>it's a "will you notice" thread

Would you, personally, notice input lag of 23ms? That's the specs for the panel I'm buying.

They say it's '1-2 frames'. Are we talking legendary eagle master or a silver-level handicap?

On my TV with 40ms delay it's unplayable, with just over half that you will stay notice, particular in games that require you git gud with any kind of pixel perfect/split second timing like clearing jumpgs, iframes or combos. It will not be immediately obvious but it will effect target tracking in FPS/TPS.

Just buy a Triniton user.

OP don't
23ms is huge
it's bigger than your dick, you'll notice it easily

Yes. It's very jarring. Enough to turn Mike Tyson into a crapshoot. Smooth movements don't stop where you'd expect them to stop, and if whatever you're playing is demanding in reaction speed, it's going to get considerably less pleasant. Double digit delay should not even be considered.

If, however, you're used to input lag, you might not notice anything wrong with it, and live in ignorant bliss until the day you gaze into a CRT, maybe one at 120Hz.

just as a comparison, enable double buffering vsync (but real double buffering vsync, not the hacky "double buffering" vsync with a slightly lower fps cap where the delay ends up being from 0 to half a frame because the fps cap forces it to always put out the swapchain since it can never keep up with the monitor)
At 60fps the delay added would be 16ms and that's already enough for anyone to notice and for pretty much everyone who plays a video game turning it off

How playable a game feels isn't a direct effect of framerate but of total system latency. Each ms more static latency from your input or output devices has to be counteracted by the same value lower frametime so the game feels the same. If you look at it this way you'll see that low latency input and output devices are the cheaper way to playability compared to highend CPUs and GPUs.

23 ms input lag means that you are seeing things happen at a rate of roughly 1.5 frames after the system has computed them, your blink lasts for about 350 ms and your reaction times are roughly 200ms, Even when playing competitive quake the disadvantage is negligible. Your monitor would be fine, mine is 13ms but I didn't buy it because of the low input lag it just had it.

That's nothing. My plasma TV has 72ms input delay.

The question is whether you would or not.
Here, I made a gif showing what 23ms of delay is like.

How does input lag work anyways? Is it only something that affects non-CRT displays?

23ms is miserable and Quake at nearly any level would be unplayable with a monitor that bad.

Prove it.

Instead of guessing your reaction speed go and actually check your reaction speed. Then convert that to milliseconds.

Guessing, that's just the average reaction speed, because I'm giving a generalization.

Holy fuck that's too long.

Buy a monitor with 23ms response time and try to play Quake with it.

Whatever you feed a CRT (unless it's a fancy crt with digital inputs) goes straight to the light cannon (CRTs are very similar to digital projectors except they work differently and are analog).
On a modern screen you have a billion signal processing/conversion phases and then you have the individual LEDs that compose a pixel getting brighter or darker after they're told what to do

can't gif only adjust frame delays in steps of 10? what's that, 20ms or 30ms? Why does input only blink once and then output repeatedly blinks alone, what does it mean?

The average reaction speed is 250ms for visual input, and it's given that way for simplicity since when measuring averages it's a curve which peaks around 240-280ms.

Not as far as I know.
There are only three frames.
It means the gif is out of sync with your monitor.

With CRTs, there is very little processing to be done with the signal. Most if not all of that work is pushed off into the input's side of things. As long as whatever the input is feeding to the monitor is to spec, it's going to work and work very quickly.

TVs and other monitors involve more processing. They have to take the whole frame's data into a buffer before it can display it, maybe transform it and scale it, if it's a TV and it's using some of that in-betweening shit to insert frames for a "smoother experience" and other extraneous processing, then that's going to take time. It might even base those on previous images and need more frames in the pipeline before it can do its magic. Also, TVs are designed for, well, TV programming. Whatever mechanisms they have in there to make sure the image remains steady if something happens to the input stream or something gets mangled along the way will also introduce a delay.
It saves money for the manufacturer to just have all inputs go through the same slow processing, so that's what they generally do. Some TVs have "game modes" which probably cut out some of the processing, but it's still going to have something done to the signal.

And then you have the display response time itself, which isn't exactly zero. Takes a bit of time before a pixel goes from one color to another, but that's generally on the order of milliseconds.

faggots are unreal.

Well I guess you tried at least, so there's the definitive answer, it's negligible.


Something like that, I keep finding different results of 250ms and 200ms

Most I find on 200ms is dubious answers outside some scientific papers to very specific tests. Since this is about latency and vidya gaems only tests involving clickers are relevant since that's generally what you're doing when playing vidya. If you're trying to be accurate and covering everybody then 215-225ms, however keep in mind 200ms is a fifth of a second and 250ms is a quarter of a second.

Thanks, a more definitive average reaction time is useful to me.

i haven't looked a lot into those studies but i assume they used off the shelf computers, here are some numbers
Overall on a modern PC you can expect ~30ms best case scenario (gsync/freesync, low latency peripherals, uncapped frame rate, lowest possible renderahead and on windows), ~35ms best case scenario but with low latency non-crt monitor, ~50ms what you will probably find on articles about computer latency since 99% of the time they forget renderahead but do everything else right ~80ms what most people should or can be getting (budget monitor, keyboard that just works on usb port, office mouse, uncapped frame rate and ~150ms your grandma's PC (triple buffering vsync, lowest price possible hardware that won't break in a day)
here's some things anyone can do to lower their latency

Will you notice my digits, OP?

