So as I'm German, I'll also try to translate it (babelfishes are not really good at this...)
Oh - and sorry for the bad English...

"Madjai and I did some detailled tests (in single player mode) on "Chance of Open Wounds" (OW), and there seems to be a considerable bug:

In the time a monster is affected by OW it's physical resistance is set to zero. However, OW does not work with ranged attacks or the Paladin's smite skill.

This leads to the following phenomena:

- Global physical resistances in hell (50%) are nearly avoided with a high enough chance of OW (during OW is in effect the chance works again and increases the OW time again and again).

- For physically immune monsters a little bit of elemental damage is enough to avoid their immunity.
For example, you carry a charm with lightning damage of only 1 to 5 and hit a PI monster. If you hit, it receives 1 to 5 damage. But this little damage is enough to trigger OW. And as long as OW is in effect you can hurt the monster by using physical attacks.

- Crushing Blow is extended to its maximum impact of 25% when hitting a monster with a resistance of 50%. On PI monsters it does not work anyway (although every other physical damage gets through).

- In PvP the effect is the same. Since there are no physical immunities you don't need elemental damage.
Additionally, it has to be mentioned that OW is decreades in PvP, but not quartered. <...>"

I hope, my translations is a little bit understandable.

If the general B.net public hears about this, I know what will happen: every single physical damage character will have their very own malice-wielding merc

______________________________________________________________________

"Aye, so I've heard! I've also heard he can kill a hundred men in battle. And if the real William Wallace was here, he'd slay the English with fire from his eyes, and bolts of lightning from his arse!"

Of course there are minor other things to overcome, like the life drain on Malice, and the loss of a slot for said merc to get some life leech so he can survive combat. Otherwise he will be a very high maintenance companion indeed.

One bigger problem with Malice is getting enough life leech for the merc to survive. It is possible with Tal Rasha's Mask and Skin of the Flayed One (I think that's the Exceptional Unique Armor with life leech), that would give him 15% life leech, I think. However, Malice only has 33% Enhanced Damage, not enough to keep a merc alive in Hell.

The best way to solve this would be to make Malice out of a Superior Ethereal 3-socketed Elite weapon and give him the previously mentioned two pieces of equipment. Hmm, a weapon like this would be just a little hard to find

EDIT: Spelling mistakes caused by lack of proofreading, must proofread before posting!
___________________________________________________________________

"Aye, so I've heard! I've also heard he can kill a hundred men in battle. And if the real William Wallace was here, he'd slay the English with fire from his eyes, and bolts of lightning from his arse!"

My HC druid, GentleBenwa, has a merc with just such a weapon, Sparkling Mail. {I fell in love with open wounds when BurritOcchi the Paladin had the Steel long sword and found Swordback hold. SHield. I just love it.} I like the SM LGT resist, and its Diablo I tie in.) I had not idea of this bug till you all brought it up.

I am not yet in hell diff, but drat it, I wish you had never told me of this. Grrrrrr. I was perfectly happy with this guy . . . and his malice merc. I made the Malice merc specifically for the Druid to make the -100 to DR help my to hit in big games or on champions.

Sigh.

My thanks for the translation; thanks to the original posters and to Flaming_June for bringing it here.

>"Madjai and I did some detailled tests (in single player
>mode) on "Chance of Open Wounds" (OW), and there seems to be
>a considerable bug:

D2 is a complex game. I think it is a common mistake for people to assume that any belated discovery implies a bug. The game is poorly documented (by Blizzard at least--though I love GFraizer's work) and to someone new to the game a whole host of discoveries we take for granted might plausibly be interpreted as unintentional (i.e. "bugs"). Often Blizzard will add a powerful game feature and later nerf it when they realize it was too good, but that does not necessarily mean it was originally a bug.

Assuming it unlikely to be able to query the D2 designers on this issue I would propose the next best thing would be to have someone (Jarulf?) code-read to see if the effect seems incidental/accidental or is obviously deliberately coded...

>In the time a monster is affected by OW it's physical
>resistance is set to zero.

I have yet to test this myself, but this seems to suggest the monster's physical resistance is restored properly after the OW wears off. This implies the likelyhood, to me at least, of a deliberate branch or deliberate save/alter/restore coding for the effect.

An obvious test, apart from code-reading, would be to see if this effect existed in D2C or was added for LOD (clearly its import is far greater in LOD). If the feature was active, but little used, in D2C, upon redisigning the game for LoD they may have overlooked its new awesome importance.

