Possible way to fix bad hitreg!

Pages: 1 … 5 6 7 8 9 10

Index »
Games »
Battlefield Series »
Battlefield 2 »
Help »
Possible way to fix bad hitreg!

2006-11-20 07:40:00

#151

Lucien: Fantasma Parastasie; +1,451|6861

Well, it was interesting to play around with your minds. You see, like in real tests, there is a placebo group: in this case, it was me withholding some information.

The interp command controls how often updates are sent to the server. The lower the value, the more often. This stresses the server, hence why it is set to 100 normally. Of course, it won't ruin it, but if 32 players set it to 10 it will cause problems.

The less players, the lower the value. 10 minimum.
Using this command is a trade-off: the lower, the more accurate, but the easier you are to hit.

I recommend between 30 to 60 on pubs. On >48 servers 60-100

2006-11-20 13:21:25

#152

Scorpion0x17: can detect anyone's visible post count...; +691|6974|Cambridge (UK)

sneeky_fluff wrote:
I may be off the mark here but if the algorithm used to calculate hit reg is wave based within a set of fixed values (say like a sine wave) then the exceedingly high values of 400 could realistically have the same effect as having a low value of 40 if it hit the same point on the wave.

If that were the case you would have to find out what is the maximum height of each wave segment is and then work out the best value within that, or use a mulitple to hit the same height on the next segment.

Again, its only guesswork but it would explain the differential between the experiences of high and low values. Of course this assumes that additional variables are not influencing (which most likely is the case based on some of the comments made)

Hmm... This kinda makes some sense. But then, why would you have any kind of wave equation with any part of the hitreg calculations? Particularly given what this variable is doing...

But any way, I've just now caught up with the latest post and seen SargeV1.4's changes...

One question:

SargeV1.4 wrote:
Well, it was interesting to play around with your minds. You see, like in real tests, there is a placebo group: in this case, it was me withholding some information.

The interp command controls how often updates are sent to the server. The lower the value, the more often. This stresses the server, hence why it is set to 100 normally. Of course, it won't ruin it, but if 32 players set it to 10 it will cause problems.

The less players, the lower the value. 10 minimum.
Using this command is a trade-off: the lower, the more accurate, but the easier you are to hit.

I recommend between 30 to 60 on pubs. On >48 servers 60-100

If you knew this from the begining, why didn't you just tell us? And don't give me that 'placebo group' bollox - we're not your guinea pigs to test as you please.

2006-11-20 13:26:41

#153

SlightlySto0pid: Member; +7|6658|New York

What about extrapolation

2006-11-20 14:41:59

#154

Lucien: Fantasma Parastasie; +1,451|6861

Scorpion0x17 wrote:
If you knew this from the begining, why didn't you just tell us? And don't give me that 'placebo group' bollox - we're not your guinea pigs to test as you please.

Despite knowing this beforehand, the results from personal experience weren't coherent. So, I decided to use the hundreds of people here to find out what the results were and in turn help them.

2006-11-20 16:56:37

#155

SlightlySto0pid: Member; +7|6658|New York

So basically, leave it on default?

2006-11-20 17:00:30

#156

Towelly: It's A State Of Mind; +399|6799|Your attic

I have mine set to 400 and I swear to god I hit more stuff than I did before, weird, screw you and your placebo effect it works for me.

2006-11-21 02:12:27

#157

SMELLYMONKEY: Member; +5|6586|SOMERSET UK

SargeV1.4 wrote:
killer21 wrote:
I don't have a problem with hit detection. I just know how to aim. Maybe you should give it a try.
http://www.clanbase.com/rating.php?&lid=2172
http://www.clanbase.com/claninfo.php?cid=1097201

yeah, I'm so terrible at infantry.. I mean, so bad we even get paid to go to LAN events!

LMFAO I want that as a sig.

2006-11-22 03:53:49

#158

Gryphen_au: Member; +5|6688

Well it works out for me I'm leaving mine at 50 amd dropping the other down to .06000.... or whatever, most of the time it is an improvement. Plus if it makes me think I'm shooting better... well... cool

2006-11-22 10:24:03

#159

ECH: Member; +50|6851|Some where near you

Ok, lets look at this from a different angle between Interpolation and Extrapolation as it relates to "TIME".
Instead of looking at the drawing time of the game itself is it possible that it has something to do with the server and client? Is it possible that these 2 commands make much more sense in time between server and client vs. drawing time on screen in a map on BF2?

If this is correct then it would make sense that the ExtrapolationTime would have something to do with client's ability to receive data from the server. Set at 1200 that allows a buffer of 1200 to properly receive and process.

