Touhou SRPG - Story of a Lost Sky [Update 1/22/11 - Version 0.6.0] (Thread 3)

[mew77]

I think the consumable items and the fixed small gain are most logical options.

Although all of the options are feasible.

[Psieye]

As an alternative to consumables, mark certain trash mobs as "gives extra SC". This way, players can plan ahead to give one character more SC than normal just by carefully choosing which mob to kill with what unit. Depends whether you want players to be fielding many or few "difficult to use" units. Also along that line, consider whether you want any dissuaders for over-using the best units - currently I get the impression if you deliberately keep Youmu away from the battle then you're playing at Challenge Mode difficulty.

[mew77]

Well you see Youmu is the jeigan character for much of the beginning becuase of how much damage dagger toss does now.

I can now one shot enemies with it without even adding focused movement. And I jsut started chapter 2

[Psieye]

By the "Jeigan" do you mean the true sense of "EXP sink that noobs waste kills on and then realise he doesn't grow in stats" or just "powerhouse at start of game you can rely on"? I haven't played through the latter missions and haven't looked at the data to tell, but has Youmu been deliberately gimped so that she'd be weak in the mid-endgame? If not, then Youmu is a regular powerhouse, not a Jeigan. She starts with much higher Attack Stat (Str) than other characters and given spells (weapons) inherently have a x2 aspect to them, it really hurts enemies.

So my point stands: do you want players to safely rely on Youmu for the entire game with no concern for the other characters, except for the rare (or will they be rare?) situations specifically tailored to take advantage of Youmu's weaknesses? Doesn't need balancing now, but it's worth thinking about as you design future missions.

[mew77]

I meant it as an early game powerhouse.

Actually I think it is because Youmu has both attack +1 as an action trait coupled with deadly strike as a support.

Blows gaping holes in enemies.

[Psieye]

Actually I think it is because Youmu has both attack +1 as an action trait coupled with deadly strike as a support.

Meaningless if her base stats were sucky. Attack+1 means at most +1 Str in earlygame, translating to +2 dmg. Deadly Strike adds 10% crit chance, where a crit gives +50% damage. These effects are dwarfed compared to how she has some 5 more attack stat than say Chen or regular enemies, resulting in some +10 dmg in comparison and making those occasional crits really hurt as opposed to being a minor bonus.

Granted, Marisa has almost as much attack stat as Youmu but people are finding her too fragile to be a tank. Likewise Ran is almost as good as Marisa in offense and hardly is the most fragile of your earlygame units but compared to Youmu who has superior offense and defence... Well, later in the game it'll balance out as you start to face strong magic attacks from enemies but in the earlygame they all have similar MDEF and Youmu has so much more HP than anyone else she can easily absorb the extra damage her inferior MDEF gives her.


I'd say wait until some Ch 4 or so before thinking seriously about balance tweaks. You can't put much weight on earlygame results.

[RuneDevros]

Regarding Youmu and other story characters:  Since the stat growth system uses a linear function of level and growth constants, we don't have a way of tuning the growth of characters down at higher levels at the moment. I think Youmu should be a reliable character throughout the game and not lose usefulness. She is after all, the one the story is being told from, so she would be like one of the Lord characters in Fire Emblem. Regarding how she plays, it seems that Psieye is right and that her HP is good enough to absorb any magic damage that she takes. It also is an important factor that we haven't really introduced any hard hitting magic using enemies. Later on when we face off against characters like Misaki and Patchouli, Youmu isn't going to be as useful. My use for Youmu is mainly to have her strike first and have everyone follow up. She can soak the initial counterattack hit and if others get the subsequent hits in, at least one will avoid the counterattack by defeating the target. That's my strategy for taking on walking trees and powering up Marisa early on.

Redesigned the interface to the predictor as shown in the screenshot below. Two arrows are shown for very Strong and very weak cases and a dash is shown for neutral and healing. This has been committed to the repository. This addresses two problems.
1. Feedback from testing indicated that people didn't really get that strengths and weaknesses were associated with spells instead of units. I got the question of why Fireball is less effective against a tree. This clearly ties the advantage disadvantage to the spell's magic type.
2. The old way of displaying strength and weakness was not very clear and unexplained. This way intuitively shows which unit has an advantage using arrows.

[mew77]

Great...the interface looks more like the fire emblem battle prediction window now.

And I can finally tell without having to constantly refer to the status screen for the chart. It gets tedious flipping between the battle and the chart.

But in truth fire should be good on walking trees...I mean they are freaking trees. Trees!!

then again swords are effective on trees too so I guess it makes sense, but I sometimes feel the system isn't intuitive. When designing you want the player to read your tutorials and have it make sense. I read through the in-game tutorial and then proceeded to fight...then in the next mission I ran into walking trees. Logically fire would be supereffective on trees. But in the system force works better. "lolwut" I thought. The system clearly ties the weakness to spell type rather than unit type, but it still is strange that fire is not effective on a tree. Maybe it's just too much time around the pokemon type chart.

