Please, please excuse the massive wall of text! If you like, just read my reply to you (under quotes). I saved the cool new stuff for a second post after this one!
I've been inactive a bit too much lately, time to change that!
SpiderFighter wrote:I was wondering about this sort of thing myself. The Difficulty Index (or Suggested Level) may be a bit misleading in some rare instances. For example, I have one room with three fairly powerful creatures that, once defeated, will grant
no xp, whereas I have a smaller critter that will grant normal xp, even though it becomes a much tougher than usual fight due to its having developed a new...um...skill, let's say.
Are we tweaking individual values, or just confident that all will even out by then dungeon's end?
The Suggested Level is an automated rating. No matter how much I try to account for with it, it won't be able to catch everything. It's more of a guide than a strict rule. Ultimately it has to be up to the designer to pick the tier their dungeon is meant for when they submit it. Although I might ask them to change it if they pick something really really off.
How good each tool is for finding the correct tier (worst to best):
SL rating < Designer Expertise < Play Tester Expertise
Ancylus wrote:I don't quite understand what you mean. My idea was a room like the following:
Edit: Image hopefully fixed now. In case it isn't, here's the same in ASCII:
S ans s are stairs up and down, T and t are teleports to actual level start and end, D and d are destinations for teleports at level start and end, + means wall and . marks empty space.
Surround the whole thing with walls or secret doors, or remove the teleporters and connect it to the rest of the level with corridors if that fits your design. Of course if you have multiple levels it's not necessary to place stairs between them here. And getting up/down stairs to match between different designers' levels only requires mirroring the room.
A room like that could function fine, but it doesn't solve the issues I was referring to in my last post. When going from one floor (above) to another (below), teleporters would have to link to teleporters and stairs would have to link to stairs. If we want to offer both possibilities, many people may implement a central room as you suggested. If the stairs end up being used, we're limited in that the entrance/exit locations are predictable. If the teleporters end up being used, they can't be placed in clever ways or in puzzles that teleporters often appear. In either case, we must reserve up to a 5x5 block (because of clipping on the sides of stair models) the designer must work around.
Having teleporters also lets us have "Scrolls of Return" which are only usable at the dungeon entrance and exit teleporters! (Making them usable anywhere makes them cheap in the "escape from any danger" sort of way.)
Ancylus wrote:Edsploration wrote:So, do you think median weapon SL is the best way to rate a "typical dungeon level" by item content? I may have to add something like this to my Custom Dungeon Analyzer...
Median may not work well if there are lots of low-SL items on the level. A better idea would be to take the highest SL found in the level, and subtract one or two from that.
I was thinking that if we only consider equipment weapons (not throwing items, missile items, etc, or any other item type), it should be the best measure of the intended level of the floor. It would be rare for a designer to give out multiple cudgels late in the dungeon, for example. And weapons are usually given out in a fairly tight range around their intended monster difficulty range because they are such a driving factor in champion power. If we look at the highest level weapon, and take off a few... how many would we take off? Some designers may put no more than item SL 6 into one floor, while another may put item SL 8 into the same floor and be just as balanced if it's not overdone. Ultimately we have to consider many cases. Perhaps a (median OR (max - 2), whichever's higher) approach would work best for the widest number of cases. Or a weighted-"median". I'd need to fiddle with many examples to find something that works.
Brodie301 wrote:I posted a 2 level dungeon called Low Levels for a play test on Nexus.
Please give give it a try and give suggestions.
This is an awesome dungeon! It plays much like the main game, but is full of new surprises. I think I was able to obtain everything except for what looked like Leather Boots in an alcove. A few observations:
- I would suggest including secret objects for the secret locations. The game keeps track of those and makes total secrets viewable from the statistics screen. It's nice knowing how many there are total.
- I saw a few clipping models, mostly dirt on the ground. Maybe the game places dirt randomly even when you have already done so manually?
- The scavengers fight may be too tough for someone who hasn't yet become fast with the controls. They probably are good if they're playing custom dungeons. But you might not want to put skill-demanding roadblocks like that so early in a dungeon unless they are optional side areas. In the main game one of the early tough rooms was the slime area, but it was optional.
