Samsung LN32D403 review: input lag, deinterlacing and upscaling using the piLagTesterPRO
This 720p TV cost about $300 in 2012. It looks good at 480i/p and does a great job of deinterlacing. It's pretty low lag, and has a fast response time, so while it's only 720p it would be a solid choice for a gamer, especially a retro gamer who can't store (or find) a CRT.
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; the results below are for the best mode I could find for each.
The 480i result looks particularly good for an LCD. Almost CRT level.
| resolution | aliasing | cropping (side, top) |
| 480p/i | almost none | about 20 off each side |
| 720p | mild, but visible, even though this is the "native" resolution of the screen! | none if you choose "screen fit". |
| 1080p | only slightly more than 720p even though this is above the "native" resolution of the screen! | none if you choose "screen fit". |
1080p is supported and looks ok. It is of course down sampled to fit 720p, but since this TV is a bit blurry and aliased at 720p, the image looks about as good in 720p as 1080p.
The actual native resolution according to the manual is 1360 x 768. The monitor does support this mode, and if you use it, it's pixel perfect with no aliasing and no blur and no cropping. Lag was the same, though.
While the TV reports that it supports 1080p at 24hz, it doesn't actually draw at that refresh rate if you select that mode, so movie playback will be jittery.
The display has 2 HDMI and 1 yPbPr input. I only tested lag on HDMI, but I did check 480i via yPbPr, and it looked just as good as 480i over 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 has a game mode, but toggling it made no difference, despite the help text on the screen "helpfully" telling me that game mode would speed things up (what scuzzy salesmanship). I toggled all the other display quality settings as well, but did not see any improvements, however the tests reported below 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.
One odd and disappointing design choice is that this TV draws from the bottom of the screen up. Thus the first pixels painted on each frame are the bottom row. This seems like a REALLY dumb choice; since the HDMI signal is sent from top to bottom this means the TV can't possibly start processing the next frame until the entire thing is sent to the TV, implying an extra 16ms of lag. At least when it starts drawing it actually scans out slightly faster than the refresh rate, taking only about 14.8ms.
Performance is actually decent, even given the upside down drawing. On average it starts drawing 21ms after receiving a signal (input lag), and pixels only take about 6ms to reach their final brightness (response time).
Interestingly, it's 480i performance is actually quite good, adding a scant 17ms to the total lag relative to 480i. That's not due to using bob deinterlacing, either. The results are good looking, sharp, and flicker free.
This TV supports the OSSC-friendly 960p resolution with minimal aliasing, so you could use the OSSC to convert 480i to 480p, and shave off 17 milliseconds of lag, if you don't mind the flicker caused by bob deinterlacing.
Results compared to other displays & conclusion
To allow quick comparison between many displays I've summarized the results
across all the 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, but for this TV the bottom); real lag is the time to the full response,
measured where drawing finishes (usually the screen bottom, here it's the top), i.e. input lag +
scan out + response time. Numbers in red denote average values that can vary by
up to 8ms between power cycles.
I've chosen to sort by 480i real lag, because I doubt anybody is buying this to play ps5 games; not at 720p certainly. While it would be a fine xbox 360 or ps3 console given it's low lag and very low response time, the real star here is the 480i performance which makes it the 3rd fastest non-plasma TV I've tested, and it does it with proper deinterlacing. That said, it's still quite a bit slower than a CRT or the Panasonic plasmas that are at the top of my list. But since CRTs are hard to find in good condition and Panasonic Plasmas are starting to disappear, this would be a pretty good choice.
Other models (same performance?)
I tested the LN32D403, which is the 32" version. The manual says it is part of "series 4" , but it's not clear there are any other models in that exact series in different sizes. According that same manual, there's a UN32D4003, which seems to be slightly lighter (LED backlight?). There's also 1080p versions listed in the manual: UN40D5003 and LN40D503. So maybe those would perform similarly, though it's a wild guess.
I'm not optimistic, however, as I've tested a couple other Samsung TVs from that era and their performance is all over the map, but generally worse. The Samsung LN32B360 is from 2010, and it's 480i performance is only 3ms slower, but in all other modes it's about 16ms slower. The Samsung LN46B610 is from 2012, and it's 480i performance averaged 24ms slower. And the Samsung LN46C630 from 2012 is a shocking 33ms slower in 480i. And last + least, the Samsung HP-T4254 from 2011 is 37ms slower in 480i. Not encouraging, to say the least. If I saw a random display for sale from the early 2010s with Samsung's name on it I'd probably avoid it unless there were documented lag test run on that specific model.
Comments