Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[IndigoFenix]

For the golems, my advice would be to do one or more of the following:

1. Don't allow the subgolems to "mature", just have them remain small golems.
2. Allow the subgolems to mature, but remove the spawning ability from the golem clones.  This would allow them to multiply to an extent, but not indefinitely.
3. Give the golem a timer that is reset every time it is hit.  If it goes too long without being hit, destroy all spawns and restore its HP.
4. This is a bit trickier, but probably what I would do if making a similar enemy: don't have the stones mature at all, instead give the golem an ability that makes it stand still and makes one of its stones jump toward it (you can loop through the sprite list to find a suitable stone, if one exists, using the code below).  The stone is destroyed and the golem heals 1 HP.  Basically it slowly puts itself back together if you leave it alone for too long.

I also made an example of an enemy type that can multiply over time from any individual, but the total number in a given "group" is restricted.  It's in the Amanita package, you can check it out if you want an idea of how this kind of thing can work.

The usual way to destroy all spawns is:

Code: [Select]

Each _room._sprites
  If #._spawner = _this
    Set #._hp to 0
  End
End

Note this only works if the spawns were created as the original object's spawns (or projectiles).  Otherwise you can do something similar by setting a custom variable to the spawn when it is created.  Also make sure that any transformations the spawns make are actual transformations; if you do the destroy/create thing then the _spawner will not transfer.

Any particular reason why you want Spids to get smaller in its second phase?  Nothing wrong with that, but most game bosses tend to do the opposite.

Spoiler (click to show/hide)

What are your plans for the secret spot in the hidden area?  I'm assuming you're planning on making some kind of thing to return to after you get an item from a different world, does that mean the player will have to complete the dungeon a second time?  Will the second time have harder bosses?

[Kamani]

I’ll try out 4. I agree in that it sounds the most interesting.

Oh, didn’t know about _spawner. I should have checked for that.

I thought about setting them to variables, but then, if there are more than one, that falls. Bright side: using method four, I have to worry less about proper transformations.

The 2nd phase has way less life (and less damaging big attacks), but they move faster (and passive move, unlike phase one, who stays still except when charging). The idea is that it is that Spids is actually running low on energy, so they are using more finesse than brute force. If you can keep near to Spids in the second phase, they shouldn’t last too long, and they don’t hit as hard, but they are harder to hit and dodge.

Spoiler (click to show/hide)

The secret things in the secret rooms are going to be optional things to explain a bit more about the world, I think. Although I am considering hardmode versions of the bosses, these won’t be necessary or anything. I just want players who look around for a lot of secrets to feel rewarded (for example, there will be threeish secret rooms in Lusuria (primarily accessible if you have a rope from the dungeon)). This is in large part because I always love finding secrets, so I figured others might as well.
Although doing a secret bonus dungeon for people that found all of the secrets would be interesting...

Edit: Okay, I think Spids boss is done(ish). I don't really feel like it is good as the original, but I took like 2 weeks on the original, and 2 and a bit days on this, so... I'd try to improve it, but I can't figure out what it is that doesn't seem right. If you play it, please tell me what is up with it. For now, I am going to work on post-dungeon Lusuria.
Edit: Oh, and Spids's dialogue. Have to do that first.

Edit: Bad glitch (halted progress entirely until fixed). Double number entries don't seem to be working (like those in the jump effect or in image scale). I click on it, and then both of them disappear and can't seem to be changed. 
Edit: After some brief testing, it appears that they CAN be changed, but it requires a workaround and should still be fixed.
Edit: After some more brief testing, I can only get one digit with the workarounds. Ugh.

[IndigoFenix]

Ah, I was worried about something like this.  I've been patching up memory leaks (turns out there are a whole lot of them) and it's possible that some of the fixes would cause other issues.  I did some testing but clearly not enough.  I'll change that part back as soon as I can.

[Kamani]

Okay. Thanks.