Like accuracy with firearms, input lag is cumulative.

So let's say you've got a 200ms reaction time and are playing on a display with 23ms delay. Your "total reaction time" is 223ms. Now let's swap that with a monitor with 4ms delay, which is quite average. Your "total reacton time" is 204ms, which is 9% faster. That is not an insignificant figure, and in shooters, especially those with very low TTKs, that will frequently be the difference between winning or losing an encounter.

Let's take it one step further. Most modern TVs have around a 40ms delay, so if a console gamer swaps from a TV to our 4ms monitor, he's going to be 17% faster.

I'm going to give you the benefit of the doubt that you aren't just shitposting and being purposefully dense, but you can't possibly argue that those numbers are "negligible". If all you play are movie "games" or singleplayer, yeah it probably doesn't matter for you and you can get used to anything Hell, consolefags are used to framerates dipping to the teens.

You're seeing two dots in a clear background, not a video game.

23ms is nothing.

i haven't looked a lot into those studies but i assume they used off the shelf computers, here are some numbers
Overall on a modern PC you can expect ~30ms best case scenario (gsync/freesync, low latency peripherals, uncapped frame rate, lowest possible renderahead and on windows), ~35ms best case scenario but with low latency non-crt monitor, ~50ms what you will probably find on articles about computer latency since 99% of the time they forget renderahead but do everything else right ~80ms what most people should or can be getting (budget monitor, keyboard that just works on usb port, office mouse, uncapped frame rate and ~150ms your grandma's PC (triple buffering vsync, lowest price possible hardware that won't break in a day)
here's some things anyone can do to lower their latency

wew lads ubisoft is here

...

...

correcting a mistake
should be

Don't you have a gold plated HDMI cable to shill?

This discussion really makes me wonder, at what point does latency become imperceptible? All I know for sure is that some people have really low standards. Games like Monster Hunter 3 Ultimate (on WiiU) and Killzone 2 had absolutely disgusting amounts of latency that I felt as soon as I was in-game, but I've found people who claim that they don't feel it.

nope, i have a gold plated RCA cable to shill

...

Imperceptible /= irrelevant
There's one phenomenom these people are experiencing and it is how they probably don't have anything to compare it with since they never saw any better or they don't even know what input lag is
I would say imperceptible sits at below 30ms because no computer that you can actually play vidya on ever reached shit that low and everything out there is enough to be noticeable, not in 1ms steps but noticeable

That sounds pretty bad. Even my previous monitor was something like 8ms. This this isn't exactly TOL but it's at least 5ms.

Not true your numbers don't mean shit if you can't give me real world studies showing the actual affects of input lag in different games you can fuck off, that's literally all I want, the actual effects, I don't care if there is a 17% gain I want to know what that 17% gain actually means. I'd be absolutely willing to concede this point, I just want you to back up your claims.

You're not the smartest of the bunch and you sound 12
Why do you discuss latency if you don't know what it means? If you want to look smart you can wear fedoras or write with feathers, but don't come shit on people discussing latency because they haven't told you what it is

I've been decently competitive (as in having a shot at the WR) in a few speedrunning categories on an IPS screen with 20ms response time and 15+ms display delay using borderless fullscreen with desktop compositing on and stock settings in my GPU control panel, membrane keyboard and laser mouse.

People that really care about small details are one of three groups, retards who can't get good and think the hardware is the problem, people who are very sensible to some specific issue (imperfect mouse tracking, blur, input lag, key travel time, uneven actuation force) which has nothing to do with performance and more to do with comfort and finally people who are good enough for better hardware to actually make a sizeable difference.

So short answer if you're good you'll still be good with a slow screen unless it triggers your autism, if you're bad an OLED / 144hz

I can see the difference. I doubt I could REACT to it, but I can SEE it.

t. 144hz monitor user

The point isn't if you can react to it or not, no person can it's just not humanely possible.
The fact that you can spot it however means it's already within human perception and very relevant.

I gave you the numbers, now if you want "real world studies", set up some grants and fund them. You're requiring an ever-escalating and unreasonable burden of proof which is how faggots argue when they know they're wrong, have backed themselves into a corner, and just refuse to admit they're wrong.

Nice try but you still didn't give an answer, I guess the CRT thread really is just a circle jerk. By the looks of things the only gain from a minimal decrease in input delay is the false air of superiority.


I just want to know what sort of actual gain there is, and if you can't give me some quantifiable performance gain then you can fuck off back to your circlejerk.

>Input lag on a video output device

If you don't know what are the gains of lower latency, you don't know anything that pertains to this thread and you should lurk more
It's just how it is

This thread is turning into "acting like a nerd but i never got the grades"

I think the rough estimate for eye response has been shown to be around 15-20ms. Probably more like, 'you might not even notice' level of difference.

