LG 48CX

kasakka

[H]ard|Gawd
Joined
Aug 25, 2008
Messages
1,446
At least on the LG C9 the "OLED Motion" setting which turns on BFI is a single toggle switch in the all too deep down submenu. I've only tried it on Netflix content and with HDR I felt there wasn't much of a difference in brightness but that's on 24 Hz content. On the CX it might depend on refresh rate.

My understanding is that the BFI on these screens is more like a rolling scan rather than full frames and that helps retain more brightness. There was a video of it somewhere but I can't find it now.
 

Pastuch

Limp Gawd
Joined
May 13, 2010
Messages
456
Dnice in AVS forum confirmed there are 3 settings for BFI: Low, medium and high. He said the high setting lowered brightness dramatically but that was almost definitely with a 60hz source. With a 120hz source like a PC the brightness reduction should be less drastic. Dnice also provided his calibration results.
 
Last edited:

Pastuch

Limp Gawd
Joined
May 13, 2010
Messages
456
Reviews have begun! https://www.flatpanelshd.com/review.php?subaction=showfull&id=1585020920&utm_source=MadMimi&utm_medium=email&utm_content=Review:+LG+CX+OLED+-+Disney++launches+in+UK,+Germany,+Spain,+more+-+In-cinema+movies+now+available+at+home+for+$20&utm_campaign=20200324_m157528335_FlatpanelsHD+-+March+24,+2020&utm_term=_0D_0AReview_3A+LG+CX+OLED

The input lag increase with BFI scares me. They didn't test with 120hz input though so it will probably cut that number in half. I don't trust their input lag results because we've seen much lower numbers demonstrated on stage. The wait for the Rtings review is on.

"Another new development this year is 'OLED Motion Pro', which is a more effective version of the BFI (Black Frame Insertion) system found in previous OLED TVs. BFI inserts black frames into the video stream to 'reset' the human eye in order to make motion appear less blurry. You can achieve similar results (plus more) by increasing the frame rate of the content (i.e. 4K120 or higher) but it is nice to have BFI as an option for lower frame rate content, too, in order to replicate more "plasma-like" motion on an OLED panel. The improved 120Hz BFI system was actually intended for LG Display's 2019 OLED panel but was pulled before release. Now it has reappeared in the 2020 OLED panel and as such it will also be available in OLED TVs from competing brands, too. FlatpanelsHD saw the 2019 implementation in action and although we did not get a change to thoroughly examine it, it seems to us that the 2020 implementation sacrifices brightness to a higher degree. In LG CX there are five levels for OLED Motion Pro (Off, Low, Medium, High and Auto). With a special test pattern we measured 'Off' to 318 nits brightness, 'Low' to 273 nits, 'Medium' to 187 nits and 'High' to 77 nits. The exact brightness values are not important so focus on the relative change in brightness: 'Low' will reduce brightness by 15%, 'Medium' by 40% and 'High' by 75%. The 'High' setting produces visible flicker and is not recommended for any type of content (it should probably be removed). 'Medium' is more effective at increasing motion resolution than 'Low' but brightness obviously takes a more significant hit. Lastly, there is an 'Auto' option that varies between 'Low' and 'Medium' but avoids 'High'. The conclusion? Well, at its two lower settings the BFI system is definitely useful now, as opposed to BFI in previous years' OLED TVs, but improved motion resolution comes at the expense of a reduction in brightness that is a little higher than we had hoped. Also note that by engaging 'OLED Motion Pro' input lag increases slightly to 22 ms. Further improvement is welcome.

Even without HDMI 2.1, LG 2020 OLED TVs are excellent performers in the area. We measured input lag to 13 ms in Game Mode - 1080p SDR, 4K SDR and 4K HDR. At this time, we cannot measure input lag in VRR and 4K120 where it will be as low as 6 ms with, according to the company. "
 
Last edited:

Lateralus

More [H]uman than Human
Joined
Aug 7, 2004
Messages
15,190
Thanks for that post Pastuch!

