/jp/ - Otaku Culture

>>17153656
Bloodshade and I have talked about it a bit before. This is unlikely to be implemented because of the way Elona's inventory code runs (the window / inventory list is completely recreated each frame).

Essentially, implementing this means we would have to iterate up to (3000*number of filterable attribute types) times per frame. I haven't implemented it to see what impact it would have on performance, but I don't imagine it's good.

If I ever think of something clever to cut down on that, maybe we'll try it. But until then...

>>17032372
>If the amount of enemies on a floor is dictated by map size

It's not, actually. There's another variable that controls both mob density and item density on the floor. Which is why in the above spiral you see no items on the floor; I have it set to 0 for testing so I don't have to deal with enemies.

Big Daddy spawn chance is also independent of normal enemy spawns, so even if map size influenced other enemy spawns, big daddy would be unaffected.

The rest is valid though, but I would end up adding a minimum that would overwrite the player's value if they put something smaller than that. Otherwise they could put something nonsensical like a 0x0 map (or even 1x1 or 3x3) which would cause a crash.

It may be worth adding some tighter controls on what a user can edit - though at the same time, people will also find ways to cheat if they want to, so should other users really be restricted because of that?

>>17032259
>I'd like to see floor layouts that look purposeful, ie a mess hall, manufactory, catacombs etc...
Hmm. It would be possible, but certainly more complicated. I'm also not great at designing good looking rooms in general. If you or someone else wants to put together a mockup sample of a room or floor like this, I can see about generating similar setups.

For customization... I guess it would be better if I gave specific examples for the types I've implemented (putting aside the obvious min/max map dimensions).

For spiral...
You can adjust the width of the spiral corridor, the angle(?) of the spiral itself (as in, making the spiral skewed more horizontal or more vertical - in the screenshot the spiral is even), the "tightness" of the spiral (space between each segment of the spiral), so on. This will have some randomness anyways to prevent getting the same map over and over, but you could play with it more if you wanted.

For the recursive division floors...
The primary two customizations are the recursion depth (how many times each segment will be split) and the minimum room/corridor size. It's a little hard to explain without pictures, but basically a lower recursion depth will likely leave larger rooms, with a high recursion depth leaving smaller corridors/rooms. Minimum room/corridor size is pretty self explanatory though I think.

So, something that people were requesting several times for Custom was additional map types for the randomly generated maps.

I finally got around to looking at the map generation code and figuring out how it works. I've implemented two additional types so far, one which is a Archimidean Spiral, and the other which is more like a maze based on recursive division. Not sure either one really makes a great map, to be honest.

So, anons, here are my questions for you:

1) Do you have any ideas for additional random map types? Details or examples of the generation would be helpful. Honestly I can't really come up with many ideas for new floor types.

2) Do you think configuration for the maps should be user editable (in a txt file)? There will be some randomness either way, but for example with the recursive division design, you could set the depth of recursion, min corridor/room size, min/max map size, etc. I've done all these right now while I'm testing them, but I'm not sure if I should keep all this open for players.

Thanks.

Image is an example of the Archimidean Spiral.

>>16851879
I don't work for Oracle directly. I work for a company that uses Oracle's ERP system (previously EBS, now transitioning to Fusion).

In EBS, I was primarily designing SQL reports and making customizations to Oracle's pre-baked modules as per our business needs. In Fusion, a lot of the ability to customize is going away, so we'll see what I'll get stuck doing.

But no, I work with Oracle's software, not for Oracle directly. I'd probably be paid significantly more if I did.

>>16850497
I am the qt brown girl, anon.

I was born and raised in the US, but my family (mother's side) comes from Ethiopia. I've never been, so I felt it was a good opportunity.

My husband and I are both American, so we will settle in America. Eventually anyways, when we're both done with our work in the country we're currently in.

>>16086850
>>16087255
Custom AI doesn't affect lay hands or multicasting. Multicast only works on dart spells though.

Lay Hands is not actually a special action, it's a unique trait that triggers when an ally drops below 0 hp. That's why it can't be set in the AI - it's not actually a spell or special action.