>- Crushing Blow is extended to its maximum impact of 25%
>when hitting a monster with a resistance of 50%. On PI
>monsters it does not work anyway (although every other
>physical damage gets through).

I read quite recently a post (probably Jarulf) mentioning that CB had it own simplified resist routine, resulting in some minor quirks, so this doesn't surprise me.

Of course, an additional question would be: does this OW effect also deprive the monster of other resists (e.g. we create a triple immune iron golem, can it be harmed elementally after OW)?

-- Crys

p.s. over a week ago DNS stopped giving me a resolve for theamazonbasin. Did they go under? If not, does someone have the IP for forums.theamazonbasin.com so I can continue to peruse there? Thanks.

I tested, SP LoD v1.09c, by socketing an Um rune into a Ribcracker (which has something like 50% chance of crushing blow fyi) and giving it to my fury druid (this is the highest rune I've ever gotten, so I hope you'll forgive me for duping it for the test).

I stripped the equip down to only add non-physical damage of 2-5 fire from a charm and went to act5 hell to find some Death Maulers.

Barehanded or with my alt weapon I could not dent them (the fire damage could be seen but was insignificant). Using the OW staff I could take them down but a lot of the movement in their HPs seemed due to OW over time, as opposed to the hit. Additionally there was *never* any life leech (when the occasional non-phys immune wandered into my fury, my HPs would skyrocket, at 3% leech pre-penalties).

On Spirea's damage page I note the formula for OW as:
Damage = (clvl * 9 + 40)/256 * 25 * 4
supposedly spread out over 4 seconds (it didn't appear to be 4 seconds, but I didn't attempt to time it). As the test druid was level 74 (which explains why I didn't socket an IKSC, clvl req 76) this would work out to about 275 points of damage per successful OW.

I can't get to Arreat Summit just now to check on the total HPs of the Death Maulers, but I strongly suspect the entire kill (taking more than a few OW hits) could, for example, be explained by OW damage *ignoring* physical resist should other damage make it through the monster's immunities.

In any event, the OW ==> 0% physical resist theory doesn't seem to fit the data from my test.

-- Crys

p.s. regretfully I failed to test if rabies transmits OW damage (at least to the initial target). Since rabies doesn't cause LEB emission but is poison damage, adding OW damage to it vs. physical immune LEBs might be rather useful (assuming you care about the lightning).

Tried a malice broad sword (100% OW) pvp (looks like OW works, probably without usual 1/4 pvp reduction--not sure). Tried the malice on hell Death Maulers under "players 8". It sure seemed like all the damage occurred over 4 seconds (and could not be "hurried" by stacking).

-- Crys

p.s. related trivia: I've wondered for some time who gets the xp when a monster dies from poison/OW but was touched (for damage that is) by someone else after the poison/OW was applied. Certain interesting xp twinking possibilities exist (as with Static Field) if xp goes to the last "touch" (e.g. poison nova, esp. preLoD).

Yes in PvP it's like Jarulf said 1/2 and it does not stack. Here are the two test values from my original thread:

clvl 85, 100% OW, 123-124 physical damage and 75% physical resistance of the target player:
One hit: 165 dam
Two hits: 193 dam

The first hit consists of 8 damage from the weapon (124/4*0.25) and 157 damage from OW (1/2 of the 314 for lvl 85 chars). After this first hit, the second one isn't affected by the target's physical resistance, so it should be 196 dam total (157+31), but the difference is very small.

EDIT: corrected explanation of my values (thx @ Madjai ).

Unless they changed in one of the 1.09 patches no one get any xp. The game does not keep track of who has "touched" a monster.

Interesting. D1 kept track. I find it interesting trivia to note changes in programmer techniques between D1/2.

However your reply raises a minor technical question nonetheless. You see, after reading your post I quickly went and confirmed that delayed monster death by poison nova and by open wounds does indeed credit the xp (tested SP and MP/lan LoD v1.09c). The problem, of course, has to do with issues of overwriting poison/OW damage, since clearly the game must remember who applied the poison/OW ongoing damage in order to credit xp if that DoT kills the monster.

Alas I didn't have the patience to do tests at, say, 5 FPS, and at normal speed I could not discern if the poison green tinge had passed before the death of a mob first OWed then poison nova'ed (the xp credit went to the poisoner--the poison was about 50% HPs strong whereas the OW was going to kill the mob all by itself near the 4 second expiration).