So, since I keep my OLED light setting so low for daily use anyway, I wonder if I can just bump it up to compensate for the brightness hit with BFI enabled and achieve the same end result in terms of nits. It will be interesting to play with this feature and see what ends up looking and performing best. Thankfully, some of that work is being done for us as we speak. :)
 

kasakka

[H]ard|Gawd
Joined
Aug 25, 2008
Messages
1,446
Shame about the input lag increase. BFI would be useful for games with fast motion and those are usually also the ones that require the lowest input lag. That said, 22 is quite alright. It's about what my Samsung KS8000 had at best and I did not really notice it as a problem even in games that require fairly precise input timing like the Souls series for example.
 

Lateralus

More [H]uman than Human
Joined
Aug 7, 2004
Messages
15,190
Shame about the input lag increase. BFI would be useful for games with fast motion and those are usually also the ones that require the lowest input lag. That said, 22 is quite alright. It's about what my Samsung KS8000 had at best and I did not really notice it as a problem even in games that require fairly precise input timing like the Souls series for example.
Yeah, according to rtings my B7 is 21.x so I would be fine with that. Plus, there's always a chance that it can be improved via firmware updates. It'll probably get better as they continue to develop and tweak the feature. There was a firmware update for the 2016 sets that dramatically reduced input lag from what I remember, though I don't know if they've made similar updates since then or just tweaked it year by year as new sets are released.
 

Pastuch

Limp Gawd
Joined
May 13, 2010
Messages
456
Shame about the input lag increase. BFI would be useful for games with fast motion and those are usually also the ones that require the lowest input lag. That said, 22 is quite alright. It's about what my Samsung KS8000 had at best and I did not really notice it as a problem even in games that require fairly precise input timing like the Souls series for example.
If you read the rest of the article they say the input lag should about 7ms with a 120hz input. What I'm searching for is the following: Windows 10 + 2000 series Nvidia card + 1440P @ 120hz or 4k @ 120hz with BFI enabled on High setting. What is the input lag in this configuration?
 

Pastuch

Limp Gawd
Joined
May 13, 2010
Messages
456
Thanks for that post Pastuch!

So, since I keep my OLED light setting so low for daily use anyway, I wonder if I can just bump it up to compensate for the brightness hit with BFI enabled and achieve the same end result in terms of nits. It will be interesting to play with this feature and see what ends up looking and performing best. Thankfully, some of that work is being done for us as we speak. :)
That's exactly what I was planning on doing. My C7 is at 40 OLED light, with BFI on ultra I suspect 100 will be required.
 

elvn

2[H]4U
Joined
May 5, 2006
Messages
3,462
You can achieve similar results (plus more) by increasing the frame rate of the content (i.e. 4K120 or higher) but it is nice to have BFI as an option for lower frame rate content, too, in order to replicate more "plasma-like" motion on an OLED panel
Nice to have another tool in the toolbox but as they said, it has diminishing returns vs high fps+highHz gameplay. The less percent of BFI and dimming you engage the gap narrows between that compared to the blur reduction benefits of raw fps+HighHz at full color brightnesses/highlight's ranges ~ vibrancy capabilities. Of course it wouldn't work correctly with HDR content either as that is based on a range of absolute values not something you dial higher and lower like SDR brightness controls, and OLED is already more limited in what color volumes it can do at what % of screen, especially for HDR material.

The nits they quoted were the average screen brightness and did not take into consideration that if ABL kicks in, the screen drops to 140nits before the BFI cuts it by up to 25%, 50%, or 75%. That is, unless you otherwise keep your screen brightness at around 250 if it's anything like the C9:
The C9 has a new Peak Brightness setting, which adjusts how the ABL performs. Setting this to 'Off' results in most scenes being displayed at around 303 cd/m², unless the entire screen is bright, in which case the luminosity drops to around 139 cd/m². Increasing this setting to 'Low', 'Med', or 'High' increases the peak brightness of small highlights. If ABL bothers you, setting the contrast to '80' and setting Peak Brightness to 'Off' essentially disables ABL, but the peak brightness is quite a bit lower (246-258 cd/m² in all scenes).
-------------------------------------------------------------
FlatPanelsHD Review: LG CX OLED 24 Mar 2020 | Rasmus Larsen
With a special test pattern we measured 'Off' to 318 nits brightness, 'Low' to 273 nits, 'Medium' to 187 nits and 'High' to 77 nits. The exact brightness values are not important so focus on the relative change in brightness: 'Low' will reduce brightness by 15%, 'Medium' by 40% and 'High' by 75%. The 'High' setting produces visible flicker and is not recommended for any type of content (it should probably be removed). 'Medium' is more effective at increasing motion resolution than 'Low' but brightness obviously takes a more significant hit.
--------------------------------------------------------------

