@NikTheSHNIK Does that mean a Saturday delivery or do I need to start thinking about asking to work from home on Monday to accept delivery? If it's not a Saturday delivery can it be upgraded?
Last edited:
Poll: ** LG 27GP950 AVAILABLE TO PRE-ORDER NOW - LIMITED AVAILABILITY **
- Thread starter NikTheSHNIK
- Start date
@NikTheSHNIK Does that mean a Saturday delivery or do I need to start thinking about asking to work from home on Monday to accept delivery? If it's not a Saturday delivery can it be upgraded?
Associate
- Joined
- 29 Mar 2021
- Posts
- 37
@shankly1985 and @Junk
I think I just had my first blackscreen...(had my monitor since last Monday)
I'm running a dual monitor setup with a USB-C to my one LG and a HDMI to the GP950 from my laptop.
Everything, like both monitors went black and then only the GP950 reconnected leaving my other LG screen black.
Is this what happens with the notorious black screen issue?
I might add that I also lost wifi from my laptop for a moment when it happened during my teams meetings - was very odd.
I think I just had my first blackscreen...(had my monitor since last Monday)
I'm running a dual monitor setup with a USB-C to my one LG and a HDMI to the GP950 from my laptop.
Everything, like both monitors went black and then only the GP950 reconnected leaving my other LG screen black.
Is this what happens with the notorious black screen issue?
I might add that I also lost wifi from my laptop for a moment when it happened during my teams meetings - was very odd.
No, this sounds like a different issue.
The normal black screen issue only seems to affect the LG monitor and it returns image within a second.
Associate
- Joined
- 29 Mar 2021
- Posts
- 37
The GP950 did comeback online after a couple or seconds or so.
Hopefully it isn't the infamous black screen issue.
Thanks for your reply though!
Soldato
Yeah like SADS said your issue seems like something else but is kinda close.
Soldato
Check of this post from Blurbusters - https://forums.blurbusters.com/viewtopic.php?f=17&t=8283
<Technical>
Frame presentation APIs such as Present() instantly triggers refresh cycles during VRR. That allows framerate and Hz to be in perfect sync to each other -- and eliminate framdrop-style stutters -- and de-stuttering smooth framerate changes -- but creates problem when framerates exits the VRR range (frametimes longer than min-Hz, for example).
Software bugs with VRR can cause blackouts on many FreeSync monitors, since it is kind of a real-time operation for software, and drivers that violate VRR ranges is sometimes more common than it should be.
Native G-SYNC chipped monitors will often be able to refresh unattended if drivers don't begin sending a new refresh cycles, but FreeSync doesn't do this -- the graphics driver is responsible for beginning to transmit a new refresh cycle on time. But oftentimes, software is not microsecond precise.
Ideally, manufacturers need at least fraction of a Hz of safety headroom below min Hz, to allow software imprecisions. For example, panels should support ~35-240Hz for a software-based 48Hz-240Hz VRR range. Unfortunately not enough manufacturers include enough min-Hz VRR headroom, and sometimes.... software-driven refresh cycle imprecision occurs.
Drivers (including Microsoft non-realtime behaviors) accidentally keeping the monitor unrefreshed too long -- is a major cause of FreeSync monitors going blank below min Hz.
VRR blackout issues are usually caused by software tolerance issues. A well-tuned Windows system may have no problems, while a flawed driver installation (on a monitor without min-Hz tolerance) may fail to keep up the refresh cycles.
Besides this, behaviours such as flicker (refresh-cycle decay, inversion algorithms, etc), also make it favourable to raise the min-Hz, to force the drivers to refresh more frequently.
LFC is simply driver-initiated automatic repeat-refreshes, but it must be done before monitor refresh cycles times out (and goes blank automatically). Even if sometimes minor stutters occur from new-refreshes colliding with monitor-busy-repeat-refreshing events. (But at 180Hz, a monitor is only busy for 1/180sec executing a repeat-refresh. And since it's chance whether a new frame may occur beyond halftime or before halftime of that refresh cycle -- so halve that. 0.5/180sec = 1/360sec average stutter caused by an LFC miss. 1/360sec stutters are very hard to see at 30fps or 40fps anyway, so a higher min-Hz is perfectly fine for wide-VRR ranges (3:1 ratio or bigger). So you're trading away those annoying blackouts for potentially un-seeable added stutter at low frame rates.
