PDA

View Full Version : Make dungeon alert better?



Dragavon
01-02-2010, 12:58 PM
A friend of mine posted this on the Codemasters forums. Some excellent ideas here IMO so I thought I should post this here, as the chance it will get some DEV attention is infinitely bigger here:


This thread assumes/accepts that Turbine has introduced Dungeon Alert to balance "zerging" and give sneak characters a reason to be viable. So, since it's here to stay, these are the things that are wrong with it, and how I think they should be:

First of all, Dungeon Alert should not exist in explorer areas. Vast areas like the Menechtarun Desert and the Vale cannot possibly be fortified in the way described by DA. The 12 rats that chase you in the South West corner simply cannot increase the Orthons' defense in the North East. Besides, it's called Dungeon Alert, not Everywhere Alert.
DA in explorer areas also hurts groups. One person who is late to the quest, or if a non-full group has already started the quest, or someone gets lost, etc, are cases where it hurts grouping. Many people have difficulties finding the quest without killing the enemies, and at the same time they can't kill the enemies to find/run to the quest.
Unintelligent enemies like Vermin and Rats should not raise the DA level. They simply cannot notify or alert anyone or anything.
DA at the moment increases slower for Green, then faster for Yellow, even faster for Orange, and almost immediately for Red. It should be the opposite. The first few monsters that you encounter in a dungeon should be able to alert a few more, but you'd need "more to notify even more". For example, you shouldn't have 12 monsters raising it to Green, and 16 raising it to Yellow. This is slightly subjective, but this is how I see it as "realistic".
DA should add up per vicinity/character/group. To explain, if player X is in room A with 10 enemies and player Y is in room B with 20 enemies, and rooms A and B are miles away from each other, the DA should not be cumulative. Player X and anyone near room A should face Green alert, and player Y along with anyone near room B should face Yellow alert (numbers used are examples).
For DA to be raised, some monsters need to actually raise the Alarm. Random monsters should stop chasing the players and run towards "alert" spots. Even for just a couple of seconds.
Harried works wrongly. It should first of all have a Dexterity or Strength save (IMO) whichever is higher, but even without one, it should occur when there are actually many enemies around you. At the moment, it happens even if you run passed one enemy, when DA is at least yellow. Even if the enemy doesn't actually hit/touch you.
Finally, Harried should be level equivalent. A level 20 character in a level 1 quest should not be Harried, and it should be a lot more difficult to "Harried" a level 16 character in a level 10 quest, than a level 18 character in a level 20 quest.

Also, if you want to be even cleverer, add the opposite of Dungeon Alert, something like Cowering... Hundreds of times, players have been going in quests overwhelming monsters. Why do they always give their lives so easily. Every single monster in this game has a Bravery +100 and Dedication to the Cause +100. If DA is raised to Red, check how quickly it dropped to Yellow (again, for example; from a Wail of Banshee, or Wall of Fire, or simply a hasted Barbarian chopping everything down in 3 seconds). If it's faster than X, then make DA disappear completely, instead of dropping it to Yellow, as the monsters are cowering, and dropping their defenses. Ideally, I'd like to see monsters actually running away from you and your overwhelming power, but just dropping the DA faster would imitate the cowering effect (scared enemies don't fight as hard).


I completely understand that the above need loads of work, but seriously, do the work DA was the worst thing you ever implemented in DDO for me, but I've accepted its need. It (sometimes) ruins our gaming experience so why not make it to (always) make it more pleasant instead.

http://community.codemasters.com/forum/dungeons-dragons-online-general-discussion-296/403024-my-suggestions-better-dungeon-alert.html

Guildmaster_Kadish
01-02-2010, 01:01 PM
Well, I'd say the biggest problem is this:


This thread assumes/accepts that Turbine has introduced Dungeon Alert to balance "zerging" and give sneak characters a reason to be viable. So, since it's here to stay, these are the things that are wrong with it, and how I think they should be:

The developers specifically stated that DA was meant to decrease lag. Now, I'm not sure how much truth there is to that (I know my lag hasn't gone down), but I highly doubt that their working goal was to make sneaking more viable.

Strakeln
01-02-2010, 01:55 PM
Nuke it from orbit. Only way and all that.

Seriously. DA is a hack to circumvent the real problem. It removes much of the fun that DDO used to have, and it doesn't seem to have any effect on the problem it purportedly was created to avoid.

Demoyn
01-02-2010, 03:13 PM
So, since it's here to stay


Let's give it another six months before we cement this. Actually, our time frame may be down to five months by now. I know this much, at least... I can't wait! ;)

chester99
01-02-2010, 03:17 PM
DA is the worst part of ddo. Given that, there's one very simple fix.......

Xyfiel
01-02-2010, 03:25 PM
Make it tied to highest level in quest and quest level.
If below quest level, make it easier to get DA, if above quest level, make it harder.

Say equal levels 12 mobs=green
For every level above/below, add/subtract one

Therefore a level 5 in a level 8 quest would only need 9 for green, while a level 20 doing a favor run would need 24 for green.

They already did this with xp.

Soldack
01-02-2010, 03:36 PM
DA is the worst part of ddo. Given that, there's one very simple fix.......

Dungeon alert is the worst part of DDO. If I don't keep playing it will be primarily because of DA.

centurionaramis
01-02-2010, 04:00 PM
Just quit hindering our ability to even TRY to fight back with the harried status, and we're good to go.

