Yo Saccho, thanks for that. My head was beginning to hurt trying to wade through all the threads on these topics.
I had a basic understanding of the terms but not why the higher numbers were better. Or, who would benefit more from which changes.
After applying these settings to my server, I noticed more warping and lag overall. Once I returned the settings back to default, the lag and warping of players went away.....
Your server probably can't handle 90 tickrate. Try this
MaxClientRate=25000(Default)
NetServerMaxTickRate=60(Default is 30)
ConfiguredInternetSpeed=20000 (Default is 10000)
MaxClientRate=15000(Default)
MaxInternetClientRate=10000(Default)
NetServerMaxTickRate=60(Default is 30)
I have tried many different setups on my NFO server atm I am using
(AAEngine.ini)
MaxClientRate=50000
NetServerMaxTickRate=90
ConfiguredInternetSpeed=50000
MaxClientRate=50000
MaxInternetClientRate=50000
NetServerMaxTickRate=90
(AAGame.ini)
TotalNetBandwidth=50000
MaxDynamicBandwidth=50000
MinDynamicBandwidth=1000
A week ago I was running this with no issues myself. I change every now and then to see if there is any huge differences.
The main thing in editing these values to higher than default is your server must be able to handle the bandwidth and it must be capable of high cpu usage than what 30 tickrate does.
For my NFO server it is a Chicago Location with 3GB of ram and 2CPU cores.
*Also when ever I use a tickrate higher than 90 the server seems to respond well at first but overtime or some matches the hit detection seems to become awful. For this reason I don't go higher than 90 anymore.
After applying these settings to my server, I noticed more warping and lag overall. Once I returned the settings back to default, the lag and warping of players went away.....
Your server probably can't handle 90 tickrate. Try this
MaxClientRate=25000(Default)
NetServerMaxTickRate=60(Default is 30)
ConfiguredInternetSpeed=20000 (Default is 10000)
MaxClientRate=15000(Default)
MaxInternetClientRate=10000(Default)
NetServerMaxTickRate=60(Default is 30)
I have tried many different setups on my NFO server atm I am using
(AAEngine.ini)
MaxClientRate=50000
NetServerMaxTickRate=90
ConfiguredInternetSpeed=50000
MaxClientRate=50000
MaxInternetClientRate=50000
NetServerMaxTickRate=90
(AAGame.ini)
TotalNetBandwidth=50000
MaxDynamicBandwidth=50000
MinDynamicBandwidth=1000
A week ago I was running this with no issues myself. I change every now and then to see if there is any huge differences.
The main thing in editing these values to higher than default is your server must be able to handle the bandwidth and it must be capable of high cpu usage than what 30 tickrate does.
For my NFO server it is a Chicago Location with 3GB of ram and 2CPU cores.
*Also when ever I use a tickrate higher than 90 the server seems to respond well at first but overtime or some matches the hit detection seems to become awful. For this reason I don't go higher than 90 anymore.
HOOAH Server never ran smoother... I was getting 30 ping, now 12 to 20 not bad!!
Thanks again!
NP what is your server provider?
Colts, I actually have 4GB of Ram and 2 Xeon CPU Cores dedicated to my VM. I'll try a tick rate of 60 later on and see what happens. I'm not sure the number of cores would matter much since the game server code is still 32bit if I'm not mistaken.
Don't forget to mention that a high servertickrate not only means the server has to be stable enough to deal with it. The connection quality AND the fps(>tickrate) of every single player connected to the server should be expected to handle the high tickrate too or you will see again interpolation-errors/packetloss/"lag" on some end. Thats one of the reasons the default tickrate in ue is set to around 32 what is considered as the bottom of playerable (fps wise) for a client/player. Ergo everybody could be assumed to reach minimum 32 fps. (below that net stability is not the biggest problem anymore xD)
So rising the tickrate "skyhigh" is not the first priority you should consider for a good experience on the server. WHEN the server is using cshd. The fun part is.. on cshd the "hitdetection" quality is not at all limited by the tickrate discussed around here (thats more sshd related) . If you aim at the your opponents head and register a hit clientside he will be dead at the next (completed) servertick. 100% - IF ...
