12 replies to this topic

[Superslicks]

its not the first time ive had problems hitting the little guys.
They just seem to lag all over the place, anybody else notice this?
regards
Josh

[jakucha]

Everybody has; there are netcode issues they're trying to fix.

[Noodlesoup]

i don't see them lag, you just have to lead ahead of them by a few steps to hit them with lasers/etc. their actual "location" in game is not being accurately reflected in the client (the in-game model is a few steps behind of where the actual person is in-game due to the server-centric/authority nature of the MWO game engine).

[pied]

Many players are also using this as an exploit and will very their ping so their hit location remains ever-changing.

[LordHarco]

Noodlesoup, on 14 December 2012 - 11:12 AM, said:

i don't see them lag, you just have to lead ahead of them by a few steps to hit them with lasers/etc. their actual "location" in game is not being accurately reflected in the client (the in-game model is a few steps behind of where the actual person is in-game due to the server-centric/authority nature of the MWO game engine).

haveing to lead with lasers (u know Light speed wpns) is LAG.

[dF0X]

The entire game has this lag, not just small mechs. It is the combination of server side processing lag and bad network code. It's just more obvious regarding light mechs.

[Dakkath]

They are working on the Netcode as quick as possible. There are some great improvements in the pipeline from what I've read in the Command Chair section of the forums.

[Sifright]

Dakkath, on 14 December 2012 - 01:19 PM, said:

They are working on the Netcode as quick as possible. There are some great improvements in the pipeline from what I've read in the Command Chair section of the forums.

Link please, The only post i can find on the subject is incredibly vague and brings no mention of when we can expect any kind of improvements.

Given how important the netcode is it's a little disconcerting to see that fixes for it don't seem to be coming any time soon.

[Dakkath]

Sifright, on 14 December 2012 - 01:46 PM, said:

Link please, The only post i can find on the subject is incredibly vague and brings no mention of when we can expect any kind of improvements.

Given how important the netcode is it's a little disconcerting to see that fixes for it don't seem to be coming any time soon.

http://mwomercs.com/...t-code-roadmap/

It's in the works!

[Sifright]

Dakkath, on 14 December 2012 - 01:49 PM, said:

http://mwomercs.com/...t-code-roadmap/

It's in the works!

I was hoping you were referencing a different post. As that particular post comes across as pr marketing spew to appease and silence complaints.... :\ It's not exactly substantive on details I guess that isn't to surprising but it's not confidence inspiring.

Quote

This is exactly what we are doing and have been doing for weeks now our network engineers have been going over the movement code and re-factoring it with everything they’ve learned so far working with CryEngines netcode and simplifying the system to make it more robust and less prone to bugs. This isn’t a fast process but as mentioned weeks ago we had meetings discussing the long term roadmap for Mech movement and weapon firing and we decided to take the time to get it right from the ground up.

yea, no the issue isn't about bugs or how 'robust' it is to unexpected scenarios. Sure that matters for when they reimplement collisions and knockdown.

But the main core of the problem Is the fact that your server at the moment only uses current server side positional data to determine whether shots hit or miss. Instead of taking into account the latency of the player involved and running a parse of where the enemy mech was position X milliseconds before hand (your ping basically) and checking if the shot would have hit if made with out latency effecting the positioning.

From reading the dev post it doesn't even seem that the problem is properly understood.
lastly the post only mentions having hopes of having something they can test internally soon...

Which means we are months and months away from getting something the end user gets to play with given that CW was supposed to be slated for 90 days away from the end of CB and isn't going to be coming out until what 2014?

Edited by Sifright, 14 December 2012 - 02:01 PM.

[Sifright]

bumpity

[Franchi]

Hit detection against lights has been terrible since the removal of knockdown.


ALL HAIL THE HALFCAT YOUR NEW KING!!!

[Stickjock]

I for one found what Matthew said in Net Code Roadmap extremely helpful... and, as the OP's question has been answered and the Official thread on the issue has been linked, I'm going to go ahead and close this one down...

*click*