSRP $229.00).
Video Performance* (2.75)*
Unfortunately, the Samsung SC-MX20 didn't get the same sensor upgrade that its high definition cousin received; the MX20 still sports the same 1/6-inch CCD that we saw in last generation's MX10. The gross pixel count is still 680,000 and everything about the lens remains the same, except for the new Schneider branding. If the SC-MX20 is to have improved video performance, let's hope that Samsung has upgraded their processing. We tested the camcorder both in our labs and out in 'real life' to see how it fared.
Inside the lab, we shot a DSC Labs Chroma DuMonde chart at an even 3000 lux, then compared the results to other similar camcorders. Shooting in auto mode with a manual white balance, the Samsung produced footage fairly comparable to most of the competition. The biggest difference you'll see is that the MX20 seems to have a darker, noisier image. This becomes more of an issue in less-than-ideal lighting conditions. Compression artifacting plagues both larger fields of color and also areas of fine detail, but the sharpness looks quite good overall. Color, in general, looks muddy, especially compared to the often oversaturated look of today's camcorders.
Unfortunately, we did not test last generation's SC-MX10, but we can compare the MX20 to similar standard definition camcorders from other manufacturers. Overall, the MX10 looks darker and muddier than competing camcorders from JVC, Canon, Sony, and Panasonic. Is this the result of a sub-par sensor? Poor manual white balance? Or disappointing processing? The likely answer is 'all of the above,' but we would give the edge to processing, since this is the same type of sensor used in the JVC GZ-MS100, Canon FS100, and the Panasonic SDR-H200—all with better results.
Even though we've come to expect a fair amount of compression artifacting from standard definition camcorders, these blocky chunks of discoloration are reduced among the competition, especially the Sony DCR-DVD910. The Samsung SC-MX20 looks even worse than the JVC GZ-MS100, which was also riddled with compression artifacting and purple fringing. The most surprising result from the SC-MX20 was how sharp the footage looked, especially compared to the JVC. It's too bad that the areas of finest detail are marred by so much noise and discoloration.
Out of the lab, the Samsung SC-MX20 verified what we saw with our in-lab testing: the sharpness is quite good, but compression artifacting is everywhere you look. The colors, at least, seemed to look a lot better with natural outdoor lighting. Before we go any further, it's important to emphasize that Samsung is marketing the MX20 as their 'Shoot and Share' YouTube camcorder. We're often spoiled by the beautiful footage we see on $2000 high-definition camcorders—we're bound to be disappointed with these standard def budget camcorders. Though there's a considerable amount of compression artifacting and motion trailing, these unpleasant side effects will get added to your video if you upload it to YouTube anyway—YouTube compresses video significantly in order to make files managable for the web. If the only destination you have in mind is YouTube, the SC-MX20 will be more than sufficient.
As for how the Samsung compares to similar camcorders from the competition, we were disappointed to find Samsung falling behind the rest: more noise, more artifacting, disappointing color balance. It seems that the improvements to Samsung's high definition line—the ones that surprised and impressed us on the SC-HMX20—simply didn't make it over to the standard definition models. The video performance on the MX20 is sufficient for YouTube, but generally mediocre.
Video Resolution* (4.875)*
To test video resolution, we shoot a DSC Labs video resolution chart at an even, bright light, then watch the playback footage on an HD monitor. The resolution is determined in line widths per picture height (lw/ph). The Samsung SC-MX20 produced a horizontal resolution of 325 lw/ph and a vertical resolution of 300 lw/ph.
These results confirm what we saw with out eyes: the SC-MX20 is just a bit better than most other camcorders in its class when it comes to resolution. These numbers are identical to the Panasonic SDR-H200, but better than competing models from JVC, Canon, and Sony.
Every camcorder that comes through our labs undergoes the same barrage of low light testing. Unfortunately, it's a rigorous series of tests that the Samsung SC-MX20 just couldn't withstand. Despite the excellent low light performance of Samsung's latest high definition model, the standard definition SC-MX20 didn't stand up to the competition.
The first stage of our low light lab testing involves shooting the DSC Labs Chroma DuMonde chart at an even 60 lux and 15 lux, then comparing it to the results of similar camcorders.
At 60 lux, the most striking shortcomings of the MX20 are brightness, color balance, and noise. The footage looks darker than much of the competition—especially the Panasonic SDR-H200 and Sony DCR-DVD910, which probably boost the gain in order to brighten the image. The Samsung isn't the darkest camcorder we've seen at 60 lux, but it's certainly near the bottom of the pack. Unfortunately, making out your subject in the darkness is even more difficult with the clutter of compression artifacting and noise. And unlike the fine-grained noise on the JVC GZ-MS100 or the more subtle grayscale noise on the Panasonic SDR-H200, the noise here on the Samsung SC-MX20 is coarse, tinted with color, and distracting.
At 15 lux, problems with darkness and noise are even more troubling than before. The camcorder also really struggles with color balance, depicting purple as a strange navy blue and giving the greens a yellow-brownish hue. This was a problem at 60 lux and even at 3000 lux, but it's made more apparent in the darkest settings.
