NVIDIA G-Sync Pulsar monitor - Asus ROG Strix XG27AQNGV

[kyube]

A few people in this thread measured it at 60Hz compared to the XG2431 and DyAc:

- DyAc adds about 5 ms
- PureXP adds about 8 ms
- ULMB2 adds about 13 ms

This seems to be about in line as the LG CX, which also adds around a frame (14 ms) of input lag with BFI enabled.

So from a purely competitive standpoint, the XG2431 do seem to perform better. At least compared to my CRT@60, I definitely noticed the increased latency of ULMB2@60.

At least on the NVIDIA forum, they may seem to try to lower the 60Hz latency, so hopefully they too can reduce to 5-8 ms.

TFTCentral claims that it adds 2,5 – 3,5ms processing latency, so I don't know where the 13 ms value has been derived from.

[tsarri]

Maybe a dumb question, but apart from <10% duty cycle (which gets a bit too dark for my taste, 50 nits I can still live with) and freely adjustable Hz strobing, does the XG2431 has any real advantage to the pulsar monitors?

I thought that maybe because of QFT, 60Hz might has less input lag than ULMB2? Given that many console only support 1080p60 anyways, that would be cool, but I guess QFT doesn’t work with consoles anyways. But at least for PC? Subjectively, 60Hz ULMB2 on the pulsar does seem quite high of an input lag, especially compared to my CRT. I knew it will be higher and even if it’s a bit lower with QFT, trading it for 1080p compared to 1440p is also not the best.

Edit: I saw in this thread that the XG2431 indeed seems to have lower input lag. Too bad you can’t properly use QFT/LVT with consoles.

It seems like, now that I tried the XG27AQNGV 60Hz ULMB2, I might simply keep the Sony W900 for 60Hz gaming, as that should have even lower lag than the XG2431 and a better brightness/clarity ratio.

—

Other topic, but I might return the XG27AQNGV after all. All in all, the motion clarity is indeed very nice, but in practice and with how I play first- and third-person games (very high mouse/controller sensitivity), the difference is not as big as I thought. I mean on paper it is, but I don’t seem to eye track things in motion but rather turn very fast to have a „still image“ again (where 165Hz OLED is already good enough).

So comparatively, I look at far more frames which are already clear enough with 165Hz OLED than frames that are in motion. And for all frames, especially „static“, good HDR with OLED (Samsung S95F, 4000 nits peak, 400 nits full) is simply far better than IPS SDR, by a huge margin.

I may simply be too far gone in the OLED rabbit hole already too break out of it. But of course that’s SDR vs HDR. Maybe with SDR content, I prefer the Pulsar, but even then OLED will look better in „static“ frames.

Given that CRTs look also really good for static frames compared to IPS, I may stick to the W900 60Hz, FW900 >60Hz SDR and S95F HDR combo…

So annoying that the only way to get a display that is good at everything equally, is a 4k 3000Hz or so OLED with proper CRT beam simulation. But I doubt we will be there any time soon. Until then, you need to make trade offs.

And then there’s VR, where we „need“ proper motion clarity and extreme resolutions, so I don’t see proper HDR here any time soon as well. Damn physics!

I think people in general underestimate how hard it is to go from OLED HDR back to IPS SDR. The reality is you have to make compromises either way. Depends what you value more.

[mango87]

...

Thanks for this. I was able to find the original post by Angel Soler. From an absolute standpoint, the input lag among XL2411P, XG2431, and the Pulsar monitors:

• DyAc: 15 ms
  PureXP: 16.4 ms
  ULMB2: 25.1 ms

According to the original post, the absolute minimum input lag at 60hz is about 8.33 ms. This would mean ULMB2 adds about 1 frame of input lag @ 60hz (25.1-8.33). Thinking about this, this makes sense as ULMB2 likely pulses near the end of a frame, and there is no rolling scan. Despite this, I find the input lag to be acceptable, and I'm not a competitive gamer. I will have to test further on HDMI.

If only Nvidia could implement something like QFT and/or rolling scan for 60hz, maybe they can lower the input lag. It should become the perfect monitor for retro and competitive gaming!