[IndigoFenix]

Should be fixed now.

[Kamani]

Oh, good. Thanks.

Edit: I guess I'll need to test how long it takes to run through the dungeon to see how long it takes so I can set up the value of tokens.

Edit: Well, instead of doing that, I made the changes to golem. Tell me what you think, if you have the time. The golemstones heal them after 10 seconds for 1.5 mana, and as long as one golemstone is present, the golem cannot die (well, you could probably reduce their health to 0, but that would take a LONG time).

That, and I figure out what jump you meant when you were talking about the flame at the bottom of the jump. Yeah, that was bad. I moved it.

Edit: Took me about 10 minutes. I'll make it 10 sparks per token, I guess. If people take longer, I can adjust. I might have gone faster than normal, since I know it well, but I also don't know how much most people will spend. I bought a sword, and it made it feel excessively easy, but I know the enemies quite well by now. Edit: I guess I'll use 15 sparks instead, thinking about it.

Edit: Well, I finished Spids and did a rough time test. Now I just have to do the post-dungeon town (instead of using a bunch of tests in everything, there is just an entirely different set of rooms that look identical excepting objects. That should be faster). This should be done tomorrow or two days at the latest (I am pretty busy this month, so I might be slowed down a tiny bit).

[IndigoFenix]

I'd advise against copying rooms - aside from being memory inefficient, it'll probably wind up being more work in the long run, especially if you go back and edit things later.  Gates will have to be re-aligned, and if you're auto-linking from another world the exit gate won't align with the entrance (so entering from one world will cause you to transport to a different one when you leave).

The easiest way to make multiple versions of a room with very different sprite arrangements is to put in both sets of sprites and have one of them destroy itself in its _create function depending on whether or not a variable is set.  (You can set a global variable in a world without using items by targeting _world.)  For characters though, I'd just use one sprite and put everything in a single flowchart with different start points.

There are other things you can do if you want to get really fancy (I made a random encounter system that uses a special "variant" helper sprite to make a room with 10 different variations) but for only 2 variations this should be easy.



Edit: I went through the dungeon - nicely done!  Took me quite a number of tries to beat, despite having full equipment, but maybe I'm just not as good of a gamer Just a few things:

Flames still need to be able to jump back.  The last one actually jumped out of range in the last room (where you have to kill all the enemies), and since the room is persistent it made the dungeon unwinnable (without cheating).

Spid's intro - nice!  Should probably be facing down when done though.

Sometimes when the second phase starts, if Spids is partially off the edge they get pushed entirely off.  Maybe they can jump to the center of the room when transforming?  A quote would be nice as well, to give a moment to prepare for the second phase.

Something seems wrong with the ending cutscene.  It works if you put "continue immediately" on the Walk command that makes Spids face downwards, though this does appear to be another cutscene bug that I need to fix.  (By the way, you can set _face to make an object face in a particular direction.  You can even {Set _face to _direction_to(sprite or point)} to make it face a particular object.)

The new golems are...interesting.  Might want to add a hint about how to beat them somewhere in the town (some things that may seem obvious when you've made them could be less clear to a new player.)  Golems along with sorcery flames are a tough combo to deal with.


Don't rush yourself!  It's a nicely-done world and the characterizations lend a lot to its charm (maybe some multi-character sidequests could go in later?).  I'm probably going to showcase it in my next video once it's done.


Now, back to sorting out these memory leaks.  It's a little bit like pruning dead branches off a tangled bush that's stuck to a ceiling, except the bush is moving, the dead branches are still growing, and the branches extend through time as well as space.

[Kamani]

Fair enough. It just means setting a lot of stuff to have destroy on create bits (most of the characters are in Remy’s tavern after the dungeon. This is because it is lunch, and a majority of Lusurians are unable to easily cook for themselves).

I’ll do that.

Makes sense.

