Good idea, I was actually wondering why it didn't work like that to begin with.

Giving this one my yearly bump a bit early, since I think it works as a nice anti-frustration feature for new players and that seems to be the focus of this upcoming release.

Wow, still a lot of good ideas buried that i overlooked. Like blasphemous I had the same thought about the scroll (maybe it's even supposed to work that way but bugged/unfinished?)., but it's one of those odd occasions where the thought of making an RFE never popped into my brain. Even if not a newbie if you're trying to get the blanket ASAP it can help avoid taking false turns on the map.

There have been a number of issues in ADOM where an RFE was submitted to improve a certain behavior only for Thomas to fix it and say "yeah, this was actually a bug - it was actually supposed to work differently." I think there's a strong chance that this is one of those cases, where this has actually been broken for the past 20 years or so and people just thought it is working as intended. :P

No, it works as I originally intended it. Because calculating the dynamic distance requires some special handling:
- Each reading will require looking through all items on the map to find "the" blanket.
- What to do if there are two or more blankets (I didn't like the idea of introducing a special marker for "the" blanket)?

I'll do it at some point but it's one of those "improvements" where I personally do find the cost/benefit ratio not very tempting.

It must be. SMC is quite popular dungeon to visit in the very begining of the game and newbies must be choosing it from time to time. Having static info causes frustration and leaves vague impression about the game since it looks unfinished.

That does sound computationally expensive, but why can't the scroll object (and every other item) store its own x,y coordinates and then you just maintain a pointer to the scroll?

What's an object? ;)

Are you making a comment about it not being OO code? It's all illusory anyway, the CPU doesn't know what an object or a data structure is. Humans do.

Making a joke about C. :) (I spent a few months earlier this year working on a project with some ancient C code, it's given me a lot of sympathy for whenever TB talks about the difficulties of implementing certain things in ADOM)

Anyway, there might be an "item" struct or something, but I've gotten the impression (from various item duplication bugs and exploits as well as weird shopkeeper behavior, etc.) that items are part of the dungeon tile they're on or the inventory they are in, and don't really exist independent of some other context. That, and TB has a PhD in Computer Science and has been working on this game for 20+ years, I'm sure if there were x,y coordinates directly associated with items (like a "blanket" struct you can reference with a pointer that has x and y ints) he would have probably thought of it. :)

I think so too. Was just wondering what part of the picture I'm missing.

You can't know how hard this would be without having even seen the code. Speculating is pretty pointless, if the Creator says it's hard, there isn't really a reason to doubt his word. And it really would be a shame if valuable time would be spent on things like this, when we're waiting for super exciting stuff, like Rolf and the Volcano.

I wasn't doubting his word. Think I already explained why I was asking.

Nevermind

LOL, you should try to put some care into reading my comments instead of lecturing me. I learned about BFS about 25 years ago. I have programmed it dozens of times and I know it pretty well (and it's "algorithm" BTW... no "y" there).

The final sentence in my last comment said: "I don't think this has a good cost/benefit ratio".

First of all it's pretty funny to think of all scrolls in ADOM as magical scrolls. Has the idea ever crossed your mind that you simply found a plain paper scroll with static text on that damn stairway? That would explain why it isn't a magical compass. Someone used bloody plain ink on bloody plain parchment to leave a message.

Second I _never_ talked about efficiency of the algorithm. This kind of stuff (for the given size of ADOMs levels) is so blazingly fast that you probably could run it a thousand times and wouldn't notice on the UI level.

Third your explanations totally miss the point... algorithmically this is not very complicated. The complications are on the logical side:

• I probably would have to special mark _the_ blanket to not confuse it with other blankets.
• I probably might confuse players if there ever happen to be two blankets on the level as the constantly changing distance might go up and down with two blankets in wildly irregular matter.

It's bad enough that I by now probably have already wasted 15 minutes on this instead of programming fun features.

I hereby declare the scroll to be a non-magical scroll. RFE rejected.

Aw. :/

ixi: I don't know if my point to auricbond wasn't clear enough, but pretty much anyone who's ever been in an undergrad Computer Science course knows about and has written a BFS. Do you not see how condescending it is to lecture someone with a PhD in the field who is literally a CEO of a tech company on how to implement an elementary algorithm?

Well anyway, it cost me a cool RFE, so thanks...

@Dogbreath: Maybe I calm down tomorrow and reopen it :-/

Ah, damn, let's talk about it again once we have the next stable Steam release finished.

Sounds good to me. :)