Code-In Input Delay

96 milliseconds. Subtract Latency from this base delay and you get the player's Input Delay. I say 96 because it's simpler in binary. 00110000

Someone with 40 ms latency and someone with 90 ms latency can play the same game. Small shared delay lends itself to easing up and thinking about what you're doing. To be clear: They both play with about 96 ms latency this way.

40ms player and 180ms player can play a similar game, improving fairness and community. To be clear: One plays with 96 ms latency, the other with 180 ms latency.

Latency checks shouldn't be exploitable; it would definitely be attempted.

Find more posts tagged with

Comments

Sapiverenus

FuryBladeborne wrote: »

Regardless of how you feel about our meaningless votes, I vote no as well.

Input delay is a great way to make a game feel bad.

I've had plenty of reasons for what I think and have referred to specific delays rather than an arbitrarily large amount; dont' characterize it as 'feeling' to suit your own.

Sapiverenus

https://www.online-stopwatch.com/

Go ahead and tell me how quickly you can start and stop this stopwatch. Get a feel for the different milliseconds.

You will all have to deal with delay either way. All of you have dealt with delay in online games in excess of 64 ms.
Having a baseline delay makes the game somewhat more fair and allows for more coherent processing of inputs from all players.
As I've suggested, individual player latency would be subtracted from the delay. If it takes about 64 milliseconds or longer to reach the server then there is no delay on updating your client.
The combat isn't tight and reactive anyway. It's a hack n slash.
But a delay synchronizes everyone's gamestate and allows for coherent combat; no "oh he was actually 40 pixels to the right but my screen showed him dead center".

Sapiverenus

Let's assume the extreme.
Server has 1 second delay. Everyone has 0 - .5s latency. With this arrangement, the server can collect everyone's inputs and place them in the correct order by calculating when their input was Actually sent; then the .5s delay individual is not always reacting late.
This is not ideal. 96 or 64 ms at least provides some correction and complete correction to those with low latency; those with higher latency are not as 'behind' as well since the delay would not apply to them.

prymortal

Sapiverenus wrote: »

96 milliseconds. Subtract Latency from this base delay and you get the player's Input Delay. I say 96 because it's simpler in binary. 00110000

Someone with 40 ms latency and someone with 90 ms latency can play the same game. Small shared delay lends itself to easing up and thinking about what you're doing. To be clear: They both play with about 96 ms latency this way.

40ms player and 180ms player can play a similar game, improving fairness and community. To be clear: One plays with 96 ms latency, the other with 180 ms latency.

Latency checks shouldn't be exploitable; it would definitely be attempted.

36-42ms but that's due to LP MXspeed keyboard & 8K polling KB + Mouse. 70MS is a more generalized number for PC input latency.
The thing is Game code should already have allowances for up 200ms offset because 100ms upload, 100ms download.
My concern is A lot of ashes is tested locally.

Azherae

Sapiverenus wrote: »

Let's assume the extreme.
Server has 1 second delay. Everyone has 0 - .5s latency. With this arrangement, the server can collect everyone's inputs and place them in the correct order by calculating when their input was Actually sent; then the .5s delay individual is not always reacting late.
This is not ideal. 96 or 64 ms at least provides some correction and complete correction to those with low latency; those with higher latency are not as 'behind' as well since the delay would not apply to them.

As someone who has done some specific tests in their Alpha-1, I will only say:

"On right track, don't worry about it."

I will hope that from your short experience here so far you understand why I am not inclined to say more.

Sapiverenus

prymortal wrote: »

36-42ms but that's due to LP MXspeed keyboard & 8K polling KB + Mouse. 70MS is a more generalized number for PC input latency.
The thing is Game code should already have allowances for up 200ms offset because 100ms upload, 100ms download.
My concern is A lot of ashes is tested locally.

36 -42 ms from keyboard and mouse? I don't think that's right. It should be 1 - 3 ms each. I'm referring to difference between input and server updating the game; so network latency.
Since many games measure inputs on release I can see how an unnecessary delay can and should be removed though. Input on press not release is important.
Yeah who knows how Ashes is handling testing.

Players could be split between 2 game state updaters. Every 32 or 48 ms or whatever half of players can be updated. Having everyone on the same game state is pretty good though; How do good PvP games usually do it lol.
Maybe 64 ms delay upload/download is good; 128 ms delay for everyone. Should be played out.

Sapiverenus

Azherae wrote: »

As someone who has done some specific tests in their Alpha-1, I will only say:

"On right track, don't worry about it."

I will hope that from your short experience here so far you understand why I am not inclined to say more.

Alright.

Sapiverenus

https://libreddit.tiekoetter.com/r/MechanicalKeyboards/comments/ekkutf/keyboard_latency_and_competitive_gaming/

Depressing stuff. Keyboards and mice apparently have a high delay.

daveywavey

Sapiverenus wrote: »

https://libreddit.tiekoetter.com/r/MechanicalKeyboards/comments/ekkutf/keyboard_latency_and_competitive_gaming/

Depressing stuff. Keyboards and mice apparently have a high delay.

