NVIDIA G-Sync Pulsar monitor - Asus ROG Strix XG27AQNGV

[BLooDS_inc]

and i find another bug with pulsar, when u go into full sleep , when power indicator goes from white to orange, after sleep is off, the pulsar always turns off, trying this several times and this goes all the time.

Already 2 bugs on MSI's version - pulsar goes off after sleep mode and unenable pulse width setting for 60hz ULMB

That is not MSI exclusive. I have the same thing with the ASUS model. I mentioned this some weeks ago. Whenever it goes to sleep for more than just a few minutes and i come back, its just ULMB2. And I cannot change this by software, not by OSD and not by nvcp like switching gsync off and back on.... it remains ULMB2 on osd. Which is weird because how would a ULMB2 mode look like with gsync turned on?!

The only thing getting pulsar back is restarting the display or rebooting windows. Annoying, but for me this has the lowest priority to be fixed

Im just turn pulsar again in osd

[Lunatic]

Hi there!

I recently bought the Acer Predator XB273U F5 (immediately updated to firmware version 1.1.4) and although I’m generally very happy with it and the motion clarity it provides, I wanted to share some feedback: viewtopic.php?p=124317

[edgintheledge]

-Regarding the double image behavior observed on G-SYNC Pulsar displays when operating within the 100–180 FPS range, a fix has been implemented and will be included in an upcoming G-SYNC Pulsar firmware update version 1.1.6.

Well now that we know there's another update coming these pursuits might be redundant but I'll leave them here anyways. Though I wonder if this means this range will look even better than what I photographed here?


Things to know about these shots

• This is 180fps capped only ingame (Engine is unreal 5 if it matters)
• 1/45sec shutter used
• My phone was held vertically during these shots, unsure if relevant but this changes the orientation of rolling shutter to be horizontal across the screen instead of vertical.
• Without a sync track the best verification I can do is with the grid of stationary lines in the background. The vertical lines are only single lines, but each subsequent strobe captured in the 1/45sec window creates that grouping of 3. The consistent spacing suggests good horizontal tracking (I thought it would be groupings of 4 since 180/4=45? Maybe that logic is for testufo which isn't vrr or smoothfrog which doesn't have frametime inconsistencies that an actual game can? Or my phone is lying about the shutter speed).
• The character runs across the monitor in ~0.9 seconds timed with a stopwatch including initial dash (Characters don't instantly run full speed in this game) this means the pursuit speed is around 3000px/sec in these photos depending on how accurate my stopwatch timing was and initial dash speed/accel ingame.
• Unlike many fighting games, the character animations in this game are decoupled from the game's 60hz logic and interpolated to the framerate. 1/45sec shutter is capturing multiple screen refreshes/frames of 180 fps animation causing image duplication-esque effects in some photos. The text above the character moves horizontally more like a smoothfrog test scenario and is more indicative of what I see.

I have attached a shot where the strobe looked best with basically no preceeding image, one where a faint vertical line in the background might be a compensation pulse captured, and one where the preceeding image was most noticeable. I don't think I'm seeing the same strobe behavior as Mawi and Brownvim have mentioned.

[brownvim]

You can recreate the ghosting behaviour doing the below, this is what I posted for Nvidia to recreate it, realised I didnt post it on this forum.

I tried recording the issue in RE9 but it's almost impossible to capture properly on video.

Instead, I recreated the exact ghosting behaviour using SmoothFrog v0.5p (simple UI version here):
https://www.aperturegrille.com/software ... it0.5p.zip

How to test it:

1. Open the program and press F to bring up the long FPS slider.
2. Gently wobble the slider left and right to simulate fluctuating frame rates (like real VRR behaviour).

On the new firmware ~180 FPS to 360 FPS it stays perfectly clean with no ghosting. However, once I drop it to around 120 FPS (or let it fluctuate in that range), I immediately see a sharp ghosted/double image for a short period as soon as the frame rate gets adjusted with the slider.

Before the new firmware: Ghosting only appeared at 90 FPS and below. The 90–360 FPS range was clean.
After the new firmware: Ghosting now starts appearing from around 90-180 FPS.

Maybe its harder to tell in shooters? I don't really play them. In games I play its noticeable. Its not like a contant ghosted image, more like when frame rate shifts up and down. I dont think the compensation pulse should cause a ghosted image either? its just so we dont perceive flicker right?

[mawi]

Maybe its harder to tell in shooters? I don't really play them. In games I play its noticeable. Its not like a contant ghosted image, more like when frame rate shifts up and down. I dont think the compensation pulse should cause a ghosted image either? its just so we dont perceive flicker right?

To me this "ghost" image on RE9 also felt like it has a different color... some kind of purple side ghost of the actual image. I dont know if this is just some kind of human eye thing or if this can really be captured by a camera. Especially in the bright white hallway of RE9 this is very noticable.
This was different on the old firmware below 90 fps with the double imaging which was introduced due to double Hz (OSD showed 180 Hz when game had 90 fps etc). And as mentioned, no ghosting at all above 90 fps.

This is not the case here now. Its some kind of ghosting with 100-160fps internally and the same Hz on OSD. No double hz etc.