While fooling around I did notice poison nova completely overwriting a poison arrow (arrow 401 poison over 10 seconds, nova slightly higher damage per frame but, of course, vastly shorter). I imagined, at this point, that having a self-dose of very strong but extremely short-lived poison would be useful, in the absence of antidotes

It would be nice to know, if you recall off of the top of your head, if OW/poison overwrite each other (and if so, always, or on the basis of damage/frame etc.).

-- Crys

p.s. I've heard people say OW doesn't work on bows, so I suppose the two LoD uniques, bow/xbow are broken, eh? Bummer.

As a minor trivia, it would seem pausible that summoning an iron golem from a Malice weapon and then Enchanting the golem ought to produce a golem doing 100% OW vs. phys immunes (presumably at the clvl of the summoner). Given the Eth bug, I'd assume the golem would always be hitting as well.

A more esoteric trivia would be to wonder if D2 uses "slots" ala D1 such that a long poison would enable a new player to join a game and immediately get xp (from the monster finally dying of poison after the poisoner had left the game). Of course under perfect conditions this might imply a rather odd event: "Welcome to the game. Congratulations, you have just killed Baal!".

Alas, with poison revamped (esp. poison nova duration) this is rather hard to do (I'm fond of my toxic hunter's bow of pestilence even though I haven't used it for real play in months, because 401/10 seconds is just so amusing).

>However your reply raises a minor technical question
>nonetheless. You see, after reading your post I quickly went
>and confirmed that delayed monster death by poison nova and
>by open wounds does indeed credit the xp (tested SP and
>MP/lan LoD v1.09c).

Interesting. I have never really looked ayt it but people seemed to say that you never got xp for poison kills. This was long ago before LoD of course. Since poison and OW seems to work by altering the life regen rate such death would be set by the function that handle the update of the character. It is possible that they added some check so that it can check who set any possible poison/OW effect and thus send that player (hireling, minion) as the "killer" onward to the death routine, or rather to the xp routine. The xp routine find the owner of any hireling/minion by the way. I really have not pondered much about this.

Perhaps it was that peopel said that poison kill did not use MF of killer since the item drop routine did not get the killer passed on. thus the xp rotuine COULD find the killer. I really can't tell. THe interesting question is then which effect affecting life regene is the the one telling who the killer is.

> The problem, of course, has to do with
>issues of overwriting poison/OW damage, since clearly the
>game must remember who applied the poison/OW ongoing damage
>in order to credit xp if that DoT kills the monster.

Yah.

>Alas I didn't have the patience to do tests at, say, 5 FPS,
>and at normal speed I could not discern if the poison green
>tinge had passed before the death of a mob first OWed then
>poison nova'ed (the xp credit went to the poisoner--the
>poison was about 50% HPs strong whereas the OW was going to
>kill the mob all by itself near the 4 second expiration).
>
>While fooling around I did notice poison nova completely
>overwriting a poison arrow (arrow 401 poison over 10
>seconds, nova slightly higher damage per frame but, of
>course, vastly shorter). I imagined, at this point, that
>having a self-dose of very strong but extremely short-lived
>poison would be useful, in the absence of antidotes

As for overruling an ongoing poison effect, the game simply compare the damage per frame value. So I suppose one can do cheesy things by having someone attack with a high damage poison effect with short duration. Not sure if there is such possible setups though.

>It would be nice to know, if you recall off of the top of
>your head, if OW/poison overwrite each other (and if so,
>always, or on the basis of damage/frame etc.).

Can't tell. Definately not based on which do most damage. I never checked the routine that implement "effects" on players/monsters. I was mostly happy to know that "this call sets the effect". The check for higher poison damage to over ride is NOT part of the "set effect" but code in the apply damage. Thus it would not be based on most damage. SInce "poisoned" and under "OW" is if I recall correctly two different effects, I see no problem of them running in parallell. No idea if they *set* the life regen value though, or simply modify it. It should be easy to verify if it nullify a high +life valeu from items. In the case of set, which ever is applied last would be in effect. For poison and OW on same attack, that would always be OW if I recall order correctly.

>
>-- Crys
>
>p.s. I've heard people say OW doesn't work on bows, so I
>suppose the two LoD uniques, bow/xbow are broken, eh?
>Bummer.

No idea about this.

>As a minor trivia, it would seem pausible that summoning an
>iron golem from a Malice weapon and then Enchanting the
>golem ought to produce a golem doing 100% OW vs. phys
>immunes (presumably at the clvl of the summoner). Given the
>Eth bug, I'd assume the golem would always be hitting as
>well.
>
>A more esoteric trivia would be to wonder if D2 uses "slots"
>ala D1 such that a long poison would enable a new player to
>join a game and immediately get xp (from the monster finally
>dying of poison after the poisoner had left the game). Of
>course under perfect conditions this might imply a rather
>odd event: "Welcome to the game. Congratulations, you have
>just killed Baal!".