[whitespider999]

This monitor is exceptional in ways that are seemingly subtle, but huge in reality. Ive been playing high skill, high motion games, and even with my 500hz oled, i would get overwhelmed and exhausted by too much onscreen. It's not that i couldn't play these games, its just that doing things like keeping things in sights, reading text, being aware of all enemies, knowing what's an explosion, what's someone moving to the left, what's a fire sprite, etc. All of those subtle things we just take for granted i would extend a microscopic amount of effort into tracking. Those 0.01% of efforts stacked up, and suddenly in 20 minutes you're like. Eh, i've had enough of this game.

But it was never a lack of interest. It was essentially you being tired of having to wait for text to settle before you can read it, predicting where an enemy is rather than observing him.

This pulsar monitor allows me to effortlessly track details without extending effort. And it means that you always feel comfortable engaging with content. Also, I really need to point out. This display is significantly less tiring than normal monitors for the eyes. Not just because of the motion, but rather because of the way the rolling scan tech works. Things look visually "Deeper" than a normal ips, and they don't start glaring even after extended use. The entire panel feels incredibly consistent. It's far easier for me to look at than an oled screen for general desktop use. The brightness is bright without feeling like it's creating glare.

Also, with this level of motion clarity you start to notice the strangest things. You can identify artwork of the models, you see specular detail on everything, you can see individual micro text on billboards, sprites and particles are so clear that you just observe them individually like they are assets. Because of the motion things just seem deeper. I can parse so much more detail than a 4k oled at 240hz.

Pulsar does make you notice issues however, stutters hit you harder, even though I feel like gsync is doing an exceptional job here at reducing those stutters. When frame pacing is off, it becomes intolerable. And because of the huge leap in motion clarity, you actually loose the illusion of smoothness. A lot of the "enjoyable" smoothness i used to get from huge amounts of fps/hz combined with something like lossless scaling (so i could cap a stable base), would feel incredibly smooth. A large percentage of that feeling was sample and hold blurring merging the frames. Pulsar can still create a huge amount of smoothness, its just not the blurring kind as a result it's more subtle and refined.

Frame pacing flaws on pulsar is really an issue, since the clarity of the display shows flaws in pacing immediately. If there's a wonky MFG implementation for example, you'll see it.You also tend to see the flaws in frame generation far more, like you can see the artifacts from frames being generated and it can often reduce the motion resolution clarity. However, if you just let games run against a cpu or gpu wall with traditional gsync, it's not always a win. Nor is capping pulsar fps at the lowest value you can get.

Playing in gsync mode without framgen does suit a large % of games, this is in large part because the motion clarity does not seemingly change that much if the framerate if variable. So you could be playing a roguelike first person game and there's a ton of entropy and explosions taking it down to 190fps from the usual 290, and unlike normal monitors, blur does not start creeping in during that, so you don't feel afraid to lose fps. Things just feel sturdy.

I would say frame generation is still a net bigger win because if its dynamic (MFG), it tends to take the edge of the limits of where the system would hit the cpu. And if you use lossless (you should be using a static base fps cap for that), then having that extra smoothness is affords is, at least to me, worth the slight degradation to motion clarity. Also, despite pulsar doing great with a changing fps, a stable locked fps will always be a bit clearer and a bit smoother, so the lossless scaling dynamic framegen mode bringing you close to you max refresh is probably the best of both worlds.

The monitor also has a richy punchy feel to it without colours feeling like they are skewed to get that punch. Things feel grounded, accurate, reds and blues and greens all feel grounded without off colourings. Some ips monitors can have a sickly, desaturated, and inaccurate feel to them. This is the opposite of that. Colours arent quite as exceptional at the very best calibrated QD oleds, but it's in that ballpark. Rather than outside of it.


This pulsar monitor is essentially the best multi tool for clarity out there. If one method does not produce the results you want, you can rapidly switch to another. And even pulsar not producing good results still often looks worlds better than competing tech producing its best results in that same area.


- Last thought for the day. I definitely do notice motion clarity and smoothness in framerates over 120 all the way to 360 specifically on this panel in pulsar mode. So the upcoming 5k model at 144hz is not for me. Intentionally dropping down to 144hz was a tagibly worse experience, despite being very clear.