[edgintheledge]

You can recreate the ghosting behaviour doing the below, this is what I posted for Nvidia to recreate it, realised I didnt post it on this forum.

I see what you mean now cause I can recreate that same behavior in smoothfrog. Maybe I just got lucky that my ingame framerate is stable enough to not trigger the issue, but it definitely isn't external framecap stable. When I ingame capped to 180, my avg is 180 and my 0.1% low is 142.

In shooters it probably depends more on if you fix your gaze on a crosshair (stroboscopic/hz limited) vs eyetrack a target (persistence limited). The game I took pursuits of is a 2.5D platform fighter so there is alot of fast eyetracked scenarios.

As for the compensation pulse, I thought it was to mitigate how noticeable changing the strobe persistence is at the cost of introducing some extra double image looking artifacts, and that the monitor weakens this pulse as framerate stabilizes. Someone more knowledgeable can correct me here but I don't believe there's anyway to not introduce double imaging artifacts by pulsing more than once per frame since your eyes will continue smooth tracking and place the extra pulse of the same frame in a slightly different place in your vision.

How I understand the compensation pulse behavior appears to line up with what I captured in the one photo. There's one extra strobe that was weaker indicated by the faint 4th line while every other group of lines only has 3. My phone's shutter was rolling horizontally across the screen during the photo, so that could explain why it showed up that way in the photo.

To me this "ghost" image on RE9 also felt like it has a different color... some kind of purple side ghost of the actual image. I dont know if this is just some kind of human eye thing or if this can really be captured by a camera. Especially in the bright white hallway of RE9 this is very noticable.

I can't say I've ever seen this in synthetic testing or real gaming scenarios.

[Discorz]

Maybe you find Smooth Frog's 3D Rotator test helpful. Press "3" or select "Smooth 3D Rotator" in menu. Once inside press "H" to toggle between Auto Rotation Test/3D Rotator. Adjust the fps variable range, period, and rotation speed and off u go.

No Sync Track here, but at least it does the auto sweep for u. Would be nice if this was a feature for Frog Pursuit test as well.

[bahamutfan]

Is the 48Hz update that was announced still in the works?

[RealNC]

Is the 48Hz update that was announced still in the works?

I believe they said no, they didn't consider the flicker below 75Hz in pulsar mode acceptable.

[liquidshadowfox]

I think the problem with flicker below 75 hz has to do with them tying the flicker to changes in fps. When it's around the flicker fusion threshold (around 72) most people might not notice the change in flicker but the moment you go below that value, variable flicker causes immense strain on your eyes vs a consistent 60 hz flicker where the scaler knows exactly when to pulse the backlight.

That's why I think the suggestion I made of allowing users to define per game/software profiles for monitor display mode (Pulsar/Gsync/ULMB2/Fixed mode) should be implemented alongside Display refresh rate mode (setting max display refresh rate to say 60 hz or 120 hz for ULMB2) without having to mess around with a physical monitor joystick and OSD or relying on some 3rd party scripts most users won't trust or understand. Cause realistically, the people (me included) who want 60 hz strobing to be included in pulsar want it so they can enjoy games that are locked to 60 fps (legacy content, elden ring or any souls series game, smash bros, etc) where it might be better to just facilitate an "nvidia only way" that automatically switches to ULMB 2 mode whenever such a program is open on the windows desktop. I also think IF they implement this, they should also allow ULMB 2 to TEMPORARILY stop flickering the moment someone tries to alt +tab out possibly because playing elden ring at 60 hz ULMB 2 and then having a white background web browser behind it when you alt + tab out is a flash bang to the face (and makes the flicker unbearable).

I don't disagree with nvidia not wanting to decrease pulsar minimum fps, I just think they should
1. Implement the profile switching idea because it leads to a much more satisfying customer experience
2. Improve Gsync pulsar at higher fps (200 fps - 360 fps doesn't look as good as lower refreshes, not sure if it's because light diffusion is too slow OR because the frame timing window is so small, that 10% (or whatever they have it set to) deviation causes it to fire compensation pulses left and right that ruins the picture quality)
3. Give us the option to use ULMB 2 proper with option of selecting rolling scan (for less eye strain and less crosstalk) or global scan (more motion clarity but more crosstalk) because Asus PG27AQN looks sharper to my eye than Gsync pulsar at 120, 240 and 360 hz ULMB (yes the PG27AQN has more crosstalk and red phophor but I expected we'd at least get more clarity at such high refresh rates).
4. Investigate what's going on with reflex + Gsync pulsar because historically I have only known Vsync + Gsync + Reflex on/Ultra low latency mode was the way to go but I get significantly better motion clarity using RTSS frontedge sync in most of the games I've tested and only a couple of games play nice with reflex + pulsar because their frames seem rock solid.
5. Allow Gsync pulsar to work at lower refresh display states so folks who own a 40 series or older GPU don't run into the DSC black screen flicker issue that's been reported on here. They invested into the nvidia ecosystem, they should also be able to have a good product experience given this product actually does more for them than anyone else (since I find that NOT using frame gen with pulsar looks better than using frame gen, the clarity of this screen makes frame gen artifacts really show because there's little to no motion blur to hide them).