Huge secret to winning: spring jump. You can use it to avoid a lot of the tough projectiles. But based on that, I might have to increase the number of sparks per token. It’s tough for me to judge difficulty since I know the entire dungeon very well.

Oops! Sorry. I also realized I have to make spore clouds from sporeghasts more visible.

I tried to make it do so, but I messed up and forgot to fix it. I’ll do that.

Oh, yeesh, missed that in testing. Thanks. I’ll do so.

Huh, weird. I didn’t have time to test it yesterday, but I’ll fix that all today. I had similar problems in the Remy cutscene, but I worked around them, and I thought they got fixed. I’ll just use _face instead (I didn’t think about it because I had been using the move commands that way before I learned about _face).

Not entirely sure if interesting is a good thing, but I’ll make sure to put a hint in town.

Ugh, I’ll make sure not to rush. I just wanted to have this done by my original target and I failed that. But, it’s better to make a good world than have it done fast. Multi-character sidequests would be interesting to do later. I’ll be publishing (if that is the word. I’m blanking on it) it without them for now, but it would be a good thing to do. Now I just have to figure out what kind of sidequests would work.

Yeesh. Good luck.

Also, thanks for all the feedback. It is highly appreciated.

Edit: Fixes done:
Flames jump back after 10 seconds of not being hit.
All Spids fixes are made (turning, falling, and cutscene fixed and added (this included making a 13 line if statement that destroyed every single boss skill object. I figured the longest function part I used in Lusuria would be dialogue. I guess that showed me)).

So lorewise, what are sparks? Stores use them, even in places that don't know about gates, so are they the Ancient's currency (like the language thing?) or something?

[IndigoFenix]

Sparks are bits of "stuff" that falls into the world from "outside" - little bits of reality that is stuck in the world of illusion.  It is soul-stuff, star-stuff, solidified light, pure wealth, the basic material that composes all things.  It is the stuff that alchemists speak of when they use the word "gold" and the broken pieces of divine light that the Kabbalists talk of (which is where the term "sparks" comes from).

It is also the stuff that floats around game worlds for no clear reason, and which is, for even less clear reasons, usually yellow.  Coins, rings, bananas... All are different manifestations of the same thing.

They may appear differently in different worlds and to different people.  Sparks transform into whatever people see as the main form of value in that world.  So stores aren't actually trading sparks, they may be trading coins or whatever.

The role of the Travelers is to find these sparks and return them to their source, from which they can create new worlds.



I think most of the memory leaks are cleaned up.  At least, I'm hoping things should be less prone to crashing.

I'm going to redo knockback physics to make them feel more natural.  I don't think any of your objects receive knockback, so it probably won't change anything significant.

[Kamani]

On sparks: Okay. Cool. Thanks.

On crashing: I didn’t crash at all during my last edit session, so I think it is probably pretty good. Thank you.

On knockback physics: Okay. I disabled them on my enemies to prevent skill cancel, so yeah, won’t effect them much, but neat.

Edit: Something that would improve flowchart convenience: when you are adding in a piece (startpoint, dialogue, function, etc.), it would be helpful if it moved to the top left corner of where you are rather than the top left corner. When you are editing something on the opposite end of the flowchart, it can be really quite inconvenient. Not something that is in any way required, it would just be handy.

Edit: I guess I am now 1/5 done with post dungeon dialogues (and that's only counting with counting Thaum and Arc, who barely change at all). Forgot how long these take to write. Plus I didn't really get anything done yesterday at all, so this'll probably take a while. But I am going to make sure I don't rush on this and just keep  getting it done, so it'll be out pretty soon.

[IndigoFenix]

Something that would improve flowchart convenience: when you are adding in a piece (startpoint, dialogue, function, etc.), it would be helpful if it moved to the top left corner of where you are rather than the top left corner. When you are editing something on the opposite end of the flowchart, it can be really quite inconvenient. Not something that is in any way required, it would just be handy.

Oh right, I kept meaning to do that.

