Search

That doc is unmodified from Zakky atm, he asked if I had an items list, and gave me it to work off.  I just figured it could be useful to you while I work on adding other info to it.

Also, exciting times, we'll be sure to do some test runs and document our adventures, though I'm sadly out of town this weekend.

I started working on the document, and started chopping up the map.  Will base my formatting on this doc, which has all the item spawn information, will then grab images, and then attempt to document the enemy spawns of every room in the game (roughly).

https://docs.google.com/spreadsheets/d/tqe3Sxhqw1h7jG5If1c3BwA/htmlview?pli=1

Talked to Zakky, he said that he had got to C10 with 1 shield 1 gun, so he thinks that most likely we would be able to finish with just the stock number randomized.  He says it would be good to have an option to have either the shield area rule, the guns area rule, both, or neither.  In which case, you can probably just not code in the rule until a later version, or not code the choice until a later version.  Doesn't matter to me.

Will work on that document either today or tomorrow.

Ok, all of that seems doable.  I plan to have the first draft up by this Friday.
"The Guardian Legend SpeedRacer." Do you think you'd have a need to limit the amount of corridors required to win? This way, you could shorter or longer races depending on how many corridors the user has to get through.

It's easy enough to do on a regular sized map, just by saying "first to beat c0-c5 wins".  I mean, a half map would be neat, but It'd be of low importance to me. 

All I want to ask from you word document with every jpeg of all the enemies, bosses, and items in the game.  I'm going to write information about each element and it would be helpful if I had the images in a word document.  Thanks.

Consider it done.  Are you wanting me to document all current spawns of items/minibosses/enemies, I'll certainly do that for shields/guns regardless.  I use a map that has most of the information already there, so separating the info shouldn't be bad.

If you just want a image with cleaned up images of just one of each items/bosses/enemies, I can do that too.  Just want to be sure on what you want before I start.

Ah, I was going to work on the logic a bit to not try to impose my own personal goals on you, but, if you want to work on it, it would make me happy.  If you have any questions/comments about anything I say, ask, or if you want hard numbers, I'll happily count the occurrences of things in the main game or things like that.  If you're finding this hard to read I'll reformat it.  I'm also willing to chat on... pretty much any messenger if that's more convenient, or skype call, or whatever.  Ask and I'll pm you whatever.



I talked to another speedrunner, and it seems like a base set of rules for our current group would be something like:

Seeding from a entered number/string would be good, so we could all generate a rom based on a working base, instead of having to transfer the rom.

Corridors
   Stock corridors are fine, especially early on.  We probably could use the practice anyway.
Separated zones with keys. 
   The game should be roughly the same size as the original, probably meaning that each zone would be somewhere between 24-40 rooms large.
   (random zone interlinking is neat, but probably shouldn't happen more then 2-3 times in a rom much, should still require key to switch zones)
   Somewhat maze like, not entirely open, though perhaps open in places, as much as you can convince random generation to fake stock design.
Items
   Our group has done enough challenge runs, as long as the items are in the game, the only thing that must be controlled is that shields and gun powerups have to happen at around the time they do in the base game. With too low of shields, the game stalls out at about red grimgrin, if our no shields no guns challenge race says anything.  Red grimgrin is beatable, we were ehhhh on continuing.
      Actually, it may just be shields that is required.
   Minibosses can be randomized however they may fall, just try to limit miniboss occurance to about 3.
      The only way this could screw you over is if many of your shields got locked behind the red seaweed bosses, or the skull bosses, I'm pretty sure we could take any other occurrence.  If you limit shields to 3 or less in these 2 spawns combined, we will be fine. (there are several extra shields)
   Item distribution
      Normal Landers, Normal Shields, Normal Guns
      4 of each weapon, to deal with that boss you don't want to backtrack for, or losing items to pick 3 shops.  If you must lose an item, we'll certainly deal with it.
      Shops, same shop number and prices is fine, and you can put them anywhere, but if it contains a shield, it's a shield from an area that matches the area the shop price came from.  You could also just ban shields as a shop spawn, this is inferior to the prior logic, but it's also less work.
   As far as items go, you lock the spawn area for your shield's/guns, then you go nuts with placing the other ones.
      20 items come from corridors
      ?? items come from minibosses, I assume you aim for the same number of minibosses as the main game
      other stuff comes from crates, again, I assume you just pull that number from the main game (though it might increase or decrease slightly due to the 4 of each item, lock spawn just about as many EEs as stock.)
Enemies
   Honestly don't care, because they're generally easy enough to evade.  Locking enemies to their areas is dandy.  Having full game chaos is fine too, only thing that would slaughter of the games normal enemy set is the re-purposed spidercrabs of a9/10, and we'd probably go around them.

I think that's about as much balance as one would need to create a labyrinth in the style of the original game that one/all of the current runner group could beat.  May want to sanitize it more to be easier, but for the people we're currently doing races for, this is probably the best balance of randomness/beatability, most of the people who would want it probably also are solid enough to handle it.


--edit-- probably best to only have one area touching the other by a single or few squares as well, just to keep the map a tad more readable.

If I were to write something that generated some sanity checked labyrinths for races, would you give me the technical info to patch the data into the roms?