Decades of playing games and I've never seen so horrid a feature.

HeavenlyCloud
01-02-2010, 04:21 PM
nuke It From Orbit. Only Way And All That.

Seriously. Da Is A Hack To Circumvent The Real Problem. It Removes Much Of The Fun That Ddo Used To Have, And It Doesn't Seem To Have Any Effect On The Problem It Purportedly Was Created To Avoid.

Qft

Lorien_the_First_One
01-04-2010, 06:34 AM
DA is a good idea...you shouldn't be able to zerg past everything without concequence, but the implimentation is stupid. Random debuffs don't make sense. Have them switch to tactics to slow us or punish us that are within the game, for example, a bunch of them switch to cripling weapons, cast slow, cast greater dispell, have the mobs ahead form a blockade with extra mobs and a barrier, etc. Plenty they can do, random "I can't move effect" is just annoying and stupid.

Razcar
01-04-2010, 06:37 AM
Nuke it from orbit. Only way and all that.

Seriously. DA is a hack to circumvent the real problem. It removes much of the fun that DDO used to have, and it doesn't seem to have any effect on the problem it purportedly was created to avoid.
My emphasis. We as players might notice game performance changes, but we have no idea if a performance change comes from adding/removing server hardware, changes in client code, changes in server code, changes in infrastucture, changes in player numbers, changes in player behaviour or whatever - unless Turbine tells us. (Clearly we know about changes on our side, i.e. PC performance). Only Turbine knows if DA has helped or not. I'm not defending DA as such - I do think it is a bubblegum fix - but we players cannot judge its effectiveness in decreasing server load.

Strakeln
01-04-2010, 10:56 AM
but we players cannot judge its effectiveness in decreasing server load.
So? We can easily judge its effectiveness in improving our gameplay (by reducing lag).

If it successfully reduced server load but had no effect on lag, it's still a complete failure in my book. If you want to go across town but can't get past the end of your driveway, that's not "close enough".

Letrii
01-04-2010, 11:32 AM
http://www.llbbl.com/data/RPG-motivational/images/nuke.jpg

Ack, they won't let me hotlink to the picture, but llinking to the page isn't showing the picture.

Razcar
01-04-2010, 12:07 PM
So? We can easily judge its effectiveness in improving our gameplay (by reducing lag).

If it successfully reduced server load but had no effect on lag, it's still a complete failure in my book. If you want to go across town but can't get past the end of your driveway, that's not "close enough".
Oh, even if we could make an accurate A - B lag comparison as players with DA or not with DA, with all other factors being exactly the same (the couple of days it was "off" is not such a test environment), there might be other benefits. Maybe it doesn't decrease lag, but e.g. makes it possible for Turbine to squeeze in more instances on a given amount of hardware. That's a win for Turbine and might be a valid reason for them to keep it. I'm of course guessing, as that is all we can do since we do not have access to Turbine's metrics.

If DA gives them neither economical gain nor increased customer satisfaction (as in decreased lag, I think very few people like to be told how to play) the only reasons for Turbine to keep it would be pride and inability to admit mistakes (quite a potent factor, but they are on the other hand a business - not some random developer's vanity project), or a power-tripish urge to steer players to a certain type of gameplay (which does sound quite conspiratorical but I guess not impossible). I can see no other "reasons" to keep such an unpopular game system - if it is not working and doing what it was supposed to do.

Rheebus
01-04-2010, 12:15 PM
I am in total agreement over the "harried" status bestowed upon the grazed hapless toon marked by a level 3 or 4 DA. I like the idea that the mobs change weapons, add buffs, become immune to firewall (supreme fire prot/res), etc.. as the dungeon alert increases. This is probably a much more difficult solution than the current patch. Maybe Turbine is working on something like this to replace the current system?

Bloodhaven
01-04-2010, 12:19 PM
I dislike DA and always will in any form.

I am just waiting for it to be removed like the XP penalties for death. The people do not like it. Listen to your customers.

Judo
01-04-2010, 12:25 PM
i really dont mind DA, in concept, what i hate is the fact that you can agro only the mobs you see, and still get red alert from the broken clipping, i.e. bastion of power, new invasion, etc.

also, if i want to zerg von3 on my capped sorc, a drow ranger that didnt bother me at level 8 shouldnt have the ability to slow me down in any possible way at level 20

tweak it, fix it, cuz right now, its not working as intended, and that is what makes the gaming unpleasant, not the DA itself

Borror0
01-04-2010, 12:54 PM
So? We can easily judge its effectiveness in improving our gameplay (by reducing lag).
You have to remember the population went up significantly, around the same time. My hits on the wiki went x6 in September compared to August, for example. It's not as easy as saying "there was lag and there is still lag". When you consider the fact that server cost goes up exponentially for better performances, it's not that hard to see how DA might be economically advantageous to Turbine.

With that said, DA really sucks (which probably has something to do with the fact it was not designed to be fun) and they should put some energy it making not suck.

Cyr
01-04-2010, 04:12 PM
With that said, DA really sucks (which probably has something to do with the fact it was not designed to be fun) and they should put some energy it making not suck.

lol Borro. Making it not suck would require custom tweaking per dungeon as there are multiple issues that make it suck.

