Thoughts on the Design of the Map System

Great post Scrotie, thanks for taking the time. You had me here:

"

ScrotieMcB wrote:

I believe these failures have occurred due to a development mindset which looks at the map system as an endgame currency and time sink first, and as replayable endgame content second, if at all.

It's important to look at the maps system as sustainable endgame content first.

..and the rest was delicious icing on top of the red velvet cake.

/signed, please implement GGG.

P.

Very good post, Scrotie. I liked everything I read in your OP; however, I also liked what LSN said on page 1 (disclaimer: I've only read the OP and page 1; running out of time to catch up on the rest of the thread).

Before I get into that, though, one thing I'd like to add as a suggestion to normalizing affix quantity: hide the affix quantity percentages from the player. If it were established that per prefix and per suffix, each modifier would be relatively as rewarding, then the expectation becomes more affix yields more near-equally rewarding quantity. Hiding the final quantity (beyond chisel quality bonuses) would make it more difficult (impossible) to assess the "correct" combination of modifiers for most efficient returns, thus reinforcing quantity of modifiers over quality of modifiers' quantities.

I don't know if this is really a good idea or not, but I thought I'd share on the chance that other minds would mull it over. The ol' gut tells me that, at some point, a degree of "staleness" would return to the system as players learn new optimal combinations, and this might help relieve the burden of further re-balance to keep the system fresh.

Now, I absolutely agree that the first step is to "fix" the map system. It is not what it was intended to be, and it has potential to be so much more; that that potential is unused is a damn shame. However, I as I said, I also really like what LSN said on page 1.

"

LSN wrote:

In my perfect PoE world maps would not at all (CHP: seldomly) drop new maps but you could obtain maps from the other content only (CHP: much more reliably) (which can be randomized very well too such as randomized questzones given from npcs in a3n (CHP: such as an Endless Ledge, only more than just Ledge, with only one portal out, and a strongbox at the end of each level; opening it would replace the exit to the next level with a portal back to town, and the deeper you go the better chances are of dropping more and/or better maps)). This way maps could become something more special instead of being the standard repetitive content of the endgame. Droprates for items and orbs could be heavily improved on these new maps and grinding the normal endgame content would reward you with maps once in a while. I like the ratio of 1:2 map:nomap content (CHP: whatever "feels right" after testing). If these new maps (CHP: were more scarce, and...) had heavily increased item dropchances it would be even fun to roll them decently and even play the lower ones just for magic find reasons while the normal repetitive content would be more for mindless (CHP: randomised alternative content would be more for fun, ...) grinding xp and maps.

One problem I feel that is missing from your OP, and cannot be addressed solely by "fixing" maps, is that a large portion of the variety is restricted to (eventually) unrewarding, low level maps that (eventually) become nothing more than long forgotten stepping stones for a much smaller pool of more-rewarding higher level maps. 90% of maps are not special because there is a definite, linear progression to the map system. This too must be broken such that lower level maps are more than just stepping stones. This is partially resolved by a less masochistic map recipe formula; however, low level maps would still never receive an orb more than a trans/alt/aug or three, maybe an alch.

Low-mid level maps need to feel special too (edit: I forgot an important point here--such that players actually want to run them as well as higher level maps), and the way to do that is to remove sole end-game expectation ("pressure") from the map system, and share it with an equally fun alternative, accessible system--once maps themselves are "fixed" of course.
(aside: I envision a vendor-craft recipe using Vaal Orbs that maintains rolled affix, and adds other crazy properties to lower level maps--including but not limited to randomly increasing the map level.)

Following this, then, are map-gated uniques. I hate the system, as is now; however it could be an engaging process were the map (and alternative content) system(s) designed with "entertainment" as first priority, "rewards" a secondary consequence of said entertainment, and masochistic grind-fest a long forgotten thing of the past--it is reliably "fun" now, after all. Specific uniques (or, God forbid, vendor crafting "materials") could then be gated (or more reliably found) behind specific maps, thus making maps of all levels special and engaging (another edit: addendum: utilizing a more-fair map vendor recipe system to "craft" maps of specific names using your pool of otherwise unwanted maps).

At any rate, those are my two or three cents.. Perhaps my solutions are not correct, but at the very least I do believe I've adequately diagnosed another problem in the system.

