Blur Busters Forums

Posted: **07 Aug 2026, 13:33**

kyube wrote: ↑
Yesterday, 08:54
New data as of today

Link: https://www.nvidia.com/en-us/geforce/fo ... opicPage=3

Nvidia has finalized the table of how ULMB2 @ 240Hz is.... and I'm already disappointed with the targetted pulse on time @ 120cd/m²...
Let's hope that 360Hz slope will be more aggressive...

Here's the interpolated slope behavior of how each Pulse Width setting scales in terms of Effective Motion Clarity (Nvidia term for the inverse of pulse on time AKA frequency-domain value of pulse-on time):

Notice how 120 Hz (8,333 ms) @ PW=25 (0,65 ms) @ 120 cd/m² has a SHORTER pulse on time than 240 Hz (4,1667 ms) @ PW=25 (0,83 ms) @ 120 cd/m²
This means that, if the claim of 1800–2000 cd/m² voltage boosting behavior is true:
PW=25 @ 120 Hz targets a ~7,8% pulse width
To be able to hit that at 120 cd/m², this implies a 1538 cd/m² voltage boosting behavior (which is roughly in-line with reported data from users here)

PW=25 @ 240 Hz targets a ~19,9199% pulse width
To be able to hit that at 120 cd/m², this implies a 602,409 cd/m² voltage boosting behavior

Why is there such a large negative discrepancy between 120Hz & 240Hz in terms of voltage boosting?
I don't know... This is ULMB1 levels of voltage boosting, akin to the PG259QN & S2417DG... Very mediocre target...

This implies that Nvidia is still being tame in pushing lower duty cycle targets on the 240Hz & 360Hz refresh rate settings.
Much performance is left under the table...

This implies that, on PW=10:
https://testufo.com/ghosting#pps=1440 would be 1px of blur
https://testufo.com/ghosting#pps=2880 would be 2px of blur

Which already invalidates it from competing with a CRT, like other models within the same category (QHD 240–360 Hz)
Don't forget that CRTs can do +3000 px/s with ≤1px of blur, which is the benchmark for any backlight strobing implementation on the market

They answered my question about it, pretty much they said if they go any lower it'll introduce double images to the top and bottom. Overall I think it's a good upgrade either way where you get more motion clarity than before without giving up crosstalk performance.

Posted: **07 Aug 2026, 13:38**

liquidshadowfox wrote: ↑
Yesterday, 13:33
They answered my question about it, pretty much they said if they go any lower it'll introduce double images to the top and bottom. Overall I think it's a good upgrade either way where you get more motion clarity than before without giving up crosstalk performance.

For some reason, I am unable to post a question on Nvidia forums from my new account.
I don't believe that explanation in the slightest.
There's a ~1000 cd/m² voltage boosting behavior difference between 120Hz ULMB2 PW=25 & 240Hz ULMB2 at PW=25 according to their own data.
The value of 0,83ms pulse on time should be achieved at ~320 cd/m², not at 120cd/m², if they were using the same amount of voltage boosting.
I don't know what they've done...

Could you ask them the following questions:
1.) Why is there such a large backlight voltage boosting discrepancy between the 120 Hz ULMB2 & 240 Hz ULMB2 mode at the same PW=25 setting? There's a large ~1000cd/m² difference based on your tabular data, by doing simple arithmetics.

Namely:
(0,65/8,333) * 100 = ~7,8 % duty cycle

Let's rely on a range of 1500–2000 cd/m². Let x be a value in between that range.
1600 * 0,078 = 124,8 cd/m² (very close to Nvidia's claim of 120cd/m²)

Now, for 240Hz:
(0,83 / 4,16667) * 100 = ~19,99% duty cycle

Let Z be the value you're looking for:
Z * 0,1999 = 120 cd/m² → Z = 120 / 0,1999 → Z = 600 cd/m²
How? Why?
If you were using the same voltage boosting behavior, we'd be looking at ~320cd/m² instead?

PS (don't add this part): They're hitting the same pulse on time interval as the KTC M27P6 does when its MPRT mode is enabled... on the 120-160Hz setting.
It does that at the full panel brightness (300–400cd/m²). On a much lower refresh rate setting. On a much slower G2G RT panel (leading to increased crosstalk)

2.) Could you give us tabular data on how 360 Hz ULMB2 behaves at PW=100,,50,25,10?

3.) Could you give us tabular data on how Pulsar's pulse on time ("MPRT") scales depending on the refresh rate setting chosen? For example: 100, 150, 200, 250, 300 Hz

liquidshadowfox wrote: ↑
Yesterday, 13:33
Overall I think it's a good upgrade either way where you get more motion clarity than before without giving up crosstalk performance.