Short list

A) Fake Mobs causing agro. As seen in the library in threnal.
B) Respawn dungeons having it enabled at all. Offering of Blood anyone? This is a problem because the quest was balanced originally without DA...
C) Mob agro radius issue's/landscape issues. Bastion is a good example of this where mobs on the floor above you can trigger this (btw this one has a double whammy of stupidity in it being enabled at all due to the respawns from portals)
D) Burst mob spawns causing DA. Coal Chamber bats are good examples here.
E) Extra computations automatically required by DA regardless of how it effects game behavior.
F) DA does not make logical sense. Ten bats spawn so now they make me move slower (no save!) and everyone else gets uber buffed. What spells are they using again? Oh yeah, no spells because they can't be debuffed with magic only by killing mobs... ie breaks immersion.
G) Harried is NOT fun.
H) Attempts at Behavioral Psychology by a software company to fix inherent flaws in their software is counter intuitive to say the least.
I) DA does NOTHING to reduce the number of operations required for similar situations in fact it increases them. In other words, it does nothing to improve the code or fix outright errors.

Borror0
01-04-2010, 05:06 PM
F) DA does not make logical sense. Ten bats spawn so now they make me move slower (no save!) and everyone else gets uber buffed. What spells are they using again? Oh yeah, no spells because they can't be debuffed with magic only by killing mobs... ie breaks immersion.
G) Harried is NOT fun.
Both of these can be fixed and are probably the biggest ones.

If the snares are changed to CC spells by casters and tactics (Trip, Stunning blow, etc.) by melee characters, it is less annoying than being snared is - somehow, being immobilized is less frustrating than being slowed - and makes logical sense that monsters would try to prevent intruders from going any further inside their lair.

H) Attempts at Behavioral Psychology by a software company to fix inherent flaws in their software is counter intuitive to say the least.
Being intuitive to you does not make it the smart decision, though. If obtaining servers capable of processing the huge load caused by zerging is unrealistic, then it is logical that game developers would change the game to not cause such an heavy load.

A few of the possible ways are:

Worse monster pathing
Removal of collision tests between monsters and players
Making zerging much more difficult (DA fits into that category)
Changing quest design to disallow zerging (adding barriers fit into that category)

Take your pick.

Personally, #4 is the lesser evil but YMMW.

I) DA does NOTHING to reduce the number of operations required for similar situations in fact it increases them.
The amount of computation required by DA is far smaller than hundreds of AIs trying to calculate the right path through a dungeon.

Razcar
01-05-2010, 04:24 AM
A few of the possible ways are:
Worse monster pathing
Removal of collision tests between monsters and players
Making zerging much more difficult (DA fits into that category)
Changing quest design to disallow zerging (adding barriers fit into that category)



They could use leashing in dungeons as well, some version of what they use in wilderness zones. However, if Turbine does have a "hidden agenda" not just to decrease server workload but also to steer playstyle and player behaviour - in other words to discourage zerging (for whatever reason, maybe make grouping newbie friendly or increase D&D fidelity or whatever), well then that would not be an option. Leashing might very well increase zerging.

But to have full leashing, i.e. that the mobs will deactivate and return after unsuccesfully chasing the players for x minutes, seems like it could achive the "offical" goal of easing server load. When the mobs give up the chase there won't be any resources wasted on pathing and collision detection for mobs that will never reach the zerger anyway.

However, this does assume that the mobs are always "there". The game might very well be programmed so that the server process that is a DDO monster is created as soon as a player becomes aware of it - before that it does not exist as far as the server is concerned. If I would programme a game like DDO I guess that's how I would do it - it is wasteful to create potentially hundreds of AI processes on quest initiation that then would just lie idle, when instead you could create them just when they are needed.

But if that is the case, then the mob could of course terminate (i.e. just like it was killed) when it has leashed back by reversing the procedure that initially created it. And if/when a player becomes aware of that mob again, well, just recreate the mob like first time. This would require full leashing though, since the server would likely not know where the monster was "terminated" but just its original, initial spawn point. It goes without saying that if the monster is killed by players, it stays dead. I don't think anyone wants the incredibly silly EQ-type respawns popping back on a schedule.

Leashing does stifle immersion somewhat (at least when Lawful monsters are concerned, the Chaotic ones might not care to dump this adventurer-intrusion problem on their colleagues deeper in their lair ;)). But so does DA, and in a greater way in my opinion.

Borror0
01-05-2010, 05:13 AM
They could use leashing in dungeons as well, some version of what they use in wilderness zones.
You have to be careful with leashing, though. Excessive leashing can result in bad lag, as well, since leashing means that the monster is destroyed and a new one is spawned at its place. If you keep the leash too small and have an high concentration of monster, you could end up increasing the lag...

Also, it changes the way dungeons are played. In many cases, you want monsters to follow the players for a long while. Those who complain about "dumbing down the game" should complain here because reducing the leash's length to have any measurable impact would really make the game much easier (and zerging much more effective as it would be easier to bypass certain mobs).

Zippo
01-05-2010, 05:23 AM
DA is the worst part of ddo. Given that, there's one very simple fix.......

And notice everyones reaction when it disappeared for that day or so. :D I for one was happy


Turbine let that be an example to you about how most everyone feels about it.

Letrii
01-05-2010, 07:01 AM
Why would leashing destroy the monster and require a new one later? Couldn't mob return to its focal area and be inactive if noone is in range?