Updated to add new nodes under the current selected one.

By the way, not sure if you figured this out already but to rearrange dialogue options you can position them in the flowchart space.  Options placed higher up will appear first in the option list.

As far as knockback is concerned, I've finally figured out where I was going wrong - I've been trying to make knockback realistic.  But realistic physics in games, counterintuitively, feels more awkward and is overall less fun than simpler, easily controlled mechanics (unless the whole point of the game is abusing the physics engine, like Just Cause or Goat Simulator).  Sort of like the uncanny valley for physics.

I'm going to do a study of how different games handle knockback and try and make something that is more easily controlled from both a world builder and a player perspective.

Edit: I added an option to change the size of the graphics in a tileset.  The room will still use 32x32 tiles, so if you use for example a 16x16 tileset the graphics will be stretched to fit.  This is mainly because a huge number of free tilesets are 16x16; they will be a bit pixellated but you can use them if you want to.

[Kamani]

Okay.

Thanks.

Yeah, just figured that out a few days back. Didn’t realize it earlier, so things won’t be quite in order if they are older than that, but newer things are usually in the right order.

Ah, okay. Makes sense.

Okay. Hope it goes well.

Okay. Makes sense.

Edit: That inescapable moment when you end up making a dialogue with two groups of people in it, and you may know any combination of them, so you have to write four different intro dialogues before starting the actual dialogue. Both frustrating and interesting.

Edit: Problem with the nodes thing for flowcharts. It doesn't work if you are scrolled past the screen. It goes to where it would be if the screen was not scrolled around.

[IndigoFenix]

Edit: That inescapable moment when you end up making a dialogue with two groups of people in it, and you may know any combination of them, so you have to write four different intro dialogues before starting the actual dialogue. Both frustrating and interesting.

Such is the price of authenticity and detail - which is why most games don't include it.  But it does make your world feel more "alive" when characters react and remember your choices.

Speaking of which, if you enter the farm after the dungeon without having visited before, the cutscene starts trying to play out.  This is because, if you run a Destroy event but don't select "Instantly", the object is not actually destroyed until the end of the frame.  (This allows you to run some final actions after the object is destroyed, and also prevents inconsistent results when destroying an object during a collision by ensuring that the order of events doesn't matter.  Maybe a bit counter-intuitive though... maybe I should change it.)

Edit: Problem with the nodes thing for flowcharts. It doesn't work if you are scrolled past the screen. It goes to where it would be if the screen was not scrolled around.

Fixed.

[Kamani]

Yeah. I'd rather my world feel more alive and I spend some extra time on it than have it feel like characters don't care about what you did 20 minutes ago, so I feel like it is worth it. Characters won't react perfectly, since I am one person making this with 0 experience (both in design and writing), but I'm still going to do the best I can.

Oops, I'll fix that. On that, perhaps setting it to be instantly with an option to wait until end of frame would work?

Thanks. Glad that is fixed.

[IndigoFenix]

Doing some more experiments with the knockback system.  Knockback accumulates poise damage, which is removed through Poise Regen.  When poise exceeds Flinch, the object cannot move.  When poise exceeds Knockback, the object is knocked back with a speed equal to the hitbox's force divided by their Mass.  When poise exceeds Knockdown, the object will be unable to move until their Poise damage returns to 0.  Hitboxes can be given a Rebound value, which causes the attacker to be knocked back based on how much the targeted sprite resisted the knockback.  Sprites can also have a Rebound value themselves, which multiplies the rebound.  It's more "gamey" than a proper physics simulation, but you can use it to simulate objects with various degrees of weight, stability, and "bounciness".

I'm still not 100% satisfied with the result, but I've added these properties to several sprites, including the gremlins, floating eyes, spiders and bats.  I'll improve on it more.  I'd also like to get "defensive" hitboxes working as well, basically allowing hitboxes to clash with and/or block attacks.