- For inclusion into the CFD, I am a bit concerned about how quickly a party can level up in a dungeon of this length. I would hope to be able to fit in at least two dungeons in each tier, but this one easily fills the first tier by how much XP is supplied.
- I think items were done pretty darn well. There is a wide variety of them. Recipes and spells included! But the party is still a bit starved when it comes to filling armor slots.
- There are a lot of torches! I got too used to them and nearly got stuck when I had to eventually do something with one. Nothing wrong with that though.
SpiderFighter wrote:
I'm waiting on item placement until we have more definitive guidelines. I'm worried we'll wind up with 10 areas in a row that all have the same weapon (for example) and leave out something a player may want/need instead. Is there a way to parse every item that appears in a dungeon, to make the stitiching easier, or should we keep a running list of what we're using to submit with our maps? I have no problem with a team leader (aka "The Stitcher"
) PMing me with "Hey, we could really use an "item X" by the time we've gotten to your area; would you mind throwing one in your map somewhere?".
I would prefer to stick to a reference chart much like the Stats-Build tab of my spreadsheet here:
https://docs.google.com/spreadsheet/ccc ... 6SGc#gid=9
Using this chart as an example, If you were working on dungeon level 7, for example, gauntlets would be a good item to consider adding to your floor. Or a throwing weapon upgrade for dungeon level 4. The chart would be filled in as people submit dungeons, but they wouldn't exactly know WHICH item was on each floor. My reason for wanting it this way is:
- It lets each designer work entirely on their own submission without rehashing based on other's work.
- Many designers may implement custom items, so doing an overview based on vanilla items would require spoilers when custom items need mentioning.
- The submissions won't be designed to fit together at all! And may have a few jarring differences when sewn together, just like our ugly monster Mr. Frankenstein. Fitting aesthetics within a submission, and jarring differences between floors fits the FrankenDungeon theme, and also draws attention to the credit of the designer who made the floor!
Szragodesca wrote:Just a thought, but if you could figure it out, the "level" LED code from Lark's elevator script would actually work pretty well to show what SL the party should be before attempting a given portal. Open the door, see a big red 13 on the back wall when you're only 8, and it's your funeral.
I wouldn't want to do this for the hub floor because the balance of the dungeon may not line up well with the SL system, and many people might not know what it means. We're looking to forcing the player to do the dungeon levels in order at this point anyway, unless they enter a special password to skip ahead. If they're doing that, they should hopefully already know what the password goes to.
That being said... doing something like this would be
AWESOME for your own custom submission to the CFD.
Speculating here... you could make a "level of challenges" with such warning systems as a preview into arena style battle rooms. As long as the challenges are optional, there's nothing wrong with making the SL for each one as high as you want. The only problem I'd have with it is not allowing the player to gain XP from the fights, or at least not so much XP to skip entire SL tiers. Providing your own items or custom super-potions for victory would be very cool prizes.
HaunterV wrote:Maybe even have a reset lever for the submissions outside of the submission teleporters if you wanted to replay the submission again.
Resetting a dungeon with a script is prohibitively difficult to do for most dungeons. If someone wants to do this, they should keep a save file before the dungeon or skip it until later with the password warp feature which is explained below.
Oh, but I would like to do a super secret password-to-drop-equipment-and-level-up-party-to-specified-power (SL), which may or may not get done before this whole thing launches. It would require me to do a lot more work and reach a future version of my Custom Dungeon Analyzer. But it would allow a player to start a new party, and warp instantly to a specific dungeon level they want to replay without having to worry about keeping a save file there.
"
In the hub floor, doors locking portals to each floor are unlocked when that floor is reached by normal progression.
), particularly if the entrance is in a central point to the rest of the map. How about if we divide the number of included levels by, say, 33.3% and include only that many? They'd still be dispersed throughout, but that way the player would be forced to conserve a bit as they go through the entire Frungeon (my truncated term for Frankendungeon