This was a bigger problem with 144Hz FreeSync monitors, where a 0.5/144sec (3.5ms) stutter may become visible during 55fps operation (18ms), since 3.5ms:18ms is an approximately 20% stutter (e.g. a frame step between adjacent frames suddenly increases by 20% in movement distance).
But as VRR ranges get wider, 240Hz, your 55fps material (21ms) versus the halftime of 240Hz repeat-refreshes (2ms) means stutter deviations are smaller -- less than a 10% stutter. This starts to become impossible to see for most people, and still looks like perfect 55Hz VSYNC ON with the stutter deviation so small that the low-framerate stutter becomes hidden in the plain display motion blur of the lowness of frame rate (as you've noticed, the lower the frame rate on VRR, the more motion blur, until it falls into the stutter-visibility region of the stutter-to-blur continuum).
TL;DR: LFC has far less penalty when VRR range is extremely wide, and LFC (hidden repeat refreshes) can be superior in image quality to extremely low native refresh rate because of various reasons. You you get the advantages of a higher min Hz (less flicker, less GtG decay, more consistent color equalling colors of max-Hz, avoid varying-ghosting effects, fewer inversion-artifacts such as scrolling chessboard effects, etc). Since picture quality at different refresh rates may be slightly different.
When LFC is absent of refresh-cycle collision events, it looks indistinguishable to well-done low-Hz (since repeat-refreshes ideally behave like no-operations, as long as they don't delay the new frame). At some point, the higher-min-Hz pros starts to exceed the LFC-cons when "busy-doing-refresh-cycle" events are very brief (1/240sec or faster).
So that's why Blur Busters now feels higher min-Hz on wide-VRR-ranges produces superior VRR experiences, visually -- not just for blackout reasons.
It is /sometimes/ ALSO the manufacturer's fault, occasionally
That said, manufacturers should include lower undocumented min Hz in more FreeSync panels just as safety headroom (even as low as 30-40Hz). This safety headroom is just accommodation for software performance -- graphics drivers too slow to refresh frequently enough risks these blackouts you saw. There should always be more undocumented min-Hz headroom below the EDID-advertised VRR range, but unfortunately many panels cheap out and have no safety margin -- creating lots of blackout complaints.
</Technical>
<Technical>
Frame presentation APIs such as Present() instantly triggers refresh cycles during VRR. That allows framerate and Hz to be in perfect sync to each other -- and eliminate framdrop-style stutters -- and de-stuttering smooth framerate changes -- but creates problem when framerates exits the VRR range (frametimes longer than min-Hz, for example).
Software bugs with VRR can cause blackouts on many FreeSync monitors, since it is kind of a real-time operation for software, and drivers that violate VRR ranges is sometimes more common than it should be.
Native G-SYNC chipped monitors will often be able to refresh unattended if drivers don't begin sending a new refresh cycles, but FreeSync doesn't do this -- the graphics driver is responsible for beginning to transmit a new refresh cycle on time. But oftentimes, software is not microsecond precise.
Ideally, manufacturers need at least fraction of a Hz of safety headroom below min Hz, to allow software imprecisions. For example, panels should support ~35-240Hz for a software-based 48Hz-240Hz VRR range. Unfortunately not enough manufacturers include enough min-Hz VRR headroom, and sometimes.... software-driven refresh cycle imprecision occurs.
Drivers (including Microsoft non-realtime behaviors) accidentally keeping the monitor unrefreshed too long -- is a major cause of FreeSync monitors going blank below min Hz.
VRR blackout issues are usually caused by software tolerance issues. A well-tuned Windows system may have no problems, while a flawed driver installation (on a monitor without min-Hz tolerance) may fail to keep up the refresh cycles.
Besides this, behaviours such as flicker (refresh-cycle decay, inversion algorithms, etc), also make it favourable to raise the min-Hz, to force the drivers to refresh more frequently.