So assuming you rule out 75% outright, the 318 nit they are using as a basis with BFI off (and assumingly with ABL enabled~not avoided) gets
..cut 40% at medium BFI setting, which does mathematically result in 190.8 nit so is around what they measured at 187nit.
..on Low BFI 25% the math result is at 238.5nit which is lower than the 273nit they quoted as being measured so perhaps it's not as aggressive as a 25% brightness reduction, perhaps interpolation is on too which would makes sense if the input lag is higher. I'm not sure if the tradeoffs would be worth it runing low BFI compared rather than no BFI with a raw fps hovering around a 120fps+120Hz graph considering the PWM effect on your eyes and the input lag increase from BFI.
..
..Neither of those quotes show the after ABL values. ABL on regular SDR brightness cuts it down to around 140nit on a C9 before BFI brightness reduction is considered so if you used medium 40% you could end up with a result of 84nit* of scene color brightness/highlights/detail in areas of color post ABL.
...
..If you run at around 250nit color brightness/color detail ceiling in SDR with the contrast at 80 and peak brightness to "Off" in order to avoid ABL kicking in you end up with medium BFI reducing 40% of 250 -> probably having a result of around 140nit of color detail/brightness (which is coincidentally the same as what you get in the previous (default) ABL scenario after ABL kicks in but before BFI brightness reduction considerations if enabled).

--------------------------------------------------------------------
So would you want to run SDR on the screen
..normally at ~ 320nit of color detail but seeing 140nit ABL kick in intermittently
..normally + medium BFI --> 190nit of color detail (320nit minus 40%) until ABL kicks in then down to as low as *84nit (40% strobe effect subtracted from 140nit ABL) . 40% blur reduction (compared to 60fps+60hz but not 120fps+120hz), PWM, input lag.

..ABL-avoiding 80contrast, peak brightness"off" ~> 250nit seeing no ABL triggering
..ABL-avoiding 80 contrast, peak brightness "off" ~> BFI active (250 nit minus 40%) lowering it to 140nit average screen brightness (and detail limit in fields of color) throughout. 40% blur reduction compared to 60fps+60Hz, perhaps 20% vs 120fps+120hz.. +PWM, input lag increase.

*That is because as I understand it - BFI's dimming effect is on the way our eyes perceive the screen meaning the LG ABL burn in safety tech would not realize the after-BFI-effecive-brightness so would still be operating normally as if BFI wasn't a factor.
 
Last edited:

elvn

2[H]4U
Joined
May 5, 2006
Messages
3,462
Why can’t they measure VRR now? Hdmi thing?
What do you mean by measure? HDMI 2.1 can do 4k 120hz at 4:4:4 chroma, HDMI 2.0 can't. You can run 1440p at 120hz on HDMI 2.0 though with full chroma, and I think you can run 4k 120hz at 4:2:0, 8bit, on the LG CX from a hdmi 2.0 gpu for now. Lower chroma affects text rendering and for color detail purists, color detail in fields of color. It can look similar to upscaling a lower resolution which looks kind of "muddy".

Nvidia g-sync compatible VRR screens typically start syncing at 40fps-Hz on the low end so going beneath that frame rate - in your frame rate graph at all, not just your frame rate average - isn't recommended. That means if a particular game's frame rate fluctuated +/- 30fps from it's average, in that particular case you might have to dial a demanding game's graphics settings in so that it runs 70fps-Hz average or better in order to stay above 40fps-Hz on the low end. In order to get appreciable benefits in motion clarity (blur reduction) and motion definition (more unique pages in a flip book flipping faster) out of higher hz, you prob should be shooting for 100fps-Hz average or better anyway on a 120hz+ monitor.

