Element ELST5016s review: TV input lag, deinterlacing and upscaling using the piLagTesterPRO
Image quality
Good upscaling is critical for retro gaming. Ideally, all pixels should appear equally sharp and bright (no aliasing), and angled lines should appear smooth, with no jagged, irregular steps. Also important is that the display shows most or all of the pixels it is sent. Often, this is not the case, with some number of pixels cropped from the bottom or top edges. Shockingly, these tests are relevant for modern gaming as well, because even at their native resolution many TVs have aliasing and cropping.
I attempted to adjust the set to minimize cropping and aliasing. On this set scaling and aspect ratio are combined under a single option called "aspect". Your options include a couple way-zoomed-in modes, a just-scan option that does zero cropping but always uses 16:9, and two moderately cropped settings, one 16:9 and the other 4:3. Interestingly, this TV has a somewhat hidden service menu, (press 0 a bunch of times on the main menu) which gives to-the-pixel control over cropping. But unfortunately it doesn't seem to save this over a power cycle, so I didn't use it for the report below.
| resolution | quality | cropping (side, top) |
| 480p/i | A. Mild over sharpening and halos, but the overall result is nearly as good as you ever outside of a dedicated scaler. | 10,13 |
| 720p | A. Pixels get rearranged such that a 1-pixel checkerboard is no longer a checkerboard, but most detail is preserved and there's no aliasing and no cropping. | 0,0 |
| 1080p | A. Even in just-scan there's a tiny bit of aliasing even though there's no cropping. Bizarre, but the effect is very mild. Possibly what I'm really seeing is over sharpening, not aliasing from resizing. | 0,0 |
| 960p | B- (4:3), A-(16:9). This resolution is a favorite with the OSSC crowd, but doesn't look that good on this TV. Might as well use 480p output from the OSSC since this TV does good upscaling at that resolution. | 0,0 |
Overall this set does a good job of upscaling. No resolution is perfect but all are handled pretty well. In fact this might be the best well-rounded TV I've seen. Since it does such a solid job in all standard resolutions, the 960p results, while disappointing, are not a big deal for OSSC users, as they can just use 480p.
The display has 3 HDMI, 1VGA, and 1 yPbPr input. I only tested HDMI.
Input Lag
I used a piLagTesterPRO to measure input lag. This device sends a frame of video over HDMI and measures how long it takes to display it. This display does not have a game mode; not even a game 'color' preset. I toggled all the display quality settings and did not see a consistent effect on lag, however the tests reported are with every "enhancement" set to off.
Input Lag Test Results
I report two kinds of values. 1st response measures how long it takes for the TV to start responding (I use a 5% change in display brightness). This overly optimistic value doesn't tell how long it takes to see anything useful, but matches what other reviewers call input lag. full response is a more realistic measure of lag, and requires the display to reach 80% of full brightness. This combines both input lag and response time, and is closer to what you would actually experience in a game.
| top | bottom | ||||
| Resolution | 1st response | full response | 1st response | full response | scan out |
| 480i | 39.6 | 47.6 | 55.5 | 63.5 | 15.9 |
| 480p | 22.7 | 30.7 | 38.4 | 46.4 | 15.7 |
| 720p | 21.4 | 29.4 | 37.1 | 45.1 | 15.7 |
| 1080p | 21.5 | 29.5 | 37.3 | 45.3 | 15.8 |
| 1080p50 | 28.2 | 36.2 | 44 | 52.0 | 15.8 |
| 1080p24 | 55.8 | 63.8 | 71.5 | 79.5 | 15.7 |
This set advertises a huge range of supported modes, and actually it probably supports any reasonable resolution and refresh rate that's sub 1080p60. For progressive sources at 60hz, the input lag (first response) is consistently 22ms, the response time is 8ms, and the scan out is 16ms. Deinterlacing adds 16ms. The TV also supports a number of other refreshes rates. In particular, 1080p PAL (50hz) and film (24hz) are handled perfectly, with no dropped frames or drifting lag. Interestingly, the scanout is the same 60hz rate; the TV just waits longer to start drawing until the required data has been sent over HDMI, but once it starts drawing it goes at the full speed.