>>15767969
Oh, if you're after individual skills for gene engineering, look at the race/class pages on the wiki. If the race or class has it naturally, it should be donatable. It's what I always did and never had a pro anyways.

>>15767236
Creature info isn't stored in seperate files. Everything is in the executable.

But it should be trivial to add creature spawning to wizard mode. I'm still on vacation (and so am being even more lazy than usual), but wait another week and I'll add this into Custom.

If the pastebin has the experimental builds I did before though, every NPC's actions and spells should be listed in the external npc files I made. It would be by action ID though, and I don't remember if I ever posted a dump of action ID to skill/spell name.

>>15760316
Move check (to see if a pet is at their preferred distance) is done before the AI routine is called. If the pet isn't at their preferred distance, then the pet moves and the AI routine never gets called. The random chance for shadowstep from tactical attack is inside the AI routine. Ergo, a pet with preferred distance of 1 will never use it from tactical attack (there might be some exceptions).

>>15754350
Don't take this as gospel, since it just comes from my own playing experience, but I believe you might have more luck finding acidproof/fireproof potions from low level magic vendors.

They are both cheap potions. Investing will (to a point) increase the average value of the items the shop generates to sell. Consequently, if you invest much, you end up creating a lower chance for them to be generated. It's the same reason that over-investing in souvenir vendors is bad if you want treasure maps. The other potential items that the shop sells have a higher base value, so by investing too much you end up making maps generate less often.

>>15743366
It already exists.

Use the * key to select an instruction, then use it again on the place you want it to go and the two instructions will be swapped.

I mentioned this in the patch notes when I added it, but I suspect most people missed it. I really need to update the blog at some point.

>>15742523
Add
Target distance > 5 move(forward)

Somewhere before your hp lines. Most likely your exile gets the target of what your little girl goes to attack, gets a true condition on one of the magic spells, but the target is actually too far away to cast on.

>>15439360
If you post a screenshot of the AI setup I can help you debug what the problem might be.

I haven't updated the wiki, but if anyone is curious on some of the technical details regarding a few of the changes...

Spellpower calculation for pets went from Casting*6 + 10 to Casting*7 + 10.

The spell training gives a flat 3000xp to every spell in that category (which is then affected by the skillexp formula, of course).

1.59 Custom on Friday (or you can shoot me)

>>15353992
I believe you are the first one I've ever seen comment on it, actually.

>>15109527
Sorry anon, but I'm much busier now than I was when I first started Custom. I still try to put out an update build on the Friday after a new release (which is my day off). The past two weeks have just been bad because we got 2 new E+ versions back to back.

But don't worry, custom will be updated for 1.57 tonight.

>>15101073
Yes, I'll update Custom to 1.57 on Friday.

>>15085958
Is this exclusive to custom or does it happen in vanilla 1.56?

>>15056241
Easy enough, but this week is already going to be a pain between the 1.56 update and updating my current work. I'll try to fit it in though.

>>15041389
Yes. Extract Elona+ 1.55, add in Custom 1.55.4 (or just take the scene2.hsp and scene3.hsp files from it) and then add the files from this zip. Then just run Elona_Experimental.exe.

>>15040704
Thanks for the plug, I was just about to do so myself.

Really, if anyone has the time, I'd really appreciate some help doing the testing here. This should open up quite a bit of potential for modifications that won't require the executable to be recompiled. That includes adding translations for flavor text. I just need to make sure everything works properly before I push this into a stable build.

Once this is working, I'll do the same for items (item descriptions) and card descriptions, making them entirely editable outside the executable.

I've even got an idea in my head for NPC dialogue and conversation choices, but I don't know for sure yet that it'll work.

But please, please don't use your save on this without backing it up first. It has serious potential to affect your file.

Sorry for shilling it here, but really, any help with the testing would be appreciated.

>>15030073
No, it's a random dialogue when you choose the option. One of them has her starting to tell you, but she catches herself before saying too much.

>>15002833
You'd have to ask Ano, not me. All I did was translate it and enable it under English settings.

>>15001330
It's beem enabled for a while. There is an option in the settings that you need to change for it to show everything, though.