PDA

View Full Version : Inpute lag and online gaming



jayhall0315
07-17-2009, 06:50 PM
I have been meaning to write a good long detailed article on Input Lag and exactly how to optimize your system for IL2 online play (Hyperlobby, Xfire, etc...) but cant seem to find the time so far. Here is a great new read though, that many of you may be interested in :

http://anandtech.com/video/showdoc.aspx?i=3601

This article is written more for the prospective online shooter (say Team Fortress) but it still applies to IL2. This is also part of the reason I advocate for maximum ergonomic setup when it comes to HOTAS button mapping (and possibly not using pedals).

Hope this helps some guys out - S!

jayhall0315
07-17-2009, 06:50 PM
I have been meaning to write a good long detailed article on Input Lag and exactly how to optimize your system for IL2 online play (Hyperlobby, Xfire, etc...) but cant seem to find the time so far. Here is a great new read though, that many of you may be interested in :

http://anandtech.com/video/showdoc.aspx?i=3601

This article is written more for the prospective online shooter (say Team Fortress) but it still applies to IL2. This is also part of the reason I advocate for maximum ergonomic setup when it comes to HOTAS button mapping (and possibly not using pedals).

Hope this helps some guys out - S!

ytareh
07-17-2009, 06:55 PM
Surely your internet connection (latency/lag)is just as important if not more so in determining lag...

jayhall0315
07-17-2009, 07:01 PM
It is Ytareh, but what about the case where you have 10 MB/sec download and I do as well and the server is exactly between us (geographically) with 100 millisecond lag for both of us. Then, if my setup is more ergonomic, I will likely have the edge.

BillSwagger
07-18-2009, 01:17 PM
It is an interesting article.
I just dont think the movements in Il2 are so fast, or require a twitch response to be competitive.
on the other hand...

Aside from lag you would have the edge of an ergo set up for the simple reason everything is at your finger tips requiring less movement to perform an action comfortably...

If i can hit every possible button or combo with out having to move my wrist or elbow, then i would have the edge over someone who needs to move their wrist or elbow, even if they have faster hardware and a connection located on the same block as the server. =)

deepo_HP
07-18-2009, 04:06 PM
i don't see, what this article has to do with an ergonomic setup of controls?
it is all about hardware-processing.

considering, that it takes 200ms as the average human reaction time (which is too slow for hight-attentioned gaming), i also don't know, why a 100ms ping to server matters.
noone will claim, that there is no advantage in having all inputs layouted best for reach, but besides the reaction time of 200ms, there is a following time needed for movement of whatever. mouse-click would be fast, but mouse movement or stick movement much slower.

so, the described input lag in the article (hardware processing the input data, nothing else) is mainly about grafics and frame-rendering - taking 20-80ms, as i understood it.
comparing that to human factors is less than negectable.
as billswagger said, having stick and controls in suitable reach and position would be a good idea. but that's obvious!
and i agree with him, that a flight sim won't need that extreme demands as a first person shooter.

BillSwagger
07-18-2009, 05:23 PM
maybe compare:

hat switch

mouse pan

head tracking

Is turning your head as fast as moving your wrist or faster than wiggling your thumb??

Then you have groups of movements. Maybe with one movement one method is superior, but if you need to make several movements in a very short time which method would be better??

The article also mentions learned behaviors, so i think a lot of what we do relies heavily on the anticipation of whats going to happen anyway.

I do sometimes wonder if head tracking is as fast and accurate as panning with a mouse, though (using mouse smoothing software).
cause i can pan right to left and right again fast as needed with out hurting my neck.

I wonder what people must look like trying to duplicate these movements using a head tracker.

deepo_HP
07-18-2009, 09:09 PM
hi billswagger,

i understand the problem of chosing the right devices, or get them set up in a good and proper way.

but that is not input lag, it all happens before the input signal is processed.

the mentioned learned anticipated behaviour is mainly why one experiences input lag. it is most noticeable in games, where one has to aim accuratly in a short time...
input lag shows up as the action on screen is delayed behind the input.

in case of a flight sim, the whole simulated reaction time of plane controls is far too slow to show a noticeable input lag. the same for firing bullets: even if there was an input lag, it would be hardly noticeable, as noone expects the cannons fire exactly at the instance of triggering (well, and if one would notice, hardly anything to do in that case).


that doesn't change the question of having all set up so that it can be reached easily. but that is not 'input lag' (what the article is about).

the only aspect of input lag in flight sims, which i know of, is the joystick. the usb-port is slower than ps2... so one will experience better and faster control, when using the keyboard for pitch, yaw and bank. maybe a similar reason prevents the jets from burning on take-off, if keys are used.

a head-tracker will probably show even more lag, as the camera-picture needs software processing... the whole chain is much longer. for the purpose of viewing around it is well enough.
however, it is indeed a fancy imagination to have multiple inputs connected for more complicated controls. one could think of muscle-sensors even, so the gamer is shaking and grimassing all time http://forums.ubi.com/groupee_common/emoticons/icon_smile.gif

BillSwagger
07-19-2009, 03:59 AM
i read the article, too.

I just thought the bigger issue with this sim is reaction time and the readiness of keys/commands.

The input lag is an interesting read but it is of no importance to this game.
I also wonder what kind of input lag is experienced on a ww2 airplane. If i hit the flaps switch, how long does it take for the electric signal of the switch to run its course and tell the motor that moves the flaps to begin pushing down.
Input lag might even be necessary for an accurate simulation of controls.


ergonomically speaking (another thing mentioned in this post)

Which is going to give me the edge :

a hat switch

mouse panning

or

head tracking

Head tracking frees up your hands so your able to concentrate better at controlling your plane.

Mouse panning, i think is faster because you can draw the mouse around the screen faster and more accurately than you could shake your head.

Hat switch:
Is probably the simplest form of panning, and works well for people who are used to it. Snap views offer the same reflex response that a mouse might offer.
The only difference is a finger or thumb moving the hat could otherwise be doing something like operating a weapon, flaps, throttle, etc...
So there is a much bigger delay in response from moving your thumb off your hat switch to a key that operates a command.

As the article mentions, learned behaviors will circumvent or diminish these delays, but in combat i would argue that the person who already has their hands where they are needed to respond will have the edge over the person who needs to make even the most subtle adjustment.



http://noveltyknees.files.wordpress.com/2007/02/cowboy.jpg

M_Gunz
07-19-2009, 06:45 AM
Jerking the plane around doesn't do much for conserving energy. As Deepo says, flight sims ain't FPS.
Of course it's important if you play Air-Quake running low alt at near stall....