Razcar
01-05-2010, 08:47 AM
You have to be careful with leashing, though. Excessive leashing can result in bad lag, as well, since leashing means that the monster is destroyed and a new one is spawned at its place. If you keep the leash too small and have an high concentration of monster, you could end up increasing the lag...
That would depend on the programming I suppose. Although terminating the mob and managing the garbage/freeing memory etc. may take some resources it might be cheaper than pathing the mob back "physically". In wilderness areas mobs are pathed back, although in a non-interactive state which is quite cheap I guess. It might look better to path them back than just making them disappear (if the zerger would look over his shoulder, which is not likely ;))



Also, it changes the way dungeons are played. In many cases, you want monsters to follow the players for a long while. Those who complain about "dumbing down the game" should complain here because reducing the leash's length to have any measurable impact would really make the game much easier (and zerging much more effective as it would be easier to bypass certain mobs).
That is true, but the zerger can say the same for DA as well, and DA does also cause problems for players that never zerg. Leashing could be programmed so the leash will only activate after the player target is out of sight for a set amount of time, to keep immersion and make zerging more fun ("did we loose them?"). This would also free the resources needed to path mobs back to their "home point". Chasing mobs, before they leash, could be made more dangerous, maybe setting off traps like spike pits and rockslides in the players way. It could also be paired with more "must kill to progress" group of mobs, something that is quite common already in game.

Made in the right way, maybe it could fulfill the goal of saving resources and also make zerging more fun for those that enjoy that gameplay. But yes implemented badly it might end up worse than DA, not save on server load and making DDO even easier than it is today. All ideas with some initial merit can be implemented badly, just look at democracy or socialism ;)


Why would leashing destroy the monster and require a new one later? Couldn't mob return to its focal area and be inactive if noone is in range?
That would depend if it is more costly to recreate the mob than keeping it idle, and I'm guessing "no mob" is cheaper than "idle mob".

Destroying it is a kind of "if a tree falls in the forest where noone hears it does it make a sound?" situation :). If no player client is aware of the mob there is no reason for it to exist. Instead you will have a radius where a monster will "spawn" if a player comes close. When noone is there you can completely terminate the monster process instead and save resources.

DownClan
01-05-2010, 09:19 AM
Lots of players used to zerg through dungeons. The long trains of MOB's would cause server lag (we are told). Turbine introduced a feature (DA) which makes players more hesitant to zerg through dungeons, so it happens less often now.

From Turbine's perspective, that probably equals "success".

--DownClan

GoldyGopher
01-05-2010, 10:14 AM
Okay, I am going to make a statement that may not make everyone the most happy....

We should assume that DA is here to stay and rather than saying Nuke it from orbit we should put some time into helping reduce the problems it is causing.

1) Put a NPC in House P or make it so Jalizco can port characters to the Delver's Canteen in Thernal.
2) Change the Thernal Adventure Area so the area around the Threnal Quests is a public area and the Area outside of Thernal Quests is the Adventure Area.
3) Put a NPC in the Twelve near the entrance to Vale that can port characters to Meredia.

Now honestly I am a fan of the Adventure Areas but lets be honest here the problem is that many times people run through adventure areas just to join the party so when they run through all they do is zerg and build agro and have a trail of 50 mobs following them. If you make it so you need to run once, great, make it so it costs say 10 GP cool. Lets not create instances just to create instances that players have to run to. There is also no reason to make instances and populate them with mobs that simply a character will be moving from a quest entrance (thinking Thernal mostly) to a tavern.

4) If you can't make it so we can port lets at least look at mob placement. There is no reason that we should activate 50 mobs during a run to Meredia along the road or in the sands along the marked path.

5) There should be another level of DA, one that tells players that at least one mob is active. Trust me the zergers will find the path of least resistance and when they learn how to not activate 50 mobs they will. Right now they don't know and don't care.

6) As players when we come across weird/atypical results of DA, take the two minutes and Bug report it. I was in A New Invasion last night solo when suddenly I got the fourth (Red) level of DA. The problem was my character was in an empty room and I had only seen two or three mobs to begin with, so no green, yellow, orange. That's a bug for several reasons. Needless to say in the next encounter (several rooms away) I was flattened like a pancake.
As a part of that Turbine must look at those quests we are highlighting and fix the system.

7) No single encounter should activate a DA of any kind (that currently exists since I said we should have another level). If ten bats burst in Col Chamber we shouldn't be in Green Alert. (Now if you are dragging ten scorpions with you).

8) The Lagro system needs to be fixed. Hounds running off for no apparent reason activating ten other mobs, archers running off, clerics making the mad dash at you and let’s not talk about air elementals, and when you are whacking away at a ogre he shouldn't be buffing himself, swing and defend themselves.

9) The effects of DA should fit the mob. I have no problem when 10 hound/wolf mobs suddenly get bonuses to knockdown (don't like it but that doesn't mean I have a problem with it) but when 10 bats start knocking me down....

10) DA shouldn't be a death sentance, but it should make it more difficult for the player. Right now you get to red solo you are dead (at least for most people). Considering how easy it is to get there in some quests....

11) Leashing won't resolve the problem that you have Stated DA is supposed to fix. So look at mob placement, it should be relativily easy to sneak around a lone kobold, but five kobolds at a door is another matter. The problem is fighting those five kobolds activates five kobolds on the other side of the door and maybe the three you just snuck passed creating a yellow alert. Is that what was intended?
Now if the kobolds could open the door and press the attack against you party I would say yes, but since they can't....

