I'm sure they would notice the .08 ms difference in the 2 servers. We are dealing with super humans here. :p
For that kind of hardware in the US would be pretty costly monthly. I remember ENSL 2 being an i7 920 from hetzner. For that kind of hardware in the US they charge a stupid amount. Unless someone wants to fork out 150+ a month just for a couple NS servers. Xtinction is off a dual core 3.4 ghz system.
Dedicated > game server company. But i would be surprised if anyone would really know or be able to tell. I know quite a few people already who can't :) But they play it, so wherever they like it the best.
Thanks for the kernel patch btw jiriki, it does work better than the RT.
Stability is an the key word here. 200 or 300 FPS stable server is equally excellent. For hitscan weapons like LMG its important that the response happens in linear manner.
00:35 <jiriki> basically about server FPS
00:35 <jiriki> is there any difference between say stable 500FPS to 1000FPS except the 1ms response time
00:36 <puzl> I wouldn't think so
00:37 <puzl> you look at source and how they ticklocked the server to 66/100, for example
00:37 <jiriki> yeah
00:38 <puzl> but 500 should be the pinnacle of responsiveness IMO
00:38 <jiriki> so basically a server with 1000FPS won't be any better than serve rwith 200FPS except the 5ms added lag
00:39 <jiriki> or in other words, does it make difference if player is playing with 40ms ping on server with 5ms response time compared to a player whos playing on a server with 44ms ping and 1ms server response time
00:39 <puzl> well, in theory 10ms
00:39 <puzl> but average 5ms
00:41 <jiriki> I mean, it does make a difference if a player is playing on a server with 20FPS (50ms response time) with 5ms ping, compared to a player whos playing on a server with 50ms ping and 5ms respone time (latter is better yes?)
00:42 <puzl> you want a higher server fps definitely
00:42 <jiriki> so basically what's the other quality factor than server response time
00:42 <puzl> because lag compensation can make up for the network
00:42 <puzl> but low server framerate is gonna shaft you
00:42 <jiriki> I mean there's some kind of quantiative quality here
00:42 <puzl> stability is the most important thing
00:42 <puzl> if I was given a choice between a stable 100FPS that was almost guaranteed
00:42 <jiriki> but I'm not sure what is it exactly
00:43 <puzl> and a server fps spiking between 500 and 700, I'd take the 100fps
00:43 <jiriki> competitive players take the same pick aswell
00:43 <puzl> if the inter-frame gap is stable, then lc is very very good
00:43 <jiriki> are the hitboxes moving unlinearly if the server fps is changing?
00:43 <puzl> well, LC is simple
00:44 <puzl> you have a position, a known velocity and an inter-frame gap
00:44 <puzl> so you just project the position along the velocity for inter-gap miliseconds
00:45 <puzl> I mean, how much player input can there be over 10-30 miliseconds?
00:45 <jiriki> not much
00:45 <jiriki> but there can be
00:45 <puzl> yeah
00:45 <jiriki> you see, I cannot playing piano in rhythm if the audio lag is more than say 5-10ms
00:46 <puzl> and if you go to 100ms it can be huge
00:46 <puzl> especially for hitscan weapons like lmg or pistol
00:46 <jiriki> yeah I see
00:47 <jiriki> so that's why FPS matters
00:47 <puzl> indeed
00:47 <jiriki> also a lot of players feel a 200FPS/333FPS server better than 100FPS
00:47 <jiriki> it might be because of the higher DPS with LMG?
00:47 <puzl> so I presume this discussion is in relation to the bizarre movement with jetpack/blink at really high fps
00:48 <jiriki> yeah
00:48 <puzl> just to be clear, lmg damage is correct at high fps, and it is reduced at low fps
