• There is NO official Otland's Discord server and NO official Otland's server list. The Otland's Staff does not manage any Discord server or server list. Moderators or administrator of any Discord server or server lists have NO connection to the Otland's Staff. Do not get scammed!

have you noticed that monsters are very dumb in tfs 1.x in comparision with older releases?

have you?

  • yes

    Votes: 8 80.0%
  • no

    Votes: 2 20.0%

  • Total voters
    10

johnsamir

Active Member
Joined
Oct 13, 2009
Messages
232
Solutions
1
Reaction score
35
Location
Nowhere
Hello Community

Lately i have moved into tfs 1.3 and have noticed that all the monsters in this distribution behave overly "dumb" and quite slow too.
Is completely strange how monsters behave in this distribution, that have been flattered by many otland users and developers under the pretext of it's perfomance, which is true. It's perfomance is unique in comparision to older distributions, also is quite cool that it doesn't consume almost any cpu usage and the syntax used in newer tfs easier is than it used to be in the old tfs 0.4.
But again barring that, have you noticed how overly "dumb" and slow are monsters in this distribution? For example i think and figured out too that monster find path, attack better and walk faster or smater in older distributions like OThire, OTX2, TFS 0.4 i have tested the same datapack monsters in the already mentioned distribution, the monsters, datapack was taken from othire, undeniably not before a thorough have done the corresponding changes to make it work.
so i did the changes in the monster folder to test it with tfs 1.3 and at least for me is disappointing view how the monster works in here, a level 8 can run away easily from a cyclops, giant spiders with speed sucks, same for orc berserker and many monster more. have do you noticed? What has been your experience with monster behavior in this distribution? is there a workarround that could be made to "solve this" ?
 
Last edited:

4Nathu4

valoria.online
Joined
Oct 27, 2009
Messages
318
Solutions
9
Reaction score
178
GitHub
4mrcn4
 

oen432

Legendary OT User
Joined
Oct 3, 2014
Messages
1,428
Solutions
49
Reaction score
1,523
Location
Poland
GitHub
Oen44
Amazing this still hasnt been fixed XDDDDDD
Entire TFS repo is a mess and should be ditched long time ago. It peaked in 2010s and is a shame in 2021.
Anyway this "fix" is not a proper fix. It affects performance significantly based on number of active monsters (not in idle state).
 

ralke

(҂ ͠❛ ෴ ͡❛)ᕤ
Joined
Dec 17, 2011
Messages
898
Solutions
21
Reaction score
428
Location
Santiago - Chile
Entire TFS repo is a mess and should be ditched long time ago. It peaked in 2010s and is a shame in 2021.
Anyway this "fix" is not a proper fix. It affects performance significantly based on number of active monsters (not in idle state).

Mark said that doesn't affects perfomance unless you have over 500 players running around at dense hunting grounds. How true is this? If you could explain how this significantly affects in a common server (0-100 players) would be great,

Regards!
 

oen432

Legendary OT User
Joined
Oct 3, 2014
Messages
1,428
Solutions
49
Reaction score
1,523
Location
Poland
GitHub
Oen44
Mark said that doesn't affects perfomance unless you have over 500 players running around at dense hunting grounds. How true is this? If you could explain how this significantly affects in a common server (0-100 players) would be great,

Regards!
With 100 players it won't be noticeable (probably around 5% increased CPU usage). 500 players hunting at the same time is a stretch imo, but around 300+ you will notice CPU usage being higher than usual. Performance hit doesn't mean that the server will start lagging as hell.
For most servers it won't make a difference at all but that doesn't mean it should be implemented.

Checked TFS repo if there were any significant optimizations for pathfinding since this fix was posted in 2015 and... nothing. So the statement is still valid.
 

dami1310

◄ Unidentified ►
Joined
Jan 27, 2013
Messages
643
Solutions
9
Reaction score
544
Location
Poland

xKrazyx

Old School
Joined
Oct 23, 2012
Messages
873
Solutions
6
Reaction score
609
With 100 players it won't be noticeable (probably less than 10% increased CPU usage). 500 players hunting at the same time is a stretch imo, but around 300+ you will notice CPU usage being higher than usual.
For most servers it won't make a difference at all but that doesn't mean it should be implemented.
ME: Creates bad code effecting CPU/Memory performance

Also me: buys more ram and higher rated CPU after 6 months due to moores law.