Just to touch upon this aspect a bit:
The PG27AQN, which was the predecessor to these panels, having similar or identical G2G RT capabilities as these new Pulsar M270DAN10.6 panels, is able to achieve this level of clarity without visible crosstalk in this region at 2400 px/s with ≤1px pixels of blur using the (supposedly) same global flash technique.

I'll quote the image here:
2400px/s PW=40 240Hz PG27AQN

So once again, I am left perplexed as to what Nvidia is doing in the background with these new panels.
It was the perfect time to reimplement this level of effective motion clarity, as this time there was no KSF limitation this time around.
Of course, I am relying on visual approximations of effective motion clarity based on the pursuit photograph, as we don't have data on PG27AQN's actual pulse on time at these settings.

To summarise:
• Do I think that the new ULMB2 240Hz update will be a major upgrade to the current implementation?
Yes, cause the current 240Hz & 360Hz ULMB2 handling is bad.

• Do I think that it's competing with competitors or competing with itself (vs its 120Hz ULMB2 mode) AKA do I think that there's more left on the table for Nvidia to squeeze out of these monitors?
Yes. Especially in the brightness department, at the very least, if they're this adamant on avoiding crosstalk behavior.

Posted: **07 Aug 2026, 14:06**

kyube wrote: ↑
Yesterday, 13:38

liquidshadowfox wrote: ↑
Yesterday, 13:33
They answered my question about it, pretty much they said if they go any lower it'll introduce double images to the top and bottom. Overall I think it's a good upgrade either way where you get more motion clarity than before without giving up crosstalk performance.
For some reason, I am unable to post a question on Nvidia forums from my new account.
I don't believe that explanation in the slightest.
There's a ~1000 cd/m² voltage boosting behavior difference between 120Hz ULMB2 PW=25 & 240Hz ULMB2 at PW=25 according to their own data.
The value of 0,83ms pulse on time should be achieved at ~320 cd/m², not at 120cd/m², if they were using the same amount of voltage boosting.
I don't know what they've done...

Could you ask them the following questions:
1.) Why is there such a large backlight voltage boosting discrepancy between the 120 Hz ULMB2 & 240 Hz ULMB2 mode at the same PW=25 setting? There's a large ~1000cd/m² difference based on your tabular data, by doing simple arithmetics.

Namely:
(0,65/8,333) * 100 = ~7,8 % duty cycle

Let's rely on a range of 1500–2000 cd/m². Let x be a value in between that range.
1600 * 0,078 = 124,8 cd/m² (very close to Nvidia's claim of 120cd/m²)

Now, for 240Hz:
(0,83 / 4,16667) * 100 = ~19,99% duty cycle

Let Z be the value you're looking for:
Z * 0,1999 = 120 cd/m² → Z = 120 / 0,1999 → Z = 600 cd/m²
How? Why?
If you were using the same voltage boosting behavior, we'd be looking at ~320cd/m² instead?

PS (don't add this part): They're hitting the same pulse on time interval as the KTC M27P6 does when its MPRT mode is enabled... on the 120-160Hz setting.
It does that at the full panel brightness (300–400cd/m²). On a much lower refresh rate setting. On a much slower G2G RT panel (leading to increased crosstalk)

2.) Could you give us tabular data on how 360 Hz ULMB2 behaves at PW=100,,50,25,10?

3.) Could you give us tabular data on how Pulsar's pulse on time ("MPRT") scales depending on the refresh rate setting chosen? For example: 100, 150, 200, 250, 300 Hz

liquidshadowfox wrote: ↑
Yesterday, 13:33
Overall I think it's a good upgrade either way where you get more motion clarity than before without giving up crosstalk performance.
Just to touch upon this aspect a bit:
The PG27AQN, which was the predecessor to these panels, having similar or identical G2G RT capabilities as these new Pulsar M270DAN10.6 panels, is able to achieve this level of clarity without visible crosstalk in this region at 2400 px/s with ≤1px pixels of blur using the (supposedly) same global flash technique.

I'll quote the image here:
2400px/s PW=40 240Hz PG27AQN

So once again, I am left perplexed as to what Nvidia is doing in the background with these new panels.
It was the perfect time to reimplement this level of effective motion clarity, as this time there was no KSF limitation this time around.
Of course, I am relying on visual approximations of effective motion clarity based on the pursuit photograph, as we don't have data on PG27AQN's actual pulse on time at these settings.

To summarise:
• Do I think that the new ULMB2 240Hz update will be a major upgrade to the current implementation?
Yes, cause the current 240Hz & 360Hz ULMB2 handling is bad.

• Do I think that it's competing with competitors or competing with itself (vs its 120Hz ULMB2 mode) AKA do I think that there's more left on the table for Nvidia to squeeze out of these monitors?
Yes. Especially in the brightness department, at the very least, if they're this adamant on avoiding crosstalk behavior.