12) My understanding of how quest instances work is that quests are run on the servers responsible for the zone, hence House D would have the Depths, the Pit and Stormcleave (and others) running in it, so in essence 45 people running Stormcleave (solo) would have a greater effect on performance for people running the Depths series than 10 parties of 4. If that is the case do more to seperate out where the quests are located and the "server" they are running on. Turbine should know what popular <cough> Shroud <cough> is and what is unpopular and attempt to achieve a better balance.
On a side note I think Turbine missed placed Meredia when it created the Vale of Twilight. Had Meredia been located in say the NE Corner of the Vale (On the Other side of the Gnoll Villages) it could have been placed as a gateway between multiple servers and explorer areas thus increasing the value of Meredia as well as helping decrease specific server loads.

13) If a Burst Mod is spawned they should be active. Don’t go through all the effort to create a mob just to sit at a portal.

dasein18
01-05-2010, 10:36 AM
Best way to make it better is to remove it. Does nothing but decrease my fun in this game. Bastion.. enter a room.. portal opens... boom DA alert goes way up... sure we kill and survive... but what is the point?

One of the worse additions to the game ever.

Tin_Dragon
01-05-2010, 11:02 AM
DO AWAY WITH DUNGEON ALERT< BRING BACK INDIVIDUAL DEATH PENALTIES!

fixed that for you.

Tin_Dragon
01-05-2010, 11:07 AM
Okay, I am going to make a statement that may not make everyone the most happy....

We should assume that DA is here to stay and rather than saying Nuke it from orbit ...



.

No, your giving in, your forgetting that many of us are, and have been, PAYING Customers for a long time.

I never saw a thread once ASKING for such a thing like DA, the game is slowly being dumbed down for people who refuse to run quests EXACTLY the way turbine thinks they should while spending MORE $ for turbine points.

So, you either support DA, or you dont. IF you do NOT support, simply say that, and keep asking that it be removed, not modified.

Again.

GET RID OF DUNGEON ALERT, BRING BACK INDIVIDUAL DEATH PENALTY!

Beherit_Baphomar
01-05-2010, 11:22 AM
The only way to help DA is to get rid of it.

The Devs should work on the poor AI rather than implementing lag-effects to cure...lag...

Aaxeyu
01-05-2010, 11:30 AM
Here's the solution:
Remove DA completely, add leashing to mobs in quests just like in explore areas.



And everytime a mob "leashes" a copy of it is created in the last room of the dungeon.
That would make coal chamber interesing :D

GoldyGopher
01-05-2010, 11:36 AM
No, your giving in, your forgetting that many of us are, and have been, PAYING Customers for a long time.

I never saw a thread once ASKING for such a thing like DA, the game is slowly being dumbed down for people who refuse to run quests EXACTLY the way turbine thinks they should while spending MORE $ for turbine points.

So, you either support DA, or you dont. IF you do NOT support, simply say that, and keep asking that it be removed, not modified.

Again.

GET RID OF DUNGEON ALERT, BRING BACK INDIVIDUAL DEATH PENALTY!

I to am a long time paying customer; however we saw thousands of threads asking Turbine to reduce lag. Regardless of how Turbine got there, we got DA as a potential solution. It's now implemented, if Turbine is going to keep it implemented make it better. My problem with DA isn't that it exists, rather that so many interconnecting things are implemented so poorly it makes DA virtually unplayable. If Turbine starts to fix the issues which affect DA than DA becomes less of an issue.

If you look at my response many of the fixes for DA have absolutly nothing to do with DA directly. But indirectly they should help reduce the sources of server side lag.

For example I think too many instances are created just to create instances. If the number of instances is an issue, create fewer instances. Some sort of Teleport to Meredia, Threnal, Sands, Titan.... reduces the number of instances that players run through. If that can't won't be done, look at the placement of mobs along the paths. Just running by while on a marked path shouldn't start the train.

I could go on but I would just be rehashing what I have already said.

Borror0
01-05-2010, 12:02 PM
/snip
Great post. +rep.

Why would leashing destroy the monster and require a new one later? Couldn't mob return to its focal area and be inactive if noone is in range?
It's the way it is coded now. This is why the monster goes back to full health, loses all debuffs, etc.

It preferrable to be that way, though. Otherwise, you'd still need the game to calculate the monster's path to where you want it to go back which would mean the leash would result in a lesser load reduction than you wanted. It's also better in that it avoids clunky AI behavior near the limits.

In wilderness areas mobs are pathed back
I define leashing as when there is yellowish circle that pops up their head and they disappear back to their spawn point.

That is true, but the zerger can say the same for DA as well
How so? DA does the opposite effect by making the game harder.

Made in the right way, maybe it could fulfill the goal of saving resources and also make zerging more fun for those that enjoy that gameplay.
That much is obvious, but the high improbability of how "getting it right' is the problem, here.

If no player client is aware of the mob there is no reason for it to exist. Instead you will have a radius where a monster will "spawn" if a player comes close. When noone is there you can completely terminate the monster process instead and save resources.
I think you misunderstood me. Regardless of if the player is around to witness it or not, the NPC is destroyed and is respawned at it spawn point immediately.

ahpook
01-05-2010, 12:42 PM
For example I think too many instances are created just to create instances. If the number of instances is an issue, create fewer instances. Some sort of Teleport to Meredia, Threnal, Sands, Titan.... reduces the number of instances that players run through. If that can't won't be done, look at the placement of mobs along the paths. Just running by while on a marked path shouldn't start the train.