I know how it works but I want to know what the actual difference made is, and since you can't give me that I can rightfully say that you are full of shit.

This. 23ms by itself is not much, but the problem is you have to add it. You have the latency to draw the frame, display it, react to it, and read the answer. Reaction time is the worst, and not much you can do there besides git gud, but the rest of the equipment can't be ignored.


It's reasonable to expect a difference in reaction times affects your outcome in a game where reaction time is important. Demanding proof for something so obvious on a fucking imageboard is ridiculous.

How much was that study affected by gaymers?

I Know it gives you a slight edge, but I suspect that that slight edge is too small to make a reasonable difference, I want you to show me that that slight edge makes a difference that I should be able to care about, and if you can't show me that then why should I care.

How about you fucking test it yourself if you want so much proof?

You can see on a gif they posted above the difference. The fact you dont want to perceive it doesnt mean its not there, aspie.

I can see it, I just don't think it makes enough of a difference while playing a game to worry about.

That's absolutely noticeable, like a cycling road signal.

...

Now you're hedging. You don't give a fuck about any of this, you just want to stir shit and be an annoying fucking faggot.

Then whats your fucking problem? go and cry to both of your moms somewhere else you raging faggot.

well fuck

CRT purists are a fucking joke.

Nothing personal but it may be related to people who started out on CRTs and may be a bit more jarring than those more accustomed to the new monitors regarding response time issues.

Now you went to far, you mediocre faggot. There's an absolute difference and your eyes have been brought up stupid.

Did you fall off bed a couple of times when you sucked your mom's titties?

...

To clarify, the model I'm considering has 23ms input lag, but its g2g time is 5ms.

Ordinarily, I'd just go with a cheapo TN, but this has really good color and contrast.

get a QNIX QX2710

XD HUE HUE HUE TROLD

No monitor just has the actual input lag in the specifications. You're looking at a spec that isn't actually input lag.

I'm not going by the spec sheet, I'm going by Tom's Hardware testing.

SFV 8 frames.. that's over 100ms input lag..

What's the point? You've already lost.

if all you play is first person shitters, it doesn't matter, you're going to be a plebe forever. If you want a good display for games get a CRT.

you've never compared it, then.

You should keep in mind that your monitor doesn't add input lag, it adds output lag. As long as you adjust your frame timings it doesn't matter

I bought it on March 2016 like a sucker
never again

I don't have a table the size of an aircraft carrier tho

Have this. Same gif but twice as slow.
Remember, the actual delay would be half of this.

Come at me you fatherfuckers

…

Something is wrong with my eyes. When I focus on the out put dot, I can see the flicker of the input but the output remains steady black. When I don't focus on either image I can see them both flicker but with no perceptible latency, they both flicker at the same moment.

Focusing between them trying to keep both dots in focus, both dots remain steady black with no flicker. Should I see an optometrist?

Your eyes have too much latency man, I'm sorry.

You're a sad fucking sperg

btw i looked it up and i was wrong gif can't only adjust frame delays in steps of 10, it can do up to 50fps. It's just that some old browsers could only do gif frame rates in 10ms steps for some reason

it means you blinked when the input was flashing

I came here to laugh at you

I don't think a human has a reaction time under 100 ms, I don't think it's physically possible to notice 23 ms of output lag.

Someone please correct me if I'm wrong though

….I see slightly slower. Slightly slower. Slightly slower. Massively slower. Slightly slower repeat

Is how this works right? **r-right?

if it was a fast or competitive game yes I'd notice. That kinda lag would fuck me in Tekken

How is it tested? If they're not using an osciliscope treat the value as meaningless. The only proper monitor review sites i know of in this regard are prad and tftcentral. There are plenty of good IPS panels with only a few ms of input lag (not the same thing as response time)

Didn't read first line. Just thought it said 1-2 full frames of delay, which I'd notice.

23ms is fine. Wouldn't hurt my game performance. My skill would be the bottleneck by then.

Post the model

Your eyes can notice any change in imagery down to 10ms and a bit lower if they're healthy and you're sharp and focused on the object. But yeah you actually processing and reactioning to it is an entirely different thing.
23 ms refresh rate delay may not seem much, but it stacks with other input delays such as your actual reaction delay, the pheripheral's input lag, network lag (usually around 50 to 150 ms if not on LAN) if on multiplayer, framerate lag if the game isn't exactly running at 60+FPS, among other things i'm missing.

What about the brain's input lag?

I guess I have a really good, albeit old monitor and above average reflexes, because this is jarringly noticeable. It's about the amount of time it takes for me to switch from target to target or fix my sights on an enemy.

Me too
What could possibly go wrong?

One job

Fuck.

it could make a difference, but that just means you need to stop sucking

It's three frames.

FUCKING CHECKED

...

...

...

So the number of the beast is just Satan's reaction speed?

As point of comparison it's worth noting that most current game fighting games sit at about 5 frames of composite lag. One frame of difference isn't very noticeable but anything above that will feel pretty shitty.