... there is no lag/packetlost on your connection or serverside (=data gets lost on the way) OR your hit wont count because he fails in a plausibility check (packetlost/lag on your side) OR your opponent had registered a confirmed hit on your head before your packet reached the server. (= your hit gets revoked)
The primary benefit with higher ticks on cshd servers is the improvement of timings , synchronization and motion changes. A generally better fluid feel and response (with interactions that require server-response) .
Problems of every other kind are more likely related to packetloss/connection problems or high pings/unstable server hitting their limits.
z!o[N]cya[;38554"]Problems of every other kind are more likely related to packetloss/connection problems or high pings/unstable server hitting their limits.
I would second that connection issues, network problems, etc are responsible for a large number of issues in many games that players end up blaming on "bad netcode" or "bad hitreg" or "bad optimization" instead of the responsible poor networking conditions.
Ok so I know, you can not please all the people all of the time
The other day 3 people where complaining I had my server set on Server Side hit detection, said they couldn't hit crap...
So I set it back to Client Side...
Today 2 guys are on and complaining I have the server set on Client Side and they can't hit crap..
LMAO!!!!!!!!!!
Ok so I am not very techy, if someone can tell me, what is the right way to set it I would appreciate that..
I was thinking, maybe they should go back to the shoot house if they can't hit crap and practice some more,
Anyway, they have peaked my curiosity! What is the general consensus?
Our servers have always been on Server Side only ever had one complaint.
Next day when he joined I told him i had put it on Client Side and he said "yes i know i can tell its better now"
The truth is i never change it.
Our servers have always been on Server Side only ever had one complaint.
Next day when he joined I told him i had put it on Client Side and he said "yes i know i can tell its better now"
The truth is i never change it.
Haha! the placebo effect
I gave my son a Rennie (for indigestion) and told him it was a good painkiller for his headache.. 20 mins later.. when asked, he said he was fine and dandy.. so this 3 hour bad headache he had was magically gone by an indigestion tablet
You'd be surprised what people act like when given a placebo, one of my close friends acted all drunk after we gave him non-alcoholic booze, telling him it was "strong stuff from abroad".. after a while he even said he felt sick and dizzy, never laughed so much in my life,
If my trollery drives you crazy, you'd better put on your seatbelt.
Hit detection is code, animation and network -related. A player's GAMING skill (or opinion) has nothing to do with it, which is why I can 'help' find bugs even if I suck as a player.
______
This has been a test of the emergency flame-fest system. Please do not adjust your set.
Guys, this was the question posted. Does anyone knows the required answers instead of railing off the topic for the sake of playing the old "CSHD vs SSHD vs the world" debate for the 166th time on the forum?
I'm a complete n00b on this, I'd think that most of them put a load on CPU and tickrates put load on the network as well. I don't know what the other numbers do, but I know that the tickrate is the frequency of server-client information exchange (so a tickrate of 30 means 30 information pack per second etc.). But as I said before, don't take my world for it, because I'm really not into server configs.
Theory and reality are not that different. In theory.
"Which resources they use" is tough to answer, but otherwise...
I don't know of an authoritative source for what MaxClientRate etc (the first section) do. Generally it's assumed that MaxClientRate plays a role in how much info is sent to clients over some unit of time (ticks or seconds). It's not clear to me how this relates to the DynamicBandwidth settings. The *TickRate settings control the rate at which the server both simulates the game time progression and communicates game changes with clients. Whether higher values are "better" has been argued in the past; similarly, I don't know of an authoritative account on whether higher tick rates increase CPU and memory load.
TotalNetBandwidth - GLOBALCONFIG. The total amount of bandwidth available for a listen server, which is split dynamically across net connections.
[Min/Max]DynamicBandwidth - The minimum and maximum amount of bandwidth dynamically set per connection.
Expect higher numbers here to potentially mean more info travelling over the network. Effects of that would depend on network conditions.
I think a better starting point may be this. Why do you want to change these settings in particular and what are you hoping to achieve from the changes? Game devs generally put some thought into the default values in terms of getting good player outcomes.
Hit detection is code, animation and network -related. A player's GAMING skill (or opinion) has nothing to do with it, which is why I can 'help' find bugs even if I suck as a player.