You want the design to be done in such a way that the player doesn't question what the developers were drinking, though in ZUN's case that point is moot. (That creature is always drunk)

Spell synthesis makes sense as is, but how about also getting recipies as mission rewards so I can make more than basic attack types. I know you're already hard at work getting this to work so no rush.


This interface upgrade is a step in the right direction.

[RuneDevros]

Hmm... would changing the basic elemental spell to something other than a fire spell help? I can see how a player would be confused by that kind of a system. I've always had in the back of my mind that we could go back and rename everything to more Touhou-y things. Most of the names were chosen based off of generic fantasy settings.

Recipes as mission rewards: Planned. You get healing crafting at the end of CH2ST3-4. The goal is to have main spell upgrades occur in missions. We might get a package of spells at the end of CH3 with tier 2 spells when Yukari and Yuyuko return. I guess the reason I haven't pushed towards making a larger set of spells to use is that trying to test against so many spells is going to be difficult at this stage. I think the final count from the old forum thread was something like 27 generic spells by the end of the game + special character Spell Cards + items.

[Psieye]

There will be fire spells somewhere in the game (HI THAR MOKOU), even if not in the earlygame. Something needs to be done to convince people that your system makes sense. It could be as simple as "the spell gives them elemental alignment - even if they were made of paper, if they're wielding a spell that's strong against fire, they'll have high resists".

Personally I have to wonder why people think wood should be weak against fire when other more flammable things aren't, but I acknowledge this game logic is very widespread.

[RuneDevros]

That's a good point. Let's try to incorporate it into the first mission's tutorial.

To do list tonight:
- CH2ST3 work
- Maybe expand the prologue to include the planned Yuyuko / Yukari scene that we had long since planned.

[mew77]

The wood weak against fire is known partly because well lets see...

megaman does it

pokemon uses it

most things that have an element/type system makes plants weak to fire...

SO that point still holds.

Perhaps if the basic elemental danmaku was charged bolts or ice ball...something that isn't fire...might make it more intuitive.

Then again we should discuss synthesis spells and what unique synthesis spells the player can make.

[RuneDevros]

I'm having a lot of trouble developing CH2ST3.

Current draft goes like this:
- Enemies show up from all corners of the map
- You hold off the bad guys for 7 turns.
- Enemies end up all bunching around you tightly.

In essence, it's not a very interesting or fun mission. The player has no motivation to actively engage the enemies and one strategy that seems to work (from Gekko's report) is to split off and run to a corner of the map. I think at its core the mission design needs to be rethought for this mission. Here is what I propose. Let me know what you think.

The map will be divided into three sections
[ ##### Graveyard #### ]
=====================
[ Island ]||[Island]||[Island]
=====================
[ ##### Exit ######## ]

= and || represent rivers. Islands are connected to lands by bridges. We can say that there happens to be a river that flows to the south of the graveyard with a number of little islands in between.

- Most of the dialogue is the same up to the part where Fuyuhana's minions come into play.
- To cover Fuyuhana's retreat, she sends a number of very high level Kodama Lords / Walking Trees / Fireflies to the top of the map. It is very clear that this entire mission was a trap. She can cover her retreat and try and take care of the greatest threat to her at the same time.
- Your objective is to retreat and make it to the exit at the bottom with at least one unit.
- We can add a treasure or other discovery event in the middle of the map.
- There are forests on the islands. As soon as you step onto the islands, some enemies will appear out of the forests. I added a new map action to support spawning of enemies mid battle. We can definitely make good use of this.
- To prevent just rushing Marisa down to the exit, fireflies patrol the central island. They have extended range (Move+) and flight.

I think this would make for a much more exciting mission to give the impression that the player is being chased by some powerful enemies. It also gives them a specific goal to aim for and progress towards (reach the exit) rather than hold on for a certain amount of time where progress is not as clear.

@Mew77: I can understand what you're saying. Maybe we can introduce some dialog that says:
- The trees are in tune with the type of magic that is associated with nature.
-  Remembering that in Japanese mythology, things that live long enough can eventually become youkai, these are Youkai trees. By manipulation of magic, they can actively resist elemental attacks like fireballs. By manipulating spiritual spells they can resist the blades that try to cut them down. Danmaku battles are very complex duels with both sides weaving elaborate sets of offensive and defensive spells.

Spell Synth: What ideas did you have for it? I'll have to pull up my excel spreadsheet that I had describing the various combinations.

[Psieye]

SO that point still holds.

Perhaps if the basic elemental danmaku was charged bolts or ice ball...something that isn't fire...might make it more intuitive.

but I acknowledge this game logic is very widespread.

