end to end latency measuring tool

[Corona Scurrae]

that doesn't exist for customers, unless you buy one of nvidia's gsync certified monitors with compatible mice and keyboard or idk what people have to do in order to receive an LDAT.

why are there no tools available for customers? this is crucial basic information that everyone should have at their disposal, because ultimately, rendering time - alone - is meaningless if end to end latency is way higher.

[_I_]

what are you talking about?

what ends?

there are ping tools and most games give ping info on how long the delay is from client-host-client for online games
and most can show the ping

frame draw time can be shown, along with fps
if fps is not limited by vsync/gsync/freesync, thats easy to calc
1000 / fps = draw time in ms

[Corona Scurrae]

_I_ eredeti hozzászólása:

what are you talking about?

what ends?

there are ping tools and most games give ping info on how long the delay is from client-host-client for online games
and most can show the ping

frame draw time can be shown, along with fps
if fps is not limited by vsync/gsync/freesync, thats easy to calc
1000 / fps = draw time in ms

from mouse to display or controller to display. we don't have any tools at our disposal. rendering time alone is meaningless if you don't have a proper tool, to measure how much time it takes from mouse-click to display.

[_I_]

those are near 0
much less than draw/display times

if polling rate on mouse is 1000hz, its 1ms, most are 250-500hz which is 4-2 ms
add that to the draw time

[Corona Scurrae]

_I_ eredeti hozzászólása:

those are near 0
much less than draw/display times

just looked at some of the reviewers that got their hands on them. rendering time was minuscule but total system latency ended up being 60ms at 60hz.

at 144hz it dropped to 30ms but the thing is that they are using high-end hardware like 13900k and 4090. I want to know how much it differs between other hardware configurations.

but you are technically correct. it's nigh zero

[_I_]

60ms frame time is 16fps
30ms = 33fps

16ms = 60fps
7ms = 144fps
6ms = 165fps

there is no real way to measure the mouse response time
as any signal sent from the mouse will take that long to get to the pc

[Corona Scurrae]

_I_ eredeti hozzászólása:

60ms frame time is 16fps
30ms = 33fps

16ms = 60fps
7ms = 144fps
6ms = 165fps

just google ldat and see the results for yourself.

[DevaVictrix]

I'm sure there was a LTT video on this a year or two ago. I can't remember the results but it's tiny. Things like the time of day you play, how much beer you've drunk and how much food you haven't eaten will have a far bigger affect. It's not worth worrying about or investing in!

Edit: This might have been the one... Ldat
https://www.youtube.com/watch?v=26DbJ3E4YKI

[Corona Scurrae]

DevaVictrix eredeti hozzászólása:

I'm sure there was a LTT video on this a year or two ago. I can't remember the results but it's tiny. Things like the time of day you play, how much beer you've drunk and how much food you haven't eaten will have a far bigger affect. It's not worth worrying about or investing in!

Edit: This might have been the one... Ldat
https://www.youtube.com/watch?v=26DbJ3E4YKI

I see well having a tool to know that for sure, would be great nonetheless

[Rod]

I just guess it based on what the measured tests by monitor reviewers say but yea would be cool to see it on the gaming oleds and the esport screens 500hz?


It would be part of a decent toolset, I would use it, Plus ingame netgraph for pings, And Latencymon as well with the frametime readouts. Basically anything that will spit out realtime numbers can be totalled up to relief system performance anxiety. Yes thats the diagnosis :p

[[N]ebsun]

I am working on a tool for this - preliminary pricing will be around $150k usd and development time around 1 year.
There needs to be around 100 orders in the next few months to make it possible though - since interest here is high, would you guys be willing to pre-order at $150k usd per unit ?

The problem with things like this is... the measurement needs to be accurate, since the actual delay is going to be pretty small - unnoticeably small... which kinda makes it not very useful

[DevaVictrix]

The monitor latency is going to be the biggest (mine is apparently 20ms) and that's quote easily tested with consumer tools. Still a waste of £150 if you already own the monitor and can't improve whatever you discover.

edit:sorry, £100, not £150.

[Corona Scurrae]

DevaVictrix eredeti hozzászólása:

The monitor latency is going to be the biggest (mine is apparently 20ms) and that's quote easily tested with consumer tools. Still a waste of £150 if you already own the monitor and can't improve whatever you discover.

edit:sorry, £100, not £150.

every lcd out there will have an end to end latency 20-30ms. only oled microled and nanorod qdot, will be faster.

currently most oled monitors sold, have 6-8ms input lag

[_I_]

newer displays are <10ms g-g time
ones 20 years ago were closer to 25+ms which were unusable for any gaming or even videos or any moving objects

[PopinFRESH]

@op the answer to your question is because the majority of monitors don’t have the hardware built in to determine this. It is not just a software problem. You’ll either be purely guessing with estimates that are worhless, or you’ll need a sensor on your mouse and a very high speed camera.

The reason why Reflex Analyzer requires specific Gsync monitors is because the new gsync module has the hardware built in to time the draw rate and the compatible input devices have updated firmware to report their switch activation timing which allows the difference to be determined.

[Corona Scurrae]

PopinFRESH eredeti hozzászólása:

@op the answer to your question is because the majority of monitors don’t have the hardware built in to determine this. It is not just a software problem. You’ll either be purely guessing with estimates that are worhless, or you’ll need a sensor on your mouse and a very high speed camera.

The reason why Reflex Analyzer requires specific Gsync monitors is because the new gsync module has the hardware built in to time the draw rate and the compatible input devices have updated firmware to report their switch activation timing which allows the difference to be determined.

yet this could be easily mitigated if LDAT was available to consumers.