Monster Randomisation

[Jedouard]

I noticed in a screenshot of "The Unburied" for the bestiary update that there were 7 or 8 identical copies of the monster in one room. The same goes for the hordes of wall-climbing monsters the barbarian fights in the first gameplay video. This reminded me of the days of the Diablo II and even Diablo I, where we were working off of the same cookie-cutter sprite.

While DI and DII were indeed fun at the time, if Blizzard would like to get the player more into the world, micro-level individualization of the creeps would be really helpfull.

I have four suggestions for randomization:

1) Divide the any given monster type's texturing into six or seven parts (leg, leg, arm, arm, head, abdomen, back, etc.) and then have five or six skins per part.

2) Make the "body" size for any given monster type vary by a given percentage, say 30%.

3) Make the movement of the creature dependent on the above. If the skin of the creature is pretty gnarled up at the legs, give him a limp. If his arms are more muscular and he is huge, then make him swing like a brutish hulk.

4) Lastly, make the monster types stats vary according to all of the above. Beaten up, infected, hobbling, tiny variants are weaker than their brutish, fast, big counterparts.

This would not only be quite enthralling in terms of "feel" and storyline, but it would also cause the player to really think heavily about strategy, prioritizing his damage-dealing toward the more threatening variants and taking into consideration movement speed, damage, etc. of each individual creep.

However, I understand that there may be hardware limitations to this.

[NioTumsSpik]

Iirc they have said that they will have some sort of monster randomization

[NASE]

Not hardware limitations, design limitations. Designing randomness like in 2 and 3 is very hard - if not impossible. So it should probably have to be hardcoded, making develop time even longer.

And 1 has the same problem. That are exactly the things that take time. Making 6 skin per monster and making sure the fit onto each other will only lengthen the develop time.

And to gain what?
very little as all the monsters will die in a few seconds anyway. it's not like many will stop to admire the monsters - I know I don't.


P.S. The developers are aware of the 'silliness' of the group of Unburied. In the final release, you won't probably see more then 2 tops.
And those wall climbers are in a very early build. Perhaps they have changed things already.

[SlechtWeerBeer]

7 parts, 5 'skins' per is already 35 'skins' for just one unit, which, as NASE stated, will spontaneously explode in a bloody mess as soon as you look angrily. I personally wouldn't mind 2 or 3 skins for a unit (Skeletons in Diablo II had this, fyi), but not random body parts, please. Too much work for too little gain.

[Krugar]

Indeed. The amount of work needed is unsurmountable. Taking Diablo 2 as a a reference, we had approximately 70 uniquely designed monsters. Assuming, only half of these would be "skinnable" (so, 35 monsters), 3 skins per monster would mean 105 skins, minimum. And a skin is made of more than one texture. A typical monster would contain... what?... 20 textures?

Each skin would then need to go through the same process of conceptualization, design, sending for approval, perform any needed changes. And I'm not even thinking the option of having different monster body parts. If there is one process that can take longer is the final approval of a model.

Code-wise, the thing is pretty straightforward. But artists would be swamped with work and they would still have to work on tress, rocks, walls, floors, placeables, items, equipment, etc...

Finally, this would weight down on the game resources, increasing the needed memory to run the game and consequently the game minimum requirements.

[Wirt]

I can live with 2 or 3 different skins/colors of monsters from the same model. Having them differ in size a bit (suggestion 2 above) should not be a development problem at all, and you might also differ their stats (suggestion 4 above) on the same scale. I like that idea.

Showing monsters with different weapons, armor and shields and making their damage and defences differ based on these, would be very nice too.


Wirt

[Wirt]

BTW, this is what diablowiki.net has to say about it:


Component System
This term referred to the way monsters of the same type in Diablo II could appear different. Skeletons spawned with different types of weapons, armor, shields, and so forth, just to give the demonic legions some variety in their appearance. The different equipment wasn't purely superficial either; it went along with logical changes to the monster's damage or defense.

This technique is being used in Diablo III, and while details haven't yet been revealed, the team has mentioned their plans to use graphical tweaks and palette shifts to give the monsters some variety.

http://www.diablowiki.net/Monster

Wirt

[Telzen]

Best way would be to have 3-4 different models for each mob and have it randomly select one when spawning it. But I really don't care and would rather them not waste time on it.

[Jedouard]

I agree with this taking up more memory (which is why I said hardware issues), but, even if this took up an extra gig or two of hardware space, how many of us would have an issue with that?

I differ on the other disqualifiers.

That it takes more time to program is not as big of a deal as we are making it. First, Blizzard takes a long time to make the game anyway, and if they feel it is not done properly, they don't release it. That said, this woud be something that would make the game proper, which I will defend in a second. We are discussing a company that earns $150 million a month just in WoW revenue. If there were 35 partial skins per monster (something would make for thousands upon thousands of combinations just over all skins) as compared to the 5-15 complete skins in Diablo 2, then they need to hire 3.5 times the artists and programers working on this. The same goes for having 4-5 movement types per monster instead of one. Given how small the design team is, forking out an additional $400-500 thousand a year for the extra manpower is not to extreme given how much the company makes.

The other issue of not caring because they are all going to die anyway is, I guess up to the player. But then why not have stick figures? I think this would add a lot to the erieness and realness of the game. And, as I first mentioned, I think it would really add a lot to strategy.

[SlechtWeerBeer]

I agree with this taking up more memory (which is why I said hardware issues), but, even if this took up an extra gig or two of hardware space, how many of us would have an issue with that?

I would. Although I have a rather new computer (half a year or so), I don't have infinite amount of hard disk, and I don't plan on expanding any time soon (would involve buying a 900W+ power supply. No thanks). Imagine if every game took up an extra gig or 2. Or more, if it's a big game and/or has high-detail textures. Hah.

The other issue of not caring because they are all going to die anyway is, I guess up to the player. But then why not have stick figures? I think this would add a lot to the erieness and realness of the game. And, as I first mentioned, I think it would really add a lot to strategy.

There's a difference between a good looking Skeleton (graphics wise >.> ) and a stick figure. I cannot figure out what would be eerie about monster variation, though.
As for strategy; Only if the differences are extreme, such as damage going from x0.5 to x5. Otherwise, it's just a Zombie with 5 hp and 0-1 damage extra, closing in on your Spontaneous Combustion skill just a little faster than the next. Big deal.
Oh, and Fallen, for an example, had variable health in Diablo II. Did you ever seriously consider taking out those with 4hp more before the rest? Or those with torches? I seriously doubt it.