The Mana World Forums

One last thing. I'm not sure if anyone has pointed this out yet, but tile constricted walking, while in the spirit of RO is really against the SoM feel. I'm not sure how the walking systems works in eAthena, so I can't say how to correct this. Still, something should be done.

After a bit of thought, I came up with this:

I'm not sure How Athena works, or if this can be done, but instead of integers for drawing position, use a double value. Pushing a direction moves the player .1 spaces in that direction. if a player moves past .5 or -.5 on his space, he is transferred to the opposite end of the next space.
This way, space moving will appear fluid without actually having to abandon the space grid. Just have the player's position at x,y while drawing his sprite at the double values.
Hit detection will still work, but players get alot more freedom.
Freedom is good.

Also, the animations are a bit overclocked on my machine. Maybe it's only me, but it looks like my character just drank about 5 espressos.

Also, the animations are a bit overclocked on my machine. Maybe it's only me, but it looks like my character just drank about 5 espressos.

LOL
yeah looks same here
hmm we allready talked a lot about movement
and about making better ways for this ...
well i forgot the result we came to.

but erm i will think about what you said

When we switch from a tile based moved to a pixel based SoM like movement it could be convenient to use floats, though I don't understand the 0.5 thingy, maybe you just mean it should be rounded back to ints when drawing the sprite.

In any case the control is a complex issue and the way the game is controlled now won't be able to stand up when more players join the game. I think I can see it lagging already. When we would do the same on a pixel level, the synchronozation problems will just increase.

So we'll be working towards a more indirect way of controlling the character. This will probably mean mouse movement is implemented but some people want to keep playing the game with keyboard as well. But the control can't be as direct as it is now.

If the controlling can be done via keyboard in 3D MMORPGs, why couldn't it be done so in 2D ones..? Perhaps mouse movement would eat away some of our uniqueness..?

Realistically speaking: if mouse movement is the only sensible way to handle it: then it's settled, no counter arguments here.

Uh the basic problem with direct control is that the client doesn't wait for the server to confirm he can actually perform the move. This is a problem because the client doesn't have the same game state as the server, and obstacles will not always be in the same places.

Real time 3D games with direct control have found ways around this, and I'm not familiar with MMORPGs here but only with how Unreal Tournament (and 03/04) solved the problem, which is by recording a history of input so that in the case of conflicts (too much difference in client and server), the server will correct the client and the client can re-play the input from there (because the correction is from the past) to reduce the amount of percieved lag by the player.

In our case I would rather prevent the client and server about ever disagreeing about the position of the player. I think we'll also safe quite some bandwidth here, because I don't see yet how direct control could be mapped on a low message frequency.

It should be fine if player characters are allowed to walk overtop of monsters and other PCs. This is how SoM did it. Besides, if we prevent people from walking over other PCs, PCs might start building walls and charging tolls for entrance to cities and dungeons. That's not cool...

Hmm yeah, in SoM you can walk overtop teammates, but not overtop enemies, you push them aside. Well I guess we still need to give this more thought before rewriting the movement system then.

In uo pass throught another player/monster cost you stamina. you cannot move without stamina.

We should think about our attributes and calculations