Hope this helps, I wasn't exactly sure what you were asking.
 
Last edited:

IdiotInCharge

[H]F Junkie
Joined
Jun 13, 2003
Messages
13,529
Nvidia g-sync compatible VRR screens typically start syncing at 40fps-Hz on the low end so going beneath that frame rate - in your frame rate graph at all, not just your frame rate average - isn't recommended.
To my understanding, this is still better than non-VRR, or at least non-VRR with V-Sync. What I'm not sure about is how quickly a follow-up frame could be transmitted, or if tearing could be allowed to limit the lag. What do you think?
 

elvn

2[H]4U
Joined
May 5, 2006
Messages
3,462
It's way better than non VRR.. I didn't mean to imply that at all. VRR itself is great.

What I was saying is that you already have a high Hz monitor that gets appreciable blur reduction and motion definition increases starting around 100fps average, in some cases equating to a 70 - 100 - 130 graph, (120 average or higher even better).... so if running high enough frame rate ranges to get appreciable benefit out of the higher hz you would already be running your frame rate roller coaster graph +/- your average high enough to avoid sinking below 40fps-Hz. That is, imo, you should be running frame rates higher than having to worry about sub 40fps already in order to get benefit out of 100 to 120hz (or 115hz capped).

Some VR headsets use a type of interpolation that cuts anything below 90fps down to 45fps and doubles it. I'd rather not suffer 45fps in motion definition (sloshy staying on same freeze-frame longer), but it would keep the blur from sinking back to full smearing and would also keep the frame rate higher than the VRR low end cutoff. Neither nvidia or LG does this afaik, I just thought I'd mention it since it exists as a tech.

I think in the future some much better interpolation without considerable input lag or artifacts will be needed to fill high refresh rates, even if you are at 80 - 100fps solid someday you could benefit from interpolation x3, x4 , x10 for a 240hz, 360hz, 480hz, 1000hz monitor in the future.

----------

I'm not sure about the next frame time. You could probably suffer the tearing but from what I read of early freesync it would actual do a stutter when going below the VRR minimum. I'm not sure if that is still the case.

Personally I would just keep the frame rate higher by dialing in the graphics settings to 100fps-Hz average or better and use a beefy gpu like a 3080ti.
If not I'd say you are better off running a lower rez, even on this monitor you could do 1:1 ultrawide rez or a 1440p rez letterbox framed all around with ultra oled black bars "off" .. and still have full sized monitor. Running graphs that span into 40fps-Hz and lower - that low you aren't getting appreciable benefits from the high Hz in the first place. To avoid going off the VRR range's rails I'd say at least 75fps-Hz average for games that swing +/- 30fps from the average. It would depend on the game, some have less variance. In that example you'd have 45fps-Hz <---- 75fps-Hz -----> 105 fps/Hz. That is still low for keeping in the 100 - 115fps-Hz range for more of the graph (to reduce blur and increase motion definition considerably) but it would keep your graph from hitting or going below 40fps-hz VRR limit for the most part.
 

MrDeas

n00b
Joined
May 11, 2012
Messages
20
The flatpanel review said “ At this time, we cannot measure input lag in VRR and 4K120 where it will be as low as 6 ms with, according to the company.”

just curious why that isn’t possible. 6ms would be amazing
 

Pastuch

Limp Gawd
Joined
May 13, 2010
Messages
456
The flatpanel review said “ At this time, we cannot measure input lag in VRR and 4K120 where it will be as low as 6 ms with, according to the company.”

just curious why that isn’t possible. 6ms would be amazing
6ms is very possible based on what we saw at CES. They can't measure it because TFTcentral probably doesn't have the new version of the Leo Bodnar input lag tester that does 4k. The original Leo Bodnar was 1080p only.
 

kasakka

[H]ard|Gawd
Joined
Aug 25, 2008
Messages
1,446
LG in some CES interview already stated they are not interested in chasing lower input lag when it is already in sub 1 frame area for supported refresh rates. You have to remember that OLED also does not have response time related input lag like LCDs do.