Making shots has everything to do with it, I can't listen to ones opinions on hit detection when he can't aim in general.
Hit detection is very noticeable when off to a good player.. it's so easy to notice guys are eating bullets.. I can tell I'm on a SSHD server in no time
Subjectivity is not science. You're talking about 'feelings' and saying that by itself is good enough.
How can you analyze hit detection without the proper tools? You have no idea what your packets are doing within the various network systems, and that's just for starters. What we visually think should be a hit....who's to say that's the fault of the game? (Not saying it isn't, either, just that we always have these discussions about "Knowing.")
Fact is, I'm not sure the proper tools exist.
I've actually put two computers and their monitors side by side and analyzed both with EXTERNAL video (IE a real camera) trying to find cases where what you see doesn't match. Haven't done it since Open Beta, but what science do YOU propose?
What makes you think hit detection changes, and what would be the cause?
______
This has been a test of the emergency flame-fest system. Please do not adjust your set.
Comments
I had a basic understanding of the terms but not why the higher numbers were better. Or, who would benefit more from which changes.
And you’re running a server out of your basement, check? hmmmm...
Are you StoneMountain Unit 64? Are you taking orders from Captain Cortez?
ARE YOU, YOLO???
Your server probably can't handle 90 tickrate. Try this
MaxClientRate=25000(Default)
NetServerMaxTickRate=60(Default is 30)
ConfiguredInternetSpeed=20000 (Default is 10000)
MaxClientRate=15000(Default)
MaxInternetClientRate=10000(Default)
NetServerMaxTickRate=60(Default is 30)
I have tried many different setups on my NFO server atm I am using
(AAEngine.ini)
MaxClientRate=50000
NetServerMaxTickRate=90
ConfiguredInternetSpeed=50000
MaxClientRate=50000
MaxInternetClientRate=50000
NetServerMaxTickRate=90
(AAGame.ini)
TotalNetBandwidth=50000
MaxDynamicBandwidth=50000
MinDynamicBandwidth=1000
A week ago I was running this with no issues myself. I change every now and then to see if there is any huge differences.
AAEngine.ini
MaxClientRate=128000
NetServerMaxTickRate=128
ConfiguredInternetSpeed=128000
MaxClientRate=128000
MaxInternetClientRate=128000
NetServerMaxTickRate=128
AAGame.ini
TotalNetBandwidth=128000
MaxDynamicBandwidth=128000
MinDynamicBandwidth=10000
The main thing in editing these values to higher than default is your server must be able to handle the bandwidth and it must be capable of high cpu usage than what 30 tickrate does.
For my NFO server it is a Chicago Location with 3GB of ram and 2CPU cores.
*Also when ever I use a tickrate higher than 90 the server seems to respond well at first but overtime or some matches the hit detection seems to become awful. For this reason I don't go higher than 90 anymore.
NP what is your server provider?
This has been a test of the emergency flame-fest system. Please do not adjust your set.
Colts, I actually have 4GB of Ram and 2 Xeon CPU Cores dedicated to my VM. I'll try a tick rate of 60 later on and see what happens. I'm not sure the number of cores would matter much since the game server code is still 32bit if I'm not mistaken.
So rising the tickrate "skyhigh" is not the first priority you should consider for a good experience on the server. WHEN the server is using cshd. The fun part is.. on cshd the "hitdetection" quality is not at all limited by the tickrate discussed around here (thats more sshd related) . If you aim at the your opponents head and register a hit clientside he will be dead at the next (completed) servertick. 100% - IF ...
... there is no lag/packetlost on your connection or serverside (=data gets lost on the way) OR your hit wont count because he fails in a plausibility check (packetlost/lag on your side) OR your opponent had registered a confirmed hit on your head before your packet reached the server. (= your hit gets revoked)
The primary benefit with higher ticks on cshd servers is the improvement of timings , synchronization and motion changes. A generally better fluid feel and response (with interactions that require server-response) .
Problems of every other kind are more likely related to packetloss/connection problems or high pings/unstable server hitting their limits.