The TV also reports that it can support some computer modes that refresh at 72 or 75 hz. This is both a lie and true: it still draws the screen 60hz and thus has to drop frames, but it will at least accept these inputs and display them. I would strongly advise against using these modes as the higher refresh rate in no way improves the lag. For anybody who cares, here's the raw data for the full response at the bottom of the screen for all supported modes. It's sorted by the average drift in lag per second, so the results at the bottom are modes where the TV cannot actually match the input refresh rate.
| mode name | res/refresh | full response | variability |
| dmt#83 | 900p60 | 38.1 | 0.02ms |
| dmt#47 | 900p60 | 37.2 | 0.04ms |
| dmt#16 | 768p60 | 37.4 | 0.05ms |
| cea#16 | 1080p60 | 37.3 | 0.06ms |
| cea#19 | 720p50 | 43.9 | 0.06ms |
| cea#4 | 720p60 | 37.1 | 0.07ms |
| dmt#39 | 768p60 | 37 | 0.07ms |
| dmt#58 | 1050p60 | 37 | 0.07ms |
| cea#5 | 1080i60 | 53.8 | 0.08ms |
| cea#17 | 576p50 | 45.3 | 0.09ms |
| cea#18 | 576p50 | 45.2 | 0.09ms |
| cea#32 | 1080p24 | 71.5 | 0.10ms |
| cea#6 | 480i60 | 55.4 | 0.11ms |
| dmt#9 | 600p60 | 37.3 | 0.12ms |
| cea#1 | 480p60 | 38.5 | 0.14ms |
| cea#2 | 480p60 | 38.3 | 0.14ms |
| cea#3 | 480p60 | 38.4 | 0.14ms |
| cea#20 | 1080i50 | 64 | 0.14ms |
| cea#7 | 480i60 | 55.4 | 0.16ms |
| cea#31 | 1080p50 | 44 | 0.16ms |
| dmt#4 | 480p60 | 38.5 | 0.16ms |
| cea#22 | 576i50 | 65.5 | 0.23ms |
| cea#21 | 576i50 | 65.6 | 0.24ms |
| dmt#8 | 600p56 | 46.3 | 1.09ms |
| dmt#5 | 480p72 | 42.7 | 2.18ms |
| dmt#18 | 768p75 | 42.8 | 2.82ms |
| dmt#36 | 1024p75 | 41.7 | 3.01ms |
| dmt#17 | 768p70 | 42.7 | 3.65ms |
| dmt#6 | 480p75 | 42.8 | 6.70ms |
| dmt#11 | 600p75 | 42.9 | 6.85ms |
| dmt#10 | 600p72 | 42.1 | 8.25ms |
Results compared to other displays
To allow quick comparison between many displays I've summarized the results
across all the mainstream displays I've personally tested with the piLagTester Pro. Min
lag is the time to the first response, measured where the screen starts
drawing (typically, the top); real lag is the time to the full response,
measured where drawing finishes (usually the screen bottom), i.e. input lag +
scan out + response time. Numbers in red denote average values that can vary by
up to 8ms between power cycles.
This list is sorted by real lag for each display's native resolution and max
refresh rate (usually 1080p60 but some sets are 720p60, and other monitors
support > 60hz).
| Display | Year made (TV?) | Native Res | native min lag | native real lag | 480i real lag | 480p real lag | 720p real lag | 1080p real lag | native response time | native scan out |
| Dell E198FPb | 2008 | 1024p | 2.7 | 20.4 | 39.0 | 35.0 | 5.00 | 12.70 | ||
| Samsung 2494sw | 2011 | 1080p | 2.8 | 22.7 | 26.5 | 26.5 | 26.5 | 8.00 | 13.30 | |
| Vizio VO370M | 2010 | 1080p | 2.5 | 23.6 | 83.0 | 49.0 | 47.0 | 24.3 | 5.47 | 15.67 |
| LG W1953T | 2010 | 768p | 2.6 | 25.6 | 28.7 | 28.7 | 10.00 | 13.00 | ||
| Dell U2410 (game) | 2010 | 1080p | 4.0 | 26.2 | 62.2 | 28.3 | 26.5 | 26.5 | 6.00 | 16.20 |
| TCL 40S325 | 2021 | 1080p | 6.5 | 27.3 | 60.6 | 29.0 | 27.9 | 27.7 | 6.00 | 14.83 |
| TCL 49s403 | 2018 | 4k | 6.1 | 30.2 | 76.8 | 30.9 | 30.3 | 30.7 | 8.00 | 16.13 |
| Dell E2211H | 2014 | 1080p | 3.0 | 33.6 | 34.7 | 34.5 | 34.1 | 33.8 | 15.00 | 15.57 |