So both input lag and response time on these TVs are issues you don't have to consider at all.
 

sharknice

[H]ard|Gawd
Joined
Nov 12, 2012
Messages
1,930
LG in some CES interview already stated they are not interested in chasing lower input lag when it is already in sub 1 frame area for supported refresh rates. You have to remember that OLED also does not have response time related input lag like LCDs do.

So both input lag and response time on these TVs are issues you don't have to consider at all.
Yep, and it's not only sub 1 frame, it's basically as low as it can go at those framerates. There really isn't a significant amount they can improve it.

Also like you said there is no pixel response time like with LCDs. So the input lag measured is for the actual final image.
Some of the pixels don't even finish transitioning before the next frame on the fastest ".5 ms" response time LCDs, so you may have 6 ms input lag on an LCD, but that's the measured input lag for displaying partially transitioned pixels.
 

elvn

2[H]4U
Joined
May 5, 2006
Messages
3,462
I'll be happy with ~6ms on a monitor with such great OLED per pixel visuals and other features (hdmi 2.1, VRR, 120hz 4k 4:4:4, etc.) while also not being limited to a slim ~13' tall or so belt model .

120hz is 8.3ms per frame and that is only if, even if ignoring a few frame rate "potholes" in your graph, you are getting 120fps solid all of the time as a common frame rate low.... which most people aren't in any kind of demanding games and settings near or at 4k resolution.

-------------------------------------------------------------------------------------------------------

120fps-Hz (solid, not average):
.................8.3ms per frame
.................50% reduction in sample and hold blur to a "soften blur" compared to baseline 60fps-Hz smearing blur (~ 8ms persistence vs 16ms persistence)
.................double the motion definition 2:1 ("# of unique pages~action states in a flip book flipping twice as fast"), incl. viewport movement, path articulation
................. 10 frames shown to every 5 at 60fps-Hz solid
................. 10 frames to every 8.3 shown at 100fps-Hz solide
................. 10 frames to every 10Hz refresh on a 120hz screen at 120fps solid (1:1)


115fps-Hz rate cap (solid frame rate, not average)
................. 8.7ms per frame
................ ~ 45% +?? reduction in sample in hold blur (8.7ms image persistence)

100fps-Hz (solid, not average):
................. 10ms per frame
................. ~40% reduction in sample and hold blur (10ms image persistence)
................. 10 frames to every 6 shown at 60fps-Hz solid (1.6:1)
................. 10 frames to ever 10 shown at 100fps-Hz solid
................. 10 frames to every 12 shown at 120fps-hz solid (1.2:1, 20fps-hz short of 120fps-hz)

75fps-Hz (solid, not average):
................. 13.3ms per frame
................. 15 ??? % reduction in sample and hold blur? (13.3ms persistence) .. still smearing blur, slightly moderated
................. 10 frames to every 8 shown at 60fps-Hz solid (1.25:1)
................. 10 frames to every 13 shown at 100fps-Hz solid (1.3:1)
................. 10 frames to every 16 shown at 120fps-Hz solid (1.6:1)

60fps-Hz (solid, not average):
................. 16.6ms per frame
................. baseline "100%" smearing blur "outside of the lines" of individual objects as well as viewport movement of the whole game world at speed
................. 1:1 motion definition vs 60fps-Hz baseline
................. 10 frames at 60fps-Hz to every 16.6 frames shown at 100fps-Hz
.................. [6 frames to every 10 shown at 100fps-Hz solid (5:3)]
..................10 frames at 60fps-Hz to every 20 frames shown at 120fps-Hz
.................. [5 frames to every 10 shown at 120fps-Hz solid (2:1, half)]