If that is correct then the InterpolationTime would have something to do with the client sending information to the server. At 100 their is less time for the client to send less information to the sever.

Imagine yourself on a propped filled stage of BF2 Karkand. You have the roads, the buildings, gun placements, tanks, APC, etc. This represents the server. All that is missing are the players or clients. The clients fill the server (AKA players fill the stage). The only information that the server sends are the actions of other players. Thus the reason for ExtrapolationTime of 1200 is set for the client to recieve. All that the players do (AKA clients) is send information to the server regarding them self as a player and their actions on the server (AKA stage). Thus requiring Interpolation of only 100. Which is less information in less time.

By reducing these both of these values you decrease the amount of "Time" that the information is sent or received. For example:

If you decrease the Interpolation you may actually decrease the time it takes the client to send date to the server. IE. clients sending information faster to the server which is dependent on your ISP connection/computer.

If you decrease the Exterpolation you may actually decrease the time it takes server to send date to the client. IE clients receiving information faster from the server. Again, dependent on your ISP/computer.

However, in some cases lowering one and not the other may procide the best results for others.

Take this with a grain of salt as this just a theory at best.

Last edited by ECH (2006-11-22 10:40:57)

2006-11-24 11:40:05

#160

Scorpion0x17: can detect anyone's visible post count...; +691|6974|Cambridge (UK)

ECH wrote:
Ok, lets look at this from a different angle between Interpolation and Extrapolation as it relates to "TIME".
Instead of looking at the drawing time of the game itself is it possible that it has something to do with the server and client?...

...If you decrease the Interpolation you may actually decrease the time it takes the client to send date to the server. IE. clients sending information faster to the server which is dependent on your ISP connection/computer.

If you decrease the Exterpolation you may actually decrease the time it takes server to send date to the client. IE clients receiving information faster from the server. Again, dependent on your ISP/computer.

However, in some cases lowering one and not the other may procide the best results for others.

Take this with a grain of salt as this just a theory at best.

Hmm... Interesting theory. And I can see your thinking, however, the interpolation and extrapolation are client side variables, whereas what you suggests requires cooperation between both client and server. Though this doesn't disprove your theory, I feel that, as they are client-side variables, they most probably control something that is purely handled on the client side. Also their names, 'interpolation' and 'extrapolation' suggest they're involved in the client side process of calculating higher-definition data from the low-def sample data recieved by the client from the server.

2006-11-24 12:24:52

#161

ECH: Member; +50|6851|Some where near you

Scorpion0x17 wrote:
ECH wrote:
Ok, lets look at this from a different angle between Interpolation and Extrapolation as it relates to "TIME".
Instead of looking at the drawing time of the game itself is it possible that it has something to do with the server and client?...

...If you decrease the Interpolation you may actually decrease the time it takes the client to send date to the server. IE. clients sending information faster to the server which is dependent on your ISP connection/computer.

If you decrease the Exterpolation you may actually decrease the time it takes server to send date to the client. IE clients receiving information faster from the server. Again, dependent on your ISP/computer.

However, in some cases lowering one and not the other may procide the best results for others.

Take this with a grain of salt as this just a theory at best.
Hmm... Interesting theory. And I can see your thinking, however, the interpolation and extrapolation are client side variables, whereas what you suggests requires cooperation between both client and server. Though this doesn't disprove your theory, I feel that, as they are client-side variables, they most probably control something that is purely handled on the client side. Also their names, 'interpolation' and 'extrapolation' suggest they're involved in the client side process of calculating higher-definition data from the low-def sample data recieved by the client from the server.

I see what you are saying. Is it then possible that the servers are some sort of PSP mirror (of sorts) and we need to factor in the other clients on that server besides just the server itself?

Last edited by ECH (2006-11-24 12:25:29)

2006-11-24 14:45:57

#162

ECH: Member; +50|6851|Some where near you

For those that have EF, SF or AF installed on your hard drive could you take a minute and check the usersettings.con file? I want to know if the folks in Canada tweaked the the settings below vs. what is found in BF2.

SettingsManager.floatSet GSDefaultLatencyCompensation
SettingsManager.U32Set GSExtrapolationTime
SettingsManager.U32Set GSInterpolationTime
SettingsManager.U32Set GSPerfLogAtPlayerCount

You can find your BF2 usersettings.con file here:
X:\Program Files\EA GAMES\Battlefield 2\mods\BF2\Settings

You can find the other maps in here:
X:\Program Files\EA GAMES\Battlefield 2\mods\xpack\Settings

Could you be so kind as to post if the settings are the same or different? If they are different could you post which command value is different?