It isn't all bad news, however. The MX20 does manage impressive whites and blacks at both 60 lux and 15 lux—surprisingly good when compared to the abysmal white balance that the camcorder performs under bright, indoor lighting conditions. The MX20 also continues to look sharper than much of the competition, which helps when trying to distinguish fine detail under these less than ideal conditions.
We would be remiss if we didn't also comment upon the SC-MX20's C.Nite modes, which are designed to improve the camcorder's performance in low light. There are three C.Nite modes to choose from: Auto, 1/30th shutter speed, and 1/15th shutter speed. At all three C.Nite settings, we found that the image was, in fact, brighter, less noisy, and more saturated. In Auto C.Nite, especially, the footage was a good improvement over the regular 15 lux results of the camcorder. Unfortunately, recording a still subject while mounted to a tripod gives a false impression of the C.Nite results: even with the auto settings for C.Nite, there was increased motion trailing when either the camcorder or your subject is moving. At 1/30th and 1/15th shutter speeds, the image was remarkably bright, but any movement made the footage practically useless.
In short, we recommend trying out C.Nite for low light situations, but suggest sticking to the auto setting to avoid the unreadable blurriness of the 1/30th and 1/15th shutter speeds. Even the User Manual suggests that C.Nite is more useful for achieving a 'slow motion like effect' than for improving ordinary low light performance. A tripod is also a must—the digital stabilization feature cannot be used when the camcorder is in C.Nite mode.
The second stage of low light testing measures sensitivity—what is the lowest light level at which the camcorder still produces readable footage? For this test, we connect the Samsung SC-MX20 to a waveform monitor, then slowly and steadily lower the light until the camcorder is outputting a maximum exposure of 50 IRE. The MX20 was able to produce a peak 50 IRE at 13 lux. This is a good score for camcorders in this class: identical to the JVC GZ-MS100 and the Canon FS11, better than the Sony DCR-DVD910 and Panasonic SDR-H200.
Why then does the Samsung appear to be darker than the competition? Our hypothesis is that the MX20 processor is good at handling blacks and whites, but less well equipped for grays and all the colors in between. While the brightest whites of a scene produce 50 IRE at 13 lux, the rest of the scene may lag far behind. This isn't good news for recording subtle nuances at twilight, but high contrast scenes will probably turn out better.
The final stage of low light testing determines color accuracy, noise, and saturation. We shoot an X-Rite Color Checker chart at an even 60 lux, then run the results through Imatest imaging software. At best, the Samsung SC-MX20 was able to produce a color error of 16.6—identical to the poor results from the Panasonic SDR-H200. In the case of the Panasonic, the color error is likely a result of undersaturation due to the three small sensors; for the Samsung, we see the processor skewing colors toward the wrong tonality. (Seeing purple constantly outputted as blue is a videographer's nightmare.)
Unsurprisingly, the Samsung also produced some of the worst noise in its class: the 1.5375 percent noise is a little higher than comparable models from Sony, Canon, and Panasonic. It's considerably worse than the results from the JVC GZ-MS100 or the JVC GZ-MG330. Saturation measured just 59.95%—about on par with the Sony DCR-DVD910 and Panasonic SDR-H200, but a bit worse than the rest of the competition.
All in all, what we saw in low light was pretty underwhelming. The Samsung SC-MX20 just isn't the low light performer that its high def big brother is; if you want an affordable standard definition camcorder for club hopping, evening barbeques, or your child's school play, the JVC GZ-MS100 or Canon FS100 would be better choices.
The SC-MX20 is equipped with Electronic Image Stabilization (EIS), just like its high definition big brother, the HMX20. EIS is generally not as effective as Optical Image Stabilization (OIS), which is featured on even standard definition Panasonics and Sonys. EIS creates a digital buffer around the frame to compensate for abrupt panning or shaking, while OIS isolates the lens itself in order to maintain resolution. Based on the disappointing results from the EIS on Samsung's HMX20, we weren't expecting much from the SC-MX20. Even so, we were disappointed.
Image stabilization is tested at two speeds using our custom-built shake emulator. Speed one is roughly equivalent to normal handheld shake, while speed two is more similar to what you would experience on a bumpy car ride. At both speeds, the SC-MX20 EIS system did minimize some vertical shake, but did almost nothing to eliminate horizontal shake. At speed one, the shake reduction was a pitiful 28.57%, while speed two saw only a 7.69% decrease. These results are so pitifully below the average that we almost wonder why Samsung even bothered.
Wide Angle* (9.6)*
To test a camcorder's maximum field of view, we set the camcorder on a tripod, pull back the zoom fully, and disengage EIS. Using a vertical laser and interpreting the video on an external monitor, we can obtain a true aspect ratio. The SC-MX20 displayed a wide angle measurement of 48 degrees—identical to similar models from Canon and Sony and better than the competition from Panasonic. The JVC GZ-MS100 displayed an impressive 56 degrees.