"

ScrotieMcB wrote:

7 (Monsters gain Endurance Charge when Hit): Monsters are capped at 3 charges, so I don't believe this one is unreasonable.

Oh. It would still be rather tough, but that helps.

"

ScrotieMcB wrote:

13 (Monsters have increased Accuracy): I was actually just guessing when I wrote 30-40%. After actually running the numbers, I've determined the appropriate values are... 50-80%. That's right, harsher. In any case, those values are carefully crafted to ensure that builds with less than 25% chance to evade feel no strong effect (9-13% increasing attack damage received), those between 25% and 60% evade feel only a mild effect on melee attack damage (at worst, 25-36% increased damage received) but a drastic decrease in effectiveness in Ondar's Guile, while finally those with over 60% chance to evade can notice an effect on melee attack damage greater than a 25-36% increase, increasing as their block chance increases, but their new chance to evade remains at 45% or higher, thus virtually zero impact on their Ondar's Guile. If I'm wrong and the whole Ondar's Guile thing becomes a problem (because it really is severely hit by this affix at certain levels), it's possible the affix might be limited to Monster melee attacks to allow Guile to remain map viable.

The problem with this is that you also need to take accuracy mobs and rares with the accuracy/crit aura into account. An accurate pack of archers can still be fairly threatening to an EV build, even with Ondar's, and having the bad luck to run into, say, a Quick, Accurate rare with the accuracy aura in a map with this mod would easily wreck most EV builds. And my point about needing something that's difficult for AR builds still holds - EV builds take extra damage from accuracy and extra damage mods, whereas AR builds only really have to worry about the latter.

"

ScrotieMcB wrote:

20 (Monsters have a chance to Avoid Stun): You are aware that the map affix which exists now gives a complete stun immunity, right?

Yes, but as far as I can tell, the goal was to give builds a certain challenge instead of a map they'd just reroll because they can't do it. Admittedly, I can't think of anything better to challenge stun builds than the Chance to Avoid Stun other than reduced stun duration/increased stun threshold, but having those as mods would at least allow stun builds to dodge bad streaks of "chance to dodge" RNG.

"

ScrotieMcB wrote:

22 (Monsters cast Summon Skeletons on death): The purpose of the affix was to replace Fracturing; the reason for the change is to eliminate corpse clutter. I don't consider the current Fracturing to be a joke mod.

Oh, that makes sense then. I still don't consider Fracturing to be a serious high-level mod, but that would probably help. It'd make Vaal DD users sad, though :P

"

ScrotieMcB wrote:

34 (Area contains Proximity Shields): It doesn't need to be a commonly appearing affix. Also, I enjoy the Evangelist design; one of my favorite monsters. Although I am the sort of weirdo who considers Sceptre to be the best area in the pre-map game...

In my opinion, the main flaw with Evangelists is how commonly they spawn. Having a prox shield every pack or so is a challenge; having to find a way to get arrows through four of the fucking things is just tedious. Plus they're the one type of mob which becomes extraordinarily hard to kill if hit by a bad bout of desync. That said, I guess it'd be manageable as an uncommon affix, although I think the number of prox shields and the areas they cover would need a bit of fine-tuning; overlaps would still really have the potential to fuck everyone over, melee builds included.

"

ScrotieMcB wrote:

45 (Block Chance is Unlucky): The interesting thing about this one is I originally had it as 20 to 30% reduced Player Block Chance (same mechanic as the Block Chance Reduction support gem), and someone pointed out to me that making it Unlucky would use a cool mechanic while just averaging it out to a 25% reduction for 75% block builds (75% of 75% is the same as 75% reduced by 25%). In any case, it's a tough affix, but it's not unbeatable.

I still think it's a little bit too rough for builds that use block but haven't managed to max it, since every point short of max block is equivalent to a more severe drop in effectiveness, but that might just be me.

"

ScrotieMcB wrote:

48 (Less Player Attack and Cast Speed): I personally find the reduced movement speed to be the only truly maddening aspect of Temporal Chains maps. I carefully avoided reducing player movement speed in any direct way (only thing close is chilled ground).

True, that would be better than temp chains, but I still don't know if it'd constitute the auto-reroll seen so frequently with temp chains. Basically, my point is that it's very similar to a mod that a lot of people hate, but I can't say whether or not most people hate it because of the movespeed loss, the attack speed loss, or both.