I would second that connection issues, network problems, etc are responsible for a large number of issues in many games that players end up blaming on "bad netcode" or "bad hitreg" or "bad optimization" instead of the responsible poor networking conditions.
NP what is your server provider?[/quote]
Lowpingservers! Mine is in Chicago..
Lowpingservers! Mine is in Chicago..
[/quote]
ah yes the lowping servers seem to respond very well to those settings
I want to try different settings for my server.
Could somone explain what the settings below do , and also wich ressources they use on the server? (ram , cpu ect).
MaxClientRate=50000
NetServerMaxTickRate=90
ConfiguredInternetSpeed=50000
MaxClientRate=50000
MaxInternetClientRate=50000
NetServerMaxTickRate=90
(AAGame.ini)
TotalNetBandwidth=50000
MaxDynamicBandwidth=50000
MinDynamicBandwidth=1000
Thanking you in advance
Jay
[EF] =Elite Fighters=
Server IP:82.255.178.154:7778
Our servers have always been on Server Side only ever had one complaint.
Next day when he joined I told him i had put it on Client Side and he said "yes i know i can tell its better now"
The truth is i never change it.
This has been a test of the emergency flame-fest system. Please do not adjust your set.
Haha! the placebo effect
I gave my son a Rennie (for indigestion) and told him it was a good painkiller for his headache.. 20 mins later.. when asked, he said he was fine and dandy.. so this 3 hour bad headache he had was magically gone by an indigestion tablet
You'd be surprised what people act like when given a placebo, one of my close friends acted all drunk after we gave him non-alcoholic booze, telling him it was "strong stuff from abroad".. after a while he even said he felt sick and dizzy, never laughed so much in my life,
#Support Comp Mode
https://www.youtube.com/channel/UCN4YhM6jUB2MxVj8i3b9rhw
This has been a test of the emergency flame-fest system. Please do not adjust your set.
Guys, this was the question posted. Does anyone knows the required answers instead of railing off the topic for the sake of playing the old "CSHD vs SSHD vs the world" debate for the 166th time on the forum?
I'm a complete n00b on this, I'd think that most of them put a load on CPU and tickrates put load on the network as well. I don't know what the other numbers do, but I know that the tickrate is the frequency of server-client information exchange (so a tickrate of 30 means 30 information pack per second etc.). But as I said before, don't take my world for it, because I'm really not into server configs.
I don't know of an authoritative source for what MaxClientRate etc (the first section) do. Generally it's assumed that MaxClientRate plays a role in how much info is sent to clients over some unit of time (ticks or seconds). It's not clear to me how this relates to the DynamicBandwidth settings. The *TickRate settings control the rate at which the server both simulates the game time progression and communicates game changes with clients. Whether higher values are "better" has been argued in the past; similarly, I don't know of an authoritative account on whether higher tick rates increase CPU and memory load.
Expect higher numbers here to potentially mean more info travelling over the network. Effects of that would depend on network conditions.
I think a better starting point may be this. Why do you want to change these settings in particular and what are you hoping to achieve from the changes? Game devs generally put some thought into the default values in terms of getting good player outcomes.
Making shots has everything to do with it, I can't listen to ones opinions on hit detection when he can't aim in general.
Hit detection is very noticeable when off to a good player.. it's so easy to notice guys are eating bullets.. I can tell I'm on a SSHD server in no time
#Support Comp Mode
https://www.youtube.com/channel/UCN4YhM6jUB2MxVj8i3b9rhw
How can you analyze hit detection without the proper tools? You have no idea what your packets are doing within the various network systems, and that's just for starters. What we visually think should be a hit....who's to say that's the fault of the game? (Not saying it isn't, either, just that we always have these discussions about "Knowing.")
Fact is, I'm not sure the proper tools exist.
I've actually put two computers and their monitors side by side and analyzed both with EXTERNAL video (IE a real camera) trying to find cases where what you see doesn't match. Haven't done it since Open Beta, but what science do YOU propose?
What makes you think hit detection changes, and what would be the cause?
This has been a test of the emergency flame-fest system. Please do not adjust your set.
For the most part the hit detection in this game is pretty good
#Support Comp Mode
https://www.youtube.com/channel/UCN4YhM6jUB2MxVj8i3b9rhw