Last edited by ECH (2006-11-24 14:46:12)

2006-11-25 02:05:52

#163

Scorpion0x17: can detect anyone's visible post count...; +691|6974|Cambridge (UK)

I have SF only installed and they're the identical for me.

2006-11-25 04:21:26

#164

[pt] KEIOS: srs bsns; +231|6861|pimelteror.de

Code:

rem
rem Settingsfile automatically saved by bf2
rem
SettingsManager.boolSet GSClPunkBuster 1
SettingsManager.floatSet GSDefaultLatencyCompensation 0.100000
SettingsManager.boolSet GSExtrapolateFrame 0
SettingsManager.U32Set GSExtrapolationTime 1200
SettingsManager.U32Set GSInterpolationTime 100
SettingsManager.U32Set GSPerfLogAtPlayerCount 30
SettingsManager.stringSet GSPlayerName "[pt] KEIOS"
SettingsManager.boolSet GSShowNetGraph 0
SettingsManager.boolSet GSUseObjectCache 1

I have SF, EF and AF but i can´t see any difference.

Last edited by [pt] KEIOS (2006-11-25 04:22:14)

2006-11-27 21:16:11

#165

Gryphen_au: Member; +5|6688

I found the best way wasn't to change the con file but to actually do it in-game using the console, that way I can modify it for each round, depending upon what server I'm on, asl ong as the game is up and running then all I have to do is after entering them for the first round is hit the up arrow and be able to change jsut the number, means I can tweak it in game to see what affect it has.

2006-12-06 07:21:59

#166

[pt] KEIOS: srs bsns; +231|6861|pimelteror.de

so what kind of experiences did you make with it?

2006-12-28 20:08:35

#167

_j5689_: Dreads & Bergers; +364|6925|Riva, MD

I might try this at some point.

2007-01-09 05:30:22

#168

B00MH3ADSH0T: Fresh NoobCaeks Here; +118|6599|Penrith,Nsw, Aus

Isnt this against the rules if it gives you a unfair advantage over another player, correct me if im wrong but if one player has bad hit reg and you dont isnt that classed an advantage?

2007-01-09 05:37:57

#169

DonFck: Hibernator; +3,227|6840|Finland

B00MH3ADSH0T wrote:
Isnt this against the rules if it gives you a unfair advantage over another player, correct me if im wrong but if one player has bad hit reg and you dont isnt that classed an advantage?

I wouldn't say that it would be considered an unfair advantage per se. Although IMHO the game coding should have been made if possible, in a way that the interpolation time would automatically be set each time you join a server, so that these kinds of "tweaks" shouldn't have to be made. You are correct that it's a tad unfair, due to not many actually performing this "modification". However, so is playing a round where one team consists of Sgt.Maj to General and the other team of privates to Gunnys.. For example..

Dunno.. I'll have to try the tweak first to see if it actually works, and how well.

Last edited by DonFck (2007-01-09 05:38:59)

I need around tree fiddy.

2007-01-28 11:32:53

#170

ssonrats: Member; +221|6854

kingcoyote wrote:
Ive gotten the dust cloud nonsense a lot since 1.4. I hit a buddy of mine 4 times with the l96a1 the other day and all I got was dust clouds off his jacket.

LOL +1

2007-01-30 00:58:15

#171

redilS: Member; +1|6506

Code:

rem
rem Settingsfile automatically saved by bf2
rem
SettingsManager.boolSet GSClPunkBuster 1
SettingsManager.floatSet GSDefaultLatencyCompensation 0.100000
SettingsManager.boolSet GSExtrapolateFrame 0
SettingsManager.U32Set GSExtrapolationTime 1200
SettingsManager.U32Set GSInterpolationTime 100
SettingsManager.U32Set GSPerfLogAtPlayerCount 30
SettingsManager.stringSet GSPlayerName "[pt] KEIOS"
SettingsManager.boolSet GSShowNetGraph 0
SettingsManager.boolSet GSUseObjectCache 1

Hi there guys, I think I got a goodie for you all...

1. Usersettings.con
Every single setting in usersettings.con is server related and does _absolutely nothing_ on client side. These settings are beeing discussed since early 2006 and nobody realized the 'GS' prefix? GS => GameServer

2. Concerning 'GSDefaultLatencyCompensation'
Someone should've checked bf2.exe or bf2_w32ded.exe - even on serverside, this setting does absolutely nothing until 'GSUseLatencyCompensation' is enabled (default: off).