I think this post is what you are looking for right? https://www.nvidia.com/en-us/geforce/fo ... -and-oled/
he posted the table for 360 hz with new firmware update there.

I don't know if they are willing to disclose much more than what they said otherwise they would have given a longer explanation but I'm sure the Gsync rep that's watching this forum post might reply at some point

I think they might be willing to disclose more details after the firmware update than before.

Posted: **07 Aug 2026, 14:20**

liquidshadowfox wrote: ↑
Yesterday, 14:06
I think this post is what you are looking for right? https://www.nvidia.com/en-us/geforce/fo ... -and-oled/
he posted the table for 360 hz with new firmware update there.

I didn't see that post! Good one, thanks for mentioning it!
That's precisely the data I was looking for, thank you!
I do wonder whether that Pulsar vs OLED graph is with new FW or with older FW? It makes more sense to me that it's with new FW based on anecdotal evidence that users have provided here.
The curve is seemingly better than other VRR+Strobing implementations on the market (e.g.: VG259QM/VG279QM's ELMB-Sync), but still far away from what good backlight strobing should be IMO.
I'll do a write-up soon.

Based on linear interpolation, we can derive the pulse on time interval curve for the ULMB2 mode:

Note how we're having a similar issue again.
360 Hz <=> 2,777778 ms
(0,88 / 2,77778) * 100 = 31,68% duty cycle
A * 0,3168 = 180 cd/m² → A = ~568,18 cd/m²

Even less voltage boosting occurring.... why? Why are we back to ULMB1 levels of voltage boosting? The beauty of these new panels is the buffed voltage boosting enabling you to target lower pulse on time...

Posted: **07 Aug 2026, 21:20**

so this monitor was never 1000hz effective motion clarity to begin with, and still wont be at max hz and max brightness even after update

i think they are wrong or comparing it to oled, because the pulsar is definitely clearer than my 600hz tn without strobing

Posted: **08 Aug 2026, 03:21**

MSIfanboy wrote: ↑
Yesterday, 21:20
so this monitor was never 1000hz effective motion clarity to begin with, and still wont be at max hz and max brightness even after update i think they are wrong or comparing it to oled, because the pulsar is definitely clearer than my 600hz tn without strobing

What this curve above measures is how sample & hold path compares to the impulsed path (the OLED curve follows the typical f=1/T shape; 8,333 ms = 120Hz etc.)
If this image above is Pulsar on FW 1.1.8, then your claim of Pulsar being clearer than 600FPS@600Hz TN might be slightly true

Only with 120 Hz ULMB2 did we break the barrier.

I don't know why they refuse to keep the curve in a 10-19% duty cycle range (from 100–360 Hz)
This would make the curve stay between 1 ms – 0,52777 ms pulse on time, which would lead to far sharper motion than the graph above mentions and be on the edge of the "danger zone" which they're desperately trying to avoid (<1440Hz effective motion clarity)
Sure, the brightness at 100 Hz would be tanked to ~200 cd/m², but what prevents them from keeping it at that value?
It's the typical approach of valuing brightness over effective motion clarity when doing backlight strobing... which makes me even more confused as to why they're targetting a 600-550cd/m² amplitude during voltage boosting on 240Hz & 360Hz ULMB2...

Posted: **08 Aug 2026, 08:36**

@kyube

I think if they were to shorten the pulse and increase voltage to maintain brightness, the residual glow would persist before the next pulse as there is a smaller window/frametime for the light to dissipate as you go up the hertz range. Similar to the artifacts of a crt at higher hertz. It would explain why Nvidia said that it would cause crosstalk although that doesn't explain why it would be unaffected in the middle. I'm assuming they made a mistake relaying information.

Someone mentioned earlier in the thread that the diffusion layer may be the bottleneck and that placing the leds closer to the panel would help ergo mini-led.

Blur Busters Forums

NVIDIA G-Sync Pulsar monitor - Asus ROG Strix XG27AQNGV

Re: NVIDIA G-Sync Pulsar monitor - Asus ROG Strix XG27AQNGV

Re: NVIDIA G-Sync Pulsar monitor - Asus ROG Strix XG27AQNGV

Re: NVIDIA G-Sync Pulsar monitor - Asus ROG Strix XG27AQNGV

Re: NVIDIA G-Sync Pulsar monitor - Asus ROG Strix XG27AQNGV

Re: NVIDIA G-Sync Pulsar monitor - Asus ROG Strix XG27AQNGV

Re: NVIDIA G-Sync Pulsar monitor - Asus ROG Strix XG27AQNGV

Re: NVIDIA G-Sync Pulsar monitor - Asus ROG Strix XG27AQNGV