In that case, add "allow GT from public instances" to your list. How many times do people create instance simply so they can GT to a different public area? quite silly really.

While your explorer suggestions are good, I would add the ability to jump directly into the quest from the quest giver if you have one completion on that quest (with some exceptions such as VOD and Hound, Von 6). Then you don't have to run through an explorer at all to rerun a quest saving even more instances (while making the game more enjoyable for thiose that don't always want to pointlessly run).


You have to be careful with leashing, though. Excessive leashing can result in bad lag, as well, since leashing means that the monster is destroyed and a new one is spawned at its place. If you keep the leash too small and have an high concentration of monster, you could end up increasing the lag...


You also have to be careful with leashing because it is actually a worse solution to problems than DA. I have been more frustrated by mobs that take one step too far and leash (just before dieing) than I have ever been frustrated by DA.

Cyr
01-05-2010, 12:43 PM
The amount of computation required by DA is far smaller than hundreds of AIs trying to calculate the right path through a dungeon.

I'm just going to pick on this part of your response. My statement was that during similar situations DA requires more operations per cycle. This is just plain true. It is an operational efficiency argument. If player A always sniffs the flowers DA being active requires more operations for his game play period. If player B always zergs then DA being active requires a lot more operations for his game play due to the monster scaling between DA levels as well as the constant DA level calculations that player B has. These are similar situations.

Your statement assumes a lot of things including that it is not similar situations. The largest assumption being that DA drastically alters player behavior. The second being that enough players' behavior as a percentage online are altered in this way to overcome the extra operations DA requires. Even if your statement was correct it would require a significant number of players to have altered their game play. In other words, the better DA works to reduce lag the more people it has altered the game play of whom are predominantly those who like to run stuff fast. This means the more people running stuff in a less fun way for them makes DA effectively reduce lag more. This basic analysis of the concept of how DA is supposed to work shows clearly that for it to work it will reduce the fun factor for those playing and for it to work better it reduces the fun factor for more people...

The issues trying to be addressed are monster AI and pathing using to much resources. Turbine is trying to make those scripts run less with DA. The real fix is to rewrite them where needed to be much more efficient and to upgrade hardware. To anyone who has any programming experience it should be obvious that some things in DDO just are coded very poorly from an efficiency standpoint. Things like the load on lag caused by the who list. The lag caused by the character sheet feats and enhancements (the reason why Master's Touch causes huge lag). Or the combat log lag which becomes crippling in places like the shroud or ToD. All of these should essentially use almost zero system resources, but due to their implementation they use a heck of a lot at times. It seems pretty obvious that mob AI and pathing is coded in the same roughshod way except instead of causing lag user side though hogging their system resources it hogs server side resources instead.

ahpook
01-05-2010, 12:44 PM
I think you misunderstood me. Regardless of if the player is around to witness it or not, the NPC is destroyed and is respawned at it spawn point immediately.
Not quite immediately. I have often seen mobs run a long way back to their spawn point before disappearing (probably causing even more pathing lag :( ).

Impaqt
01-05-2010, 12:55 PM
lol Borro. Making it not suck would require custom tweaking per dungeon as there are multiple issues that make it suck.

The Devs have already stated they CAN tweak per dungeon. THis is why REPORTING the issues are important. so they can tweak the system soit works better.



Short list

A) Fake Mobs causing agro. As seen in the library in threnal.
B) Respawn dungeons having it enabled at all. Offering of Blood anyone? This is a problem because the quest was balanced originally without DA...

Solo'd OOB yesterday, had no issues at all. Hit Green a couple times.. Yellow maybe once.


C) Mob agro radius issue's/landscape issues. Bastion is a good example of this where mobs on the floor above you can trigger this (btw this one has a double whammy of stupidity in it being enabled at all due to the respawns from portals)

THis quest has also been tqweaked significantly. Last couple runs I have not gone past yellow. At first, Red was almost a guarantee.


D) Burst mob spawns causing DA. Coal Chamber bats are good examples here.

ALso Solo'd CChamber. Again. Didnt get past Yellow.


E) Extra computations automatically required by DA regardless of how it effects game behavior.

Devs have also stated that DA Overhead is considerably less that the overhead caused by hundreds of extra aggor'd mobs.


F) DA does not make logical sense. Ten bats spawn so now they make me move slower (no save!) and everyone else gets uber buffed. What spells are they using again? Oh yeah, no spells because they can't be debuffed with magic only by killing mobs... ie breaks immersion.
G) Harried is NOT fun.

THen Kill Stuff. As I solo CChamber, I simply Kill the stuff that is stationary or spawns behind me. THen when 6 bats spawn, it doesnt effect the DA at all.


H) Attempts at Behavioral Psychology by a software company to fix inherent flaws in their software is counter intuitive to say the least.
I) DA does NOTHING to reduce the number of operations required for similar situations in fact it increases them. In other words, it does nothing to improve the code or fix outright errors.
You have no basis for this theory. If DA GOes off after 10 or 30 mobs ar aggroed its Far fewer computations than having 100+ Mobs aggroed.

Strakeln
01-05-2010, 01:21 PM
I to am a long time paying customer; however we saw thousands of threads asking Turbine to reduce lag. Regardless of how Turbine got there, we got DA as a potential solution.
For the longest time now, I've been saying that interacting with Turbine is like dealing with the Devil... you have to watch what you ask for.