3. How to fix the hit-detection at least on unranked servers (dunno if allowed on ranked ones)
Hit-detection is handled on client side by default - maybe to take load off of the servers.
Add 'settingsmanager.boolset GSUseClientSidePrediction 0' to your servers usersettings.con, save the file, make the file read-only (otherwise it will get overwritten when shutting down the server) and restart the server.
--->Mark my words: You DO NOT need to enable 'GSUseLatencyCompensation' at all!

Voila... model & hitbox are sync. Try it, you'll definately notice the difference. For all disbelievers: rename your bf2.exe or bf2_w32ded.exe to .txt, open it with a text editor and search for the relevant string(s).

Feel free to spread this information - bf2s.com is only the second board I posted this information.

Greetings,
redilS

//EDIT
Oh come on guys, so many views and no feedback until now? Not even a single harassment? o_O Did this "tweak" work for you?

Last edited by redilS (2007-01-31 15:41:25)

2007-02-05 04:13:35

#172

Redback00: Member; +51|6782

redilS wrote:
Code:
rem
rem Settingsfile automatically saved by bf2
rem
SettingsManager.boolSet GSClPunkBuster 1
SettingsManager.floatSet GSDefaultLatencyCompensation 0.100000
SettingsManager.boolSet GSExtrapolateFrame 0
SettingsManager.U32Set GSExtrapolationTime 1200
SettingsManager.U32Set GSInterpolationTime 100
SettingsManager.U32Set GSPerfLogAtPlayerCount 30
SettingsManager.stringSet GSPlayerName "[pt] KEIOS"
SettingsManager.boolSet GSShowNetGraph 0
SettingsManager.boolSet GSUseObjectCache 1
Hi there guys, I think I got a goodie for you all...

1. Usersettings.con
Every single setting in usersettings.con is server related and does _absolutely nothing_ on client side. These settings are beeing discussed since early 2006 and nobody realized the 'GS' prefix? GS => GameServer

2. Concerning 'GSDefaultLatencyCompensation'
Someone should've checked bf2.exe or bf2_w32ded.exe - even on serverside, this setting does absolutely nothing until 'GSUseLatencyCompensation' is enabled (default: off).

3. How to fix the hit-detection at least on unranked servers (dunno if allowed on ranked ones)
Hit-detection is handled on client side by default - maybe to take load off of the servers.
Add 'settingsmanager.boolset GSUseClientSidePrediction 0' to your servers usersettings.con, save the file, make the file read-only (otherwise it will get overwritten when shutting down the server) and restart the server.
--->Mark my words: You DO NOT need to enable 'GSUseLatencyCompensation' at all!

Voila... model & hitbox are sync. Try it, you'll definately notice the difference. For all disbelievers: rename your bf2.exe or bf2_w32ded.exe to .txt, open it with a text editor and search for the relevant string(s).

Feel free to spread this information - bf2s.com is only the second board I posted this information.

Greetings,
redilS

//EDIT
Oh come on guys, so many views and no feedback until now? Not even a single harassment? o_O Did this "tweak" work for you?

I don't understand what this means. Are you saying that you have to make the changes at the server and any local changes don't help?

2007-02-05 10:39:46

#173

redilS: Member; +1|6506

Exactly

Every GSxxxxxx command is only relevant on server side. Moreover you have to differentiate between a dedicated and ingame hosted server, because some commands are only relevant on dedicated servers (e.g. GSShowNetGraph which switches a dedicated servers console between default game imformation and detailed network statistics).

2007-02-07 11:37:21

#174

GbeTech: Member; +9|6706

I don't get it. You say the lower the value, the more frequently the server is updated = more chances to actually hit and thus fix your bad hitreg.

But someone at the first page posted he changed it to 400... That means making the hitreg 4 times worse than default!

So what does increasing and decresing do.

Last edited by GbeTech (2007-02-07 11:39:39)

2007-02-08 18:10:04

#175

joleonline: Member; +0|6496

Hmm strange.
I checked my server files and there is no usersettings.con ?
Just : banlist.con
maplist.con
mm_bans.xml
modmanager.con
reservedslots.con
serversettings.con

Where can I add 'settingsmanager.boolset GSUseClientSidePrediction 0' ?

Pages: 1 … 5 6 7 8 9 10

Index »
Games »
Battlefield Series »
Battlefield 2 »
Help »
Possible way to fix bad hitreg!

Board footer

Jump to

BF2S Forums Possible way to fix bad hitreg!

sneeky_fluff wrote:

SargeV1.4 wrote:

Scorpion0x17 wrote:

SargeV1.4 wrote:

killer21 wrote:

ECH wrote:

Scorpion0x17 wrote:

ECH wrote:

Code:

B00MH3ADSH0T wrote:

kingcoyote wrote:

Code:

redilS wrote:

Code:

Board footer