I was checking out this thread in UMSMA the other day when I realized that so many maps could use EUDs to detect burrowing instead of that 1 pixel + DT system. I've assembled a list of pros and cons for you guys.



Cons win out.

[/discuss]

Post has been edited 5 time(s), last time on Jan 12 2010, 3:22 pm by ImagoDeo.

http://farty1billion.dyndns.org/EUDDB/?pg=ref&a=unitnode

Down a little bit is the offset for burrowing. One con removed.

If someone found the unitnode start address for macs, that would be another con removed(you would have to have some way for the player to choose whether they are on mac or not)

I believe you would only need 1700 triggers to detect every unit's current burrow status, but for AoS's or RPGs with a fixed amount of lives, it would be a far simpler matter of having some preplaced units.

If you're doing the normal detection method correctly it would be completely invisible and no DT would show up.

It is essentially only viable for preplaced units. You also didn't list the fact that EUD addresses can change with patches, essentially mortalizing your map if you use them.

Thanks for that. Apparently unearthing it wouldn't be so hard after all.

Yeah, that's a simple matter of a beacon-and-civ setup.

Yes, but you have to make sure that no units are spawned before the game detects that you have died and creates a placeholder for the unit ID. Your entire system could be messed up by being applied to a spawned monster instead of the player's hero.


The current method still alters physical reality in the game. You'll see the distortion of the cloaked DT even if it belongs to another player. And it still takes up space in-game. AND it can attack, which means that you probably won't be able to use that unit for anything else. HS gets around it by calling them 'Land Mines', but that still takes up an entire unit that could be put to better use if an EUD system could be implemented.

I'll update the OP.

Post has been edited 1 time(s), last time on Jan 11 2010, 6:48 pm by ImagoDeo.

Like I said, preplaced units. You move a new one in and give it to the player and set a DC, then a different EUD would start detecting burrow.

Imagine the following scenario.

Your unit dies. However, you are player 2, and player 1's trigger(s) spawn a unit (or units). All of a sudden, the system begins to work on whatever unit was spawned by player 1's trigger(s). You respawn, but all of a sudden <insert spell here> doesn't work when you burrow like it's supposed to.

Nevermind, I see your point.

As said in the shoutbox, I think the established method is much more effective and less cumbersome to implement into a map. It's also more versatile in terms of situations it will work in, referring to things like preplaced vs not preplaced units.

Or you could ignore me. Hero Sanctuary is hardly a representation of professional mapmaking, just because it has those defects doesn't mean they are inherent with the method.

The EUD system would have a few consequences and would require a few preparations, but it would be more versatile in the long run. I understand that the current system works, but as I mentioned earlier, it still modifies the physical game world, which, while workable, is still not preferable to having an entirely virtual system.


My apologies. I hadn't thought about it in those terms.

HP detection is worthwhile due to the variable nature of HP (there's 32 bits of data). Burrow detection is 1 bit, making it increasingly difficult to detect via EUD, and there is already a well established method for detecting burrowing, whose only limitation is that it works by detecting the effects of burrowing, not the state.

If a unit stands on top of the burrowed unit, the game will think the unit isn't burrowed.

Ideally, systems never use physical units and instead resort to detecting only virtual things. However, there are some things which are just much easier to accomplish by using units (shuffling a deck of cards for example).

Also, I've never got around to testing it, but placement IDs are re-used after a time. I'm not sure of that time, but you should be able to figure out how long it takes for a new unit to be placed in the slot, then prevent any units from being created after a death until the slot has been filled by the proper unit. As long as you're using EUDs though, you can also detect a certain amount of HP and set that to be the new "zero life" so that the unit never actually dies.

There's a very easy fix: Move the DT or whatever onto the unit you want to detect the burrow status of, then see if All Players brings At Least 2 [any unit] to the location. This way, if the DT can't be moved but the unit is burrowed, it will still detect the burrow correctly.

Can you even detect two units in a pixel location though? Some units would eat up the whole location.


Oh okay, I forgot to factor in that the burrowed unit counts as a unit to the location. I thought we were detecting two units on top of the burrowed unit.

Post has been edited 2 time(s), last time on Jan 11 2010, 9:16 pm by Taylor Swift.

The point is that one of them is burrowed and any unit being on top would make '2' units brought to the location, meaning the unit is burrowed.

So I guess you'd need to waste 2 locations, or make sure you use a perfectly square unit?

As you pointed out, the current system has flaws and all kinds of things that need to be worked around.

Increasingly difficult? What do you mean by that? I would think it would be a simple matter of finding the address and modifying it by a specific amount of deaths per placement ID.

I thought placement IDs were reused as soon as a new unit was available to fill the space. You're telling me that ID #0200 can be empty and the game will continue to go up from, say, #0564, for a certain period of time - even if it could fill #0200 immediately?

Flaws and such? What are you talking about? Do you understand how the established method works? The only downside to it is that it detects a single effect of something being burrowed; two grounds units can fit into a tiny location if one is burrowed. There isn't really anything that needs to be worked out or even could be worked out. You mentioned that it modifies the game non-virtually, but this can be done within a single trigger cycle which makes it rather negligible.

As for the 1 byte vs 2, vs 3, vs 4 difficulty disparity, this post might help you a little: http://www.staredit.net/177864/

Yes, I understand how it works.

I guess the general vibe I'm getting here is that no one wants to spend the time to do it, seeing as we have a system that works perfectly already. I'll update the OP.

when a unit dies, its death animation keeps it from being removed from memory until the animation is finished, I think.

Cloak, invincible, speed upgrade, cooldown upgrade all mess with the value for burrow. I'm not entirely certain one of these won't change frequently, rendering the eud virutally useless.

Ah. That would make sense.

Yes, that would tend to put a damper on things. I might update the OP again, we'll see...

Hmm this could be good for a corpse collecting game or something. OR maybe you could be able to rescue someone as long as their death animation is still there.