You ask the Devil to make the world's most beautiful woman love you unconditionally... *POOF* your mom is hot.

You ask Turbine to reduce/eliminate the lag... *POOF* DA drops.

GoldyGopher
01-05-2010, 01:41 PM
For the longest time now, I've been saying that interacting with Turbine is like dealing with the Devil... you have to watch what you ask for.

You ask the Devil to make the world's most beautiful woman love you unconditionally... *POOF* your mom is hot.

You ask Turbine to reduce/eliminate the lag... *POOF* DA drops.

Now that I have had my "Bill and Ted" Flashback "Remember when she was a senior and we were freshmen?" "Shutup Dude."

I too have explained to people it is not what you are asking for from Turbine that matters the most it is "why" you are asking for it, which you need to convey.

Letrii
01-05-2010, 01:49 PM
Great post. +rep.

It's the way it is coded now. This is why the monster goes back to full health, loses all debuffs, etc.

It preferrable to be that way, though. Otherwise, you'd still need the game to calculate the monster's path to where you want it to go back which would mean the leash would result in a lesser load reduction than you wanted. It's also better in that it avoids clunky AI behavior near the limits.

I define leashing as when there is yellowish circle that pops up their head and they disappear back to their spawn point.

How so? DA does the opposite effect by making the game harder.

That much is obvious, but the high improbability of how "getting it right' is the problem, here.

I think you misunderstood me. Regardless of if the player is around to witness it or not, the NPC is destroyed and is respawned at it spawn point immediately.

Instead of making a path back to spawn point, why not have it warp back like hirelings do?

sephiroth1084
01-05-2010, 02:05 PM
DA sucks. Period.

Whatever its intent, we've had time to experiment with it, and it blows. Hard.

It makes favor-running lower level quests even more of a chore than it already was. It randomly punishes you severely for relatively minor mistakes you may make. DA, as implemented, makes little logical sense or mechanical sense.

All that said, rodents and such can warn others of their kind, and implementing snare-type spells won't necessarily have any effect on a lot of characters.

Borror0
01-05-2010, 02:15 PM
You also have to be careful with leashing because it is actually a worse solution to problems than DA. I have been more frustrated by mobs that take one step too far and leash (just before dieing) than I have ever been frustrated by DA.
I think the same, too, but a post is as persuasive as its weakest argument. ;)

My statement was that during similar situations DA requires more operations per cycle.
Not all processes require the same load. Some things, like pathing or collision tests, load the servers far more.

You're saying that two hamsters equals one elephant.

This basic analysis of the concept of how DA is supposed to work shows clearly that for it to work it will reduce the fun factor for those playing and for it to work better it reduces the fun factor for more people...
Obviously. This is the biggest flaw with Dungeon Alert: it sucks and it's not fun.

This is why the biggest changes that should be done to Dungeon Alert are to make it not suck and to make it fun.

The real fix is to rewrite them where needed to be much more efficient
Impossible goal is impossible.

[...] and to upgrade hardware.
As you might know, Blizzard never increased the maximum number of players on a World of Warcraft server. Now, Blizzard has a quasi-endless amount of funds when compared to Turbine and more players is a good thing on an MMO: it means more like-minded person to find and less times trying to fill a PUG even during downtimes. So, why has Blizzard chose to add more servers instead?

It's because you just can't double the hardware and expect to host twice as many players. It does not work that way.

Comes a time where "perfect solutions" are just not realistic.

To anyone who has any programming experience it should be obvious that some things in DDO just are coded very poorly from an efficiency standpoint.
Yes. Like about every MMO on the market. That is what happens when you've got a program that is several gigabytes large, 6-7 years old and has to ship a large update every once in a while, while still being financially profitable. Bad codes is easy; bad code takes less time, so you have more time to work on your other task.

Yes, bad code sucks but good codes means less new features. Again, pick the one you prefer.

Cedwin
01-05-2010, 02:23 PM
Obviously. This is the biggest flaw with Dungeon Alert: it sucks and it's not fun.

This is why the biggest changes that should be done to Dungeon Alert are to make it not suck and to make it fun.


Why didn't I think of that?? :D

Tin_Dragon
01-05-2010, 02:37 PM
The only way to help DA is to get rid of it.

The Devs should work on the poor AI rather than implementing lag-effects to cure...lag...
QFT
+1

And so many people don't realize this is exactly why it came about, poor shoddy workmanship created more poor shoddy workmanship, to try and deal with later poor shoddy (untested?) workmanship.

ahpook
01-05-2010, 02:53 PM
ALso Solo'd CChamber. Again. Didnt get past Yellow.


I was having the same experience solo'ing CChamber the other day. Then a guildie jumped in. Despite him attempting to kill the bat respawns as he went, we did hit red alert quite quickly. Unfortunately, that was just as i was fighting the 6 gnolls at the first key :(.

They need to do some work to handle the separated party scenario better. It is wrong when splitting the party is deadly not because of the dungeon threats but because of DA.

Strakeln
01-05-2010, 03:02 PM
And so many people don't realize this is exactly why it came about, poor shoddy workmanship created more poor shoddy workmanship, to try and deal with later poor shoddy (untested?) workmanship.

Agreed. This has been a snowballing problem for almost 4 years now. The hacks that were put on top of hacks to keep the other hacks from hacking your hack, got hacked.