| Panasonic TH-58PE75U | 2008 | 720p | 28.0 | 34.0 | 34.0 | 34.0 | 34.0 | 34.0 | 6.00 | 0.00 |
| Corprit D157 (hdmi) | 2021 | 1080p | 3.1 | 34.5 | 34.9 | 34.8 | 34.6 | 33.9 | 16.25 | 15.13 |
| Samsung S27C230 | 2014 | 1080p | 2.9 | 36.0 | 36.6 | 36.1 | 36.1 | 18.10 | 14.97 | |
| Vizio E470VL (vga) | 2011 | 1080p | 22.0 | 39.0 | 39.0 | 39.0 | 39.0 | 9.00 | 8.00 | |
| Samsung LN32D403 | 2012 | 720p | 20.9 | 41.2 | 58.9 | 42.4 | 40.7 | 40.7 | 5.50 | 14.83 |
| TCL50s423 | 2021 | 4k | 14.0 | 42.0 | 75.0 | 42.0 | 42.0 | 42.0 | 13.00 | 15.00 |
| sony XBR 43X800D | 2017 | 4k | 24.5 | 44.3 | 46.5 | 46.0 | 44.6 | 44.7 | 5.00 | 14.83 |
| Element elst5016s | 2017 | 1080p | 21.4 | 45.1 | 63.5 | 46.4 | 45.1 | 45.3 | 8.00 | 15.73 |
| RCA L40FHD41 | 2010 | 1080p | 20.3 | 46.6 | 65.0 | 48.0 | 47.0 | 46.0 | 9.68 | 16.63 |
| Sony 40VL130 (game) | 2008 | 1080p | 22.8 | 47.3 | 66.3 | 49.0 | 47.3 | 47.3 | 9.08 | 15.43 |
| Philips 42PFL3603D/F7 | 2009 | 1080p | 29.0 | 50.0 | 84.0 | 50.0 | 50.0 | 50.0 | 5.00 | 16.00 |
| LG 42LC2D | 2006 | 720p | 28.3 | 50.6 | 54.6 | 50.8 | 50.4 | 6.30 | 15.95 | |
| Sony KDL-46EX400 | 2010 | 1080p | 28.0 | 52.0 | 87.0 | 52.0 | 52.0 | 52.0 | 8.00 | 16.00 |
| Toshiba 40L2200U | 2014 | 1080p | 30.0 | 56.0 | 74.0 | 56.0 | 56.0 | 56.0 | 10.00 | 16.00 |
| Vizio E261VA | 2012 | 720p | 19.3 | 59.0 | 61.1 | 60.4 | 59.2 | 58.9 | 25.00 | 14.67 |
| Samsung LN32B360 | 2010 | 720p | 37.6 | 60.0 | 62.1 | 61.8 | 60.5 | 60.1 | 8.00 | 14.40 |
| Samsung P2570HD | 2010 | 1080p | 37.0 | 62.0 | 62.0 | 62.0 | 62.0 | 62.0 | 10.00 | 15.00 |
| Sharp LC-C3234U | 2009 | 720p | 33.0 | 64.6 | 83.6 | 66.6 | 64.6 | 15.00 | 16.60 | |
| Samsung LN46B610 | 2012 | 1080p | 53.0 | 66.0 | 82.0 | 66.0 | 66.0 | 66.0 | 5.00 | 8.00 |
| LG 42PT350 | 2012 | 1080p | 63.5 | 67.7 | 85.9 | 68.9 | 67.7 | 67.7 | 4.20 | 0.00 |
| Samsung LN46C630 | 2012 | 1080p | 54.5 | 72.1 | 90.7 | 90.3 | 88.5 | 72.3 | 10.00 | 7.63 |
| SANYO DP50749 | 2010 | 720p | 67.0 | 75.0 | 103.0 | 94.0 | 79.0 | 75.0 | 7.00 | 1.00 |
| Samsung HP-T4254 | 2011 | 1080p | 69.7 | 75.7 | 94.1 | 76.0 | 75.7 | 5.00 | 1.00 | |
| LG 47LW6500-UA | 2012 | 1080p | 66.6 | 80.7 | 149.7 | 149.0 | 81.7 | 80.9 | 2.27 | 11.83 |
| Vizio xvt4735v | 2011 | 1080p | 67.6 | 88.6 | 88.8 | 89.2 | 88.6 | 88.6 | 9.00 | 12.00 |
By that standard this TV is in the middle of my results from testing TVs and computer monitors. But when compared against TVs only, it actually is quite a bit above average. And, the ranking alone hides the fact that the faster TVs are not that much faster (at most about 16ms), and the slower TVs are way, way slower (up to 40ms). This is a solid showing from a company that to-date I considered to be a purveyor of generic and middling TVs. It's deinterlacing is also pretty good, using an adaptive algorithm that produces good looking results with only 16ms of extra lag.
Conclusion
If you really care about input lag to the exclusion of all else you can definitely do better with the right TV, and of course all TVs are completely crushed by their smaller cousins the computer monitor. But this is a sold reasonable choice and I would be perfectly happy to use it as my main gaming TV.
Other models
I tested the ELST5016s, which is the 50" version. There is also the ELST4316s, which is the same specs but in a 43" size. Frankly, I expected more size options, but I didn't find anything else in the --16s line.
Comments