Mitsubishi LT-46144: input lag, deinterlacing and upscaling using the piLagTesterPRO
This 120hz 1080p TV from 2007 is a mixed bag, with variable and relatively high input lag, but fast deinterlacing and decently good upscaling While the marketing says LCD, the weight says plasma (72lbs!) as does the power consumption, of ~300 watts, and the infinite viewing angles. Given that plasma was a selling point at the time, I doubt it's really an unadvertised plasma, but it's definitely an odd duck of an LCD.
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. You get pixel perfect, non-zoomed images if you set the HDMI input name to PC. That means that 720p only fills 2/3rds of the screen. In this "mode" 480i is not supported. If you set the input to game 480i works, and that's what's reported below. Note that this has game as an input name and game as a color correction mode, but the "input name" seems to be the decider for upscaling etc.
| resolution | aliasing | cropping (side, top) |
| 480p/i | some (game) | 35,25 |
| 720p | some (game) | 40,30 |
| 1080p | none (game/pc) | 0,0 |
 |
| 480i, sharpening =0 |
The TV defaults to over sharpening in all modes, causing ugly halos. For best 480i, you need to turn sharpening down to zero, which honestly doesn't look too soft, see photo. The diagonal lines are quite jagged, I think this may be a deinterlacing artifact as it flickers a bit like it's using some kind of modified bob deinterlacing. Similarly, notice the checkerboard becomes vertical lines, another hallmark of bob deinterlacing. But I don't think it's only bob as the horizontal lines are flicker free. I suspect the deinterlacing engine is the same that Panasonic uses in their plasma TVs of the 2000s, where regions of high motion get "bob" and everywhere else gets "lace". This is about as good as you can get if you want high speed deinterlacing without constant flicker.
In higher resolution modes turn down sharpening by about half to get the best image quality.
The display has 3 HDMI, and 3 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 has a game mode, and perhaps more significantly, an option to label the HDMI input as "game". With these off input lag was higher for sure. For instance 480i was almost half as fast without game settings. I toggled all the other display quality settings as well, but did not see any further improvements, however the tests reported below with every "enhancement" set to off.
Complicating things significantly, this is yet another display that does not actually sync refresh to the input signal - instead it fills its own internal frame buffer from the video input and then draws that with a fixed additional delay that is randomly determined each time you turn on the set or switch inputs. To take one example, input lag for 1080p can vary from 43ms to 59ms. What you get seems to be entirely up to chance. It can't even properly synchronize the display of this buffer with the refresh rate, at least for 60hz signals. It rapidly drifts out of sync every 10 seconds or so and has to drop a frame. For 59.94 (aka NTSC) refresh rates it is able to keep up, however, with no drifting lag (but still random changes in lag each time you change video mode or switch inputs, or do a power reset).
I've elected to report the average lag values here, since that seems fairest, but there's no right answer; for more discussion of this issue see the above link. It's worth noting that this kind of changing lag is the hardest for your brain/hands to adapt to. I'd much rather have an extra 16ms of fixed lag added to the display than one like this where sometimes the extra lag is 0, and sometimes it's 16ms.
And there's an additional complication: this TV occasionally gets mad at
your input source and declares that it's a DVR (as in setting the name of the
input to DVR) and disabling game mode. If you think you know better, too bad.
Once it's made this choice the only way to get back to game mode is to
physically unplug the console, disable the input port, and then plug the HDMI cord back in,
at which point the TV will detect the new source and ask what type of input it
is. This happens pretty easily, just in an hour of testing I saw it happen a
couple times. Maybe it is something about the PI video signal, but I suspect not
since no other TV has behaved this way.
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.
Interestingly, this TV appears to have roughly the same lag in all modes except possibly 480i. The 1st response averages 51ms at the top of the screen, and the full response adds just 9ms to that. At the bottom of the screen everything is just 8ms slower. You'd expect it to be 16ms slower, but this is a 120hz TV and once it starts drawing it goes twice as fast as a 60hz screen. Unfortunately, you can't feed it a 120hz signal, but at least you get rid of 8ms of lag this way.
Now, what about 480i? Shouldn't I be able to say with confidence if it's the same speed as every other mode? There are several factors that make that conclusion hard to make; first is that lag varies by up to 16ms each time you change video modes, and then there is the issue that the TV randomly decides to boot you out of game mode, as described above. I had this happen to me too many times and gave up on getting enough measurements for 480i in particular. It looked like 480i might be a few ms slower than 480p based on the samples I took, but if so it's not much, at absolute worst it's 8ms slower. That's a great result, but since the TV is pretty high in lag overall there must have been time in the processing pipeline to do deinterlacing with no extra delay. Note that the Panasonic plasma's also do this, but have a much preferable 34ms of total lag (top and bottom). Again, it seems like this TV has some plasma like features....
Results compared to other displays
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); 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 480i performance, since that's the only place this set is
competitive.
This TV is only average for 480i, but it's usable, which means half of the TVs I've tested aren't, when it comes to 480i content. Combine that with the better than average upscaling, and this is a decent choice for retro gaming, if your tastes tend toward less twitchy games.
Other models (to avoid?)
I tested the LT-46144, which is the 46" version. There appears to be two versions: Based similarities in their names, I suspect that the LT-52144 would perform similarly, just with bigger pixels. However, I've made no efforts to check if their specs exactly match. Interestingly, all of the following model numbers have the same schematic layout for their motherboard: LT-40133 / LT-46133 / LT-52133 / LT-40134 / LT-46144 / LT-52144 / LT-46244 / LT-52244. Only the x144 and x244 are 120hz so those are your best bet for identical performance.
Comments