40fps-Hz (solid, not average):
................. 25ms per frame
................. baseline smearing blur (or worse) page-y choppy motion , sluggish
................. 10 frames to every 15 shown at 60fps-Hz solid (1.5:1)
................. 10 frames to every 25 shown at 100fps-Hz solid (2.5:1)
................. 10 frames to every 30 shown at 120fps-Hz solid (3:1)

30fps-Hz (solid, not average):
................. 33.3ms per frame
................. smearing blur, page-y/choppy animations ~ "molasses" movement and FoV movement ~ motion definition
................. 10 frames to every 20 shown at 60fps-Hz solid
................. 10 frames to every 30 shown at 100fps-Hz solid
................. 10 frames to every 40 shown at 120fps-Hz solid
 
Last edited:

elvn

2[H]4U
Joined
May 5, 2006
Messages
3,462
I know you could always run a 2560x1440 rez upscaled even on the C9's - but I wonder if you could run a ultrawide rez or other letterboxed rez on 1:1 on a LG CX at 4:2:2 chroma 120hz off of a 1000 series or 2000 series nvidia gpu (with VRR) as long as it was within the bandwidth limit of hdmi 2.0. I think at least one person in this thread mentioned they already have a 65" LG CX so maybe someone could try it out. Just curious.
 

sharknice

[H]ard|Gawd
Joined
Nov 12, 2012
Messages
1,930
I know you could always run a 2560x1440 rez upscaled even on the C9's - but I wonder if you could run a ultrawide rez or other letterboxed rez on 1:1 on a LG CX at 4:2:2 chroma 120hz off of a 1000 series or 2000 series nvidia gpu (with VRR) as long as it was within the bandwidth limit of hdmi 2.0. I think at least one person in this thread mentioned they already have a 65" LG CX so maybe someone could try it out. Just curious.
Unfortunately only the 2000 series can do VRR over HDMI. I have a 1080ti and I want to wait for a 3000 series card before I upgrade so I probably won't buy a 48" CX until then.
 

Pastuch

Limp Gawd
Joined
May 13, 2010
Messages
456
Unfortunately only the 2000 series can do VRR over HDMI. I have a 1080ti and I want to wait for a 3000 series card before I upgrade so I probably won't buy a 48" CX until then.
I'm in the same boat with the 1080ti. I haven't been that thrilled with it honestly, I hate seeing the 5700xt top our card when it clearly should not but Nvidia just optimizes for current gen. My R9 290 got faster as it aged, the 1080 ti is the opposite.
 

kasakka

[H]ard|Gawd
Joined
Aug 25, 2008
Messages
1,446
Starting prices in Finland: 1599€ (including 24% VAT)
Available 12.5.2020
View attachment 235223
Dang, that's lower than I expected. Very tempted to order one now.

I know you could always run a 2560x1440 rez upscaled even on the C9's - but I wonder if you could run a ultrawide rez or other letterboxed rez on 1:1 on a LG CX at 4:2:2 chroma 120hz off of a 1000 series or 2000 series nvidia gpu (with VRR) as long as it was within the bandwidth limit of hdmi 2.0. I think at least one person in this thread mentioned they already have a 65" LG CX so maybe someone could try it out. Just curious.
I'd say no, but maybe at 4:2:0. The CX supports 4K 120 Hz 8-bit 4:2:0 over HDMI 2.0 so that would be the max before HDMI 2.1 GPUs are available. Basically the display itself does not support anything but select 16:9 resolutions on its own scaler, but with GPU scaling you can have ultrawide resolutions that are then upscaled to 4K so the TV knows no better. This means the bandwidth limits apply. It might be a bit tricky to get working right and you would probably be best setting your desktop resolution to the custom one before launching games. DisplayFusion is great for changing the resolution with a hotkey but it doesn't allow setting 4:2:0 and I have not found any other app that would do that either. I wish Nvidia had display presets in its own functionality.
 
Last edited:

Skott

2[H]4U
Joined
Aug 12, 2006
Messages
4,053
I'm in the same boat with the 1080ti. I haven't been that thrilled with it honestly, I hate seeing the 5700xt top our card when it clearly should not but Nvidia just optimizes for current gen. My R9 290 got faster as it aged, the 1080 ti is the opposite.
I'm waiting for the 3080Ti as well before upgrading. Not sure whether I'll get the CX 48" or one of these new 38/49 inch monitors we're seeing now. Either way I want a 3080Ti first.
 

