[ 3 / biz / cgl / ck / diy / fa / ic / jp / lit / sci / vr / vt ] [ index / top / reports ] [ become a patron ] [ status ]
2023-11: Warosu is now out of extended maintenance.

/vr/ - Retro Games

View post   
File: 8 KB, 587x130, input-lag.png [View same] [iqdb] [saucenao] [google]
4793040 No.4793040 [Reply] [Original]

Redpill me on input lag /vr/. What causes it? If I don't notice the difference, is it really bad either way? Do controllers cause it? etc.

>> No.4793048

>>4793040
>If I don't notice the difference
For you to notice a difference you have to be able to compare with and without.
People are used to the lag and then say "I don't notice anything" because they've never played without.
So at least make sure to compare.
And yes, controllers can cause it, particularly wireless ones.

>> No.4793056

>>4793040
All different things can cause it. There are a multitude of articles on the internet.

Freakin' google it.

>> No.4793732

>>4793048
I never owned an HDTV until 6 months ago, and when I made the switch I still didn't notice any input lag

The real way to test is to have a CRT and HDTV side by side and somehow output the video to both at once

>> No.4793749
File: 281 KB, 600x600, 1445946355452.jpg [View same] [iqdb] [saucenao] [google]
4793749

>>4793040
For something like an actual SNES, the flow of control is something like this

...
Display a screen <= pretend there is some event to react to here
While screen is displaying, compute next frame
When processing is done, wait
V-blank hits
Read controllers <= first opportunity to read a button press in reaction to event
Write graphic registers
Start processing next frame <= first opportunity to compute a frame based on reaction
V-blank ends
Display a screen
While screen is displaying, compute next frame
When processing is done, wait
V-blank hits
Read controllers
Write graphic registers
Start processing next frame
V-blank ends
Display a screen <= first screen that can display results based on your reaction
...

>> No.4793757

Input lag matters in games with precise timing, so stuff like racing games or fighting games.

Most people who think input lag doesn't matter only play games where their timing doesn't matter. It doesn't matter in Final Fantasy when selecting an item takes an extra twentieth because your HDTV is creating lag. It really matters when you're trying to thread a turn in F-zero.

>> No.4793951

>>4793040
>If I don't notice the difference, is it really bad either way?
Yes.

>> No.4794108
File: 73 KB, 239x163, 1524861732654.gif [View same] [iqdb] [saucenao] [google]
4794108

input lag isnt nice but there are a lot of games you can play with it and have an okay experience, however since I am poor and cant afford good gear I cant properly play fighting games which is an annoyance

>> No.4794110

>>4793951
Why?

>> No.4794119

>>4793732
I switched from an HDTV to a BenQ monitor recently. The difference was huge for me. I found Windjammers much easier to play when I tested it, I could react to things I couldn't before.

If you have a Samsung HDTV, you can switch to game mode which should help.

List of displays and their lag:
https://displaylag.com/display-database/

>> No.4794125

>>4794119
Sorry, to clarify, it's not just Samsung HDTVs have game mode.

>> No.4794696

>>4793040
People doing things retardedly causes lag. You don't notice any "difference" because you've always been done that.

>> No.4795283

>>4793757
Reminder that Street Fighter V has six frames of lag. Most other modern fighting games have several frames of lag. Online play is extremely popular and adds more like.

This whole notion of "fighting gamers refuse to play in anything but 0 lag situations" is silly. As is the accusation that anyone who says 1 or 2 frames of lag is an acceptable compromise must be someone clueless about the genre. You have no way of actually knowing that. This is an anonymous image board, it doesn't work that way.

>> No.4795285

>>4795283
Standards have lowered with modern games, eh? Tell me something new.

>> No.4795294

Use retroarch to reduce lag by running parallel versions of the game a few frames in advance, taking branches depending on your input.

>> No.4795304

>>4793040
Anything under 26-40ms is REALLY hard to notice.
Under 25ms is nearly impossible. It's 2 frames of lag.
16ms and under is 1 frame. Nobody would notice this without tests.

That said, my first TV was a shitty Visio with 80+ms of lag. It was highly noticeable.

I replaced it with a panasonic plasma with less than 12ms of lag.

now I have an OLED with less than 16ms of lag.

>> No.4795412

>>4795304
how do i test it?

>> No.4795427

>>4795412
Short answer is you don't unless you have specialized equipment. OSSC, Leo Bodnar device, signal splitter and a CRT to compare with, etc.