[hammelgammler]

Thanks for this. I was able to find the original post by Angel Soler. From an absolute standpoint, the input lag among XL2411P, XG2431, and the Pulsar monitors:

• DyAc: 15 ms
  PureXP: 16.4 ms
  ULMB2: 25.1 ms

According to the original post, the absolute minimum input lag at 60hz is about 8.33 ms. This would mean ULMB2 adds about 1 frame of input lag @ 60hz (25.1-8.33). Thinking about this, this makes sense as ULMB2 likely pulses near the end of a frame, and there is no rolling scan. Despite this, I find the input lag to be acceptable, and I'm not a competitive gamer. I will have to test further on HDMI.

If only Nvidia could implement something like QFT and/or rolling scan for 60hz, maybe they can lower the input lag. It should become the perfect monitor for retro and competitive gaming!

Yeah it’s unfortunate. Maybe the Chief has the knowledge, but what would be the theoretical best input lag you could achieve with this monitor and still keep the same motion clarity? Because wouldn’t rolling scan be a bit worse?

Going from ~8.5 to ~25ms is just a bit too much of an increase for my liking, especially since I have a good CRT. Anyways, I’m glad that I got to experience such awesome 60Hz/120Hz motion clarity!

[wiseude]

I think people in general underestimate how hard it is to go from OLED HDR back to IPS SDR. The reality is you have to make compromises either way. Depends what you value more.

If only they could fix OLED VRR flicker :/
I wonder if its even a thing they can actually do.I've been hearing about this specific issue since OLED/VRR became a thing.

I think I saw a thread talking about possible ways to address the issue and the only way to counter/minimise VRR flicker is trough an actual g-sync module.

I don't know how people can play games without VRR.

[hammelgammler]

I think people in general underestimate how hard it is to go from OLED HDR back to IPS SDR. The reality is you have to make compromises either way. Depends what you value more.

If only they could fix OLED VRR flicker :/
I wonder if its even a thing they can actually do.I've been hearing about this specific issue since OLED/VRR became a thing.

I think I saw a thread talking about possible ways to address the issue and the only way to counter/minimise VRR flicker is trough an actual g-sync module.

I don't know how people can play games without VRR.

To be fair, at least from my experience with Pulsar and OLED, for both you need a stable frametime, but it’s even worse with Pulsar if not.

So yes, VRR flicker is annoying, but only in dark scenes. Pulsar hitches on the other hand are always annoying and to me, even more annoying.

This means in both cases you need to limit your FPS ideally to your 1%/0.1% lows for the best experience and even without both issues, I can only recommend doing that. A stable frametime is awesome. But of course it goes a bit against what VRR actually tried to solve.

[whitespider999]

I think people in general underestimate how hard it is to go from OLED HDR back to IPS SDR. The reality is you have to make compromises either way. Depends what you value more.

If only they could fix OLED VRR flicker :/
I wonder if its even a thing they can actually do.I've been hearing about this specific issue since OLED/VRR became a thing.

I think I saw a thread talking about possible ways to address the issue and the only way to counter/minimise VRR flicker is trough an actual g-sync module.

I don't know how people can play games without VRR.

To be fair, at least from my experience with Pulsar and OLED, for both you need a stable frametime, but it’s even worse with Pulsar if not.

So yes, VRR flicker is annoying, but only in dark scenes. Pulsar hitches on the other hand are always annoying and to me, even more annoying.

This means in both cases you need to limit your FPS ideally to your 1%/0.1% lows for the best experience and even without both issues, I can only recommend doing that. A stable frametime is awesome. But of course it goes a bit against what VRR actually tried to solve.

Frametime spikes are the thing I notice most of all. Because of that i've developed a lot of ways to minimize them without investing too much time and effort. That came in handy when i immediately noticed that stuttering is more abrupt and obvious on the pulsar. I switched between pulsar mode and ulmb 2, and the former radically reduces their impact, more I would say, than normal "gsync certified" oled screens ive used. But the massive increase in clairty means my sensory system "halts" when i see irregularities. It's not gamebreaking for me, but the stronger sample and hold blurring on oled does mask them better.

Thankfully im very ok with using lossless scaling + fps caps in the % of games with noticeable spikes. it does not eliminate them, because a traversal stutter does not care about min framerate. But ultimately, only 1-2% of games ive tested so far have not had any solution. And me finding that solution is generally about 20 seconds, since ive developed keen senses for knowing exactly what the problem is and what kind of targets i should be aiming for.