If I must rephrase my previous statement: Warning! TVTropes Link!

I can go on trying to point out why it makes sense in Megaman and Pokemon but not in this game, but this'd be a pointless debate done for fun only. Trying to sidestep the issue with "let's not use fire for elemental spells" doesn't help because Mokou MUST use fire attacks at some point and people will bring this issue up again. We are going to have to teach the players to accept this game's elemental system of "what you equip determines what alignment you are, not who you are".

-----------------

Right, as for this CH2ST3... I see, yes that's a meaningless mission indeed if all you're tasked to do is survive 7 turns. As I recall, all "survive for X turns" missions I've seen always had a "and defend this while you keep yourself alive" added to it. Otherwise stuff happens like in Gekko's report - you can run around however you like.

Your new "escape from this map" setting is more appropriate for this "we got ambushed" scenario. Best to have that as "get every unit to the bottom of the map" though, even if we have the strange "you can get KO'd to escape the map" loophole. Actually that's a question: what happens when a unit reaches 0 HP? Is there any penalty for letting a unit die?

Anyway, problem with allowing only 1 unit to leave is that you then get players who clear out all the mobs at the bottom of the map, then leave 1 useless unit there to end the map with, while they go tackle everything else on the map in complete knowledge they have an 'insurance' of 'get out of this sticky hole I dug myself into'. We are approaching the mid-game by this point. I know there will be players out there (because I'm one myself) who will power level say, Youmu by sacrificing entire maps to her, and then have her beat up all those high-level mobs you put at the top of the map for even more EXP. Better yet if she can bring a team of support units with her that can spam heals and distract the enemies so she can fight ever longer. Unless you intentionally make those high-level mobs unkillably high in level.

If you don't mind such behaviour, then the mission is fine. If you want to discourage it, then add "everyone must escape" and possibly also "within X turns". The map layout of that mission seems sensible though.

[mew77]

Alright alright already...I get your point psieye. You don't need to bludgeon me with it.

I'll post again after playing the game a bit more.

[RuneDevros]

Mew77, we will have to integrate some clearer explanations of the battle system. It is good that you raised those points because many players may have the same confusion.

I think everyone escaping would be too harsh on the player. I don't really like the idea of having to restart the mission after losing one unit. I do think that get to the goal by X turns will work very well. I hadn't thought about what you had said so you brought up really good points about needing to rush the players along.

As far as penalties for units being defeated goes, I think the only one is they lose the EXP they may have gained for the rest of the battle. Would you suggest something like an EXP penalty that I've seen done in MMORPG games?

[Psieye]

I think everyone escaping would be too harsh on the player. I don't really like the idea of having to restart the mission after losing one unit. I do think that get to the goal by X turns will work very well. I hadn't thought about what you had said so you brought up really good points about needing to rush the players along.

As far as penalties for units being defeated goes, I think the only one is they lose the EXP they may have gained for the rest of the battle. Would you suggest something like an EXP penalty that I've seen done in MMORPG games?

Wahaha, restarting after losing 1 unit (on any map) was the norm in FE games, but you choose not to be Spartan. That's fine.

Not being able to earn EXP they could have isn't much of a penalty to players who suicide their units at the very end of the map in a calculated way. But if you don't mind that sort of behaviour, then it's fine as is. You could go for "they aren't available next map as they're healing" for non-plot-essential units or some 'financial' penalty or maybe "items they were holding take durability/ammo penalty".

[Elcura]

Do units need a penalty for being defeated?  I would think that losing the unit for the mission was enough, losing EXP at the same time is MMO tactics for added grindan.  And not usually well recieved in that environment.  I think it would be pointless, too low is the same as nothing and too high (or even the perfect mid-ground) is frustrating.

After all, not everyone is a tactics god like Psieye, some of us make mistakes lol

[Psieye]

Yeah I've just been exposed to (and pulled off a couple) so many "we gotta do this even though the developer never intended it - they didn't specifically forbid it!" stunts and feats from the early SRW and FE games, I've gotten used to thinking "is it ok if they do something ridiculous like this?"

Perfectly fine if we're happy to let stunt artists pull off those feats.

[RuneDevros]

I don't really mind not harshly penalizing the player. I actually found restarting missions due to losing one character in FE to be more frustrating than enjoyable. I think I agree with Elcura that our current system works as in.

CH2ST3 Gameplay draft has been committed to the repository. This illustrates most of the idea I'm going for here. Script has not been edited to reflect the new objective.


Mission testers: mission can be launched by editing line 96 in mission_launcher.py and modifying the characters listed a little below that.

To launch CH2ST3, change line 96 to

Code: [Select]

mission_name = "CH2ST3"
(Realized I had forgotten to set the default to Ch2St3.)