MaZa

2[H]4U
Joined
Sep 21, 2008
Messages
2,855
Starting prices in Finland: 1599€ (including 24% VAT)
Available 12.5.2020
View attachment 235223
Damn, that is expensive, especially considering we can get last year models for less than that, at 55" size mind you.

But as of this moment my plans for desktop OLED are on hold because I got a good deal on Samsung S34J550 ultrawide, which was overclockable to 96hz. Slow as molasses pixel response time but picture quality is good and this was one of the few PC VA panel monitors with contrast ratio above 4000:1. Maybe 48" OLED will find its way on my desktop but now the actual living room TV is a priority, that will be changed into OLED the next Black friday.
 

Sancus

Gawd
Joined
Jun 1, 2013
Messages
998
Damn, that is expensive, especially considering we can get last year models for less than that, at 55" size mind you.
That's totally normal though. MSRP of 2020 models is always designed to be well above the sale price of previous years models, and then gradually come down over ~6 mo until rock bottom around holiday season sales. It's pretty clearly intentional price discrimination("If you want the newest and best right away, pay up, otherwise wait."). And probably supply chain management as well.
 

SixFootDuo

Supreme [H]ardness
Joined
Oct 5, 2004
Messages
5,580
Shame about the input lag increase. BFI would be useful for games with fast motion and those are usually also the ones that require the lowest input lag. That said, 22 is quite alright. It's about what my Samsung KS8000 had at best and I did not really notice it as a problem even in games that require fairly precise input timing like the Souls series for example.

Yeah, 22 is ....pretty bad. My 2018 Samsung NU8000 is around 8.9ms?

Question, can this BFI be turned off?

I just need the OLED, 120hz at 1440p out of this monitor which I planned on buying until now and super low input lag .... I need to read a bit more.

But I need super low input lag period ... this is a deal breaker if I can't turn this off. Thought I read the input lag was around 12 or 13. Maybe I just need to stick with my amazing Samsung.
 

Nordlands

n00b
Joined
Apr 3, 2020
Messages
1
Starting prices in Finland: 1599€ (including 24% VAT)
Available 12.5.2020
Found it at the same chain website in Norway, and have put in my order to receive it by mid-May.

A bit more reasonable here at 15990 NOK = 1397 EUR at present exchange (incl. 25% VAT). It seems like LG set the prices in different currencies before the most recent fluctuations, as in January 15990 NOK was worth more than 1599 EUR.
 

Pastuch

Limp Gawd
Joined
May 13, 2010
Messages
456
Yeah, 22 is ....pretty bad. My 2018 Samsung NU8000 is around 8.9ms?

Question, can this BFI be turned off?

I just need the OLED, 120hz at 1440p out of this monitor which I planned on buying until now and super low input lag .... I need to read a bit more.

But I need super low input lag period ... this is a deal breaker if I can't turn this off. Thought I read the input lag was around 12 or 13. Maybe I just need to stick with my amazing Samsung.
BFI can be disabled but here's the thing, none of the input lag tests that have been done so far have been done at native resolution at 120hz so we still have no idea what the actual input lag numbers for PC use will be. Wait for the Rtings review.

We suspect that this display can do around 5 to 6ms input lag at 4k at 120hz because that was what LG claimed at CES and they have been honest about that number for the last few years.
 

elvn

2[H]4U
Joined
May 5, 2006
Messages
3,462
BFI can be disabled but here's the thing, none of the input lag tests that have been done so far have been done at native resolution at 120hz so we still have no idea what the actual input lag numbers for PC use will be. Wait for the Rtings review.

We suspect that this display can do around 5 to 6ms input lag at 4k at 120hz because that was what LG claimed at CES and they have been honest about that number for the last few years.
^ This.

