We now know unequivocally that DSC does indeed add input lag.
DSC turns off Multi-plane Overlay.
Windows uses MPO to communicate directly with the display.
When MPO is off, Windows adds its own compositor layer called Compose Flip, which adds delay.
So if you purchased a high refresh rate OLED and your GPU is older than the new 5000 series, DSC processing is so damn heavy it eats up your GPU's internal display heads, leaving no resources left for MPO. And even if you're within dp1.4's bandwidth, MPO still won't be active and you'll be incurring delay from Compose Flip.
DSC Problems & Misinformation [DSC IS NOT A HOAX]
-
evmadic
- Posts: 59
- Joined: 20 Feb 2022, 04:13
- Contact:
Re: DSC Problems & Misinformation [DSC IS NOT A HOAX]
5800x3d, RTX 4070 ti, Acer Nitro XV252QF 390Hz
https://www.twitch.tv/evmadic
https://www.twitch.tv/evmadic
- dervu
- Posts: 437
- Joined: 17 Apr 2020, 18:09
Re: DSC Problems & Misinformation [DSC IS NOT A HOAX]
The thing is that using presentmon uses one more overlay, but you have only one slot with 4xxx cards & DSC, so if you don't use overlay but only use shortcut to record data with presentmon and you check in file after recording, you will find independent flip was active.evmadic wrote: ↑20 Mar 2026, 04:35We now know unequivocally that DSC does indeed add input lag.
DSC turns off Multi-plane Overlay.
Windows uses MPO to communicate directly with the display.
When MPO is off, Windows adds its own compositor layer called Compose Flip, which adds delay.
So if you purchased a high refresh rate OLED and your GPU is older than the new 5000 series, DSC processing is so damn heavy it eats up your GPU's internal display heads, leaving no resources left for MPO. And even if you're within dp1.4's bandwidth, MPO still won't be active and you'll be incurring delay from Compose Flip.
Was the difference even measured?
Ryzen 7950X3D / MSI GeForce RTX 4090 Gaming X Trio / ASUS TUF GAMING X670E-PLUS / 2x16GB DDR5@6000 G.Skill Trident Z5 RGB / Dell Alienware AW3225QF / Logitech G PRO X2 SUPERSTRIKE / SkyPAD Glass 3.0 / Wooting 60HE / DT 700 PRO X || EMI Input lag issue survivor (source removed)
- kyube
- Posts: 905
- Joined: 29 Jan 2018, 12:03
Re: DSC Problems & Misinformation [DSC IS NOT A HOAX]
I hope you know that this youtuber uses LLM slop without any external tools to "evaluate" (spread disinformation) to laymen like you.
See: viewtopic.php?t=12730&start=30#p118897
evaluating xhci controller performance | audio latency discussion thread | "Why is LatencyMon not desirable to objectively measure DPC/ISR driver performance" | AM4 / AM5 system tuning considerations | latency-oriented HW considerations | “xhci hand-off” setting considerations | #1 tip for electricity-related topics | ESPORTS: Latency Perception, Temporal Ventriloquism & Horizon of Simultaneity | good lcd backlight strobing implementation list | display vs gpu scaling
-
Terepin
- Posts: 9
- Joined: 02 Dec 2022, 06:54
Re: DSC Problems & Misinformation [DSC IS NOT A HOAX]
Yeah, no. I have Acer Predator XB273UNX with DSC on (1440p 240 Hz 12bpc) and MPO works just fine.evmadic wrote: ↑20 Mar 2026, 04:35We now know unequivocally that DSC does indeed add input lag.
DSC turns off Multi-plane Overlay.
Windows uses MPO to communicate directly with the display.
When MPO is off, Windows adds its own compositor layer called Compose Flip, which adds delay.
So if you purchased a high refresh rate OLED and your GPU is older than the new 5000 series, DSC processing is so damn heavy it eats up your GPU's internal display heads, leaving no resources left for MPO. And even if you're within dp1.4's bandwidth, MPO still won't be active and you'll be incurring delay from Compose Flip.
-
pigzera
- Posts: 24
- Joined: 29 May 2021, 09:29
Re: DSC Problems & Misinformation [DSC IS NOT A HOAX]
Did a test on my acer xv240f6 (600hz dsc, rtx 4070ti) and dxdiag report MPO disabled. Then did the C.R.U trick to disable DSC but keep 600hz and now dxdiag is reporting correctly MPO enabled.evmadic wrote: ↑20 Mar 2026, 04:35We now know unequivocally that DSC does indeed add input lag.
DSC turns off Multi-plane Overlay.
Windows uses MPO to communicate directly with the display.
When MPO is off, Windows adds its own compositor layer called Compose Flip, which adds delay.
So if you purchased a high refresh rate OLED and your GPU is older than the new 5000 series, DSC processing is so damn heavy it eats up your GPU's internal display heads, leaving no resources left for MPO. And even if you're within dp1.4's bandwidth, MPO still won't be active and you'll be incurring delay from Compose Flip.
- Chief Blur Buster
- Site Admin
- Posts: 12266
- Joined: 05 Dec 2013, 15:44
- Location: Toronto / Hamilton, Ontario, Canada
- Contact:
Re: DSC Problems & Misinformation [DSC IS NOT A HOAX]
DSC and MPO can be used together.
It's the OS/driver/display fault if it does not.
DSC can be (in theory) essentially lagless *if* fully *and* properly implemented.
Unfortunately displays and OS aren't always fully optimizing the workflow...
DWM compositing is a fallback when MPO isn't doing the free compositing. On a practical matter, on an RTX 5090 going almost 2 terabytes per second, DWM compositing should theoretically be only ~9 microseconds (0.009ms lag) at the RTX 5090 memory bandwidth on a 1080p framebuffer when MPO is not active.
Now, if your GPU memory is slower, especially laptop memory, DWM compositing takes much longer... You can also use full screen exclusive to bypass that, and thus eliminate most need of lag-reducing benefits of MPO. (Nonwithstanding the ugly rabbit hole of Windows' FSO)
The real world? Kinda a mess.
Head of Blur Busters - BlurBusters.com | TestUFO.com | Follow @BlurBusters on: BlueSky | Twitter | Facebook
Forum Rules wrote: 1. Rule #1: Be Nice. This is published forum rule #1. Even To Newbies & People You Disagree With!
2. Please report rule violations If you see a post that violates forum rules, then report the post.
3. ALWAYS respect indie testers here. See how indies are bootstrapping Blur Busters research!