No, no such things as slot (although each monster, player, missile and such has a unique idnumber that increases with each newly created (not sure if it inserts new ones if one is removed or if it simply go on). Each unit type (player, monster, items, missile and so on) has its own linked list set up for its data. These lists are held in the general game data structure. When a new unit is created, the game dynamically allocate memory for the needed data. In addition, several of the data for a specific unit itself is dynamically allocated structrues as needed. Much data is also stored as lists of data. Stats for example, including not only strength and such but all modifiers form items and many other things (in total almost 300 different ones), are stored as stat number and stat value for those stats the unit has. Most will be 0 or not used and that way memory for all those are not needed to be stored. It makes accessing the data slightly more time consuming though. References to monsters, players, items and such are in game always handled by pointer and never by any id number or such (although I think at times it may be checked to avoid something attacking itself and such), still this usually is handled by the pointers any way.

In addition to the game holding lists of all units, a room (a subsection of the map) holds its own lists of all units within that room and such. Hmm, not sure if all the above makes sense, it is a bit hard to explain in an easy to understand way. Hope you understood some though.

>Alas, with poison revamped (esp. poison nova duration) this
>is rather hard to do (I'm fond of my toxic hunter's bow of
>pestilence even though I haven't used it for real play in
>months, because 401/10 seconds is just so amusing).

Will take a look at her with this in hell on Realms this weekend.

Rip Hook has OW on it . . .

I don't think so, because the monster's health goes down with each hit and OW does not stack (take a look at my two PvP values above). Also OW does it's damage over 4 seconds, but according to my tests, if I don't hit the monster, the health bar remains nearly the same, even with working OW.

The same with 50% physical resistant monsters. Their health only decreases remarkable with each hit, not between. Monsters which took 4X hits with 123-124 physical damage took only 2X hits with 100% OW. And I hit them a few times per second, so the OW damage of 78 per second is imho not enough to explain that.

Also my PvP results indicate that. The second hit - the first hit = 28 dam, and without physical resistance of the target the correct value would be 30-31.

Maybe, but I tested only in singleplayer. The first feedback I received from realm players indicates that this bug or whatever does not occur there.

Once my brother accidently put some malicious script on on my PC. It kept running a process in the background that ate up much of the systems internal bandwith. Diablo2 running in single player was getting ping times of over 4000 and the frame rate for the game dropped to 1-4 fps from its typical 60+ that it usually runs at.

So in single player you are not immune to these delayed update effects.

You shold also check some of my comments on the life bar itself in the poison thread on this page. The main point is that it does NOT display the life of a monster in an even proportional manner. This makes the life bar a very poormethod for measuring the life drop of a monster.

For example from a test I was doing to determine the life scaling on monster when using the players X command (this site was down and I couldn't look up the previous posts on this point), I observed the following. Using a sorceress with a Ice Bolt that I could easily set specific damage settings (min dam=max dam), I could get a monster with 100 hit points and kill it with 1-100 point hit. If I used 4-25 point shots it would die also. But after the third shot the life bar was usually less than 1/10 its full length. If I used a 99 point damage shot it would not kill the monster. If I used 3-33 point shots or 4-24 point shots, these would not kill the monster either. My point is the life bar is extreamly inaccurate in displaying the amount of life that a monster has and is a poor method for measuring the damage that you do to a monster.

The only way that I have found that is accurate in determining these states is to use modded games for tests. To run the test both with the monster being able to barely survive the damage and again with the same type of monster set with just the exact hit points to die from the effects of what you are testing. Anything else leaves the margin for error in your observations in the area of 20+% of the monster total hit points. P.S. in a mod you can set the monster hit points to be an exact value and not a range as you would have to deal with in a open game.

Thanks for your hints .

Hm, perhaps I should better stick to weapon speed measurings and calculator programming than doing such exotic testings .

forums.theamazonbasin.com. has address 216.40.226.19

Turns out, ironically, that I could get through via DNS for the first time in over a week right after I read your post. On the off chance anyone reading this has had the same intermittent problem a post there discusses altering your Windows LMHOSTS file...
amazon basin dns foo fix post

-- Crys

For the record, OW is not a physical attack. Actually, its damage is undefined since it is not done through any normal atack routines, it simply lowers the life regeneration of the target.

Den