5 to 6ms is well under the fluctuating wave of frame rates most people are going to be rolling at 4k resolution when using VRR and a moderate frame rate average where the lower ranges of their actual frame rates aren't above 100fps-Hz.... (10ms per frame).

I'll be happy with ~6ms on a monitor with such great OLED per pixel visuals and other features (hdmi 2.1, VRR, 120hz 4k 4:4:4, etc.) while also not being limited to a slim ~13' tall or so belt model .

120hz is 8.3ms per frame and that is only if, even if ignoring a few frame rate "potholes" in your graph, you are getting 120fps solid all of the time as a common frame rate low.... which most people aren't in any kind of demanding games and settings near or at 4k resolution.

-------------------------------------------------------------------------------------------------------

120fps-Hz (solid, not average):
.................8.3ms per frame
.................50% reduction in sample and hold blur to a "soften blur" compared to baseline 60fps-Hz smearing blur (~ 8ms persistence vs 16ms persistence)
.................double the motion definition 2:1 ("# of unique pages~action states in a flip book flipping twice as fast"), incl. viewport movement, path articulation
................. 10 frames shown to every 5 at 60fps-Hz solid
................. 10 frames to every 8.3 shown at 100fps-Hz solide
................. 10 frames to every 10Hz refresh on a 120hz screen at 120fps solid (1:1)


115fps-Hz rate cap (solid frame rate, not average)
................. 8.7ms per frame
................ ~ 45% +?? reduction in sample in hold blur (8.7ms image persistence)

100fps-Hz (solid, not average):
................. 10ms per frame
................. ~40% reduction in sample and hold blur (10ms image persistence)
................. 10 frames to every 6 shown at 60fps-Hz solid (1.6:1)
................. 10 frames to ever 10 shown at 100fps-Hz solid
................. 10 frames to every 12 shown at 120fps-hz solid (1.2:1, 20fps-hz short of 120fps-hz)

75fps-Hz (solid, not average):
................. 13.3ms per frame
................. 15 ??? % reduction in sample and hold blur? (13.3ms persistence) .. still smearing blur, slightly moderated
................. 10 frames to every 8 shown at 60fps-Hz solid (1.25:1)
................. 10 frames to every 13 shown at 100fps-Hz solid (1.3:1)
................. 10 frames to every 16 shown at 120fps-Hz solid (1.6:1)

60fps-Hz (solid, not average):
................. 16.6ms per frame
................. baseline "100%" smearing blur "outside of the lines" of individual objects as well as viewport movement of the whole game world at speed
................. 1:1 motion definition vs 60fps-Hz baseline
................. 10 frames at 60fps-Hz to every 16.6 frames shown at 100fps-Hz
.................. [6 frames to every 10 shown at 100fps-Hz solid (5:3)]
..................10 frames at 60fps-Hz to every 20 frames shown at 120fps-Hz
.................. [5 frames to every 10 shown at 120fps-Hz solid (2:1, half)]

40fps-Hz (solid, not average):
................. 25ms per frame
................. baseline smearing blur (or worse) page-y choppy motion , sluggish
................. 10 frames to every 15 shown at 60fps-Hz solid (1.5:1)
................. 10 frames to every 25 shown at 100fps-Hz solid (2.5:1)
................. 10 frames to every 30 shown at 120fps-Hz solid (3:1)

30fps-Hz (solid, not average):
................. 33.3ms per frame
................. smearing blur, page-y/choppy animations ~ "molasses" movement and FoV movement ~ motion definition
................. 10 frames to every 20 shown at 60fps-Hz solid
................. 10 frames to every 30 shown at 100fps-Hz solid
................. 10 frames to every 40 shown at 120fps-Hz solid
(source of image below: back2gaming.com)

dota2-2160_18136_image001.png

-----------------------

No vrr on 1000's ack.... Never had to worry about that with g-sync monitors. I agree with most of the 3080ti mentions -- I'll upgrade from 1080ti to a 3080Ti whenever that is. Not worth upgrading gpu until I can get 120hz 4k 4:4:4 chroma with VRR on a die shrink.
 
Last edited:
Top