Didn't realize until recently that Zer0 made some edits to the wiki. I really like the revised script so I'll try and integrate them into the revised version. Some more story stuff may follow in the wave tomorrow. One idea I've been toying with is giving the 4 kodama lords that appear in this mission personalities and names. We can have them reappear later on rather than use generic and flat enemies for the Kodama Lords.

[RuneDevros]

Heavy revisions to CH2ST3. I've integrated most of Zer0's edits and added some more stuff into the revised script. Check it out and let me know what you think. I also added personalities for some of the other Kodama Lords that show up. I've documented here what their personalities are and their histories: Wiki. Haruna's is the most developed since I was building her for a side mission a little later on that I've been brainstorming on. It'll be added to the mission design wave.

Lego posted her rendition of the shop theme to the Art wave. Please check it out and let her know what you think of it.

[RuneDevros]

Mew77 requested this: Spell Synth Database. This was an old version from last summer that I never got around to finishing.

Added a new song that Lego made to the game. It is the shop theme. I figured that the best place to evaluate it is in game, so I added it to the internal version enabled at the Treasures menu, the Spell Synthesis Menu, and the Item Trading menu. We have a thread over on the wave so let her know what you think of the song. It is the first one she has composed for us and I think it turned out nicely.

[mew77]

It prevents access to my gmail account. Can we fix that. And I just tested CH2ST3. I believe I MSNed you yesterday about the details.

Ah well...

(add to this post later)

[RuneDevros]

What prevents access to your gmail account? I'm a little confused.

[mew77]

We're sorry, ninjaducklord@gmail.com does not have permission to access this spreadsheet.
You are signed in as ninjaducklord@gmail.com, but that email address doesn't have permission to access this spreadsheet. (Sign in as a different user or request access to this document)

Find out more about this topic at the Google Docs Help Center.

As I was saying. In the missions I can't seem to beat them without losing at least one unit during the process. In CH2ST3, I beat it with 4 units remaining after using heals a lot.

I think it's designed well, but I was considering adding another bridge in the center so I could split my group up into more than 2 groups to retreat.

Why do the faeries have more HP than the walking trees?

[RuneDevros]

Okay I guess I hadn't set my permissions right when I made the spreadsheet. You should have access to it now and I set it to be viewable without needing to log in.

Yeah this is intended to be a really intense battle that sets the player up for the rest of the game. I don't think it is unexpected that you lose units. I think I got through with one unit lost when I was playing (Mission tester is with level 5s). Lots of healers are a good thing. I'd like to maybe add a treasure box cherry tree along the way to add to the challenge. The intent was to split up into three but adding a second bridge would open up the battle more. That seems to be a good idea.

They have lower defense than the trees I think. We can tweak their HP values. I've given some though to redesigning how enemies are declared in the mission data. I think it would be better rather than to redeclare every enemy, we have a set of enemies to template from in a separate data file and then load them individually. Something like:

Code: [Select]

<enemy_unit name="Fairy A">
    <template>Fairy</template>
    <level>5</level>
</enemy_unit>

This way if we tweak one enemy we don't have to fix all the enemies in the entire game.

[Psieye]

I've given some though to redesigning how enemies are declared in the mission data. I think it would be better rather than to redeclare every enemy, we have a set of enemies to template from in a separate data file and then load them individually.

Yes, a template which can be overriden for individual instances (e.g. an unusually high HP fairy but otherwise normally statted) is a big step in a very good direction.

[mew77]

What I see is that templtes will make it so much easier for you to code pre-mission scripts. Just initialize all enemeis based on the templates. Make the template a seperate class that can still be modified in the main game engine, just in case you need special enemy stats. Otherwise this seems fine.

I'm probably just repeating wht psieye was telling you anyway.

In the spell synth munu, is the order you select the synth items important?

What about spells that require three synth items?

Lastly...could you allow traits to be automatically set to standby when you learn them. I find it annoying to sift through menus to equip new traits. Just a little nitpick here.

mew77
Now returning to real life.

[RuneDevros]

Spell synth is just select a recipe and go. There's no order to it.

Three synth items the current system can support but nothing's been drafted up on them yet.

That's a good idea to automatically assign newly learned traits.

I updated CH2ST3 with more story in battle demonstrating the ability to trigger conversations in battle. I also added a brief ending. Next up is fixing the prologue and CH2ST4 which is the meeting back at Akyu's place. I think any new features on the engine level will have to wait till 0.4.3. I'd like to finish up CH2ST5 do our regular alpha testing and then launch. Ch2ST5 is a quick skirmish / boss fight to wrap up the chapter, so it should be a not too difficult battle.

[RuneDevros]

Merged Zer0's Yuyuko and Yukari script into the prologue mostly unchanged. Next up, Ch2St4, in which we learn about the past of the world. It might be a little difficult considering that we don't have Fuyuhana drawn up yet.