So, we have to stop using keyboard&mouse now, too?

Hinotori

daveywavey wrote: »

Sapiverenus wrote: »

https://libreddit.tiekoetter.com/r/MechanicalKeyboards/comments/ekkutf/keyboard_latency_and_competitive_gaming/

Depressing stuff. Keyboards and mice apparently have a high delay.

So, we have to stop using keyboard&mouse now, too?

Youre not using a hotas system for mmorpgs? Cringe!

Noaani

daveywavey wrote: »

So, we have to stop using keyboard&mouse now, too?

Nah, you just have to pick the right one for your situation.

If you are spending time trying to minimize your latency between the client and server, and are not spending any time on your input or display latency, then you are kind of doing the same as running a Ryzen 9 7950x with a GTX1050 - any gains you make in one area are lost due to not paying attention to the other area.

That said, in an MMORPG, the same is true in reverse. If you optimize your input and display latency, but not your client/server latency, your gains from the former aren't going to really be noticed due to the latter.

Liniker

how about no? next.

Sapiverenus

If you are spending time trying to minimize your latency between the client and server, and are not spending any time on your input or display latency, then you are kind of doing the same as running a Ryzen 9 7950x with a GTX1050 - any gains you make in one area are lost due to not paying attention to the other area.

That said, in an MMORPG, the same is true in reverse. If you optimize your input and display latency, but not your client/server latency, your gains from the former aren't going to really be noticed due to the latter.

Neither of those are really true. If your system creates 50 ms of latency sending out the signal and 30 ms processing and displaying the ones recieved, you'll benefit either way from reducing it.
What are these useless platitudes dude.

Liniker wrote: »

how about no? next.

You're getting over 100ms of delay either way so eat it son lol. Nothing you will ever do about it either so suck it up.

Sapiverenus

daveywavey wrote: »

Sapiverenus wrote: »

https://libreddit.tiekoetter.com/r/MechanicalKeyboards/comments/ekkutf/keyboard_latency_and_competitive_gaming/

Depressing stuff. Keyboards and mice apparently have a high delay.

So, we have to stop using keyboard&mouse now, too?

Why are you acting like I'm telling you to do something?

Reality is kicking you in the balls lol stop looking at me like I did it.

Do you always start something by acting like others are telling you what to do?

Liniker

Sapiverenus wrote: »

You're getting over 100ms of delay either way so eat it son lol. Nothing you will ever do about it either so suck it up.

you so dumb I'm not even gonna bother... dude made his account in the forums 3 days ago and thinks he is teaching the world something new lmao

Sapiverenus

Liniker wrote: »

Sapiverenus wrote: »

You're getting over 100ms of delay either way so eat it son lol. Nothing you will ever do about it either so suck it up.

you so dumb I'm not even gonna bother... dude made his account in the forums 3 days ago and thinks he is teaching the world something new lmao

Do you think I was born 3 days ago? You confused?

People already play with over 100 ms delay. If you haven't already done something about it then you won't. If anything I've said is news to you and you do something about it then I've been teaching you.

I prefer the honorary, "Master". Master Sapiverenus.

Noaani

Sapiverenus wrote: »

Neither of those are really true. If your system creates 50 ms of latency sending out the signal and 30 ms processing and displaying the ones recieved, you'll benefit either way from reducing it.
What are these useless platitudes dude.

Yeah, but if you fix both, the gains are greater than the sum of the two.

This is why I used the example of a high powered CPU paired with a low.powered GPU. Most people would understand what the results of this are in relation to balancing the two out.

Sapiverenus

Noaani wrote: »

Sapiverenus wrote: »

Neither of those are really true. If your system creates 50 ms of latency sending out the signal and 30 ms processing and displaying the ones recieved, you'll benefit either way from reducing it.
What are these useless platitudes dude.

Yeah, but if you fix both, the gains are greater than the sum of the two.

This is why I used the example of a high powered CPU paired with a low.powered GPU. Most people would understand what the results of this are in relation to balancing the two out.

"any gains you make in one area are lost due to not paying attention to the other area."
It's a bit exaggerated.

daveywavey

Sapiverenus wrote: »

daveywavey wrote: »

Sapiverenus wrote: »

https://libreddit.tiekoetter.com/r/MechanicalKeyboards/comments/ekkutf/keyboard_latency_and_competitive_gaming/

Depressing stuff. Keyboards and mice apparently have a high delay.

So, we have to stop using keyboard&mouse now, too?

Why are you acting like I'm telling you to do something?

Reality is kicking you in the balls lol stop looking at me like I did it.

Do you always start something by acting like others are telling you what to do?

As if I'd ever let someone like you tell me what to do. Grow up. And since you're not going to be playing Ashes, piss off and let us enjoy the Ashes forum without you.

Sapiverenus

For abilities and moves that are slow to respond to input anyway; as in high intertia; the devs can [and maybe already coded in] an input priority change where those high inertia moves get put as lower prio to process and some "inertia reduction" flag so a different version of the ability gets used that reacts quicker but behaves the same from the player's viewpoint.
Hopefully these kinds of optimizations will be throughout the game.

Child Item