Forged Alliance Forever Forged Alliance Forever Forums 2019-04-09T23:06:15+02:00 /feed.php?f=45&t=17376 2019-04-09T23:06:15+02:00 2019-04-09T23:06:15+02:00 /viewtopic.php?t=17376&p=173318#p173318 <![CDATA[Re: FAF Client-Server]]>
H-master wrote:
Alright, I was just wondering.

Do you know when ICE will be implemented?


Feel free to join our ICE test games (check out the channel #ice-testing on the Discord) to get more infos / help with testing :)

Statistics: Posted by Katharsas — 09 Apr 2019, 23:06


]]>
2019-04-06T17:55:20+02:00 2019-04-06T17:55:20+02:00 /viewtopic.php?t=17376&p=173244#p173244 <![CDATA[Re: FAF Client-Server]]> Statistics: Posted by Geosearchef — 06 Apr 2019, 17:55


]]>
2019-04-06T17:21:25+02:00 2019-04-06T17:21:25+02:00 /viewtopic.php?t=17376&p=173241#p173241 <![CDATA[Re: FAF Client-Server]]>
Katharsas wrote:
H-master wrote:
PhilipJFry wrote:the game is not written to support server side simulation and changing that would require a lot of work
not to mention that doing all calculation on the server side would shift 99% of the computation effort on the server which would result in a massive cost increase

as gieb already mentioned - improvements regarding connectivity (especially reconnecting) are on the way


Let's say it was magically possible to change the code to run server side, wouldn't it be possible then to use one of the players pc in a game a the server?

The game isn't that heavy anymore and with a good internet connection it might be possible.


Whats the point? You would still have connections problems (between host and others), which means you still would need ICE in the long run, which we will get anyway.

Only it would be worse because the hosting player would be able to kick players from running games by closing his connection to them and he would be also able to cheat (like proper simulation altering cheats), since the game simulation of the host would be the only source of thruth.

Then on top of that you would have to deal with the nightmare that a host migration would be (when host leaves the game), leading to probably MORE problems than there are right now.

There is a reason why competitive games with this networking model get shit on regularly (see some call of duty games).


Alright, I was just wondering.

Do you know when ICE will be implemented?

Statistics: Posted by H-master — 06 Apr 2019, 17:21


]]>
2019-04-06T16:02:54+02:00 2019-04-06T16:02:54+02:00 /viewtopic.php?t=17376&p=173238#p173238 <![CDATA[Re: FAF Client-Server]]>
H-master wrote:
PhilipJFry wrote:the game is not written to support server side simulation and changing that would require a lot of work
not to mention that doing all calculation on the server side would shift 99% of the computation effort on the server which would result in a massive cost increase

as gieb already mentioned - improvements regarding connectivity (especially reconnecting) are on the way


Let's say it was magically possible to change the code to run server side, wouldn't it be possible then to use one of the players pc in a game a the server?

The game isn't that heavy anymore and with a good internet connection it might be possible.


Whats the point? You would still have connections problems (between host and others), which means you still would need ICE in the long run, which we will get anyway.

Only it would be worse because the hosting player would be able to kick players from running games by closing his connection to them and he would be also able to cheat (like proper simulation altering cheats), since the game simulation of the host would be the only source of thruth.

Then on top of that you would have to deal with the nightmare that a host migration would be (when host leaves the game), leading to probably MORE problems than there are right now.

There is a reason why competitive games with this networking model get shit on regularly (see some call of duty games).

Statistics: Posted by Katharsas — 06 Apr 2019, 16:02


]]>
2019-04-05T16:27:31+02:00 2019-04-05T16:27:31+02:00 /viewtopic.php?t=17376&p=173223#p173223 <![CDATA[Re: FAF Client-Server]]>
PhilipJFry wrote:
the game is not written to support server side simulation and changing that would require a lot of work
not to mention that doing all calculation on the server side would shift 99% of the computation effort on the server which would result in a massive cost increase

as gieb already mentioned - improvements regarding connectivity (especially reconnecting) are on the way


Let's say it was magically possible to change the code to run server side, wouldn't it be possible then to use one of the players pc in a game a the server?

The game isn't that heavy anymore and with a good internet connection it might be possible.

Statistics: Posted by H-master — 05 Apr 2019, 16:27


]]>
2019-03-31T17:38:52+02:00 2019-03-31T17:38:52+02:00 /viewtopic.php?t=17376&p=173128#p173128 <![CDATA[Re: FAF Client-Server]]> not to mention that doing all calculation on the server side would shift 99% of the computation effort on the server which would result in a massive cost increase

as gieb already mentioned - improvements regarding connectivity (especially reconnecting) are on the way

Statistics: Posted by PhilipJFry — 31 Mar 2019, 17:38


]]>
2019-03-31T17:07:42+02:00 2019-03-31T17:07:42+02:00 /viewtopic.php?t=17376&p=173126#p173126 <![CDATA[Re: FAF Client-Server]]>
ICE has been WIP for quite some time now, once finished and publicly released should help connectivity issues remarkably. Latest ICE tests I've done have been a bliss compared to the current state of connectivity.

These are fairly old but offer some insight into ICE if you are interested:
- Killing connection problems - once and for all?
- Progress on ICE implementation

Statistics: Posted by Viba — 31 Mar 2019, 17:07


]]>
2019-03-31T16:57:59+02:00 2019-03-31T16:57:59+02:00 /viewtopic.php?t=17376&p=173125#p173125 <![CDATA[FAF Client-Server]]>
What's the reason behind *not* doing client-server with a possibility to reconnect if player loses connectivity?

Statistics: Posted by johndoe — 31 Mar 2019, 16:57


]]>