As I'm unreliable, I'd write the generator first so it made playable maps out of game to a level that seemed OK, and then worry about figuring out how to bring them into the game.  If I don't write the generator, I have no reason to waste your time probing for details.

I can give you data information about the Labrynth. Infact there's already a thread for it on this forum.  But if I'm reading this right, you want to make random races? As in a competitive race through the labyrinth?

Races for speedrunners and that like of which there's a few now.  Give everyone the same modified rom, everyone goes into the rom blind, you see who finishes the game first.  You need some consistency in how things are generated just so the races have a predictable difficulty and time frame.

--edit-- I would probably just leave the corridors the same.  Trying to keep those balanced would be a good bit of work.  Keeping the labyrinth balanced is pretty easy, to the extent of "you can lock the items to being the same area as the original, they just fall in random spots in the new labyrinth" and you'd still have a pretty fun race.  After that you kindof just add options to add variance to the number of areas things can spawn in.

If I were to write something that generated some sanity checked labyrinths for races, would you give me the technical info to patch the data into the roms?

As I'm unreliable, I'd write the generator first so it made playable maps out of game to a level that seemed OK, and then worry about figuring out how to bring them into the game.  If I don't write the generator, I have no reason to waste your time probing for details.

Double TGL would be what, 256k?  Under the 512k that current flash carts max out on.

Actually, has this been tested on a flashcart, I should probably do that.

Russian guy who's also TGL runner said that the max damage you can do to the boss per hit is 5 HP no matter of the strength of the bullet/subweapon.

Yep, It's why we use fireball and the saber lasers throughout the game.  Fireball first because it always does max damage, then move to saber laser for the most part as it applies the damage more consistently.  It seems as if bullets and special weapons are counted under a separate cap as well due to certain quirks with bosses (taking bullets to the face seems to increase damage on bombardiers, but major bosses don't take collision damage, and your saber would kill them anyway, this implies that your bullets are not hitting them, or that bullets hitting you above a boss hitbox causes damage), but I haven't set up a test to verify that myself. 

--edit-- If bullets do damage, it goes against something here, www.gamefaqs.com/boards/587320-the-guardian-legend/64314055 which is amazingly detailed and accurate.  So I really don't know what makes it so you can do certain bombardiers a phase faster by eating bullets.


Oh right, we also use sabers because for whatever reason hitting a boss with it deactivates their body hitbox for hitting you, so you can sit inside of them.

--edit-- also does the Russian runner have any method that I can follow him, even if it's in Russian I'd have fun seeing what I can gather from his runs.

I just watched that whole thing. Took a long time to download, but was worth it. Closest thing I've seen to a Guardian Legend gaming party. I liked how they knew everything about the original game and it's quirks and nuances. The reactions from Shiner are priceless. He used a lot of tactics involving the double saber and the backshot weapon.

Glad you enjoyed it.  I really wish I had posted something here earlier.  We TRY to know anything there is to know (especially what can save us time), and we've even found out a few quirks for ourselves.   There were a few times when I had pondered trying to ask you about some oddity we were trying to figure out, like how boss spawn RNG works, or various nonsense like that, as you you'd be more likely to know then anyone else (or more accurately, have a better idea of where to start to look), but decided not to waste your time.  If you did find oddities about the game when you were playing with it, I'd love to know any of them.  Partially for fun, partially for science. 

This was just over 5, didn't realize he was dropping frames early on as I wasn't there.  I will probably try a blind-ish attempt at it sometime soon.  Blind-ish because I've seen a fair bit of video of it.

I'm modifying the data more so than the game logic, so I can report on data findings that I haven't seen on the forum,  but as far as game logic goes, its hard to modify it successfully so I don't see the need to keep track of it as of yet.

That actually tells me a fair bit of how this games code is organized, which is interesting in itself. 

There was a big blind race of this about a week ago.  5 people raced it blind, mostly TGL runners.

http://www.twitch.tv/shinerccc/c/6217173

ShinerCCC was the only person to finish it (due to length mostly, 4+ hour races tend to die), so I'll only link to his VOD, but it also featured both the WR holder for 100 and any%.  Sadly I couldn't compete in it as I was working, and also sadly no one had posted here before the race, as it was fun times (for what I could watch), ah well.

I'll certainly try this out after it's tuned a bit.  I kindof like how crazy it is, though tuned to be beatable may be good for random racing.

Since you're getting into the assembly a lot for something like this, is there anything odd you've seen that people wouldn't know about the game logic.   Partially because I like to know whatever odd trivia I can know, partially as a (bad) speed runner in hopes that you'll tell me something potentially useful.

I probably should learn 6502 assembly at some point, and start trying to figure out the games logic, but it's hard to convince myself to do.

And I don't think anyone would want to make another hack anyway. Also I don't see any point in it. Why not organize a project with modern software?

--edit-- This is directed to Andrea, not ars.  I'm just quoting him to add to the statement.

This may be a little mean, but it's something that I tend to say in topics like this.  There's not going to be many programmers with the skill to make awesome things by themselves, who are really going to be sold on a project because you offer management.  Most programmers would probably prefer to self manage, and are probably fairly decent at it.  If you want to convince people to work on a project for you, you have to convince them that you will also be producing equally valuable work... or money, money works too.