It's like lesser restoration potions, but game-wide!

This is why you don't always take the cheapest/fastest route.

gwlech
01-05-2010, 03:09 PM
DA is fine. It adds challenge and consequence to zerging. Not that I do not like zerging, don't get me wrong, I absolutely love a good zerg. To me, DA red with no deaths is a sign that I am doing a good job.

And to those complaining that it is a pain, and it impedes progress, I see it as a really good balancing factor. If you are going to zerg, you better be good enough to handle it, or play the content as it is intended, period. No Easy-Button.

Letrii
01-05-2010, 04:38 PM
DA is fine. It adds challenge and consequence to zerging. Not that I do not like zerging, don't get me wrong, I absolutely love a good zerg. To me, DA red with no deaths is a sign that I am doing a good job.

And to those complaining that it is a pain, and it impedes progress, I see it as a really good balancing factor. If you are going to zerg, you better be good enough to handle it, or play the content as it is intended, period. No Easy-Button.

I don't zerg and DA still needs to die.

Cyr
01-05-2010, 04:44 PM
Not all processes require the same load. Some things, like pathing or collision tests, load the servers far more.

You're saying that two hamsters equals one elephant.


No Borro actually you made the argument about the process requirements in your original response to me. My original post and my response made it pretty clear that I was stating that DA was code that had to execute in addition to any existing code and did not make an argument about it's system requirements or savings, because given Turbine's coding practices it could be anywhere from very little to outrageous. In my response I did go into some detail, but it was more like 2000 hamsters may = 1 elephant. DA always has to run (the hamster). Crazy zerging outside of certain dungeons (quickly fixed by your barrier comment and implemented in monastery for example) is very rare compared to normal questing (the elephant).

There are two ways to make an overall process run faster in coding.
(1) Write better, more efficient code.
(2) Have the sub-processes that are system resource hogs or choke points run less often.

Turbine used number 2 here, but to do so they actually made number 1 worse.

And are being incredibly naive if you think that writing more efficient code is somehow impossible to do. I write micro-controller code for work and higher level stuff for fun. When you get into the nitty gritty of a process it's surprising how much more efficient you can make it run even when it was not written sloppy in the first place. You have already acknowledged that there is tons of bad coding in DDO (but let Turbine off the hook by saying everyone else does it and otherwise we wouldn't get any content). With bad coding comes much more room for improvement. You don't have to fix everything in coding. In fact, it turns out that one of the easiest things to do is figure out what sub-processes are using up the most resources and if any seem like they are using more then they should. Then streamlining (or outright rewriting) those. It takes some work yes, but it certainly is not an overhaul of the whole code.

Oh and I know you think DA is not fun, but your disagreement with me is that somehow a concept that intentionally tries to make something less fun to work can be fixed to make it both fun and serve it's purpose. I think it could be made to suck less, but it will always be a negative to the game.

Borror0
01-05-2010, 05:36 PM
In my response I did go into some detail, but it was more like 2000 hamsters may = 1 elephant.
I know this is what you mean, but do you see the difference between how you originally worded it and how I've got you to word it now?

You started off by saying it was fundamentally broke: "DA does NOTHING to reduce the number of operations required for similar situations in fact it increases them." Now, you're adopting a more nuanced position that acknowledges that, if Dungeon Alert is worth 2000 hamsters, then it does not reduce the load like it advertised. However, if load caused by Dungeon Alert is meaningless compared to the load caused by a an instance with hundreds of mobs trying to get to the players, Dungeon Alert is doing its job.

This is far weak position in that it recognizes that there has to be "2000 hamsters" to replace the elephant. But, now, here's the problem: you do not know whether there really is 2000 hamsters (or even if 2000 hamsters weights as much as one elephant, or if 2000 evenly distributed hamsters causes the same trouble as an elephant) and saying that there is would be presumptuous; you don't have that information.

And are being incredibly naive if you think that writing more efficient code is somehow impossible to do.
I was being facetious when I wrote my reply. I know it's not impossible, but rewriting the pathing to take less load does not fall into the "easy" category and there is a limit to how much you can improve your code versus how much you need to reduce the load.

I think it could be made to suck less, but it will always be a negative to the game.
You have yet to justify that position.

Matsu_Ieyasu
01-05-2010, 09:22 PM
I don't zerg and DA still needs to die.

Help me out, I am a little slow. I realize that I am not an ultra-hard core player who used to run past every mob in a quest to the end fight and call myself uber. I am in a small guild where we have to fight practically every mob in the quest just to make it through cause we aren't wearing two GS Items and carrying for GS Weapons. But other than the buggy !@*#&$ we see occasional how does DA bother me? There are occasions that DA comes up and bites are behinds, yes, but mostly that is either obvious bugs or really stupid play on our part.

Borror0
01-05-2010, 09:26 PM
Help me out, I am a little slow. I realize that I am not an ultra-hard core player who used to run past every mob in a quest to the end fight and call myself uber. I am in a small guild where we have to fight practically every mob in the quest just to make it through cause we aren't wearing two GS Items and carrying for GS Weapons. But other than the buggy !@*#&$ we see occasional how does DA bother me? There are occasions that DA comes up and bites are behinds, yes, but mostly that is either obvious bugs or really stupid play on our part.
I think it can be summarized by "snares are not fun, no matter the reason." The fact that Dungeon Alert feels cheesy makes it worse, of course.