LFC is simply driver-initiated automatic repeat-refreshes, but it must be done before monitor refresh cycles times out (and goes blank automatically). Even if sometimes minor stutters occur from new-refreshes colliding with monitor-busy-repeat-refreshing events. (But at 180Hz, a monitor is only busy for 1/180sec executing a repeat-refresh. And since it's chance whether a new frame may occur beyond halftime or before halftime of that refresh cycle -- so halve that. 0.5/180sec = 1/360sec average stutter caused by an LFC miss. 1/360sec stutters are very hard to see at 30fps or 40fps anyway, so a higher min-Hz is perfectly fine for wide-VRR ranges (3:1 ratio or bigger). So you're trading away those annoying blackouts for potentially un-seeable added stutter at low frame rates.
This was a bigger problem with 144Hz FreeSync monitors, where a 0.5/144sec (3.5ms) stutter may become visible during 55fps operation (18ms), since 3.5ms:18ms is an approximately 20% stutter (e.g. a frame step between adjacent frames suddenly increases by 20% in movement distance).
But as VRR ranges get wider, 240Hz, your 55fps material (21ms) versus the halftime of 240Hz repeat-refreshes (2ms) means stutter deviations are smaller -- less than a 10% stutter. This starts to become impossible to see for most people, and still looks like perfect 55Hz VSYNC ON with the stutter deviation so small that the low-framerate stutter becomes hidden in the plain display motion blur of the lowness of frame rate (as you've noticed, the lower the frame rate on VRR, the more motion blur, until it falls into the stutter-visibility region of the stutter-to-blur continuum).
TL;DR: LFC has far less penalty when VRR range is extremely wide, and LFC (hidden repeat refreshes) can be superior in image quality to extremely low native refresh rate because of various reasons. You you get the advantages of a higher min Hz (less flicker, less GtG decay, more consistent color equalling colors of max-Hz, avoid varying-ghosting effects, fewer inversion-artifacts such as scrolling chessboard effects, etc). Since picture quality at different refresh rates may be slightly different.
When LFC is absent of refresh-cycle collision events, it looks indistinguishable to well-done low-Hz (since repeat-refreshes ideally behave like no-operations, as long as they don't delay the new frame). At some point, the higher-min-Hz pros starts to exceed the LFC-cons when "busy-doing-refresh-cycle" events are very brief (1/240sec or faster).
So that's why Blur Busters now feels higher min-Hz on wide-VRR-ranges produces superior VRR experiences, visually -- not just for blackout reasons.
It is /sometimes/ ALSO the manufacturer's fault, occasionally
That said, manufacturers should include lower undocumented min Hz in more FreeSync panels just as safety headroom (even as low as 30-40Hz). This safety headroom is just accommodation for software performance -- graphics drivers too slow to refresh frequently enough risks these blackouts you saw. There should always be more undocumented min-Hz headroom below the EDID-advertised VRR range, but unfortunately many panels cheap out and have no safety margin -- creating lots of blackout complaints.
</Technical>
Soldato
So from what I understand at 144hz or 160hz the min frame rate for these monitors should be 55hz
fwiw since swapping out my GP950 for the M28U with everything remaining the same (games, cable, 144hz etc.), no black screens
Soldato
Thanks for the update, definitely something fishy going on with the GP950.
No worries and agreed regarding the GP950. It'd be great if the 2.0 cable idea pans out but otherwise there is a firmware issue to address for LG.
Associate
- Joined
- 29 Mar 2021
- Posts
- 37
Can I please pick your brains to see if I am doing anything incorrectly.
As mentioned yesterday, I am running a dual monitor setup, LG 27UN83A & 27GP950, connected to my laptop. (WFH)
The 27UN83A is connected via USB-C and the 27GP950 is connected via HDMI (cable I received with monitor).
The 27UN83A runs at 60hz, but the 27GP950 is limited to 30hz for some reason.
I've done some googling but haven't found a way to change the 950's hz higher than 30hz.
Is it because my laptop hardware is unable to output two screens at 4k 60hz?
Or is there some setting I am missing.
The 950 easily reaches 120hz on my next-gen consoles...so not sure why I can't change it on the Laptop.
Any tips or help would be appreciated.