"

ScrotieMcB wrote:

51 (Players gain a Bleed Charge when Hit by Physical Damage) and 54 (Monsters cast Storm Call when Hit): Rather than granting a chance, I'd rather they trigger every time for less damage. Get the damage values right and a 100% chance is still fine.

Good point. I still think that the Bleed Charges would be a little rough just due to how Bleed works, but if the damage is manageable it should be okay.

A little off topic:



Keep going. :P

+1 to a good post and good set of ideas, Scrotie.

but about that risk/reward part well, you remember the "RNG onion" model right?
your suggestion properly uses "good" RNG to generate content, re-playability and challenge, but it replaces only a couple of layers with a risk/reward concept.
the game still has plenty of layers of high-variance "bad" RNG in its core, to ruin that.

I'm not saying GGG shouldn't follow what you suggested. on the contrary.
I am saying though, a risk/reward-based change has to be universal rather than just applied to one part of the game and not the other.

"

CanHasPants wrote:

one thing I'd like to add as a suggestion to normalizing affix quantity: hide the affix quantity percentages from the player. If it were established that per prefix and per suffix, each modifier would be relatively as rewarding, then the expectation becomes more affix yields more near-equally rewarding quantity. Hiding the final quantity (beyond chisel quality bonuses) would make it more difficult (impossible) to assess the "correct" combination of modifiers for most efficient returns, thus reinforcing quantity of modifiers over quality of modifiers' quantities.

This wouldn't work long-term. Players would collect mountains of data in a community effort and eventually work out the quantities anyway, albeit without precision. It's kind of like how we have a pretty good idea of the chance of 5L or 6L with Fusings, or the chances of off-color with Chromatics. It's not necessarily a bad idea to conceal the information, but I wouldn't consider it something which we could count on, either. I guess I consider it a trivial detail.

My entire hope is to make the map system as non-repetitive, non-boring, and fun as possible. Challenging sometimes; also not challenging sometimes; the degree of challenge is also an element which should be non-repetitive, unpredictable, and thus non-boring and fun. After all, intense challenge works best when there are respites before and after; the game design term for this is "pacing," and since it's a random system, we need to rely on some form of random pacing.

@johnKeys: The last time I've ran across you using the term "RNG onion" was in my thread entitled "RNG is good." I don't fully understand what that term means to you; I think I have an inkling from context, but perhaps elaboration is in order. However, my gut reaction is that I believe that any system of "bad RNG" can, and should, be converted to one of good RNG, and that good RNG is a fucking awesome, powerful force for fun.

"

ScrotieMcB wrote:

I believe these failures have occurred due to a development mindset which looks at the map system as an endgame currency and time sink first, and as replayable endgame content second, if at all.

Endgame players should at least be able to continuously sustain a level of map in which the highest maps (78s) are droppable.

Totally agree, but if GGG followed your advice, they'd no longer have a pay2play map economy, and we'd likely see dozens of parties on the in-game board that required no entry free at all. How much fun would that be?

"

RogueMage wrote:

Totally agree, but if GGG followed your advice, they'd no longer have a pay2play map economy, and we'd likely see dozens of parties on the in-game board that required no entry free at all. How much fun would that be?

I thought the purpose of the in-game party board was to list WTB and WTS. Don't see how anything I'm suggesting here would change that. :/

I still think the problem is not the current map system, but player's expectations.

You can theoretically run 78 maps all day.
You can theoretically reach level 100.

But these maps will most likely not drop, and buying them will result in a net loss.
Level 100 will force you to run the same two maps for ages.

Result in feedback:
"Map system is boring."
"Map system is unrewarding."
"Maps are unsustainable."

If you just adapt to the fact that this game is not designed to gate you smoothly to level 100, all the problems are gone.

You finish the game with merciless dominus, then you run some maps.
You'll probably get to 75 maps without much of an issue, then you can run some 76 maps, a couple of 77s, and on rare occasions, you'll even run a 78 map, while progressing to level 85-90.
If this is fine to you, then the map system is actually very rewarding, diverse, balanced and interesting.

Level 100 is an endless grindfest and forces you to literally torture yourself.
So: why would someone even want to do that?