But for real though its crazy how many times ive run into something where faster hardware just saves me when writing bad code haha
I know that's not the correct approach, but sometimes when asking myself should I spend 40-80 hours trying to figure out a better solution, or just buy 32 gigs of more ram, I will buy the RAM xD'

Im not sure at all if it applies in this case, your post just made me think of that.
 

oen432

Legendary OT User
Joined
Oct 3, 2014
Messages
1,428
Solutions
49
Reaction score
1,523
Location
Poland
GitHub
Oen44
ME: Creates bad code effecting CPU/Memory performance

Also me: buys more ram and higher rated CPU after 6 months due to moores law.

But for real though its crazy how many times ive run into something where faster hardware just saves me when writing bad code haha
I know that's not the correct approach, but sometimes when asking myself should I spend 40 hours trying to figure out a better solution, or just buy 32 gigs of more ram, I will buy the RAM xD
There is a limit to what your hardware can handle.
Memory leaks? Just get 128GB RAM. Until those leaks will fill that RAM :p
 
OP
OP
J

johnsamir

Active Member
Joined
Oct 13, 2009
Messages
232
Solutions
1
Reaction score
35
Location
Nowhere
ME: Creates bad code effecting CPU/Memory performance

Also me: buys more ram and higher rated CPU after 6 months due to moores law.

But for real though its crazy how many times ive run into something where faster hardware just saves me when writing bad code haha
I know that's not the correct approach, but sometimes when asking myself should I spend 40-80 hours trying to figure out a better solution, or just buy 32 gigs of more ram, I will buy the RAM xD'

Im not sure at all if it applies in this case, your post just made me think of that.
it's a workaround that looks good to me hahaha
Post automatically merged:

sorry for double post admins and thank @4Nathu4 i've tested the code and with it, the monster are still extremly dumb and slow
check this xD if you guys don't want to see it entirely, watch it from sec 50 onwards.
video
 
Last edited:
OP
OP
J

johnsamir

Active Member
Joined
Oct 13, 2009
Messages
232
Solutions
1
Reaction score
35
Location
Nowhere
Isn't this how they behave even on RL Tibia?
don't know but in comparision at least with old engines, how monster behave is a lame, monsters are crippled or what
ex: in othire with a level 8 running away from a cyclops is quite hard and you feel like you must press the button harder to run away
in this engine i can even stop walking few secs
 
Last edited:

Terotrificy

Well-Known Member
Joined
Oct 18, 2020
Messages
231
Solutions
9
Reaction score
75
Location
Santiago, Chile.
don't know but in comparision at least with old engines, how monster behave is a lame, monsters are crippled or what
ex: in othre with a level 8 running away from a cyclops is quite hard and you feel like you must press the button harder to run away
in this engine i can even stop walking few secs
Still he have a point, you can notice the dance between the bushes, where the cyclops get stucked in a loop of 2 sqms movement, i think @kay said in some thread that rl tibia monster AI had a "queue" when getting the fastest route to the player, making it impossible to fool it in 2 sqm movement or around big rocks. So far no engine have this working as it should that i know.
 

zbizu

Legendary OT User
Joined
Nov 22, 2010
Messages
3,146
Solutions
20
Reaction score
2,299
Location
Poland
GitHub
Zbizu
Amazing this still hasnt been fixed XDDDDDD

It isn't 5 minutes job. A new implementation has to be written. Spawns loop, aggro trigger, onThink, pathfinding and possibly more.

This was fixed, but the fix isn't public and both of us know who might be having the code.
 
OP
OP
J

johnsamir

Active Member
Joined
Oct 13, 2009
Messages
232
Solutions
1
Reaction score
35
Location
Nowhere
It isn't 5 minutes job. A new implementation has to be written. Spawns loop, aggro trigger, onThink, pathfinding and possibly more.

This was fixed, but the fix isn't public and both of us know who might be having the code.
iryont? have read something about it at github somwhere, just today hah
 
Last edited:
Top