As mentioned yesterday, I am running a dual monitor setup, LG 27UN83A & 27GP950, connected to my laptop. (WFH)
The 27UN83A is connected via USB-C and the 27GP950 is connected via HDMI (cable I received with monitor).
The 27UN83A runs at 60hz, but the 27GP950 is limited to 30hz for some reason.
I've done some googling but haven't found a way to change the 950's hz higher than 30hz.
Is it because my laptop hardware is unable to output two screens at 4k 60hz?
Or is there some setting I am missing.
The 950 easily reaches 120hz on my next-gen consoles...so not sure why I can't change it on the Laptop.
Any tips or help would be appreciated.
I haven't seen any of these black screen issues, I have heard with the GN950 that people struggled to hit 160hz with the included cable?
FWIW my GPU is a 3090. I have previously noted when connected to my TV by a 10m HDMI cable, that where I did get occasional dropouts at 4k60 4:4:4 on the old 1080ti, these went away with the 3090, which obviously has a more modern HDMI controller supporting 2.1 rather than 2.0 (although the TV is only 2.0) so maybe GPU comes in to it as well?
FWIW my GPU is a 3090. I have previously noted when connected to my TV by a 10m HDMI cable, that where I did get occasional dropouts at 4k60 4:4:4 on the old 1080ti, these went away with the 3090, which obviously has a more modern HDMI controller supporting 2.1 rather than 2.0 (although the TV is only 2.0) so maybe GPU comes in to it as well?
Soldato
It definitely seems like you have hardware limitations happening with the laptop.
HDMI 2.1 is required for 4k 120hz
Associate
- Joined
- 29 Mar 2021
- Posts
- 37
Thanks for the reply @shankly1985
Bit odd, as the monitor I had previously (LG 27GL850) was fine outputting at 60hz on the laptop - be that at 1440p though.
Not the end of the world - as I use it for WFH - but you do notice the difference when moving your mouse from the 60hz display over to the 30hz display.
you probably only have a HDMI 1.4 port,, you need 2.0 for 4k 60hz
Associate
- Joined
- 29 Mar 2021
- Posts
- 37
good breakdown of HDMI standards here https://www.tripplite.com/products/hdmi-cable-types
Associate
Anyone excited for the long awaited Friday? Let's hope no more delays appear and we receive the dispatch email today. Nothing on my end at the moment. Thanks
Associate
Well, I'd like to update on the black screens, eventually my setup with two different wall sockets after playing for two weeks without black screens started giving black screens but extremely randomly, like 1 in first 5 minutes, than 4 hours without, and then two in a row... I do not know what it is. I should point out that FPS is capped at 155 and it never falls below 100, so I doubt that it is about a lower VRR threshold. But maybe there are some instantaneous and unnoticeable FPS dips...
Also, I've figured out the best way to play CS GO on this monitor. Make sure enhanced sync (or an Nvidia similar tech) is turned off, set fps_max to 320 (double of the 160Hz) and just play with raw frames input without any syncing. Input lag is as low as technically possible. Tearing is not noticeable unless you strafe with face up to a wall with vertical textures and then you must look closely to bottom of the screen where there will be some micro tearing. As an experiment 60Hz at 120 fps tears like a hell, but 160hz at 320 fps is absolutely tearing free for normal gameplay (of course if your PC can maintain stable 320 fps). Also I should note that amd antilag is on. And I never ever had a black screen in such mode which proves that the black screens on 27GP950 are all about VRR.
Also, I've figured out the best way to play CS GO on this monitor. Make sure enhanced sync (or an Nvidia similar tech) is turned off, set fps_max to 320 (double of the 160Hz) and just play with raw frames input without any syncing. Input lag is as low as technically possible. Tearing is not noticeable unless you strafe with face up to a wall with vertical textures and then you must look closely to bottom of the screen where there will be some micro tearing. As an experiment 60Hz at 120 fps tears like a hell, but 160hz at 320 fps is absolutely tearing free for normal gameplay (of course if your PC can maintain stable 320 fps). Also I should note that amd antilag is on. And I never ever had a black screen in such mode which proves that the black screens on 27GP950 are all about VRR.
